Text History of the Greek Ecclesiastes: Introduction to the Göttingen Septuagint Edition of Ecclesiastes [1 ed.] 9783666560736, 9783525560730

Citation preview

De Septuaginta Investigationes (DSI) Edited by Anneli Aejmelaeus, Felix Albrecht, Kristin De Troyer, Wolfgang Kraus, Michael Segal In Co-operation with Kai Brodersen (Erfurt, Germany), Cécile Dogniez (Paris, France), Peter Gentry (Louisville, USA), Anna Kharanauli (Tbilisi, Georgia), Armin Lange (Wien, Austria), Alison Salvesen (Oxford, UK), David Andrew Teeter (Cambridge, USA), Julio Trebolle (Madrid, Spain), Florian Wilk (Göttingen, Germany) Volume 17

Peter J. Gentry

Text History of the Greek Ecclesiastes Introduction to the Göttingen Septuagint Edition of Ecclesiastes

Vandenhoeck & Ruprecht

Bibliographic information published by the Deutsche Nationalbibliothek: The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data available online: https://dnb.de. © 2022 by Vandenhoeck & Ruprecht, Theaterstraße 13, 37073 Göttingen, Germany, an imprint of the Brill-Group (Koninklijke Brill NV, Leiden, The Netherlands; Brill USA Inc., Boston MA, USA; Brill Asia Pte Ltd, Singapore; Brill Deutschland GmbH, Paderborn, Germany; Brill Österreich GmbH, Vienna, Austria) Koninklijke Brill NV incorporates the imprints Brill, Brill Nijhoff, Brill Hotei, Brill Schöningh, Brill Fink, Brill mentis, Vandenhoeck & Ruprecht, Böhlau, V&R unipress. All rights reserved. No part of this work may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without prior written permission from the publisher. Typesetting: Datagrafix Vandenhoeck & Ruprecht Verlage | www.vandenhoeck-ruprecht-verlage.com ISSN 2197-0912 ISBN 978-3-666-56073-6

Table of Contents Guide to the Reader����������������������������������������������������������������������������������������������������11 A.. The Textual Witnesses������������������������������������������������������������������������������������������13 I.. The Greek Witnesses��������������������������������������������������������������������������������������13 1.. Majuscule Manuscripts�����������������������������������������������������������������������������13 2.. Minuscule Manuscripts�����������������������������������������������������������������������������13 3.. Manuscripts Not Collated for the Edition����������������������������������������������17 4.. Papyri and Papyri-Fragments�������������������������������������������������������������������18 5.. Copies, Derivatives, and Descendants����������������������������������������������������19 II.. The Ancient Translations�������������������������������������������������������������������������������21 1.. The Old Latin Translation (La94 95 160 Hi)������������������������������������������������21 1.1. The Old Latin Fragments (La94 95 160)�����������������������������������������������22 1.1.1.La94-95�����������������������������������������������������������������������������������������22 1.1.2. The Hexaplaric (?) Recension of La160����������������������������������22 1.2. The Ecclesiastes Commentary of Jerome (Hieronymus = Hi)�����23 1.3. The Indirect Old Latin Tradition (Latin Patristic Citations of Ecclesiastes)������������������������������������������������������������������������������������35 1.4. The Textual Character of the Old Latin Translation���������������������45 2.. The Syro-Hexapla Translation (Syh)�������������������������������������������������������46 2.1. The Witnesses/Editions of the Syro-Hexapla Translation������������46 2.2. The Use of Aristarchean Signs����������������������������������������������������������50 2.2.1. Asteriskoi in Syh����������������������������������������������������������������������50 2.2.2. Obeloi in Syh����������������������������������������������������������������������������53 2.2.3. Antisigmata or Lemniskoi in Syh������������������������������������������54 2.2.4. Asteriskoi in V�������������������������������������������������������������������������57 2.2.5. Asteriskoi in 788����������������������������������������������������������������������57 2.2.6. Obeloi in 788����������������������������������������������������������������������������58 2.3. The Character of the Syro-Hexapla Translation����������������������������59 3.. The Christian Palestinian Aramaic Translation (CPA)������������������������60 3.1. The Witnesses/Editions of the Christian Palestinian Aramaic Translation��������������������������������������������������������������������������60 3.2. The Textual Character of the Christian Palestinian Aramaic Translation��������������������������������������������������������������������������61 4.. The Coptic Translations (Co)�������������������������������������������������������������������61 4.1. The Sahidic Translation (Sa)�������������������������������������������������������������61 4.1.1. The Witnesses/Editions of the Sahidic Translation������������61 4.1.2. The Textual Character of the Sahidic Translation��������������62 4.2. The Fayyumic Translation (Fa)��������������������������������������������������������64 4.2.1. The Witnesses/Editions of the Fayyumic Translation��������64 4.2.2. The Textual Character of the Fayyumic Translation����������65

6

Table of Contents

5.. The Ethiopic Translation (Aeth)��������������������������������������������������������������66 5.1. The Witnesses/Editions of the Ethiopic Translation���������������������66 5.2. The Textual Character of the Ethiopic Translation�����������������������67 5.2.1. Agreements with the Catena Tradition��������������������������������67 5.2.2. Agreements with the O Group����������������������������������������������68 5.2.3. Agreements with Early Witnesses of the Egyptian Form of the Text�����������������������������������������������������68 6.. The Old Georgian Translation (Geo)������������������������������������������������������68 6.1. The Witnesses/Editions of the Old Georgian Translation������������68 6.2. The Textual Character of the Old Georgian Translation��������������70 7.. The Armenian Translation (Arm)�����������������������������������������������������������73 7.1. The Witnesses/Editions of the Armenian Translation������������������73 7.2. The Textual Character of the Armenian Translation��������������������73 8.. The Arabic Translation (Arab)����������������������������������������������������������������76 8.1. The Witnesses/Editions of the Arabic Translation������������������������76 8.2. The Textual Character of the Arabic Translation��������������������������76 8.2.1. Agreement with One Group��������������������������������������������������77 8.2.2. Agreement with Two Groups������������������������������������������������78 8.2.3. Agreement with Three Groups (Examples Only)���������������79 9.. The Old Church Slavonic Translation����������������������������������������������������79 9.1. The Witnesses/Editions of the Old Church Slavonic Translation�������������������������������������������������������������������������������������������79 9.2. The Textual Character of the Old Church Slavonic Translation�������������������������������������������������������������������������������������������79 III.  The Indirect Greek Tradition (Patristic Citations)�����������������������������������80 1.. Commentaries on Ecclesiastes by Greek Church Fathers (chronological)�������������������������������������������������������������������������������������������81 1.1. Origen (Or)�����������������������������������������������������������������������������������������81 1.2. Dionysius of Alexandria (Dion)�������������������������������������������������������81 1.3. Gregory Thaumaturgos (GregTh)���������������������������������������������������81 1.4. Evagrius of Pontus (Evag)�����������������������������������������������������������������82 1.5. Didymus the Blind (Did)�������������������������������������������������������������������82 1.6. Gregory of Nyssa (GregNy)��������������������������������������������������������������82 1.7. Pseudo-Chrysostom (PsChr)������������������������������������������������������������83 1.8. Olympiodorus of Alexandria (Ol)����������������������������������������������������84 1.9. Metrophanes of Smyrna (Met)���������������������������������������������������������86 2.. The Greek Church Fathers (alphabetical)����������������������������������������������87 IV.  Printed Editions (chronological)�����������������������������������������������������������������90 1.. Aldina (Ald)������������������������������������������������������������������������������������������������91 2.. Complutensis (Compl)������������������������������������������������������������������������������91 3.. Sixtina (Sixt)�����������������������������������������������������������������������������������������������91 4.. Grabe (Gra)�������������������������������������������������������������������������������������������������91

Table of Contents

7

5.. Holmes-Parsons (HP)�������������������������������������������������������������������������������92 6.. Rahlfs (Ra)��������������������������������������������������������������������������������������������������92 B.. The Text History: Text Groups and Textual Relations of the Greek Witnesses (A.I), the Ancient Translations (A.II) and the Indirect Tradition (A.III)�������������������������������������������������������������������������������93 I.. The Hexaplaric Recension�����������������������������������������������������������������������������93 1.. The Witnesses of the Hexaplaric Recension�������������������������������������������93 2.. The Critical (Aristarchean) Signs Employed by Origen�����������������������93 3.. The Character of the Hexaplaric Recension�������������������������������������������94 4.. Singular Readings in the Four MSS. V 253 475 637������������������������������97 II.. The Egyptian Text-Type of Codices B-S and Their Derivatives�������������102 1.. The Witnesses of the Egyptian Text-Type (E)��������������������������������������102 2.. The Provenance of the Egyptian Text���������������������������������������������������103 3.. Brief Characterisation of the Egyptian Text�����������������������������������������103 4.. More Distant Relatives����������������������������������������������������������������������������104 III.  The A-Text����������������������������������������������������������������������������������������������������105 1.. The Witnesses of the A-Text�������������������������������������������������������������������105 2.. Brief Characterisation of the A-Text�����������������������������������������������������105 IV.  The Lucianic Recension������������������������������������������������������������������������������106 1.. The Provenance of the Lucianic Text����������������������������������������������������107 2.. The Witnesses of the Lucianic Recension���������������������������������������������107 3.. The Peculiar Character of the Lucianic Recension�����������������������������107 4.. Detailed Analysis of Witnesses Attesting the Lucianic Recension����109 Brief Characterisation of Individual Witnesses�����������������������������������110 MS. 106������������������������������������������������������������������������������������������������������110 MS. 125������������������������������������������������������������������������������������������������������110 MS. 130������������������������������������������������������������������������������������������������������111 MS. 261������������������������������������������������������������������������������������������������������112 MS. 545������������������������������������������������������������������������������������������������������114 MS. Pair 125´��������������������������������������������������������������������������������������������115 5.. The Antiochian Fathers���������������������������������������������������������������������������116 John Chrysostom (died 407)������������������������������������������������������������������116 Antiochus (i.e. Church Father)��������������������������������������������������������������126 V.. The Catena-Text��������������������������������������������������������������������������������������������128 1.. The Various Catenae of Ecclesiastes and the Catena-Groups�����������129 2.. The Catena Groups����������������������������������������������������������������������������������135 3.. Text-History of the Catena-Groups������������������������������������������������������137 VI. The d-Group������������������������������������������������������������������������������������������������137 VII. The k-Group�����������������������������������������������������������������������������������������������138 VIII.  The Manuscript Pair 248´�����������������������������������������������������������������������139 IX.  The Codices Mixti (al)��������������������������������������������������������������������������������140 X.. Individual Manuscripts—Unclassified������������������������������������������������������140

8

Table of Contents

XI..The Later Greek Translators/Revisors�������������������������������������������������������141 1.. Introductory Observations���������������������������������������������������������������������141 2.. The Witnesses of the Later Greek Translators�������������������������������������142 2.1. Marginal Notes in Greek Bible Manuscripts��������������������������������142 2.2. Patristic Sources��������������������������������������������������������������������������������146 2.3. Catena Traditions�����������������������������������������������������������������������������147 2.4. The Syro-Hexapla�����������������������������������������������������������������������������150 3.. Comparison of Recent Research and Sources Used by Field�������������152 4.. On the Textual Character of the Three Later Translations����������������153 5.. Text Critical Value for the Hebrew Bible����������������������������������������������154 C.. Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes��������������������������������������������������������������������������������������������������155 I.. General Assessments of Translation Technique���������������������������������������155 1..Overview���������������������������������������������������������������������������������������������������155 2.. The Hebrew Vorlage of the Translator and Translation Technique�����������������������������������������������������������������������������157 2.1. The Masoretic Text (𝔐/MT)����������������������������������������������������������157 2.2. The Dead Sea Scrolls (𝔔)�����������������������������������������������������������������157 2.3. The Syriac Peshitta (Pesh/𝔖)�����������������������������������������������������������158 2.3.1. Examples of Polygenesis�������������������������������������������������������159 2.3.2. Examples of Common Vorlage��������������������������������������������161 2.3.3. Examples of Common Exegetical Tradition����������������������161 2.3.4. Dependence of the Syriac Version on the Septuagint������162 2.4. The Cairo Genizah Fragment (𝔎)��������������������������������������������������163 2.4.1. The Witnesses/Editions of the Cairo Genizah Fragment��������������������������������������������������������������������������������163 2.4.2. On the Textual Character of the Cairo Genizah Fragment��������������������������������������������������������������������������������164 2.5. The Latin Vulgate (𝔙)����������������������������������������������������������������������166 3.. Translation Technique: On the Character of the Translator of Ecclesiastes�������������������������������������������������������������������������������������������166 II.. Basic Principles for the Constitution of the Text��������������������������������������168 1.. Inner Greek Scribal Errors���������������������������������������������������������������������168 2.. Variants Preserved Almost Entirely in Patristic Sources��������������������179 3.. Interdependence of LXX Text-History and Text-History of the Jewish Revisors������������������������������������������������������������������������������180 4.. Jerome Correcting the Old Latin on the Basis of his Hebrew Text�����������������������������������������������������������������������������������195 5.. Further Illustration of Text-Critical Principles in Establishing the Text������������������������������������������������������������������������������������������������������199

Table of Contents

9

6.. Relation to Early Ancient Versions (e.g. Peshitta)�������������������������������234 7.. Diachronic Development of Greek (Hellenistic / Byzantine Periods)�����������������������������������������������������������235 . Excursus���������������������������������������������������������������������������������������������������������239 1.. List of Witnesses for Eccl. 1:17���������������������������������������������������������������239 2.. Morphology and Meaning of ‫הֹוללֹות‬ ֵ in Eccl. 1:17�������������������������������240 2.1.Morphology���������������������������������������������������������������������������������������240 2.2.Meaning���������������������������������������������������������������������������������������������241 3.. The Reconstruction of the Old Greek���������������������������������������������������242 3.1. The Evidence of Translation Equivalents��������������������������������������243 3.2. Analysis of Translation Technique�������������������������������������������������244 3.3. The External and Internal Evidence����������������������������������������������246 3.4. Goldman’s Proposal for the OG�����������������������������������������������������247 4.. The Significance of the Peshitta’s Reading for the OG�����������������������248 . Conclusion������������������������������������������������������������������������������������������������251 5.. The Hebrew Text��������������������������������������������������������������������������������������251 5.1. External Evidence�����������������������������������������������������������������������������251 5.2. Internal Evidence������������������������������������������������������������������������������252 5.3. Goldman’s Conjecture����������������������������������������������������������������������253 Conclusion������������������������������������������������������������������������������������������������254 D.. On the Arrangement of the Edition, Its Sigla and Abbreviations����������������255 I.. The Introduction������������������������������������������������������������������������������������������255 II.. The Critical Text�������������������������������������������������������������������������������������������255 III..Kopfleiste (List of Witnesses)�����������������������������������������������������������������������255 IV..Apparatus I����������������������������������������������������������������������������������������������������256 V.. Apparatus II���������������������������������������������������������������������������������������������������256 VI..The Colometry or Stichometry of the Edition�����������������������������������������256 E.. Sigla / Symbols and Abbreviations�������������������������������������������������������������������259 F.. Differences Compared to Rahlfs�����������������������������������������������������������������������267 G.. Corrections to the Published Edition���������������������������������������������������������������271 H..Appendices�����������������������������������������������������������������������������������������������������������275 1.. Computer Generated Stemma for Ecclesiastes�����������������������������������������275 2.. Complete List of Patristic Quotations for Ecclesiastes����������������������������289

Guide to the Reader This Text History of the Greek Ecclesiastes consists of a full Einleitung or Introduction in English of which the Einleitung or Introduction in the Göttingen Edition represents an abbreviated, shortened version in German. The arrangement or order here is the same as in the Edition. The discussion and lists in full are normal in a text history volume. Nonetheless, many case studies have been added to the data/lists to assist readers in observing the method and principles of textual criticism employed in establishing and reconstructing the earliest attainable form of the text.

A.  The Textual Witnesses I.  The Greek Witnesses The edition of Eccl.1 is based on the following manuscripts,2 which were carefully collated afresh by the Septuaginta-Unternehmen.

1.  Majuscule Manuscripts A B C S V 3010

London, British Library, Royal 1 D.V.; V. Cent. “codex Alexandrinus”. Rome, Bibl. Vat., Vat. gr. 1209; IV. Cent. “codex Vaticanus”. Paris, Bibl. Nat., Gr. 9; V. Cent. (rescr. XIII. Cent.) “codex Ephraemi Syri rescriptus”. Missing 115-218 μοχθω due to a lacuna.3 London, British Library, Add. 43725 (earlier Leningrad, State Public Library, Gr. 259); IV. Cent. “codex Sinaiticus”. Venice, Bibl. Marc., Gr. 1; VIII. Cent. “codex Venetus”. Turin, Biblioteca Nazionale Universitaria ms. C.V.25; VII. Cent. rescr. Contents: 217d-320a (extremely fragmentary and burned in parts; readings may not be assumed e silentio!).4

2.  Minuscule Manuscripts 46 68 106 125

125II

Paris, Bibl. Nat., Coisl. 4; XIII.-XIV. Cent. Venice, Bibl. Marc., Gr. 5; XV. Cent. Ferrara, Bibl. Comun., 187 I; XIV. Cent. At 815 106(2°) indicates a doubled instance of the lemma. Moscow, State Hist. Museum (GIM), Syn. Bibl., Gr. 30; XIV. Cent. Eccl. is on Fol. 262a-264b with large omissions. Contents: 11-223 317-48 517-612 713-84a 97918 104-end. Moscow, State Hist. Museum (GIM), Syn. Bibl., Gr. 30; XIV. Cent. On Fol. 265a268a Eccl. (here complete, for the second time) added subsequently.

1  Abbreviations for biblical books are based on A. Rahlfs, ed., Psalmi cum Odis, Septuaginta. Societatis Scientiarum Gottingensis auctoritate editum, Bd. X, Göttingen 1931, 9. 2  For detailed description of the manuscripts (MSS.) here listed, see A. Rahlfs, Verzeichnis der griechischen Handschriften des Alten Testaments. Die Überlieferung bis zum VIII. Jahrhundert, ed. D. Fraenkel, Septuaginta. Vetus Testamentum Graecum auctoritate Academiae Scientiarum Gottingensis editum, Suppl. I/1, Göttingen 2004, and A. Rahlfs, Verzeichnis der griechischen Handschriften des Alten Testaments, MSU 2, Berlin 1914. 3  For this MS. see F. Albrecht, Codex Ephraemi Syri rescriptus. 4  MS. 3010, badly preserved and difficult to read, was collated for the Edition by Chiara Faraggiana di Sarzana and Felix Albrecht.

14

The Textual Witnesses

130 139 147 149 155

Vienna, Nat. Bibl., Theol. gr. 23; XII.-XIII. Cent. Milan, Bibl. Ambr., A 148 inf.; X.-XI. Cent. (Cat.). Oxford, Bodl. Libr., Laud. gr. 30 (A); XII. Cent. (Cat.). Vienna, Nat. Bibl., Theol. gr. 11; XI. Cent. Copy of 260. Oxford, Bodl. Libr., Auct. T. 2. 4; XI. Cent. Contents: 216e αποθα]νηται—57 επ᾽ αυ[τους 812c γινωσ]κω εγω—121d οις. 157 Basel, Univ.-Bibl., B. VI. 23; XII. Cent. Lacking 56b οτι-516b και θυμω 816c της γης-95a οτι αποθα[νουνται due to missing folios. 159 Moscow, Russian State Archive of Ancient Documents, Fonds 1607, Inv. 1.22 (earlier Dresden, Sächsische Landesbibliothek, A. 107); XII. Cent. (Cat.).5 161 Moscow, Russian State Archive of Ancient Documents, Fonds 1607, Inv. 1.7 (earlier Dresden, Sächsische Landesbibliothek, A. 170); XIV. Cent. 248 Rome, Bibl. Vat., Vat. gr. 346; XIII. Cent. 252 Florence, Bibl. Laur., Plut. VIII 27; X. Cent. 253 Rome, Bibl. Vat., Vat. gr. 336; XI. Cent. 254 Rome, Bibl. Vat., Vat. gr. 337; X. Cent. 260 Copenhagen, Royal Library, Old Royal Collection (GKS) 6; X.-XI. Cent. (Cat.). 261 Florence, Bibl. Laur., Plut. VII 30; Written 1323. 295 Rome, Bibl. Vat., Ottob. gr. 56; XV. Cent. (Cat.). Missing Beginning (11-311). 296 Rome, Bibl. Vat. Palat. gr. 337; XI. Cent. 299 Rome, Bibl. Vat., Vat. gr. 1694; Written 1202. (Cat.). Contents: Chs. 1-7. This MS. transmits 11-11 in two versions: first a continuous Bible text; then a catena text is inserted, in which 11-11 is repeated. In the apparatus of the edition the first version is cited as 299A, and the second as 299B; when both of them transmit the same text, only 299 is used. 311 Moscow, State Hist. Museum (GIM), Syn. Bibl., Gr. 354; XII. Cent. 336 Athos, ᾽Ιβήρων 555; XIV. Cent. 337 Athos, ᾽Ιβήρων 615; XIV. Cent. 338 Athos, ᾽Ιβήρων 676; XIV. Cent. (Cat.). Lacking 11-14 and—due to loss of one folio—56 φοβοῦ-511. 339 Athos, Κουτλουμουσίου 8; XI. Cent. 342 Athens, Historical and Ethnological Society Museum, 200; XIII. Cent. Lacking 729 πολλούς-end. (Cat.). 357 Bologna, Bibl. Univ., 3640; XII. Cent. The first column 11-8 ἐμπλησθήσεται ὀφθαλ[μός is barely legible. The last column ends with 1211c οἳ παρά. The words following this to the end of the text were added during the XIV.-XV. Cent.

5  For the date of MS. 159, see P. Gentry and F. Albrecht, The Amazing History of MS Rahlfs 159.

The Greek Witnesses

359

371 390 411 415 425 443 471 475 503 504 522 534 539

15

Cambridge, Trinity Coll., O. 1. 53; XIII.-XIV. Cent. Preserved are 97a Δεῦ]ρο φάγε-15b καὶ διασώσει; 1010b ἐτ]άραξεν-end (lacking Eccl 915b-1010b due to loss of a bifolium). Escorial, Real Bibl., R-I-3; XVI. Cent. (Cat.). Genoa, Missione urb., 2; Written 1075. (Cat.). Jerusalem, Patr.-Bibl., Τάφου 370; XVI. Cent. Leiden, Univ.-Bibl.; Vulc. 50; XV.-XVI. Cent. (Cat.). London, British Library, Royal 1 A. XV.; XIV. Cent. Milan, Bibl. Ambr., B. 68 sup.; X. Cent. Moscow, State Hist. Museum (GIM), Syn. Bibl., Gr. 147; XIII.-XIV. Cent. [ex CCSG 24]. Derivative of 260. Moscow, State Hist. Museum (GIM), Syn. Bibl., Gr. 355; XIII.-XIV. Cent. Text and Commentary (but not Cat., Ralhfs’ Verzeichnis is incorrect). Oxford, Bodl. Libr., Auct. E. 2. 16; XII.-XIII. Cent. (Cat. until 89). Oxford, Bodl. Libr., Auct. E. 2. 17 und 18; XVI. Cent. (Cat.). Cambridge, Trinity Coll., B. 7. 3; XIII. Cent. Eccl. with Catena Excerpts. Paris, Bibl. Nat., Coisl. 18; XI. Cent. Paris, Bibl. Nat., Coisl. 193; XI. Cent. Eccl. Textual excerpts with hexaplaric readings and catena excerpts associated with respective textual excerpts. The lemmata of the biblical texts are as follows: 11a ῥήματα—Δαυιδ; 12b ματαιότης 2°—fin; 16c κυκλοῖ—πνς 1°; 18b οὐ 1°—λαλεῖν; 19a τί 1°—γενησόμενον; 19b τὸ ποιηθησόμενον; 111a οὐκ 1°—πρώτοις; 113b καί 2°—κατασκέψασθαι; 113e ὅτι— ανου; 115ab διεστραμμένον—ἀριθμηθῆναι; 22a γέλωτι—περιφοράν; 24a ἐμεγάλυνα—μου; 25b καί 2°—καρποῦ; 26b τοῦ—ξύλα; 210abcde καί 1°—μου 4°; 211d καί 3°—πνς; 212ab καί 1°—περιφοράν; 214a init—αὐτοῦ 2°; 222b καί—αὐτοῦ 2°; 225; 226bc ἔδωκε 1°—εὐφροσύνην; 226de καί 3°—συναγαγεῖν; 310ab εἶδον— ανου; 312b καί—αὐτοῦ; 314ab ἔγνων—αἰῶνα; 314d καί 1°—ἀφελεῖν; 315c καί 2°—διωκόμενον; 318ab ἐκεῖ—ανου; 318d-19a καί 2°—συνάντημα 1°; 321c καί 2°—κτήνους; 322d-41b ὅτι 3°—συκοφαντίας; 42ab καί—ζῶντας; 43c ὅς—πονηρόν; 44a καί 1°— μόχθον; 45a-6b ὁ ἄφρων—μόχθου; 48ab ἔστιν—αὐτῷ; 49b οἷς—αὐτῶν; 411b-13ab καί 3°—ἄφρονα; 414a ὅτι 1°—βασιλεῦσαι; 417d ὅτι—κακόν; 51bc καρδία—θυ; 53abc καθώς—ἄφροσιν; 56a ὅτι 1°—πολλῶν; 57ab ἐάν—πράγματι; 510bc καί—αὐτοῦ; 512a-13a ἔστιν—πονηρῷ; 514a καθώς—γυμνός; 517abc ἰδού—ἀγαθωσύνην; 518abc καί 1°—αὐτοῦ 1°; 519a-62c ὅτι—αὐτοῦ 1°; 62f ὅτι—αὐτόν; 67b καί—πληρωθήσεται; 68b-10a διὰ τί—αὐτοῦ; 71a-2b ἀγαθόν 1°—πότου; 72d-3ab καί 1°—καρδία; 77a ὅτι—σοφόν; 78a ἀγαθή—αὐτοῦ; 79b-10ab ὅτι—ταύτας; 711a-12a ἀγαθή—ἀργυρίου; 714c καί 2°—τούτῳ; 715b-16a ἔστιν—πολύ; 717ab μὴ ἀσεβήσῃς—σου; 717b ἵνα— σου (sic); 718ab ἀγαθόν—σου; 725b καί 3°—περιφοράν; 727b μία—λογισμόν; 82a init—φύλαξον; 87b ὅτι 2°—fin; 810bc καί 2°—ἐπορεύθησαν; 810d ὅτι—ἐποίησαν; 811a init—ἀντίρρησις; 811cd διά—ἐπληροφορήθη; 812a init—πονηρόν; 812cd ὅτι 1°—θν; 814abc init—ἀσεβῶν; 91a init—μου 1°; 93e-4a καὶ ὀπίσω—ζῶντας; 910a init—ποιῆσαι; 910c ὅτι—γνῶσις; 911bcd ὅτι 1°—ἄρτος; 911fg-12a καί 5°—αὐτοῦ;

16

The Textual Witnesses

913b καί 2°—14a πόλις; 914c καί 4°—χάρακας; 917a init—ἀκούονται; 101a init— ἡδύσματος; 104c ὅτι—fin; 109a ἐξαίρων—αὐτοῖς 1°; 1011b καί—fin; 1014bc οὐκ—ἐσόμενον; 1015b ὅς—fin; 1017b καί 1°—δυνάμει; 1019c καί 2°—fin; 113c ἐάν 2°—νότῳ; 115a init—πνς; 116c ὅτι—στοιχήσει; 118bc ἐν—σκότους; 119b καί 1°—σου 2°; 119c καί 2°—σου 4°; 1110ab init—σου 2°; 121c ἕως—κακίας; 123b καί 1°—δυνάμεως; 123d καί 3°—fin; 126b καί 1°—ἀνθέμιον; 1210abc init—λόγους 2°; 1211ab init—πεφυτευμένοι. 540 Paris, Bibl. Nat., Coisl. 194; XIII. Cent. (Cat.). 542 Paris, Bibl. Nat., Gr. 10; IX. Cent. 543 Paris, Bibl. Nat., Gr. 11; Written 1186. 545 Paris, Bibl. Nat., Gr. 18; XIII. Cent. 547 Paris, Bibl. Nat., Gr. 35; XIII. Cent. 548 Paris, Bibl. Nat., Gr. 36; XIV.-XV. Cent. 549 Paris, Bibl. Nat., Gr. 57; XI. Cent. 560 Paris, Bibl. Nat., Gr. 151; XIII. Cent. (Cat.). 561 Paris, Bibl. Nat., Gr. 152; XIII. Cent. (Cat.). 563 Paris, Bibl. Nat., Gr. 154; XII. Cent. (Cat.). 571 Paris, Bibl. Nat., Gr. 172; XVI. Cent. (Cat.). 574 Paris, Bibl. Nat., Gr. 176; XVI. Cent. (Cat.). Only Chapters 1-2 were collated; the MS. belongs clearly to C and has no distinctive features. 601 Paris, Bibl. Nat., Gr. 2509; XV. Cent. 602 Paris, Bibl. Nat., Gr. 2511; XV. Cent. 613 Patmos, ᾽Ιωάννου τοῦ θεολόγου 209; XIII. Cent. Within Chs. 4 and 5 the text is transposed as follows: 11-416c 514d-16a 416d-514c 514d-16b και 516b-end. 514d-16a is also extant twice. The duplicated texts are cited as 613I and 613II. 631 Raudnitz, Lobkowitz’sche Bibl. VI. E. f. 19; XIV. Cent. 637 Rome, Bibl. Casan. 241; XI. Cent. 645 Rome, Bibl. Vat., Barber. gr. 388; XIII. Cent. (Cat.). 698 Rome, Bibl. Vat., Vat. gr. 1974; X.-XI. Cent. 705 Athens, Nat. Bibl. 2410; XIII.-XIV. Cent. (Cat.). Contents: 11-312. 706 Athens, Nat. Bibl. 2641; Written 914. Previously: Serres, ᾽Ιωάννου τοῦ Προδρόμου 3. 728 Venice, Bibl. Marc., Appendix I 13; XI.-XII. Cent. 732 Venice, Bibl. Marc., Gr. 21; X.-XI. Cent. (Cat.). 752 Vienna, Nat. Bibl., Theol. gr. 115; XVI Cent. (Cat.). Contents: 11-312 in continuous Bible text; from 312 to the end is Catena, in which 312 is repeated. 754 Vienna, Nat. Bibl., Theol. gr. 147; XI Cent. Contents: 11-48 Cat. Proc. in Eccl. 49-fin libri only Bible text. 766A Athos, Λαύρα Γ 115 (355); XII. Cent. Contents: MS. 766 contains Eccl. twice and both texts are arranged in two columns. 766A is on Fol. 216a-224b and 766B is on Fol. 189b-197b. 766A is apparently older, probably the remains of a damaged codex, which someone bound to 766. One folio (25b παν-224c αγα)

The Greek Witnesses

17

is missing. The second column of 224a (121-125), esp. 125, and the last folio (125-end), esp. 128-1211 is only partially legible owing to the damage. 766B Athos, Λαύρα Γ 115 (355); XII. Cent. Contents: see 766A. 770 Athos, Λαύρα Β 114 (234); XII. Cent. Contents: init libri—118b προστιθη[σιν 788 Tirnavos, Δημοτικὴ Βιβλιοθήκη 25; X. Cent. (Cat.). 795 Athos, Λαύρα Γ 51 (291); XII.-XIII. Cent. 797 Athens, Benaki-Museum, Echangeables 72; XIII.-XIV. Cent. 798 Rome, Coll. gr. 16; XIV. Cent. (Cat.). 3011 Rome, Vallic. gr. 67 (E 21); XIV. Cent.

The following lost MS. was collated from H.-P.: 298

Ukraine, Codex Eugenii II; XII. Cent. (See Rahlfs, Verzeichnis, 332).

3.  Manuscripts Not Collated for the Edition The following younger MSS. were not collated for the edition, because no significant textual value or variants can be expected from them:6 122 348 352 353 436 437 442

462 464 485 486 609 632 720 753 755 3007

Venice, Bibl. Marc., Gr. 6; XV. Cent. Complete Bible. Copy of 68. Basel, Univ.-Bibl., A. VII. 6; XVI. Cent. (Cat.). Berlin, Staatsbibl., Phill. 1411; Written around 1540. (Cat.). Berlin, Staatsbibl., Phill. 1412; Written around 1540. (Cat.). Madrid, Bibl. Nac. 4749; Written in 1556. (Cat.). Madrid, Bibl. Nac. 4781; XVI. Cent. (Cat.). Madrid, Bibl. de la Univ. Central, E. 1, nº 10; XV.-XVI. Cent. The Eccl. portion was burned by fire, see N. Fernández Marcos, Un manuscrito complutense redivivo: Ms. griego 442 = Villa-Amil 22, Sefarad 65 (2005), 65-83. Modena, Bibl. Est., Gr. 64; Written in 1505. Modena, Bibl. Est., Gr. 155; Written in 1550. Eccl. with marginal scholia. Munich, Staatsbibl., Gr. 131; Written 1549. Eccl. with Catena Excerpts. Munich, Staatsbibl., Gr. 292 and 294; XVI. Cent. (Cat.). Paris, Bibl. Nat., Suppl. gr. 500; XVI. Cent. Eccl. with scholia. Rome, Bibl. Angel., Gr. 113; XVI. Cent. (Cat.). Copy of 732. Turin, Bibl. Naz. B. II. 8; Written in 1583. (Cat.). Vienna, Nat. Bibl., Theol. gr. 128; XIII.-XIV. Cent. Only excerpts from Job. Rahlfs’ Verzeichnis has incorrect information. Vienna, Nat. Bibl., Theol. gr. 199; XVI. Cent. (Cat.). Sinai, Cod. gr. 311; Written in 1510. (Cat.). Copy of 299.

6  Generally speaking, manuscripts copied after the invention of the printing press contain no significant variants and may, in fact, be copied from printed materials.

18 3012 —— —— —— —— —— ——

The Textual Witnesses

Rome, Coll. gr. 7; XV. Cent. Contents: 15-25 211-57 (om 220d 315 318-c 322 44-8 412c 415-16 55) Cat. Olymp. in Eccl. 58-fin libri alia Cat. Berlin, Staatsbibl., Phill. 1422; XVI.-XVII. (Cat.). Berlin, Staatsbibl., Phill. 1484; XVII. (Cat.). New Haven, Yale University Library, gr. 273; Written in 1583. (Cat.) Rome, Bibl. Casanatense, 203 (G.III.9); XVI. Cent. (Cat. from 312-end). Copy of 295. Rome, Bibl. Vat., Vat. gr. 1521; XVII. Cent. (Cat.) Rome, Bibl. Vat., Ottob. gr. 112; XVI. Cent. (Cat. from 312-end). Copy of 295.

4.  Papyri and Papyri-Fragments 818

870

969 992

998

(A): Milan, Univ. del S. Cuore, P. Med. Inv. 151; (B): Ann Arbor, Univ. Libr., P. Michigan, Inv. 27; End III. Cent. Fragmentary. Contents: (A): Eccl 317-18 21-22; (B): Eccl 63-5 8-11. Editions: (A): editio princeps by R. Roca-Puig, P. Med. Inv. n. 151 (Eccl. III, 17-18, 21-22), Aegyptus 32 (1952), 215-222 + Plate with Photographs of (A) + (B). See pp. 215-219 for evidence that both fragments belong to one codex. See Daris, Papiri Milanesi, vol. 1, fasc. 2, Milan 1966, Nr. 13 (in cooperation with Roca-Puig). (B): P. Mich. III, Nr. 135, pp. 8-9 (H. A. Sanders). Barcelona, Seminario de Papirologia, P. Inv. 225r (= PLit. Palau Rib. 3); IV.-V. Cent. Contents: 15 6cd 8 9c 10 11 15 215d 118b 214ab 311aα 37bβ 8ab 15ab 42 5 6ab 9 10 723b 24a 814bcde. Verses are given in the sequence found in the papyrus. It is striking that 118b 214ab follow 215d, instead of being placed before 215d. The beginning of the lines are almost unreadable; this applies especially to 15a 6c 8a 8c 9c 10a 10c 11b 11d 15a 18b 37b 42a 5a 6a 9a 10b 723b 24aα 814bcd. Edition: José O’Callaghan, Frammenti antologici dell’ Ecclesiaste, del Cantico dei Cantici e dell’ Ecclesiastico (P Palau Rib. Inv. 225r), Atti del XVII Congresso Internazionale di Papirologia 2, Napoli, 1984, 357-365. Florence, Bibl. Laur., PSI 1297; VI.-VII. Cent. Contents: 217 311 41-3 17 51 4. Edition: PSI 13, Nr. 1297, pp. 12-16 (G. Mercati). Oxford, Sackler Library (earlier Ashmolean Mus.), P. Oxyrhynchus XVII 2066; V.-VI. Cent. Extremely fragmentary. Contents: 66-8a (recto) 612bcde-71 (verso). As in the case of 818, 992 is also badly damaged; only individual words are preserved or partially preserved: 66a και ει εζ[ησε … ε] | των 66b και αγα[θωσυνην ου] | κ ειδε[ν 66c μη ουκ | πορε[υεται 67a πας μο[χθος … ] | εις σ[τομα 67b και γε η [ψυχη 68a οτι πε[ρισσεια …] | υπερ 612c αυτα 71a υπερ 71b θανατου. Edition: P. Oxy. XVII, Nr. 2066, pp. 2-3 (A. S. Hunt). Hamburg, Staats- und Univ.-Bibl., P. bil. 1; ca. 300. Contents: Fol. 15-21 Eccl (Greek) Fol. 22-28 Eccl (Coptic).

The Greek Witnesses

19

5.  Copies, Derivatives, and Descendants 122

Venice, Bibl. Marc., Gr. 6; XV. Cent. Complete Bible. Copy of 68.

125II

Moscow, State Hist. Museum (GIM), Syn. Bibl., Gr. 30; XIV. Cent. On Fol. 265a-268a Eccl. (complete text, for the second time) added later to the manuscript.

MS. 125II is a copy of 542 or a “sister” manuscript. They alone share a half-dozen or so singular readings, e.g.: 57b 517h 519a 612b 94b 911f 109a

ἐν χώρᾳ] > 125II 542 ὅτι] + μερις 125II 542 πολλά] πολλων 125II 542 ἀριθμόν] αριθμος 125II 542 αὐτὸς ἀγαθός] > 125II 542 γινώσκουσιν] ‑σκωσιν 125II 542 ἐν αὐτοῖς] επ αυτοις 125II 542

Normally 125II follows the marginal reading in 542 when it exists: 23e 224b 35b

τοῦ ἀνθρώπου 125II 542mg (γρʹ)] των ανθρωπων B S 998 542txt alii ὅ 1°] ει μη ο C C´’ 125II 155 542mg alii περιλήμψεως] περιληψεως 125II 542mg (ληψεως 542txt) alii

There are approximately fourteen places where 125II differs from 542. Most are common scribal errors, but they are included in the apparatus because they are instructive for the text history, e.g.: 321d

αὐτό] αυτω 295-797 125´ 125II* (non 542) 766A OlΑ Syh

The dative pronoun αυτω attested by the original hand of 125II (but not by 542!) is certainly an otacism (change of quantity between ο and ω lost in Hellenistic Greek). Nonetheless, Syh attests to this otacism as well even though 542 is normally the closest manuscript to the text of Syh. It is impossible to know if the otacism in Syh was in its Greek parent text or due to the process of translation. Nonetheless, the witness of 125II in relation to 542 and Syh is instructive for determining errors in Syh. Therefore 125II is cited in the apparatus only when it differs from 542.

20 149 471

The Textual Witnesses

Vienna, Nat. Bibl., Theol. gr. 11; XI. Cent. Copy of 260. Moscow, State Hist. Museum, Syn. Bibl., Gr. 147; XIII.-XIV. Cent. (Cat.). Descendant of 260.

Felix Albrecht. Eine Randbemerkung zur „Bibel des Niketas“ im Lichte der Textüberlieferung der Psalmi Salomonis, Rivista di studi bizantini e neoellenici N.S. 55 (2018), 81-83.

149 is an exact copy of MS. 260. The scribe always followed 260c. Nonetheless, 149 was collated by the Septuaginta-Unternehmen. It offers only one copyist error: in 103 the scribe of MS. 149 wrote αφων for ἄφρων. 149 is cited in the apparatus, only when the reading in 260 is noted as ‘vid’. This gives an opportunity to compare his reading of 260 with ours. Otherwise 260 always also represents the dependent witness of 149. 471 belongs to the cII-Group but is not cited in the apparatus. For a complete stemma of the descendants of 260 see F. Albrecht, ed., Psalmi Salomonis (Septuaginta Vetus Testamentum Graecum 12.3; Göttingen: Vandenhoeck & Ruprecht, 2018), 170-175. 602

Paris, Bibl. Nat., Gr. 2511; XV. Cent.

MS. 602 is a close sister of 613. Nonetheless, the scribe had access to other sources and in over a dozen places deviated from 613. There are even instances where 602 is closer to the Egyptian Group than 613. Thus 602 is only cited when it differs from 613. 752

Wien, Nat. Bibl., Theol. gr. 115; XVI Cent. (Cat.). Contents: 11-312 in continuous Bible text; from 312 to the end is Catena, in which 312 is repeated.

S. Lucà. Anonymus in Ecclesiasten Commentarius qui dicitur Catena Trium Patrum. CCSG 11. Brepols, Turnhout, 1983.

Lucà, in his description of manuscripts for the textual tradition of the Catena Trium Patrum states that for Eccl 312-1214 752 is a copy of 295. For the portion from 11-312 the scribe copied the biblical text from the Commentary on Ecclesiastes by Gregory of Nyssa which immediately precedes the Catena Trium Patrum in MS. 295. Nonetheless, in 11-312 there are a number of readings which do not agree with the biblical text of Gregory of Nyssa and also occur in other manuscripts. Therefore 752 is cited in the apparatus for the portion 11-312 and only in the portion 312-1214 when 752 deviates from 295.

The Ancient Translations

21

II.  The Ancient Translations 1.  The Old Latin Translation (La94 95 160 Hi) The following witnesses were used: La94

Marginal Notes: Escorial, Biblioteca de San Lorenzo 54. V. 35. Written 1561. Edition: Ma Ángeles Márquez7

La95

Marginal Notes: Madrid, Academia de la Historia Aemil. 2-3. XII. Cent. Edition: Ma Ángeles Márquez

La160

Sankt Gallen, Stiftsbibliothek 11. VIII. Cent. Edition: C. P. Caspari, Das Buch Hiob (1,1-38,16) in Hieronymus’s Uebersetzung aus der alexandrinischen Version nach einer St. Gallener Handschrift saec. VIII, Christiania Videnskabs-Selskabs Fordhandlinger 4, 1893. Digital images are also available: www. stiftsbibliothek.ch.

Hi

Hieronymus, Commentarius in Ecclesiasten [CCSL 72, 247-360]. Elisabeth Birnbaum, Der Kohelet-Kommentar des Hieronymus: Einleitung, Revidierter Text, Übersetzung und Kommentierung. Revision des lateinischen Textes durch Michael Margoni-Kögler. CSEL Extra Seriem. Berlin, de Gruyter, 2014.

Literature Pierre-Maurice Bogaert, “Caelorum ratio, ratio sub sole,” L’emploi de ratio chez les traducteurs latins de la Bible et notamment chez saint Jérôme. In Ratio: VII Colloquio internazionale del Lessico intellettuale europeo, Roma, 9-11 gennaio 1992 (Lessico intellettuale europeo, 61), Florence, Olschki, 1994, 69-83. W. W. Cannon, Jerome and Symmachus: Some Points in the Vulgate Translation of Koheleth, ZAW 4 (1927), 191-199. Peter J. Gentry, Propaedeutic to a Lexicon of the Three: The Priority of a New Critical Edition of Hexaplaric Fragments, Aramaic Studies 2.2 (2004): 145-74. Peter J. Gentry, The Role of the “Three” in the Text History of the Septuagint: Aspects of Interdependence of the Old Greek and the Three in Ecclesiastes,” Aramaic Studies, 4.2 (2006): 153-192. Sandro Leanza, Le tre versioni geronimiane dell’Ecclesiaste, Annali di storia dell’esegesi 4 (1987), 87-108. Sandro Leanza, Il Traduttore è d’umor nero. Notazioni sulla Vulgata dell’Ecclesiaste. In L. Holz und J.-Cl. Fredouille, eds., De Tertullien aux Mozarabes. Mélanges offerts à Jacques Fon7  Special thanks to Ma Ángeles Márquez for kindly allowing me use a copy of her critical edition prior to publication.

22

The Textual Witnesses

taine. Antiquité tardive et christianisme ancien (IIIe-VIe siècles). Paris, Institut d’Études Augustiniennes, 1992, 107-110. M. Revilla, El Códice Ovetense de los Evangelios y la Biblia de Valvanera, La Cuidad de Dios 117 (1919), 393-399; 118 (1919), 23-28; 120 (1920), 48-55; 122 (1922), 190-210. Alberto Vaccari, ‘Recupero d’un lavoro critico di S. Girolamo.’ In A. Vaccari, Scritti di Erudizione e di filologia, II (Storia e Letteratura, 67), Roma, 1958, 83-146.

1.1  The Old Latin Fragments (La94 95 160) 1.1.1 La94-95 Unadulterated witnesses to the Old Latin of Ecclesiastes are extremely scant. Codex 94 refers to a collection of glosses from the now lost Valvanera Bible (X.-XI. Cent.) in an incunabulum Vulgate Bible printed in Venice, 1478. The same glosses are found on the margin of a two-volume Vulgate Edition from the 12th century. Since these glosses of the Old Latin added to the margins of Vulgate Bibles transmit approximately only twelve verses, the text type cannot be adequately characterised. 1.1.2  The Hexaplaric (?) Recension of La160

According to Vaccari, La160 already represents the hexaplaric Recension of Jerome. For Job, Ziegler states, “Nur ihm (= Hieronymus) verdanke wir die getreueste Wiedergabe der hexaplarischen Rezension des Iob” (Ziegler, Iob, p. 39, 64). Apparently, La160 also transmits the hexaplaric recension for Canticum, but the evidence for Ecclesiastes is sparse (only some 26 verses are extant). In variation against the critical text, La160 is specifically unaligned with the O group in 31, 410, 72a, 72b, and aligned with O only in 713 and 114. 31

οὐρανόν 998 La160 Hi Fa1 SaI Metlem et com III.1,3,26,49 III.2,1 GregNy 372,21 373,15 Or X 11,16 Jb.Cat Κ 192 PsChr LatAmb Ps 118 19 Cass Co 21,12 = Compl Sixt Vulg 𝔐] ηλιον Sc O-411 C´’–(390) 574 260 Aeth Syh Dam Ollem et com An Jb 509 BrevGoth 218 Pel 1 Cor 7,31 SedScot Misc 64,2 = Ald: cf 113c

Places where hexaplaric materials and La160 are both extant are scarce, but in 82 La160 agrees with the Old Latin against hexaplaric revision in Hi (Jerome’s Ecclesiastes Commentary; Page and line numbers for Hi in the edition refer to CCSL 72, 247-360 although the new edition was consulted.). 82

στόμα βασιλέως φύλαξον] pr εγω παραινω 260 ↓ ; ego os regis custodio Hilem 314,34 (sed hab os regis custodi HiLXX 314,40s La160) ↓ 8

8  The down arrow in the Edition means see the Second Apparatus.

The Ancient Translations

23

Nonetheless, the kind of revision one sees in the Ecclesiastes Commentary, e.g. correction towards the Hebrew text, is already apparent in La160: 72

ἀγαθόν 2° ([ἀγαθόν] Gra) = Pesh] αγαθος 371; αγαθων 609; > 336 338 Hi La160 (sed hab bonum Pel 1 Cor 7,31) Syh = Ra 𝔐 𝔔 ↓

713 om ἄν 253 Hilem et com 305,182 307,231 La160

In a number of places, La160 is useful in demonstrating the Old Latin against revisions in the Lemma Text of Jerome’s Commentary: 79

θυμοῦσθαι] irascaris Hilem (sed hab irasci Hicom Amb Exh virg 11,77 Spec 603,4 La160); ad irascendum Vulg

114

ἐν ταῖς νεφέλαις] nubes Hi (sed hab in nubibus La160) = Vulg

1.2  The Ecclesiastes Commentary of Jerome (Hieronymus = Hi)

Jerome has transmitted more material by far than all other sources surviving in Latin. This was to be expected, because he edited the text three times: (1) the hexaplaric Recension in 387, (2) the lemma text of the Ecclesiastes Commentary in 388, and (3) the translation directly from the Hebrew (ex hebraica veritate— later adopted for the Vulgate) in 398. The lemma text of the Ecclesiastes Commentary is problematic, because it contains a mixture of (1) sporadic and non-systematic corrections by Jerome of the Old Latin based directly on the Hebrew text, (2) transmission of the current Old Latin Texts (N.B. plural—designated in the Commentary by Jerome as LXX)— partly in the form of a hexaplaric recension—and (3) correction of the Old Latin using the translations of the Three Jewish Revisors. It is important to note, that even when the Lemma Text is a good witness to the Old Latin, we cannot speak of a monolithic or single translation, but rather of a variety of translations. The siglum Hi by itself refers to the lemma of the Commentary. When the text of the lemma and citations in the Commentary differ, this is designated by Hilem and Hicom respectively. HiLXX is employed to denote citations of the Old Latin made by Jerome in the Commentary. In the Second Apparatus, Hi designates citations of the Three explicitly given as such in the Commentary. First, the Lemma Text entails spontaneous and non-systematic correction on the basis of Jerome’s Hebrew text: 113d 117b

ὅτι] hanc Hi = 𝔐 περιφορὰς καὶ ἐπιστήμην] errores et stultitiam Hi = 𝔐 Vulg ↓

24 24a 25a 25b 212c 316 316 319f 49a

412a 414a 414a 51c 53d 510c 515b 516a

519a 519b 66a 72c 712a 712bc 713c 714e

The Textual Witnesses

ποίημά μου] opera mea Hilem et com 263,54 264,90 = 𝔐 Vulg κήπους καὶ παραδείσους] hortos et viridaria La94 95; hortos et pomaria Hi Vulg (sed hab hortos et paradisos Ruf Cant 94,23) ξύλον πᾶν καρποῦ] omne lignum fructuosum La94 95; lignum omne fructiferum Hi ὀπίσω τῆς βουλῆς] post regem Hi (sed hab La94 95) = 𝔐 Vulg ↓ ὁ ἀσεβής 1°] impietas Hi (sed hab Lucif Athan 1,35) = 𝔐 Vulg τοῦ δικαίου] iustitiae Hi (sed hab Lucif Athan 1,35) = 𝔐 Vulg ↓ καὶ τί ἐπερίσσευσεν ὁ ἄνθρωπος παρὰ τὸ κτῆνος] et amplius homini a pecore nihil est Hi (sed hab La94 95 Aldh Metr 7) = 𝔐 ἀγαθοί] absc 998 Didlem et com; αγαθον 609* (-θων) Anton 1108 Epiph I 27 (sed hab Dionlem 227 Met IV.1,2 Ol Antioch 1676 Bas III 1228 Clim 712 Dam Or ap IX 248,13 PsChr); melius Hilem et com 286,115 287,137 Ruf Eus 11,14 Reg S Bas txt ap 174,7 = Vulg; melior Hilem 286,115; meliores Hilem 286,115 Ep 76,1 Pac Ep 3,27,1; optimi Amb Ep 81,3.8 ter Chrom Matth 22,3,5 ἐπικραταιωθῇ ὁ εἷς] invaluerit super eum unus Hi = 𝔐; praevaluerit contra unum Vulg ↓ ἐξελεύσεται] egreditur Hitxt (sed hab egredietur Hiap) τοῦ βασιλεῦσαι] in regem Hi πρὸ προσώπου] in conspectu Hilem 291,1 (sed hab ante faciem Spec 522,9); coram Aug Spe 8 Hicom 291,7 = Vulg σὺ οὖν ὅσα] quaecumque Hi (sed hab tu itaque quae An Scrip 1,22 Fulg Ep 1,11 Spec 556,9) = ‫;א ָּת ֲא ֶׁשר‬ ַ συν οσα = Ra. 𝔐 (‫)את ֲא ֶׁשר‬ ֵ ὅτι ἀρχὴ τοῦ ὁρᾶν] nisi ut videat Hi: ex ‫ם־ראֹות‬ ְ ‫( ִּכי ִא‬cf 𝔐) ἀπελεύσεται] vadit Hi (sed hab abiit Amb Nab 6,28) καί γε πᾶσαι αἱ ἡμέραι αὐτοῦ ἐν σκότει καὶ ἐν πένθει] et omnibus diebus suis in tenebris comedet Hi (sed hab et quidem omnes dies eius in tenebris et luctu Amb Nab 6,28 Ps 1,28,4) = 𝔐; cunctis diebus vitae suae comedit et in tenebris Vulg πολλά] multum Hi = 𝔐 αὐτόν] > Hi = 𝔐 Vulg ἔζησεν χιλίων ἐτῶν καθόδους] uixerit mille annos duplices Hilem 297,13s = 𝔐 𝔔; duobus milibus annis vixerit Hicom 297,20 = Vulg καθότι τοῦτο τέλος] in quo finis est Hi (sed hab quia hic finis La160; hic enim finis Pel 1 Cor 7,31) = 𝔐 ὅτι—ἀργυρίου] quia quomodo umbra sapientiae sic umbra argenti Hilem 304,143s; quia ut umbra sapientiae sic umbra pecuniae Hicom 303,149s: cf 𝔐 καὶ περισσεία γνώσεως τῆς σοφίας ζωοποιήσει τὸν παρ᾽ αὐτῆς] et quod plus est, scientia sapientiae vivificabit habentem se Hi: cf 𝔐 ὃν ἂν ὁ θεὸς διαστρέψῃ αὐτόν] quem Deus perverterit Hilem et com 305,182 307,231 La160 μηδέν] quicquam Hi = 𝔐

The Ancient Translations

25

σκληρός] stultus Hi (sed hab durus Cassiod 1 Cor 11,30 Fac Def 8,5,17) = 𝔐 Vulg 717b ἵνα μή] cur Hilem 308,279 = 𝔐; ne Cassiod 1 Cor 11,30 Fac Def 8,5,17 Hicom 308,285 = Vulg 718a τὸ ἀντέχεσθαί σε] retinere te Hilem 308,292 (sed hab parcere te Fac Def 8,5,17); sustentare Hicom 308,295 = Vulg 719a βοηθήσει = 𝔔] βοηθειση 797*; βοηθηση 571-797c 766A OlΑ; βοηθεισει 645*; βοηθησι Didlem et com 218,22 219,6 219,11 219,12; φοβηθησει 295 443 534; auxiliabitur Hi Or in Jer hom 335 (sed hab βοηθήσει Or III,56) PsHi Ep 8; confortabit Hilem 309,305 = 𝔐: cf confortavit Hicom 309,322 (sed hab adiuvit HiLXX 309,323) = Vulg 721c ὅπως μὴ ἀκούσῃς] quia non audies = 𝔐 722abc ὅτι—ἑτέρους] etenim frequenter scit cor tuum quia et tu maledixisti aliis = 𝔐 82b καί—σπουδάσῃς] et loquelam iuramenti Dei. Ne festines Hilem 314,34s = 𝔐; et de iuramento et de verbo Dei ne festines HiLXX 315,53 (sed hab et verbo iuramenti dei ne festinaveris La160) 85b κρίσεως] et iudicium Hi 315,68 = 𝔐: cf (6a) 86b ὅτι γνῶσις] afflictio quippe Hilem 316,78 (sed hab quia scientia HiLXX et θʹ 316,89): cf et afflictio Vulg = 𝔐 ↓ 88d ἐν ἡμέρᾳ πολέμου = Pesh] in bello Hilem et com 316,102 317,117 = ἐν τῷ πολέμῳ Ra. 𝔐 ↓ 815e ἡμέρας] εν ημεραις O–637 SaI 2; diebus Hi = 𝔐: cf in diebus Vulg 817e ὅσα ἂν μοχθήσῃ = Pesh] in quo laboravit Hi (sed hab quantumcunque (quantumcumque Demte; quantacumque Demap; quaecunque enim Pel 1,39) laboraverit Hi Pel 1,39 2,5 Pros Dem 12) = 𝔐 92a ματαιότης = Pesh] > Hi = 𝔐 99e om σου 4° Hi = Vulg 𝔐 910b ὡς ἡ δύναμίς σου] in virtute tua Hi = 𝔐 914b ἔλθῃ] uenit Hi 330,309 = 𝔐 Vulg 914c χάρακας μεγάλους] machinam magnam Hi (munitiones = Vulg) = 𝔐 101a θανατοῦσαι] θανουσαι 411 390-415-425-504-522-540-601-732 357 161-248c 645 Anast 641 Constit II 17,4 OlΙΚ omne lat. (praeter Hi) Aeth SaII = Compl; mortis Hi = 𝔐 101b τίμιον—μεγάλης] pretiosa est super sapientiam et gloriam stultitia parua Hi = 𝔐 103c ἅ] > Dam Hi = 𝔐: cf Field 2:399 103c πάντα—ἐστίν] omnis insipiens est Hilem 333,22 (sed hab omnis insipientia sive insipiens Hicom 334,47) = 𝔐 105b ἀκούσιον] ignorantia Fa1(lib) Hilem et com 335,80 335,85 (sed hab ως ακουσιον HiLXX 335,87); per ignorationem Hicom 335,91; ignoratio Hicom 335,107 = 𝔐 ↓ cf: per errorem = Vulg 105b ἐξῆλθεν] προηλθεν DamK; egrediens Hilem et com 335,81 335,85 = 𝔐 Vulg 717a

26 1010a 115a 116d 126a 126b 126d

129c

The Textual Witnesses

ἐὰν ἐκπέσῃ] si retusum fuerit Aug Spe 8 Hilem et com 337,156 338,164 (sed hab si exciderit Cyp Quir 3,86) SaI = Vulg 𝔐 ἐν οἷς—γινώσκων] quomodo non cognoscis tu Hi = 𝔐 ἐπὶ τὸ αὐτό] quasi unum Hi = 𝔐 ἀνατραπῇ] rumpatur Hilem et com 356,246 350,54 356,257 (sed hab vertatur Hi Pach 145,1) = 𝔐 Vulg; non pulsetur Hi Ep 140,13,3 ↓ τὸ ἀνθέμιον 539] vitta Hilem et com 356,246 350,56ap = Vulg; taenia Hicom 350,56te; ornamentum Hi Ep 140,13,3; monile Hi Pach 145,1 ἐπὶ τὸν λάκκον = Pesh] in lacu Hi Ep 140,13,3 (sed hab super lacum Hi Pach 145,1 Ps hom 61,202 PsHi Brev 76 Hilem et com 358,248 350,58 356,263) = 𝔐; super cisternam = Vulg καὶ 2°—παραβολῶν] et audire eos fecit et scrutans composuit proverbia Hi = 𝔐

Jerome completely rewrote 812a-e : 812a-e

ὃς ἥμαρτεν—αὐτοῦ] Quia peccator facit malum centies et elongate ei. Ex hoc cognosco ego, quod erit bonum timentibus Deum, qui timebunt a facie eius. Hilem 318,156ss (sed hab pro (12c-e) HiLXX 319,193s et ego recognosco, quia erit bonum timentibus Deum, ut timeant a facie eius) = 𝔐

Second, the lemma of the Ecclesiastes Commentary contains many hermeneutical and stylistic improvements of La: 13a 15b 16c

18ab 115a 216a 32a 55b 59a

τίς περισσεία] quid superest Hi (sed hab quae abundantia Amb Ps 118 12,21 Aug C D 20,3 Jul 6,24,78 Ps 118 s 12,1,11 12,2,49 Cassiod Rm 8,23) εἰς] ad Hi = Vulg κυκλοῖ κυκλῶν πορεύεται] pr και 130 613; gyrando (grando 94; girando 95) gyrat (girat 94 95) La94 95 Geo Amb Ps 118 12,22; gyrans (girans 160) gyrando (girando 160) vadit La160 Hilem et com 254,156 255,162s 317,115s Bed Rat 8 Hi Ep 100,10 Ezech I 1 Zach II 6 Isid Nat 27,2; gyrans gyrando peragrat PsMar Scr 5; gyrando gyrans vadit Syh πάντες—λαλεῖν] omnes sermones graves non poterit vir loqui Hi διεστραμμένον] perversi perversum La94 95; perversus Hi (sed hab perversum Aug Par 2,35); perversi = Vulg ἔστιν] erit Hilem 269,272 (sed hab HiLXX 269,288) = Vulg, contra 𝔐: cf 111c τοῦ τεκεῖν (pariendi) Amb Exam V 30 Tob 13,43 Cass Co 21,12 Hi Ad Iovin 1 Mi 2 Or Matth 188 La160 Hi = 𝔐] nascendi Vulg ἵνα μή] ne Hilem et com 292,46 294,97 (sed hab ut non Spec 522,12 Syh); ne forte Aug Spe 8 ConcilLat 215,2 = Vulg οὐ πλησθήσεται] ουκ εμπλησθησεται O-411 cII-260 443 766 Anton 1460 OlΑΓΖΗ (sed hab Bas II 213 Cyr VII 989 Met V.1,2 OlΒΔΕΘΙΚ); satiabitur Amb Cain I 5,21 Nab 6,28 Cyp Quir 3,61 Gaud 13,29 Spec 392,8 = La; implebitur Hi 294,115 = Vulg: cf 18c 48d 63c

The Ancient Translations

27

ἐν πλήθει] divitias Hi = Vulg: cf 𝔐 εἰ 1°] αν τε Chr II 585 III 44; καν Didcom 155,3 Hi (sive) SaI = Vulg; et si Amb Nab 6,29 Ps 1,28,4 Syh VA E 511b καὶ εἰ Antioch 1805 Dam–A V Ol] ειτε και Anton 1065; αν τε Chr II 585 III; καν Didcom 155,3 Hi (sive) SaI 6 = Vulg; et si LatAmb Nab 6,29 513a ἀπολεῖται] perierunt Hi; pereunt Vulg 514c ἐν μόχθῳ αὐτοῦ] laboris sui Hi; εκ μοχθου αυτου Aeth; ex labore suo An Scrip 1,10 Spec 392,11; de labore suo Vulg: ex (18d) 515a πονηρὰ ἀρρωστία = Pesh] languor pessimus Hilem 295,155 (sed hab pessimus languor Amb Nab 6,28): cf 62g; languor est pessimus Hicom 296,165 Syh: cf (12a) 517c ἀγαθωσύνην] iucunditatem Hilem 296,186; bonitatem Hicom 297,204 517h ὅτι αὐτό] haec quippe Hi; et haec Vulg 518a om αὐτῷ Didcom 164,2 Hi = Vulg 518d ἐν μόχθῳ αὐτοῦ] de labore suo Hi = Vulg 519a ὅτι οὐ] non enim Hi 519b ὅτι ὁ θεὸς περισπᾷ αὐτὸν ἐν εὐφροσύνῃ καρδίας αὐτοῦ] quia Deus occupant in laetitia cor eius Hi (𝔐 = ‫ ;)כי האלהים מענה בש ֹמחת לבו‬quod Deus occupet deliciis cor eius Vulg 61b πονηρία] malum Hi (sed hab malitiam Amb Ps 37,35,3) = Vulg 62a ἀνήρ, ᾧ δώσει αὐτῷ ὁ θεὸς πλοῦτον] vir cui dedit Deus divitias Hi = Vulg 62b ὑπάρχοντα] υπαρξιν 545mg: cf 518b; substantiam Hi 62c om καί 3° Hi Fa1 SaI II 62c ἔστιν ὑστερῶν] deest Hi 62e ἐξουσιάσει αὐτῷ] dedit ei potestatem Hi SaI: cf Vulg 62e ἀπ᾽ αὐτοῦ] ex eo Hi = Vulg 62f φάγεται] comedit Hi; vorabit Vulg 62f αὐτόν] illud Hi = Vulg 62g ἀρρωστία πονηρά] languor pessimus Hi: cf 515a; miseria magna Vulg 63c οὐ πλησθήσεται] non repleatur Hi 297,9 Hi Eph 3 63d καί γε ταφὴ οὐκ ἐγένετο αὐτῷ] et quidem sepulcrum non fuit ei Hicom 298,37; et sepultura non est (sit Eph 3) illi (ei Eph 3) Am Luc 10,140 Hi Eph 3; nec sepulcrum fuerit ei (illi Hiap) Hilem 297,9 63e ἀγαθόν—ἔκτρωμα = bonum super eum abortum Or Matth 71] bonus est aborsus magis quam ille Ruf Num 42,3; melior est super hunc abortum An Scrip 1,10 Spec 392,12; melius est super eum abortivum Hi Eph 3; melius ab eo esse abortivum Hi 297,9; melior illo abortus Amb Jb 2,4,15: cf Vulg (melior illo sit abortivus) 64a ὅτι ἐν ματαιότητι = quia (nam Or; qui Ruf) in vanitate Hi Eph 3 Or Matth 71 Ruf Num 7,3; in vanitate quippe Hi 297,10 69a ἀγαθόν] melior Hi; melius Vulg 610a εἴ τι ἐγένετο] quid est Hi: cf Vulg 59a 511b

28

The Textual Witnesses W

P

72b om ὅτι πορευθῆναι Hi Anast 593 Chr XI 446 XIV 131 XVI 574s Dam–VV M Dion 988 Isid 1241 Max II 881 PsChrcom 74,51 Amb Ep add 14 Fuga 3 An Jb 3 Aug Spe 8 Hi Ez 8,27 PaulN Ep 25 PsIsid Test 18 Ta Ecl pr 10 = Vulg; 73b ἐν κακίᾳ προσώπου] in maerore vultus Hi (sed hab in malicia vultus La160); in vultu An Scrip 1,22 Spec 557,6 73b ἀγαθυνθήσεται] emendabitur Hi; laetificat La160; oblectabitur An Scrip 1,22 Spec 557,6s; corrigitur Vulg 75a ἀγαθόν] melior est An Fris 404; melius est Amb Ps 1,25,4 Aug Spe 8 Bed Schem 2,1,12 Hi (sed hab bonum est La160) = Vulg 75a ἐπιτίμησιν] correptionem Hi (sed hab increpationem Amb Ps 1,25,4 increpationes La160); ‫𝔔 גערות‬ 75b ᾆσμα] carmen Hi (sed hab canticum Amb Ps 1,25,4 La160) 75b ἀφρόνων] stultorum Hi (sed hab insipientium La160) = Vulg 76c τῶν ἀφρόνων] stulti Aug Spe 8 Hi (sed hab stultorum Amb Exh virg 11,76 BenA Conc 1126 Eugip Reg 28,74) = Vulg 79a θυμοῦσθαι] ut irascaris Hilem 303,115 (sed hab irasci Hicom 303,118 Amb Exh virg 11,77 Spec 603,4 La160); ad irascendum Vulg 79b ἀφρόνων] insipientis Caes Serm 156,6 PsCaes Virg 1162 (sed hab insipientium Cass Co 16,27,4 Inst 8,1,2 Hicom 303,122 (-tiae) La160); stultorum Amb Exh virg 11,77 An Scrip 1,28 Spec 603,5 Hilem et com 303,116 303,124; stulti = Vulg 716a μὴ γίνου] noli esse Hilem et com 307,250 307,254 308,272 (sed hab ne sis Bed Luc 1 Hi Ad Iovin 1 Am 2 Ep 49,2 Is 2,52,34 (ter) Luc 25 Mal 2 Pel 1,39 Zach 3 Or IX,162) = Vulg 718a ἐν τούτῳ] istud Hi 308,292 (sed hab in hoc Fac Def 8,5,17) 725a τοῦ γνῶναι 1°] ut cognoscerem Hicom 311,374 (sed hab ut scirem Amb Mort 7,28 Aug Lib 2,95 Mus 6,4,7 Hilem 311,367) = Vulg 725a τοῦ γνῶναι 2°] ut cognoscerem Hi (sed hab ut scirem Amb Mort 7,28) = Vulg 726e πρὸ προσώπου τοῦ θεοῦ] coram deo Hi 727b λογισμόν] numerum Hilem 312,406; rationem Hicom 312,416 312,434 = Vulg 728c ἐν πᾶσι τούτοις] in his omnibus Chry II 1223 Hi Ad Iovin 1,29 Mi 2,7 La160 (sed hab in omnibus his Hi) 729a ὅ] quia Hi (sed hab quod Apon 1,29 Hi Ad Iovin 1,29 = Vulg) 87a ὅτι οὐκ ἔστιν γινώσκων] quia nescit Hilem et com 316,78s 316,94 (quia ignorat = Vulg); nemo scit Hi Pel 2,5 87a τί τὸ ἐσόμενον] quod futurum est Hilem 316,79; quid factum sit Hicom 316,95; quid futurum sit Hi Pel 2,5 87b ὅτι καθὼς ἔσται] sicut enim erit Hilem 316,79; et quid futurum sit post eum Hicom 316,95s («de verbo ad verbum nunc ex Hebraeo sermone transtulimus») 88e τὸν παρ᾽ αὐτῆς] το παραπαν Dam; habentum se Hi 316,103: cf 510b 712c ↓ 89de τὰ—ἀνθρώπῳ] et dominatus est homo homini Hi: cf Vulg 811a ἔστιν γινομένη] est Hi (sed hab fit Cass Co 2,11,4 7,8,3) Arm 811b ἀπὸ τῶν ποιούντων] facientibus Hi; ab his qui faciunt Cass Co 2,11,4 7,8,3

The Ancient Translations

29

ἔσται = 𝔐] εστι(ν) O 609 613 OlΑ; sit Hilem 319,183 (sed hab erit HiLXX 319,194): cf (12d) 813b μακρυνεῖ] prolonget Hilem 319,183 (sed hab prolongabit HiLXX 319,195) 814a πεποίηται] γινεται DamK; fit Hilem et com 319,210 320, 221 = Vulg; facta est Aug C D 20,3 814b ὅτι φθάνει πρὸς αὐτούς 1°] ad quos perveniunt (-venit Hi Pel) Hi 319,211 Hi Pel 2,5; super quos venit Aug C D 20,3 814d ὅτι φθάνει πρὸς αὐτούς 2°] ad quos perveniunt (-venit Hi Pel) Hi 320,212 Hi Pel 2,5; super quos venit Aug C D 20,3 815c ὅτι 2°—φαγεῖν] nisi comedere Hi 320,228; nisi quod manducabit Aug C D 17,20: cf Vulg (nisi quod comederet) 816c τὸν πεποιημένον] των πεποιημενων 609(-νον) Antioch 1472; quae facta est Hi (quae versatur Vulg) 817g καί γε ὅσα ἂν εἴπῃ] siquidem et si dixerit Hi; et (> Hi Pel 1,39) si dixerit Hi Pel 1,39 2,5 (sed hab et quodcumque dixerit Pros Dem 12); etiam si dixerit Vulg 817h (17h) comma] invenire non poterit Hi Pel 2,5 (sed hab non poterit (potest Pros) invenire Hi Pros Dem 12); reperire non poterit Hi Pel 1,39: cf Vulg (non poterit reperire) 91a om ὅτι Hi, contra 𝔐 91b καί 1°—τοῦτο] ut considerarem universa Hi ↓ : cf 𝔐 Vulg 91d τοῦ θεοῦ] domini Hi (sed hab Pros Dem 12 Voc 1,24) 94a πρὸς—ζῶντας] in omnes viventes Hi 94a ἔστιν ἐλπίς] est confidentia Hi (sed hab spes est Ambrst Ques 39) 94bc ὅτι 2°—νεκρόν] αγαθος ο (> Did PsCaes) κυων ο (> PsCaes) ζων υπερ τον (> PsCaes) λεοντα τον (> PsCaes) νεκρον (τεθνηκοτα Did PsCaes) Did Trin 352 Epiph I 27 PsCaes 876 (sed hab Antioch 1524 Metlem et com VIII.5,2s VIII.5,39); (+ quia Fa1) melior est canis vivens (vivus La160) (+ super Hi) leone (-ni La160) mortuo Fa1 Hi La160 Vulg 95b οὔκ 1°—οὐδέν] necsciunt quicquam Hi 95d ὅτι 3°—αὐτῶν] in oblivione enim venit memoria eorum Hi 97a δεῦρο φάγε] veni manduca Amb Exh virg 11,75; vade (+ et Hilem) comede Hilem et com 324,112 326,164 (sed hab veni comede HiLXX 326,184 Hi Ep 96,1 PsAug Pal 17) = vade ergo et comede Vulg 97c τὰ ποιήματά σου] opera tua Hilem et com 324,113 324,165 Hi Ep 96,1 PsAug Pal 17 (sed hab facta tua Amb Exh virg 11,75 Inst virg 17,110) = Vulg 98b ἐπὶ κεφαλήν] de capite Hilem et com 324,115 327,188 327,198 (sed hab in capite Amb Exh virg 10,62 10,63 Inst virg 17,110) = Vulg 911b τοῖς κούφοις] velocium Hilem et com 328,241 325,147 (velox) Cassiod Rm 9,16 (sed hab levibus Cass Co 13,18 Hicom 328,249 (levis) Tyc Reg 7) = Vulg 911b καὶ οὐ] ουδε Anast 620 Antioch 1740 GregNa II 300 Cass Co 13,18 Cassiod Rm 9,16 Hi (sed hab non Tyc Reg 7) Syh = Vulg 813a

30

The Textual Witnesses

καὶ οὐ] ουδε Anast 620 Antioch 1740 GregNa II 300 Cass Co 13,18 Cassiod Rm 9,16 Hi (sed hab non Tyc Reg 7) Syh = Vulg 911c τοῖς δυνατοῖς] fortium Cassiod Rm 9,16 Hi (sed hab fortibus Cass Co 13,18 Tyc Reg 7) = Vulg 911d καί γε οὐ 1°] non Hicom 329,271; neque Tyc Reg 7; nec Cass Co 13,18 Hilem 328,241 Syh = Vulg 911d τοῖς σοφοῖς] sapientium Hilem 328,241 (sed hab sapientibus Cass Co 13,18 Hicom 329,271) = Vulg 911e τοῖς συνετοῖς] prudentium Hi 328,242 (sed hab prudentibus Cass Co 13,18) 911f τοῖς γινώσκουσιν] scientium Hilem 328,242 (sed hab Cass Co 13,18 Hicom 329,273) 912b οἱ θηρευόμενοι] qui retinentur Hilem 330,285 qui tenentur Hicom 324,117 capiuntur Hicom 326,157 330,297 = Vulg 912b ἐν ἀμφιβλήστρῳ κακῷ] in captione pessima Hi 912c om καί 2° Hi, contra 𝔐 c 912c ἐν 2°] > Dam–H T Hilem 330,286 (sed hab Hicom 324,118) 912d παγιδεύονται] corruent Hilem 330,286 (sed hab capientur Hicom 324,118; capiuntur = Vulg) 917a ἀναπαύσει] in silentio et quiete Ruf Ex 167,7 (sed hab in requiem Spec 469,10 in quiete Hi 332,361); in silentio = Vulg 917a ἀκούονται] discuntur Ruf Ex 167,7; audientium Spec 469,10 (sed hab audiuntur Hi = Vulg) 101a σαπριοῦσιν] polluunt Hi 103b ὑστερήσει] minuitur Hilem 333,22 (sed hab indiget Hicom 334,46) 103c λογιεῖται] dicit Hilem et com 333,22 334,46 104a τοῦ ἐξουσιάζοντος] potestam habentis Hi Ep 22,26 130,8,5 Eph 2 3 Ezech 11,38 Joel 1,6,7 Is 16,23,41 Ruf 2,7,24 Zach 1,5 Hilem et com 334,54 336,111 (sed hab dominantis Fil 131,6) = Vulg 104b om σου Hicom 336,112 104b μὴ ἀφῇς] ne dimiseris Hilem 334,55 Hi Ep 130,8,5 Joel 1,6,7 Is 16,23,42 Ruf 2,7,25 Zach 1,5 PsHi Ep 1,24 = Vulg; non dimittas Ruf Num 276,4; ne dimittas Apon 2,25 Cass Co 1,19,3 Ruf Or princ 245,22; ne deseras GregM Rg 5,16 5,64, 5,108; ne dederis Hi Ep 22,26 Eph 2 Hicom 336,112; non demus Hi Eph 3 (sed hab non derelinquas Amb Jb 4,7,28; ne derelinquas Hi Ezech 11,38 Pas 33,1 Ruf Cant 211,19 non relinquas Amb Ps 37,35,2 ne relinquas Ruf Or princ 250,30 noli relinquere Amb Ep 81,13) 104c ἴαμα] sanitas Ruf Cant 211,19 Om S Bas 1724 Or princ 245,22 250,30 Hilem 334,55; sanitatem uel curationem Hicom 334,66; curatio = Vulg (sed hab cura Amb Ps 37,35,2) 104c καταπαύσει] compescit Ruf Cant 211,19 Om S Bas 1724 Or princ (con-) 245,22 250,30; mitigabit Amb Ps 37,35,2; requiescere facit (-ciet Hiap) Hite 334,55s; quiescere facit Hi Ep 130,8,5; cessare faciet (-cit Vulgap) Vulgte 911c

The Ancient Translations

105a 107a 108b

1010b

1010c

1010d

1011a 1013a 1013a 1014b 1015b 1016a

1016b 1017c 1018b 1019c 1020a 1020b 1020c 1020c 113b

31

πονηρία] malum Hi = Vulg (sed hab malitia Ruf GrN 1,50) ἐφ᾽ ἵππους] in equis Hilem 335,83 (sed hab super equos Hicom 335,101) = Vulg E O ὄφις] pr η DamKV ; pr ο DamV ; coluber Aug Spe 8 Hi Ezech 11,38,1/23 Is 16,22,42 PsMel P 7,56 V 7,35 9,75,1 SedScot Misc 64,8 (sed hab serpens Amb Ep 39,1 Inst virg 9,60 Hilem et com 336,120 336,127 Vinc Com 21,2) = Vulg καί 1°—ἐτάραξεν] et hoc non ut prius sed conturbatum erit Hilem 338,164s (sed hab et faciem eius turbauerit Hicom 337,157); et hoc non ut prius sed hebetatum fuerit Aug Spe 8 = Vulg καὶ δυνάμεις δυναμώσει] virtutibus corroborabitur Hilem 338,165; fortitudinibus corroborabitur HiHEB 338,179 (sed hab et fortitudine confortabitur HiLXX 337,161) καί 3°—σοφία] et reliquum fortitudinis sapientia est Hilem 338,166; abundantia autem hominis sapientia eius est Hi Pach 80,7 (sed hab et superflua robusti sapientia incipiet HiLXX 337,161s); et post industriam sequitur sapientia = Vulg ἐν οὐ ψιθυρισμῷ] in silentio Aug Spe 8 Bed Aet 17 Hi Ep 125,19,1 Is 18,1,68 Hi 338,183 (sed hab non in sibilo Cass Co 2,11,6 18,16,10) = Vulg στόματος 1°] > Hi (sed hab oris ConcilLat 195,12) = Vulg ἀφροσύνη] insipientia Hilem 339,209; stultitia ConcilLat 195,12 Hicom 339,216 = Vulg οὐκ ἔγνω] pr και 139-147-159-425mg-560-798-cII–561 = 𝔐mss; ου και γνω 155; ignorat Hi = Vulg οὐκ ἔγνω] nesciunt Hi (sed hab non cognoverunt Cass Co 24,24,6) νεώτερος] adolescens Hite; adulescens Aug C D 17,20 Hi Is 4,3,72 Hiap; iuvenis La160 An Fris 282 Can Hib 25,5 Euch Form 2 5 Hi Is 2,8,17 2,46,19 Isid Sent 2,43,4 Ruf GrN 1,73,1 Salv Eccl; puer PsCyp Ab 9 = Vulg ἐσθίουσιν] manducant La160; comedunt An Fris 282 Aug C D 17,20 Euch Form 2 Hi Is 2,8,18 2,46,19 PsCyp Ab 9 Isid Sent 2,43,4 Salv Eccl Hi = Vulg οὐκ αἰσχυνθήσονται] in confusione Aug C D 17,20 Hilem et com 340,255 341,293; confundentur Hi Pach 113,18: cf 𝔐 ἐν ἀργίᾳ] in (> BrevGoth) infirmitate BrevGoth 464 Hi (sed hab in segnitia Cass Co 6,17,1) ἐπακούσεται] oboediunt Hilem 342,311 (sed hab Hicom 343,345) ἐν συνειδήσει] in mente Hi (sed hab in conscientia La160 Amb Inst virg 1,7); in cogitatione = Vulg ἐν ταμιείοις κοιτώνων] in secreto cubilis (cubiculi Hiap) (sed hab in penetralibus cubiculorum Amb Inst virg 1,7) Hite; in secreto cubiculi = Vulg πετεινόν] volucres Hilem 343,350te; avis Hilem et com 343,350ap 343,356te = Vulgte; aves Hilem et com 343,350ap 343,356ap = Vulgap ἀποίσει] portabunt Hilem 343,351 = Vulgap (sed hab portabit Vulgte); auferet Hicom 343,357) ἐκχέουσιν] effundent Amb Ep 36,4,46 Hi = Vulg; effundunt Apr 1

32 113c

113c

113d

114a 114b 114b 115b 116a 116c 118a 118d 118e 121c 123a 123c 124a 124a 125f 124b 124c 124d

125a 125a

The Textual Witnesses

ἐν τῷ νότῳ] sive ad austrum GregM Iob 8,15,13 Jul-T Pro 2,14 Ta Ecl 26 Sent 3,54 Hicom 345,44 (sed hab ad austrum GregM Iob 12,4,19s Hilem et com 345,40 346,61 = Vulg) Fa1 καὶ 2°—βορρᾷ] aut ad aquilonem GregM Iob 12,4,20 Hilem et com 345,41 346,61 = Vulg; sive ad aquilonem GregM Iob 8,15,13 Jul-T Pro 2,14 PsMel P 7,5 V 7,1,5 Ta Ecl 26 Sent 3,54; sive ad boream Hicom 345,45 τόπῳ, οὗ πεσεῖται] in loco ubi ceciderit Hilem et com 345,41 346,61s; ubicumque ceciderit PsMel P 7,5 V 7,1,5; in quoqumque ceciderit Ta Ecl 26 Sent 3,54 = Vulg c τηρῶν] pr ο DamR; κηρων DamL ; qui observat CPA Hi (sed hab observans La160) = Vulg βλέπων] pr ο 125´-545; qui aspicit Hi (sed hab aspiciens La160) CPA ἐν ταῖς νεφέλαις] nubes Hi (sed hab in nubibus La160) = Vulg ὡς] pr et Hi ἐν πρωΐᾳ] in matutino Hilem 347,107; mane Hicom 348,151 = Vulg: cf 1016b ποῖον στοιχήσει] quid placeat Hi ἔτη / πολλά] tr Didcom 331,19 332,1 (sed hab Didlem et com 331,14.19) Hicom 348,128 (sed hab Hilem 347,111 = 𝔐 Vulg) πολλαί] πολλα OlΒ; plurimae Hi τὸ ἐρχόμενον] quod venturum est Hi κακίας] pessimi Hi Pach 144,13 (sed hab malitiae Hi); ᾗ ἐάν] cum Hilem 352,125te; quo Hilem 352,125ap (sed hab qua Hi Ep 140,13,3 Pach 144,20 Hicom 356,256) ἤργησαν] cessent Hi Pach 144,20; cessant Hi Ep 140,13,3; cessabunt Hilem 353,141 κλείσουσιν] clauduntur Hi Ep 140,13,3 (sed hab claudent Hi Pach 144,25 Hi = Vulg) ἐν ἀγορᾷ] in platea Hi (sed hab in foro Hi Ep 140,13,3 Pach 144,25) = Vulg: cf ἐν ἀγορᾷ] in platea Hilem et com 355,204 350,52 353,166 (sed hab in foro Hi Ep 140,13,3 Pach 144,27) = Vulg ἐν ἀσθενείᾳ] in humilitate Hi (sed hab in infirmitate Hi Ep 140,13,3 Pach 144,25) = Vulg τοῦ στρουθίου] volucris Hilem 353,160 = Vulg (sed hab volucris sive passeris Hicom 354,172; passeris Pach 144,25); avis Hi Ep 140,13,3 ταπεινωθήσονται] obmutescent Hilem 353,161; obmutescere … sive, ut melius habet in hebraeo, surdescere Hicom 354,193; obsurdescent (sed hab humiliantur Hi Ep 140,13,3te; humiliabuntur Hi Ep 140,13,3ap Pach 144,25) = Vulg ἀπὸ ὕψους] ab excelsis Hilem 354,198; ab excelso Hicom 350,43 (sed hab ab alto Hi Ep 140,13,3 Pach 144,27) ὄψονται] timebunt Hilem et com 354,198 350,43 (sed hab aspicient Hi Ep 140,13,3 Pach 144,27)

The Ancient Translations

125a

125e 125e 125f 126a 1210a 1211cd

33

θάμβοι] θαμβ 766B (|); εκθαμβοι 336´; θαμβος C 411 L C´ d 443 613 698 Hi Ep 140,13,3te Aeth Arm = Ald; formidabunt Hilem et com 354,198 350,44 (sed hab pavores Hi Ep 140,13,3ap Pach 144,27) ἐπορεύθη] ibit Euch Form 1 SedScot Misc 64,12 Hi = Vulg; abibit Hi Ep 140,13,3ap (sed hab abiit Hi Ep 140,13,3te Pach 144,27) αἰῶνος] saeculi Hi Pach 144,27 (sed hab aeternitatis Euch Form 1 Hi Ep 140,13,3 SedScot Misc 64,12 Hilem et com 355,204.209 = Vulg) ἐν ἀγορᾷ] in platea Hilem et com 355,204 350,52 353,166 (sed hab in foro Hi Ep 140,13,3 Pach 144,27) = Vulg: cf 124a ἕως—μή] antequam Hilem et com 356,246 350,54 356,257 (sed hab quoadusque Hi Ep 140,13,3 Pach 145,1) = Vulg πολλά] multum Hilem 357,285 (sed hab multa Hilem et com 357,285 359,241) ἐδόθησαν ἐκ ποιμένος ἑνός] data (dati Hi Or in Ez hom 328 Is 9,13,41) sunt a pastore uno Hi Or in Ez hom 328 Is 9,13,41 Hi = Vulg; dati ab uno pastore Or Matth X 281; unius pastoris EpiphSchol Cant 197

Third, Jerome corrected the text of La on the basis of the Three—usually from Symmachus. Cannon lists 56 instances where Hieronymus borrowed renderings from Symmachus for the Vulgate in Ecclesiastes. In an unpublished work, J. Ziegler increased the number to 61. Many of these borrowed renderings are already present in the Lemma Text of the Commentary. Examples: 416c 𝔐 Edition Hi App II Field

‫רּוח‬ ַ ‫ִּכי־גַ ם־זֶ ה ֶה ֶבל וְ ַר ְעיֹון‬ ὅτι καί γε τοῦτο ματαιότης καὶ προαίρεσις πνεύματος. sed et hoc vanitas et praesumptio spiritus sed et hoc aura et pastio venti Sym. apud Hicom σʹ ἀλλὰ καὶ τοῦτο ἀτμὸς καὶ βόσκησις ἀνέμου.

93c 𝔐 Edition Hi App II Field

‫א־רע‬ ָ ‫י־ה ָא ָדם ָמ ֵל‬ ָ ֵ‫וְ גַ ם ֵלב ְּבנ‬ καί γε καρδία υἱῶν τοῦ ἀνθρώπου ἐπληρώθη πονηροῦ, sed et cor filiorum hominum repletum est malitia sed et cor filiorum hominum repletur malitia Sym. apud Hicom σʹ sed et cor filiorum hominum repletur malitia

55b τοῦ θεοῦ = Pesh] angeli Hilem et com 292,45 294,96 = Vulg ↓ App II τοῦ θεοῦ] αʹ σʹ θʹ ‫ ܕܡܐܠܟܐ‬Syh (= τοῦ ἀγγελοῦ) ὅτι Ἄγνοιά ἐστιν] quia ignorantia est Hilem 292,45 Spec 522,12 (sed hab quoniam ignoratio est Hicom 294,96) ↓ App II ὅτι ἄγνοια] σʹ (αʹ Hi) ὅτι (> Hi) ἀκούσιον Hi (litt gr) Syh 55b

34

The Textual Witnesses

510b παρ᾽ αὐτῆς] εχοντι αυτην O–V 766 Hi Fa1 ↓ : cf 512b App II τῷ παρ᾽ αὐτῆς] τῷ ἔχοντι (σχοντι 260cat) αὐτό (αυτην OlΒ; αυτον Met OlΗ) 260cat Metcom Ol) 248´ 252 (s nom) 260cat Metcom Ol Syh καί 2°—ὑπνῶσαι] et saturitas divitis non sinit eum dormire Hi (sed hab et satiato (etsi quis satiatus fuerit Ps 1,28,4) divitiis non est qui sinat eum dormire Amb Nab 6,29 Ps 1,28,4) = Vulg ↓: cf 𝔐 App II τοῦ δούλου] σʹ θʹ ὁμοίως τοῖς οʹ Syh (= τοῦ δουλεύοντος ↑) | καί—ὑπνῶσαι] σʹ ἡ δὲ πλησμονὴ τοῦ πλουσίου οὐκ ἐᾷ καθεύδειν 248´ 511cd

Other Examples: 110a 110a

113d 61b 68a 68b 77b 710c 714a 725b 725d 729a 810a 98b 917b 917b 918a 106a 1011b 1013b

ὃς—ἐρεῖ] estne verbum de quo dicatur Hi ↓ (sed hab Pesh) = 𝔐 ὃς λαλήσει] si quis quid loquetur Ruf Or princ 1,4,5 ↓ ; si qui loquetur Ruf Or princ 3,5,3 ↓ περισπασμόν] occupationem Hilem et σʹ (sed hab Hicom 259,301.321 Anast 577 Antioch 1713 Ath III 936 (‑μους) Dam Hi Ad Iovin 1,13) = Vulg ↓ ἐπὶ τὸν ἄνθρωπον] παρα τοις ανθρωποις Hilem 297,2 (apud homines) = Compl Vulg ↓ ὅ τι περισσεία τῷ σοφῷ] τι ουν περισσον τω σοφω Olcom Hi (quid enim est amplius sapienti) ↓ ; quid habet amplius sapiens Vulg διὰ τί—πορευθῆναι] quid pauperi nisi scire ut vadat Hi ↓ τὴν καρδίαν εὐτονίας] cor fortitudinis Hi ↓ ; robur cordis Vulg ὅτι 2°—σοφίᾳ] non enim sapienter Hi ↓ ; quia non sapienter Hi Ad Iovin 1,29 (sed hab quoniam non in sapientia Hi Am 3,6 PsBas Is 1,4) ζῆθι] esto Hi = 𝔐 ↓ ψῆφον] rationem Hi (sed hab numerum Amb Mort 7,28 Aug Lib 2,95 Mus 6,4,7) = Vulg ↓ καὶ ὀχληρίαν καὶ περιφοράν] et molestiam et iactationem Amb Mort 7,28; et imprudentium errorem Hi ↓ : cf et errorem inprudentium Vulg πλήν] solummodo Hi (sed hab verumtamen Hi Ad Iovin 1,29) = Vulg ↓ εἰς—εἰσαχθέντας] sepultos et venerunt Hi = 𝔐 ↓ μὴ ὑστερησάτω] non deficiat Hilem 324,115 (sed hab desit Amb Inst virg 17,110 Hicom 327,188 327,198) ↓ = Vulg ἐξουσιαζόντων] potestatem habentis Hi ↓ : cf 104a 5b (principis = Vulg 𝔐) ἐν ἀφροσύναις] in stultis Hi ↓ = 𝔐 (inter stultos = Vulg) ὑπὲρ σκεύη πολέμου] quam arma bellica (sed hab super vasa belli Hi) = Vulg ↓ ἐδόθη ὁ ἄφρων] positum stultum (sed hab dari stultum Hi) = Vulg ↓ τῷ ἐπᾴδοντι] habenti linguam Hi (sed hab incantatori Bed Aet 17 Cass Co 2,11,6 18,16,10) = 𝔐 περιφέρεια πονηρά] circuitus maligna ConcilLat 195,12; error pessimus Hi ↓ = Vulg

The Ancient Translations

121c 122a 126a 126b 1213c

35

ἕως ὅτου μή] antequam Hilem et com 349,2 351,91 356,254 (sed hab donec Hi Pach 144,13) = Vulg ↓ ἕως οὗ μή] antequam Hilem et com 351,97 349,21 352,21 = Vulg ↓; quando Hi Ep 140,13,3 (sed hab donec Hi Pach 144,20) ἀνατραπῇ] rumpatur Hilem et com 356,246 350,54 356,257 (sed hab vertatur Hi Pach 145,1) = 𝔐 Vulg; non pulsetur Hi Ep 140,13,3 ↓ συνθλιβῇ] recurrat PsMel P 13,73 V 4,25,4 Hilem et com 356,246s 350,55 (sed hab conteratur Hi Ep 140,13,3 Pach 145,1) = Vulg ↓ ὅτι τοῦτο] hoc enim est Hite (sed hab quia hoc est Aug C D 20,3; quoniam hoc est Hi Pach 147,23 Spec 345,13) ↓ ; hoc est enim Hiap = Vulg ↓

Jerome regularly corrected Latin reflexes of καί γε to sed et based on Symmachus’ rendering ἀλλὰ καί. For an exhaustive analysis, see Peter J. Gentry, The Role of the “Three” in the Text History of the Septuagint. 226g ὅτι καί γε] sed et Hi = Vulg: ex σ΄ 416c ὅτι καί γε] sed et Hi = Vulg: ex σ΄ 518a καί γε] sed et Hi ↓; et Vulg: ex σ΄ 62f ὅτι] αλλ Antioch 1805 DamR Hi (sed) 69b καί γε] sed etiam Aug Ps 118 s 12,3,7; sed et Hi = Vulg 76c καί γε = 𝔐mss] sed et Hi; ‫𝔐 גם‬mss 𝔔 810e καί γε] sed et Hi ↓ 93a καί γε] sed et Hi; quia Arm; > Dam–T Fa1 93d καί 2°] absc Fa1; sed et Hi ↓ ; + γε 254´ 96a καί γε 1°] sed et Hi; om γε Arm Fa1 3 Sa2 96a om γε 2° Arm Fa1 3 SaII 2 Hi 96b καί γε 3°] > Fa1 Hiap; om γε 3° Arm Hitxt 913a καί γε] sed et Hi ↓ 103a καί γε] sed et Hi = Vulg 125a καί γε] sed et Hi (sed hab et quidem Hi Ep 140,13,3 Pach 144,27) ↓

1.3  The Indirect Old Latin Tradition (Latin Patristic Citations of Ecclesiastes) Roger Gryson, Répertoire général des auteurs ecclésiastiques latins de l’antiquité et du haut Moyen Âge. 5e édition mise à jour du Verzeichnis der Sigel für Kirchenschriftsteller commencé par Bonifatius Fischer, continué par Hermann Josef Frede. 2 Bde. Freiburg im Breisgau: Herder, 2007.

The basic witnesses interesting for the LXX (disregarding Jerome, for whom see above) are as follows: Ambrosius (20×); Cassiodorus (8×); Rufinus (6×); Augustine (5×) and Pseudo-Augustine Speculum/Excerpts (=AN scrip) (4×). In addition, 27 further writers provide at most one or sometimes two to three citations.

36

The Textual Witnesses

Since the citations of Ambrosius and Cassiodorus are more numerous, the Latin text they transmit should be characterised if possible. Amb The citations of Amb preserve no hexaplaric readings and show no revision towards the Vulgate in the following sixteen instances: 26b, 44c (ἀνδρός), 413b, 53a, 511a (τοῦ δούλου), 511b (εἰ), 511b (καὶ εἰ), 515c, 65b, 612c, 76a, 725a (ἐκύκλωσα—καρδία μου), 725c (ἀφροσύνην), 97a (δεῦρο φάγε), and 97c (ἤδη). Of these sixteen citations, Amb agrees with no other Greek witnesses in the following five instances: 53a, 511b (εἰ), 511b (καὶ εἰ), 725a (ἐκύκλωσα—καρδία μου), and 97a (δεῦρο φάγε). The remaining eleven instances agree with L (4×), C´ (5×), cII (5×), d (5×), and k (5×). Amb = Vulg at 410a, but probably does not represent revision towards Vulg because of the presence of the L group. This agreement might be coincidence, but in the following variation in 410a, Amb and L share a unique reading (+ ο ετερος). The only O witness for the two variants is 411, a manuscript which only partially belongs to the O group. Thus Amb (and Vulg) may be preserving the Old Latin given that L, which is known for preserving early Latin variants, also transmits the reading. Cassiod Cassiod reads with O at 717a (σκληρός; = Lemma) and Vulg at 717b (ἵνα μή). The textual affiliations of Cassiod do not differ greatly from that of Amb, but the evidence is too scant for any conclusions. Abbreviations for the Latin patristic writers: The abbreviations in parentheses in the following list are those used by Gryson, Répertoire Général, where one may find all biographical and bibliographical information for these sources. Square brackets indicate editions employed for easy quick reference. A-SS (A-SS) = Acta vel Passiones vel Vitae Sanctorum Helia (Helia) = Vita S. Heliae [G. Antolín, Boletín de la R. Academia de la Historia 54 (1909) 122-128. 204-246. 265] Amb (AM) = Ambrosius (of Milan) Cain (Ca) = De Cain et Abel [CSEL 32,1] Ep (ep) = Epistularum libri 1-10 [CSEL 82,1 82,2 82,3] Exh virg (exh) = Exhortatio virginitatis [F. Gori, Sancti Ambrosii episcopi Mediolanensis opera 14/2, Milan/Rome 1989, 198-270] Exam (ex) = Exameron [CSEL 32,1] Fide (fi) = De fide ad Gratianum Augustum [CSEL 78]

The Ancient Translations

37

Fuga (fu) = De fuga saeculi [CSEL 32,2] Helia (fu) = De helia et ieiunio [CSEL 32,2] Isa (Is) = De Isaac vel anima [CSEL 32,1] Inst virg (inst) = De institutione virginis et S. Mariae virginitate perpetua ad Eusebium [F. Gori, Sancti Ambrosii episcopi Mediolanensis opera 14/2, Milan/Rome 1989, 110-194] Jac (Jac) = De Iacob et vita beata [CSEL 32,2] Jb (Jb) = De interpellatione Job et David [CSEL 32,2] Jul (Jul) = Contra Iulianum (PL 44, 641-874] Luc (Lc) = Expositio Evangelii secundum Lucam [CCSL 14] Mort (mort) = De bono mortis [CSEL 32,1] Myst (my) = De mysteriis [CSEL 73] Nab (Nab) = De Nabuthae [CSEL 32,2] Noe (Noe) = De Noe [CSEL 32,1] Paenit (pae) = De paenitentia [CSEL 73] Ps (Ps) = Explanatio XII psalmorum [CSEL 64] Ps 118 (118 Ps) = Expositio de Psalmo CXVIII [CSEL 62] Sacr (Sa) = De Sacramentis [CSEL 73] Sat (Sat) = De excessu fratris Satyri [CSEL 73] Tob (Tb) = De Tobia [CSEL 32,2] Ambrst (AMst) = Ambrosiaster [CSEL 81,1-3] Eph (Eph) = Commentarius in Epistulam ad Efesios Rom (Rm) = Commentarius in Epistulam ad Romanos Ques (q) = Quaestiones Veteris et Novi Testamenti [CSEL 50,13-416] An (AN) = Anonymus Agap (Agap) = Epistula cuiusdam ad Agapium episcopam [W. Gundlach, Monumenta Germaniae Historica Epistolae 3 (1892) 685-686] Fris (Fris) = Florilegium Frisingense [CCSL 108D] Jb (Jb) = Commentarius in Iob [PG 17, 371-522] Ps M (Ps M) = Psalter-Glossen in der Handschrift VL 307 [F. Unterkircher, Die Glossen des Psalters von Mondsee, Freiburg 1974] Ps sen (Ps sen) = Glosa psalmorum ex traditione seniorum [H. Boese, Anonymi glosa psalmorum ex traditione seniorum I (Ps 1-100). II (Ps 101-150), Freiburg 1992] Scrip (scrip) = Testimonia divinae scripturae (et patrum) [CCSL 108D] Apon (APO) = Ap(p)onius [CCSL 19] Apr (APR) = Apringius (Bishop of Beja in Portugal, 531-548), Tractatus in Apocalypsin [PLS 4, 1222-1248]

38

The Textual Witnesses

Aug (AU) = Augustinus An (an) = De natura et origine animae [CSEL 60] C D (ci) = De civitate Dei libri 22 [CCSL 47 48] Conf (cf) = Confessionum libri 13 [CCSL 27] Ep (ep) = Epistulae [CSEL 34,1 34,2 44 57 58] Ep Div (ep Div) = Epistulae [CSEL 88] Gen ad litt (Gn li) = De Genesi ad litteram libri 12 [CSEL 28/1] Ioh Evang (Jo) = In Iohannis Evangelium tractatus 124 [CCSL 36] Lib (lib) = De libero arbitrio libri 3 [CCSL 29] Mus (mus) = De musica libri 6 [PL 32, 1081-1194] Par (Par) = Contra epistulam Parmeniani libri 3 [CSEL 51] Pel (Pel) = Contra duas epistulas Pelagianorum libri 4 [CSEL 60] Perf (perf) = De perfectione iustitiae hominis [CSEL 42] Ps (Ps) = Enarrationes in Psalmos [CCSL 38-40] Serm (s) = Sermones 562 [PLS 2] Spe (spe) = Liber qui appellatur Speculum (CSEL 12, 3-285) Spir (sp) = De spiritu et littera liber [CSEL 60] Trin (tri) = De Trinitate libri 15 [CCSL 50-50A] Bach (BACH) = Bachiarius Fid (fi) = Libellus de fide [PL 20, 1019-1036] Bas (BAS) = Basilius (the Great) Hom 1-2 (h 1-2) = Early translation of both Homilies in PG 31, 217-261 [D. Amand, Revue Bénédictine 57 (1947) 61-81] Bed (BED) = Venerable Bede Aed (aed) = De templo libri 2 [CCSL 119A] Cath (cath) = In epistolas septem catholicas [CCSL 121] Cant (Ct) = In Cantica Canticorum libri 6 [CCSL 119B] Hom (h) = Homeliarum Evangelii libri 2 [CCSL 122] Luc (Lc) = In Lucae Evangelium expositio [CCSL 120] Prv (Prv) = In Proverbia Salomonis libri 3 [CCSL 119B] Rat (rat) = De temporum ratione [CCSL 123B] Sam (Sam) = In primam partem Samuelis libri 4 [CCSL 119] Schem (sche) = De schematibus et tropis [CCSL 123A, 142-171] Tab (tab) = De tabernaculo et vasis eius ac vestibus sacerdotum libri 3 [CCSL 119A] BenA (BEN-A) = Benedictus of Aniane [PL 103, 423-1380]

The Ancient Translations

39

BrevGoth (Brev. Goth.) = Brevarium Gothicum [PL 86] Caes (CAE) = Caesarius of Arles Serm (s) = Sermones [CCSL 103-104] Can (CAN) = Canones Hib (Hib) = Collectio canonum Hibernensis [H. Wasserschleben, Die irische Kanonensammlung, Leipzig 21885] Cass (CAn) = Johannes Cassianus of Marseille Co (co) = Conlationes Patrum 24 [CSEL 13] Inst (in) = De institutis coenobiorum et de octo principalium vitiorum remediis libri 12 [CSEL 17] Cassiod (CAr) = Cassiodorus Senator An (an) = De anima [CCSL 96] 1 Cor (1 Cor) = Expositio in Epistulas S. Pauli [PL 68, 413-686] Hist (hist) = Historia tripartita [CSEL 71] Ps (Ps) = Expositio Psalmorum [CCSL 97-98] Rm (Rm) = Expositio in Epistulas S. Pauli [PL 68, 413-686] Chry (CHRY) = Chrysostomus [Opera D. Ioannis Chrysostomi I-V, Ed. J. Froben, Basel 1558] Chrom (CHRO) = Chromatius Aquileiensis Matth (Mt) = Tractatus in Matthaeum [CCSL 9, 371-447; 9A] s (s) = Sermones [CCSL 9A; CCSL 9A Supplementum] ConcilLat (CO-Lat) = Concilium Lateranense [R. Riedinger, Concilium Lateranense a. 649 celebratum, Berlin 1984] Cyp (CY) = Cyprianus (of Carthage) Quir (te) = Ad Quirinum (Testimoniorum libri 3) [CCSL 3] EpiphSchol (EP-SC) = Epiphanius Scholasticus Cant (Ct) = Translation of Commentary on Canticum by Philo of Carpasia [Corona Patrum 6; PG 40] Euch (EUCH) = Eucherius (Bishop of Lyon) Form (int) = Formulae spiritalis intellegentiae [CSEL 31; PL 50, 727-772]

40

The Textual Witnesses

Eugip (EUGI) = Eugippius, Abbot of Castellum Lucullanum at Naples Reg (reg) = Regula [CSEL 87] Eus-A (EUS-A) = Eusebius of Alexandria Eus-G (EUS-G) = Eusebius Gallicanus [CCSL 101; 101A; 101B] Ser (s) = Sermones extravagantes 9 Eust (EUST) = Eustathius [E. Amand de Mendieta, S. Y. Rudberg, TU 66, Berlin 1958]

Fac (FAC) = Facundus (Bishop of Hermiane) Def (def) = Pro defensione trium capitulorum libri 12 [CCSL 90A] Ferreol (FEol) = Ferreolus (Bishop of Uzès [V. Desprez, La Regula Ferrioli: Texte Critique, Revue Mabillon 60 (1982) 117-148] Fil (FIL) = Filastrius of Brescia [CCSL 9] Fris (AN Fris) = Florilegium Frisingense [CCSL 108D] Fulg (FU) = Fulgentius of Ruspe Ep (ep) = Epistulae [CCSL 91; CCSL 91A] Inc (inc) = Ad Scarilem de incarnatione [CCSL 91] GregM (GR-M) = Gregorius Magnus Dial (dia) = Dialogorum libri 4 [SC 251; 260; 265] Evang (ev) = Homiliae in Evangelia 40, libri 2 [PL 76, 1075-1312] Ezech (Ez) = Homiliae in Hiezechihelem prophetam [CCSL 142] Iob (Jb) = Moralium libri sive Expositio in librum Job [CCSL 143-143B] Past (past) = Regula pastoralis [CCSL 141] Rg (Rg) = In librum primum Regum expositionum libri 6 [CCSL 144] Hes (HES) = Hesychius (Presbyter in Jerusalem) Lev = Commentarius in Leviticum [PG 93, 787-1180] Hi (HI) = Hieronymus Ad Iovin (Jov) = Adversus Iovinianum libri 2 [PL 23, 211-338] Alt (alt) = Altercatio Luciferiani et Orthodoxi [PL 23, 155-182] Am (Am) = Commentariorum in Amos prophetam libri 3 Did (Did) = Didymi Alexandrini liber de Spiritu Sancto [SC 386] Ep (ep) = Epistulae [CSEL 54-55]

The Ancient Translations

41

Eph (Eph) = Commentarii in epistulam ad Ephesios libri 3 [PL 26] Ezech (Ez) = Commentariorum in Hiezechielem libri 14 [CCSL 75] Gal (Gal) = Commentarii in epistulam ad Galatas libri 3 [PL 26] H (h) = De diversis locis homiliae 10 [CCSL 78] Is (Is) = Explanationum in Esaiam libri 18 [Vetus Latina. Aus der Geschichte der lateinischen Bibel 23, 27, 30, 35, 36; Ed. R. Gryson et al.] Joel (Jl) = Commentariorum in Ioelem prophetam ad Pammachium liber [CCSL 76] Luc (Lc) = Origenis in Lucam homiliae: [GCS 49 (35)] Mal (Mal) = Commentariorum in Malachiam prophetam ad Minervium et Alexandrum liber [CCSL 76A] Marc (Mc) = Tractatus 10 in Marci Evangelium [CCSL 78] Matth (Mt) = Commentariorum in Matheum libri 4 [CCSL 77] Mi (Mi) = Commentariorum in Michaeam prophetam libri 2 [CCSL 76] Or in Ez hom (Ez h) = Origenis in Ezechielem homiliae 14 [GCS 33] Or in Is hom (Is h) = Origenis in Isaiam homiliae 9 [GCS 33, 242-289] Or in Jer hom (Jr h) = Origenis in Ieremiam homiliae 14 [SC 232, 238] Pach (Pach) = Pachomiana [A. Boon, Pachomiana Latina, Löwen 1932] Pel (Pel) = Dialogi contra Pelagianos libri 3 [CCSL 80] Ps (Ps) = Commentariolus in Psalmos [CCSL 72] Ps hom (Ps h) = In Psalmos homiliae 61 [CCSL 78] Ruf (Ruf) = Epistula adversus Rufinum [CCSL 79] Zach (Za) = Commentariorum in Zachariam prophetam ad Exsuperium Tolosanum episcopum libri 3 [CCSL 76A] Hil (HIL) = Hilarius von Poitiers Ps (Ps) = Tractatus super Psalmos [CSEL 22] Isid (IS) = Isidore of Seville Dt (Dt) = Quaestiones in Vetus Testamentum seu Mysticorum expositiones sacramentorum [PL 83, 207-424; 438-442] Ex (Ex) = Quaestiones in Vetus Testamentum seu Mysticorum expositiones sacramentorum [PL 83, 207-424; 438-442] Jud (Jud) = Quaestiones in Vetus Testamentum seu Mysticorum expositiones sacramentorum [PL 83, 207-424] Nat (na) = De natura rerum [J. Fontaine, Isidore de Séville: Traité de la nature, Bordeaux 1960] Sent (sent) = Sententiarum (sc. Augustini, Gregorii Magni, al.) libri 3 [PL 83, 537-738] Syn (syn) = Synonymorum de lamentationes animae peccatricis libri 2 [PL 83, 827-868]

42

The Textual Witnesses

JohMax (JO-M) = Johannes Maxentius Aug (Aug) = Capitula S. Augustini in urbem Romam transmissa [CCSL 85A, 251-273] Jul-T (JUL-T) = Julianus (Archbishop of Toledo) Ant (ant) = Antikeimenon [PL 96, 595-704] Pro (pro) = Prognosticorum futuri saeculi libri 3 [CCSL 115] Lic (LIC) = Licinianus (Bishop of Cartagena), Epistulae 3 [PL 72] Lucif (LUC) = Lucifer of Cagliari Athan (Ath) = De Athanasio libri 2 [CCSL 8] Nil (NIL) = Nilus Ancyranus Tr (tr) = De octo vitiis generalibus [an anonymous early Latin Translation; E. Bigot, Palladii episcopi Helenopolitani de vita S. Johannis Chrysostomi dialogus…, Paris, 1680) Or (ORI) = Origenes [GCS 38] Matth (Mt) c. Book and Section Number = Matthew Commentary; The τόμοι contained in Greek and Latin Editing Matth (ser) designated only by page number = Matthew Commentary; the Translation of the Commentary Series Pac (PAC) = Pacianus (of Barcelona) Ep (ep) = Epistulae 3 ad Sympronianum contra Novatianos [SC 410] PaulN (PAU-N) = Pontius Meropius Paulinus (of Nola) Ep (ep) = Epistulae 51 [CSEL 29] Pel (PEL) = Pelagius (Briton) 1-2 Cor (1-2 Cor) = Expositiones XIII Epistularum S. Pauli [A. Souter, Pelagius’s Expositions of Thirteen Epistles of St. Paul, II: Text and Apparatus Criticus, Cambridge 1926, Texts and Studies 9/2] PetrChr (PET-C) = Petrus Chrysologus (Bishop of Ravenna) s (s) = Sermones [CCSL 24B] Philipp (PHI) = Philippus Presbyter Commentarius in Iob [PL 23; 26]

The Ancient Translations

43

Pos (POS) = Possidius (Bishop of Calama) Vit (vi) = Vita S. Augustini [M. Pellegrino, Possidio Vita di S. Agostino, Alba 1955] Pot (POT) = Potamius of Lisbon Ep Ath (Ath) = Epistula ad Athanasium [A. Wilmart, RB 30 (1913) 280-282] Ep subst (subst) = Epistula de substantia Patris et Filii et Spiritus Sancti (PL 1, 202-216) PsAug (PS-AU) = Pseudo-Augustinus Ful (Fu) = Libellus adversus Fulgentium Donatistam [C. Lambot, RB 58 (1948) 190-222] Mir (mir) = De mirabilibus S. Scripturae [PL 35, 2149-2200] Pal [Pal] = Florilegium im cod. Vatic. Palat. lat. 556 [Graziano di S. Teresa, Ephemerides Carmeliticae 14 (1963) 210-41] s (s) = Sermones supposititii [PL 39, 1735-2354] SermCai (s Cai) = Sermones von A. B. Caillau ediert [PL 2, 1029-1123] s erem (s erem) = Sermones ad fratres in eremo [PL 40, 1235-1358] PsBas (PS-BAS) = Pseudo-Basilius Adm (adm) = Admonitio ad filium spiritalem [P. Lehmann, Sitzungs-berichte der Bayer. Akademie der Wiss., phil.-hist. Klasse 1955, Heft 7] Is (Is) = Expositio super Ysaye prophete [Bibliotheca Casinensis IV, Monte Cassino 1880, pars 1, 302-305; pars 2 391-434] PsBed (PS-BED) = Pseudo-Beda Prv (Prv) = In Proverbia Salomonis allegoricae interpretationis fragmenta [PL 91, 1051-1066] PsCaes = Pseudo-Caesarius Virg = S. Caesarii de decem virginibus [PL 67, 1161-1163] PsHi (PS-HI) = Pseudo-Hieronymus Brev (bre) = Breviarum in Psalmos [PL 26, 821-1270] 1 Cor (1 Cor) = Commentarii in Epistolas S. Pauli [H. J. Frede, Ein neuer Paulustext und Kommentar II. Die Texte, Freiburg 1974 = Vetus Latina. Aus der Geschichte der lateinischen Bibel 8] Ep (ep) = Epistulae [PL 30, 15-308] Mc (Mc) = Kommentar zu Marc [PL 30 589-644] PsIgn (PS-IGN) = Pseudo-Ignatius [F. X. Funk/F. Diekamp, Patres Apostolici, II, Tübingen, 1913]

44

The Textual Witnesses

PsMar (PS-MAR) = Pseudo-Marius Victorinus Scr (scr) = De verbis scripturae [J. Wöhrer, 24. Jahresbericht des PrivatGymnasiums der Zisterzienser in Wilhering, Wilhering 1927, 4-8] PsMel (PS-MEL) = Clavis Melitonis P (P) = recensio primitiva [J. B. Pitra, Analecta Sacra Spicilegio Solesmensi parata II (1884) 6-127] V (V) = recensio vulgata J. B. Pitra, Spicilegium Solesmense II (1855) 1-519 (Cap. 1-8); III (1855) 1-307 (Cap. 9-13)] PsPros (PROS) = Pseudo-Prosper of Aquitaine Dem (Dem) = Epistula ad Demetriadem de vera humilitate [K. C. Krabbe, Epistula ad Demetriadem De Vera Humilitate, Washington 1965 = Patristic Studies 97] Voc (voc) = De vocatione omnium gentium [PL 51, 647-722] PsRuf (PS-RUF) = Pseudo-Rufinus Fi (fi) = Liber de fide [M. W. Miller, Rufini Presbyteri Liber De Fide, Washington 1964 = Patristic Studies 96] PsVig (PS-VIG) = Pseudo-Vigilius of Thapsus Trin (tri) = De trinitate libri 12 [Buch 1-9: CCSL 9; Buch 10-12 M. Simonetti, Pseudoathanasii de Trinitate ll. X-XII, Bologna 1956, cf V. Bulhart, CCSL 9] Reg (REG) = Regulae monasticae Mag (Mag) = Regula Magistri (SC 105-106) Ruf (RUF) = Rufinus Cant (Ct) = Commentarius Origenis in Cant (GCS 8, 61-241) Eus (Eus) = Eusebii Historia Ecclesiastica [GCS 9,1-2] Ex (Ex) = Origenis in Ex homiliae 13 (GCS 6, 145-279) GrN (Gr) = Gregorii Nazianzeni orationes 9 [CSEL 46] Jdc (Jdc) = Origenis in librum Iudicum homiliae 9 (GCS 9) Lev (Lv) = Origenis in Leviticum homiliae 16 (GCS 6, 280-507) Num (Nm) = Origenis in librum Numeri homiliae 28 (GCS 7, 3-285) Om S Bas (Bas) = Omeliae S. Basilii (PG 31, 1723-1794) Or princ (pri) = Origenis libri 4 De principiis (GCS 22; Ed. P. Koetschau) Pamph (Pa) = Apologeticum Pamphili martyris pro Origene [PL 17,541-616] Ps (Ps) = Origenis homiliae 9 in Ps XXXVI-XXXVIII [E. Prinzivalli, Origene: Omelie sui Salmi. Homilae in psalmos XXXVI-XXXVIIXXXVIII, Florenz 1991]

The Ancient Translations

45

Reg S Bas (reg) = Regula S. Basilii ad monachos (CSEL 86) Rom (Rm) = Origenis Commentarius in Epistulam ad Romanos [Vetus Latina. Aus der Geschichte der lateinischen Bibel 16, 33, 34; Ed. C. P. Hammond Bammel] Salv (SALV) = Salvianus, Presbyter in Marseille Eccl (eccl) = Ad ecclesiam libri 4 [SC 176] SedScot (SED-S) = Sedulius Scottus Eph (Eph) = Collectaneum in Apostolum [PL 103,9-270] Misc (misc) = Collectaneum miscellaneum [CCCM 67] Rom (Rm) = Collectaneum in Apostolum [H. J. Frede, H. Stanjek, Sedulii Scotti Collectaneum in Apostolum, 2 vol., Freiburg 1996-1997 = Vetus Latina. Aus der Geschichte der lateinischen Bibel 31-32] Spec (PS-AU spe) = Liber de divinis scripturis sive Speculum quod fertur S. Augustini (CSEL 12, 287-700) Ta (TA) = Taio (Bishop of Saragossa) Ecl (Ecl) = Kommentar zu Ecclesiastes [A. C. Vega, España Sagrada 56, Madrid 1957, 353-372 [PLS 4, 1754-1772]] Sent (sent) = Sententiarum libri 5 [PL 80, 727-990; 1,24-25 und 5,33-35 E. Anspach, Taionis et Isidori nova fragmenta et opera, Madrid 1930, 6-22 [PLS 4, 1670-1678]] Tyc (TY) = Tyconius Reg (reg) = Liber regularum (F. C. Burkitt, The Book of Rules of Tyconius, Cambridge, 1894 [Texts and Studies 3,1]) = (SC 488)

1.4  The Textual Character of the Old Latin Translation

It is not easy or simple to describe the textual character of the translations (NB plural) represented by the witnesses to the Old Latin. According to the stemma, Chr, Dion and GregNy are close cousins. In any particular passage, provided one actually has witnesses for La, and provided one can also eliminate corruption from Jerome and hexaplaric sources, one has an early testimony to the Old Greek. The witness of La appears to be four or five generations from the autograph. An example may help to show how to assess the Old Latin in a particular passage: Example of How to Use the Old Latin Witnesses—126

46

The Textual Witnesses

The witnesses for 126 are dated to the following years: 387 Hi 392 Hi Ps hom (= Translation from Origen) 398 Vulg (= Bed Ps-Mel PsMoz ) 404 Hi Pach (Translation of a Greek translation of the Coptic text) 417/18 Hi Ep 140 7/8 Cent. PsHi Brev (dependent on An Ps sen; Hil Ps; Hi Ps and Ps hom (also specified as independent witnesses, but confirm VL, e.g. revolvetur)

Reconstruction of Old Latin: quoadusque evertatur funiculus argenti et conteratur monile auri et confringatur hydria ad fontem et convolvatur/involvetur rota super lacum The text of the Old Latin can be reconstructed largely from Hi Ep 140, Hi Ps hom and Hi Pach. Jerome cites in his works—apart from the commentaries—also according to his translation ex hebraica veritate, usually the text according to the LXX: here in Pach in which convolvatur in Hicom 350,57 / involvetur in Hi Ps hom = Origen. In his hexaplaric recension, which ought to be the most preferable word-forword rendering, he has adopted the following corrections: (1) non pulsetur for evertatur (= σ΄ μὴ / κοπῆναι), (2) impediatur instead of convolvatur (a free interpretation: the creation wheel is restrained by the coiled rope), (3) ‘in’ instead of ‘super’ (which results from the forementioned): here Jerome attempts to make sense out of the text. Hi, the lemma text of the Eccl Commentary, word for word taken over in the translation ex hebraica veritate = Vulgate, presupposes Jerome’s interpretation for the whole verse: the thread of life will be cut (ancient image: fates), the soul returns at the death of man. So far as possible, he follows the Hebrew text and the LXX; recurrat is possibly based on αʹ δράμῃ; instead of ἀνθέμιον Jerome perhaps read—or conjectured—ἀνάθημα = vitta / taenia; rumpatur and conteratur / confringatur Jerome suited to the accusative objects respectively. This example indicates how in numerous places the Latin texts must be read and evaluated. Most of the variants, however, arose solely within the Latin textual transmission process.

2.  The Syro-Hexapla Translation (Syh) 2.1  The Witnesses/Editions of the Syro-Hexapla Translation Milan, Bibl. Ambrosiana, C. 313. Inf.; VIII. Cent. “Codex Syro-Hexaplaris” G. Mercati, Nuove Note di Letteratura Biblica e Christiana Antica. Studi e Testi 95 (Città del Vaticano: Biblioteca Apostolica Vaticana, 1941), 43-46.

The Ancient Translations

47

W. Wright, A Catalogue of Syriac Manuscripts in the British Museum, I (London: Trustees of the British Museum, 1870).

Literature Codex Syro-Hexaplaris Ambrosianus photolithographice ed. A. M. Ceriani: Monumenta sacra et profana VII, Mediolani 1874. H. Middeldorpf, ed. Codex Syriaco-Hexaplaris Liber Quartus Regum e Codice Parisiensi Iesaias Duodecim Prophetae Minores Proverbia Iobus Canticum Threni Ecclesiasticus e Codice Mediolanensi. Berlin: Enslin, 1835. Ignatius Aphram I Barsoum, The Scattered Pearls: A History of Syriac Literature and Sciences, 2nd ed. (Gorgias, 2003).

The text of the Syro-Hexapla Translation is fully preserved in one manuscript: Milan, Bibl. Ambrosiana, C. 313. Inf.; VIII. Cent. “Codex Syro-Hexaplaris” The Syro-Hexapla is a translation of copies derived from the Fifth Column of Origen’s Hexapla or more accurately from his Tetrapla into Syriac.9 Included in the margins of this scholarly translation are notes on the text including translations of selected readings from Theodotion, Aquila, and Symmachus. This translation was made by Paul, Bishop of Tella, in 617 CE at the Antoninus Monastery at Enaton, a relay post on the coastal road 14 kilometres west of the city of Alexandria. Tella d’Mauzalath (between Edessa and Mardin) was the centre of a diocese of the Syrian Orthodox Church. Paul, Bishop of Tella, probably fled to Alexandria due to the Persian invasion led by Khusraw II Parwez Hormezd. Nonetheless, Patriarch Aphram Barsoum in The Scattered Pearls (pp. 39, 313) claims that Patriarch Athanasius I had invited Paul of Tella to come to the Antoninus Monastery to do the translation. Codex Syro-Hexaplaris originated in the Syrian monastery of the Nitrian Desert in Egypt known as Deir as-Suryan and is written in a late eighth or early ninth century hand (cf. colophon, Fol. 193v). This area, also known as the Desert of Scete (Wadi El-Natrun), is some 80-90 kilometres south of Alexandria. Consequently, this manuscript is not far from its source in both place and time. The colophons in Codex Syro-Hexaplaris are extremely important for our understanding of the history of the text. As the colophon for Ecclesiastes is dependent upon the one for Proverbs, both are cited here:

9  For a discussion of the relationship between the Syro-Hexapla, Hexapla, and Tetrapla, see Peter J. Gentry, “Origen’s Hexapla,” in The Oxford Handbook of the Septuagint, 553-571.

48

The Textual Witnesses

Colophon to Ecclesiastes in Codex Syro-Hexaplaris:

‫ܪܫܝܡ ܗܘܐ ܒܟܬܒܐ ܝܘܢܝܐ ܿܗܘ‬ ‫ܕܡܢܗ ܐܬܦܫܩ ܟܬܒܐ ܗܢܐ‬ ‫ܫܘܠܡܗ‬ ‫ܕܩܘܗܠܬ ܠܣܘܪܝܝܐ ܒܬܪ‬ ܼ ܿ ‫ܒܕܡܘܬܐ‬ ܼ ‫ܗܟܢܐ܀ ܩܘܗܠܬ ܒܗ‬ ‫ܕܨܚܚܐ܂‬ ‫ܐܬ ܼܢܣܒ ܡܢܗ ܟܕ ܡܢܗ‬ ܼ ܿ ܵ ‫ܒܗܘ ܕܐܦ ܗܠܝܢ‬ ‫ܕܕܡܝܢ ܒܬܪܟܢ‬ ܵ ܵ ‫ܒܐܝܕܐ‬ ܼ ‫ܡܛܟܣܢ ܗܘܝ܀ ܘܬܘܒ‬ ‫ܦܡܦܝܠܘܣ ܗܠܝܢ܀‬ ‫ܕܝܠܗ ܕܩܕܝܫܐ‬ ܼ ‫ܘܐܘܣܒܝܘܣ ݁ܬܪܨܢܢ܀‬ ‫ܦܡܦܝܠܘܣ‬ ܼ

It was noted in the Greek book from which this book of Qoheleth was translated into Syriac, after its end, as follows: Qoheleth was similarly taken from the same manuscript in which also those that are similar to it afterwards were arranged. And again: by the hand of the Holy Pamphilus are these (books). Pamphilus and Eusebius, we corrected. ᾽Εκκλησιαστὴς ὁμοίως μετελήφθη ἀπὸ τοῦ αὐτοῦ ἀντιγράφου, ἐν ᾧ καὶ οἱ λοιποὶ [τὰ λοιπὰ] ἐφεξῆς παρετέθησαν, καὶ πάλιν χειρὶ τοῦ ἁγίου Παμφίλου ταῦτα· «Πάμφιλος καὶ Εὐσέβιος διωρθώσαμεν» (Retroversion by Mercati, p. 45). This colophon is carefully divided into four parts by a standard punctuation sign in Syriac (‫)܀‬. This punctuation sign has been represented in the translation by a colon and three periods set in bold type. Retroversions by Middeldorpf into Latin and by Mercati into Greek have not been careful about this punctuation, which is the key to grasping the four part structure of the colophon. The first part indicates that the last three parts constitute a colophon or colophons in the Greek Vorlage of the Syriac translation. The second affirms that the text of Qoheleth was taken from exactly the same manuscript of the Septuagint Translation as Proverbs was taken. The third affirms that this text was produced by Pamphilus, i.e. by scribes under his supervision. The fourth declares that Pamphilus and Eusebius corrected the text produced in this way. The second statement in the colophon is grammatically and semantically dependent upon the colophon to the Book of Proverbs which immediately precedes Ecclesiastes in Codex Ambrosianus. This colophon is as follows: Colophon to Proverbs in Codex Syro-Hexaplaris:

‫ܪܫܝܡ ܗܘܐ ܒܟܬܒܐ ݂ܝܘܢܝܐ‬ ‫ܕܡܢܗ ܐܬܦܫܩ ܠܣܘܪܝܝܐ ܟܬܒܐ‬ ܵ ‫ܕܦܐܠܬܐ܃ܒܬܪ ܫܘܠܡܐ‬ ‫ܗܢܐ‬

The Ancient Translations

‫ܕܝܠܗܝܢ ܗܟܢܐ ܀ ܐܬ ܼܢ ܵܣܒܝܢ‬ ܵ ܵ ‫ܬܐ܂ ܡܢ ܨܚܚܐ‬ ݂ ‫ܘܐܬܦܚܡܝܢ ݁ ܦܐܠ‬ ‫ܘܐܬܟ ܼܬܒܘ‬ ‫ܚܬܝܬܐ ܕܐܬܣܝܡ‬ ݂ ܵ ‫ܒܐܝܕܐ‬ ‫܂‬ ‫ܘܠܝܐ‬ ݂ ‫ܒܗ ܡܢ ܠܒܪ ܣܟ‬ ܼ ‫ܕܦܡܦܝܠܘܣ ܘܕܐܘܣܒܝܘܣ܂ ܕܒܗ‬ ‫ܵܕܫܝܡܢ ܵܗ ܼܘܝ ܘܗܠܝܢ ܀ ܐܬ ܼܢ ܵܣܒܝܢ‬ ܵ ‫ܡܢ ܫܬܝܬܝ‬ ‫ܦܨܐ ܕܐܘܪܓܢܝܣ‬ ‫ܗܠܝܢ ܕܐܫܟܚܢܢ ܀ ܀ ܀‬ ‫ܘܬܘܒ ܀ ܒܟܝܪܐ ܕܝܠܗ܂ ܦܡܦܝܠܘܣ‬ ‫ܘܕܐܘܣܒܝܘܣ ݁ܬܪܨܘ ܀‬ ܼ

49

It was noted in the Greek book from which this book of Proverbs was translated into Syriac, after the end of them, as follows: The Proverbs were copied and collated from an accurate copy that was made and in which scholia were written in the margins by the hand of Pamphilus and Eusebius, in which were noted also these things: These things that we found were taken from the Hexapla Version of Origen. And again: in their own handwriting «Pamphilus and Eusebius corrected.» Once again, punctuation in Syriac, not always noted by scholars, carefully divides the colophon into five sections marked in the translation by two cola, a period, a colon, and a period set in bold type. The first statement indicates that the following colophon and marginal notes come from the Greek Vorlage of the Syriac translation. The second statement indicates that the Greek Vorlage was an accurate copy containing marginalia. Two of these marginal notes are translated into Syriac with the expression «and again» and a punctuation sign used to set the second off from the first. The first marginal note indicates that the Vorlage of the copy was the Hexapla of Origen. The second marginal note affirms accuracy by citing the personal attestation that the copy was corrected by Pamphilus and Eusebius. What these colophons tell us is that the text of the Syro-Hexapla is not derived directly from the Fifth Column of Origen’s Hexapla itself, but from copies of the Fifth Column or better of the Tetrapla which were produced and corrected under the supervision of Eusebius and Pamphilus. This information corresponds well with conclusions derived from analysis of the textual data: the text of the Syro-Hexapla is, in fact, a derivative of the text in the Fifth Column corrected by Eusebius. Thus, in terms of the Septuagint text, it represents a distant member of the O group and does not attest the text of the Fifth Column as closely as do the Greek manuscripts which constitute the O group. Nonetheless, in terms of the readings attributed to Theodotion, Aquila, and Symmachus, as well as in terms of the Aristarchean signs employed by Origen, it represents a reliable copy of what was in the Tetrapla, more so than manuscripts in the O group which only represent the Hexapla or even in marginalia elsewhere in the Greek manuscripts.

50

The Textual Witnesses

2.2  The Use of Aristarchean Signs Literature Epiphanius, Liber de Mensuris et Ponderibus, PG 43: 248-249. Also Ηλια Δ., Μουτσουλα, ed. (1973). ΤΟ «ΠΕΡΙ ΜΕΤΡΩΝ ΚΑΙ ΣΤΑΘΜΩΝ» ΕΡΓΟΝ ΕΠΙΦΑΝΙΟΥ ΤΟΥ ΣΑΛΑΜΙΝΟΣ. ΘΕΟΛΟΓΙΑ 44.1-2 (1973): 157-98. Gudeman, “Kritische Zeichen,” Paulys Realencyclopädie der Classischen Altertumswissenschaft 11.2, 1916-1927. F. Osannus, Anecdotum Romanum (Giessen, 1851).

2.2.1 Asteriskoi in Syh

For description of the use of the Aristarchean signs in the textual tradition of LXX Ecclesiastes, first the text is cited according to the edition followed by evidence from the First Apparatus and brief analysis. 116e-117a καὶ καρδία μου εἶδεν πολλά, σοφίαν καὶ γνῶσιν. καὶ ἔδωκα καρδίαν μου τοῦ γνῶναι σοφίαν καὶ γνῶσιν, παραφορὰς καὶ ἐπιστήμην ἔγνων, ----116e καί 2°—γνῶσιν] post (17a) γνῶσιν tr C´(–797) Arm; > 797 117a καί 1°—γνῶσιν] sub ※ V 788 Syh; > 253 L (–106 125) 130 d–254´ k 68 248´txt (= Compl; καὶ ἔδωκα—τοῦ γνῶναι 248´mg) 296´ 311 338 443 547 705 Clem II 37 Dion 212 Ol Geo ↓ : homoiot

V. 17a (καὶ ἔδωκα καρδίαν μου τοῦ γνῶναι σοφίαν καὶ γνῶσιν) stands in C´ following 16d and in 475 766 following 17b. MS. 253 from the O group, however together with 261-545 from the L group (plus 342-357, k, and 10 minuscules), lack v. 17a. Therefore, in the Vorlage of Origen this text was omitted due to homoioteleuton and was restored by Origen under the asteriskos. 46a ἀγαθὸν πλήρωμα δρακὸς ἀναπαύσεως ————— c mg 46a δρακός] δρακου 545c; + μετα (μετ Anton OlΒ ) Anton 880 DamWV Mi c OllemΒ et com (lib; sed hab Ollem–Β) Hi Syh (sub ※ σʹ) ↓ | ἀναπαύσεως] pr και 609; αναπαυσεις 311

In 46a μετά is under the asteriskos in Syh and without asteriskos in the biblical c text of Anton, Dam and OllemΒ et com. Naturally Jerome added this word from comparison with the Hebrew and under the influence of Symmachus.

The Ancient Translations

51

68a ὅ τι περισσεία τῷ σοφῷ ὑπὲρ τὸν ἄφρονα; ————— 68a ὅ τι 992] ητις 336; + τίς Sc A C 411 L cII k 248´ 252 296´ 311 338 339 543 549 613 645 698 706 766 788 Ol (sed hab Fa1 SaI = Pesh) Syh (sub ※ αʹθʹ) = Gra. Ra ↓ ; ‫( 𝔔 כמה‬sed hab ‫)𝔐 כי מה‬

The text of the edition for 68a is ὅ τι περισσεία τῷ σοφῷ. ὅ τι is an adequate rendering for ‫ כי מה‬in 𝔐. Most scribes, however, in the reception history of the text understood ὅτι. This explains the addition of τίς in Sc and 411, members of the O group, along with the derivative L and a portion of the textual transmission influenced by them. This addition was appropriately marked by an asteriskos in Syh. 71b καὶ ἡμέρα τοῦ θανάτου ὑπὲρ ἡμέραν γεννήσεως αὐτοῦ. ————— M PM 71b αὐτοῦ] sub ※ αʹ 788; > B-S*-68´ 998 C´ 336´645 DamM M An Jb 3 Or Matth te lem et com 71 Aeth Geo (sed hab Did 196,11 197,10 Met VI.3,3,43 Ol Hi Arm Fa1 Syh (sub ※ αʹ) = Ra 𝔐 𝔔? ↓)

It is difficult to determine the earliest form of the text in 7:1b. In view of the extreme literalism of the translator it is likely that he rendered the 3 m. s. pronoun in 𝔐 by αὐτοῦ. The omission in the Egyptian Text (B-S*-68´ 998) is probably a stylistic correction. So Origen inherited a text without αὐτοῦ and added it from Aquila as a hypercorrection. This shows how manuscripts may attest the original text in a secondary way. Both 788 and Syh attest the asteriskos in this case. 73b ὅτι ἐν κακίᾳ προσώπου ἀγαθυνθήσεται καρδία. ————— L c 73b καρδία] pr η 645 766 DamM H ; > B-68´ 998 Didlem et com 199,21 199,27 200,16 An Scrip 1,22 AGeoO Fa1 SaI II 2 (sed hab Arm Hi Syh (sub ※ αʹθʹ) = Ra 𝔐 𝔔): haplogr

It seems that καρδία was omitted in the Egyptian Text due to haplography (καρδία ends v. 3 and also begins v.4). This was the Vorlage for Origen, who rectified the omission by adding the missing word from Aquila and Theodotion. Only Syh preserves the asteriskos. We must note that the text received by Origen is not equivalent to the earliest form of the text. It was a text already defective and which he corrected in this case in the right direction. Thus a witness bearing a secondary correction actually preserves the original text in this instance.

52

The Textual Witnesses

74b καὶ καρδία ἀφρόνων ἐν οἴκῳ εὐφροσύνης ————— 74b καὶ καρδία = 𝔐] οἱ γʹ ※ καὶ ↙ καρδία δέ Syh ↓ ; και καρδια δε 252 ↓ ; καρδια PM δε O-411 cII 534´ Antioch 1709 Chr II 1055 DamKVRM T Met VI.7,3 SaI (sed hab Fa1); om καί 601-609 106 542 645

The edition has καὶ καρδία in 74b whereas the O group, followed by cII, has καρδία δέ. The latter reading is contrary to the translation technique of the translator. It is not surprising that Syh inherited the reading of the O group, but it also has the καί along with 252. In Syh, δέ should have been marked with an obelos as missing in 𝔐. 76a ὡς φωνὴ τῶν ἀκανθῶν ὑπὸ τὸν λέβητα, ————— 76a ὡς B-S-68´’ 998 C´ 357 296´ 311 338 443 645 706 795 Ammon Antioch 1724 c a V Bas III 961 Dam (ωσπερ DamKVRMH TL A Max II 996) Amb Exh virg 11,76 BenA Conc 1126 Eugip Reg 28,74 Spec 557,8 (et sicut Reg Mag 179,183) Fa1 SaI 6] pr ὅτι (sub ※ Syh) rel (Did Metlem et com VI.8,3,28 PsChr Hi Arab Arm Syh = Ra 𝔐 𝔔 Peshmss Vulg) ↓

The fact that ὅτι is preceded by an asteriskos in Syh is a clear indication that this word was not part of the text received by Origen, but was added from one of the Three revisors. Rahlfs incorrectly puts ὅτι in the Text. See Peter J. Gentry, “Special Problems in the Septuagint Text History of Ecclesiastes,” XIII Congress of the International Organization for Septuagint and Cognate Studies, Ljubljana, 2007, ed. Melvin K. Peters (Leiden, 2008), 138-139. 76b

οὕτως γέλως τῶν ἀφρόνων· ————— 76b γέλως] pr σʹ ※ καί ↙ Syh ↓ ; pr ο A C L(–125) d k 248´ 252 260 296´ 311 338 339 443 543 549 613 698 706 795 788 Ammon Antioch 1724 Bas III 961 Chr II 1055 Dam Metlem et com VI.8,4,29 VI.9,21 Max II 996 OlΔΕΙΚΜ Fa1 SaI 6; + ο A C k M 248´ 252 296´ 311 338 339 443 547 706 795 788 DamM Ol–ΒΗ Met com VI.8,29 VI.9,21 = Gra

Here in 7:6b, Syh alone marks the addition of καί in the Fifth Column by Origen and correctly notes that the source is Symmachus. 𝔐 has ‫כן ש ֹחק הכסיל‬. Could it be that the Hebrew text received by Origen had ‫( כן וש ֹחק הכסיל‬hence our MT

The Ancient Translations

53

would involve haplography based on similarity of final nun and waw)? This is unlikely. Probably, he considered οὕτως καί a better equivalent for 𝔐 than just οὕτως. In this case, his correction was contrary to both translation technique and the earliest form of the text. 91e καί γε ἀγάπην καί γε μῖσος οὐκ ἔστιν εἰδὼς ὁ ἄνθρωπος· ————— 91e ὁ ἄνθρωπος] ὁ sub ※ αʹ Syh (mend asteriscus ante ἄνθρωπος); ανθρωπου 336´; om ὁ S C´’–(157) 260 299 (503) 338 645 698 Ol PsChr, contra 𝔐

The Greek translator follows the articulation in the Hebrew text contrary to the rules of Greek grammar. Since the article is in 𝔐, our manuscripts in Greek that have the article attest to the original text. Apparently the Vorlage of Syh was similar to that of S C´’ alii and the article was added from Aquila. Since the form of the article in Greek and Syriac are radically different, the asteriskos is placed (wrongly, but unavoidably) before the Syriac word for ‘man’. 1013b καὶ ἐσχάτη στόματος αὐτοῦ περιφέρεια πονηρά· ————— 1013b πονηρά] sub ※ Syh

Why πονηρά is under the asteriskos in 1013b in Syh is unclear. Possibly the Vorlage of Syh omitted this word due to homoioteleuton, but no manuscript attests the omission.

2.2.2  Obeloi in Syh

11b βασιλέως ᾽Ισραὴλ ἐν ᾽Ιερουσαλήμ. ————— 11b ᾽Ισραήλ] sub ÷ Syh = 𝔐

Here in 11b the obelos correctly marks a word in the Septuagint which is not in the Hebrew text. 319c συνάντημα ἓν αὐτοῖς· ————— 319c αὐτοῖς] pr τοις πασιν Sc O-411 766 DamR Armte Hiap: ex (19e); > Fa1; + (sub ÷ Syh; mend ad αυτοις) πασιν Met III.20,8 Syh

54

The Textual Witnesses

The text inherited by Origen had συνάντημα ἓν τοῖς πᾶσιν αὐτοῖς. The words τοῖς πᾶσιν constitute a conflation from 319e. Origen marked these words with an obelos since they were not in his Hebrew text. Syh has mistakenly marked the word αὐτοῖς instead of τοῖς πᾶσιν, because in the word order in Syriac, αὐτοῖς precedes τοῖς πᾶσιν. Thus the scribe put the sign in the corresponding place in the manuscript, but on the wrong word. 726a καὶ εὑρίσκω ἐγὼ αὐτήν, καὶ ἐρῶ πικρότερον ὑπὲρ θάνατον, ————— 726a αὐτήν] sub ÷ et «Origenes» sup lin Syh; > Hi = Ra 𝔐

Again, Syh employs an obelos in 726a to mark a word in the text from the Fifth Column of the Hexapla but not in the Hebrew text. Naturally, Jerome omitted this word from his revision of the Old Latin. The name Origenes is above the pronoun in Syh. This may mean that some of these signs come from the work of Eusebius while others can be traced back to the work of Origen himself. No one is sure of the meaning at this point in time.

2.2.3  Antisigmata or Lemniskoi in Syh Literature F. Field, “Appendix I ad Cap. VII: De Obeli pictura () versionis Syro-hexaplaris peculiari,” Origenis Hexaplorum quae supersunt, 1:lxiv-lxvii. V. Gardthausen, Griechische Palaeographie, 2nd ed. (Leipzig, 1913), 2:414. P. J. Gentry, Hexaplaric Materials in Ecclesiastes and the Rôle of the Syro-Hexapla, Aramaic Studies 1 (2003): 20-22. F. Osannus, Anecdotum Romanum de Notis Veterum Criticis Inprimus Aristarchi Homericus et Iliade Heliconia (Giessen, 1851). A. Rahlfs, Studie über den griechischen Text des Buches Ruth (MSU 3), 61-62. T. Skat Rørdam, Libri Judicum et Ruth secundum versionem Syriaco-Hexaplarem (Copenhagen, 1861). F. Schironi, Tautologies and Transpositions: Aristarchus’ Less Known Signs. Greek, Roman, and Byzantine Studies 57 (2017): 607-630. J. Ziegler, ed., Ezechiel, 2nd ed. (Septuaginta Vetus Testamentum Graecum 16.1; Göttingen: Vandenhoeck & Ruprecht, 1952, 1977), 41-44. J. Ziegler, hg., Susanna, Daniel, Bel et Draco, 2nd ed., Versionis iuxta LXX interpretes textum plane novum constituit O. Munnich, Versionis iuxta “Theodotionem” fragmenta adiectit D. Fraenkel (Septuaginta Vetus Testamentum Graecum 16.2; Göttingen: Vandenhoeck & Ruprecht, 1954, 1999), 41-42. J. W. Wevers und D. Fraenkel, Studies in the Text Histories of Deuteronomy and Ezekiel (MSU 26; Vandenhoeck & Ruprecht, 2003), 100.

The Ancient Translations

55

Besides asteriskos, obelos, and metobelos, another critical sign is used in Syh, called a lemniskos (λημνίσκος or λιμνίσκος, so Epiphanius, de Mensuris et Ponderibus, PG 43, 248). Better known as antisigma (), it is a sigma rotated counter-clockwise, i.e. minus ninety degrees.10 Descriptions and discussions of the use of this sign in the Syro-Hexapla by Gardthausen, Field, Rahlfs, Ziegler, Munnich and Wevers demonstrate that scholars are unclear and uncertain about the identity and purpose of this critical sign. Since the most extensive study is by Field, his presentation can be briefly summarised. Field calls the symbol a pictorial obelos and describes its use under five points: (1) it is put in front of words adopted into Syh from the Peshitta (e.g. Num 26:4, Judg 5:22, Prov 27:14); (2) it introduces a pericope or section from another part of Scripture (e.g. Hos 14:3, Prov 16:1, 27:16), (3) it marks one of two translations when the LXX has a doublet for the Hebrew (e.g. Prov 9:7, 11:26, 31:29); (4) it marks words not in the Hebrew or in B (Codex Vaticanus) (e.g. Prov 19:9, 20:11, 24:12); (5) the forms of the obelos (÷ [horizontal line with one dot or with two] and ) do not have absolutely the same significance (e.g. Gen 39:17, Judg 16:14, Prov 17:21, 22:14). In this last category Field seems to confuse the ἀντισίγμα with another Aristarchean sign (called ἀντισίγμα περιεστιγμένον) as he admits he does not really know these signs. Wevers and Ziegler believed the sign was some sort of index but were unsure. One may find a clear and simple solution by going back to the sources—the fragments of Aristarchus—the one who invented the signs. And this I did not do in my earlier publication. According to the Anecdotum Romanum, the best collection of the sources for the Aristarchean signs, τὸ δὲ ἀντισίγμα καθ’ ἑαυτό, πρὸς τοὺς ἐνηλλαγμένους τόπους καὶ ἀπᾴδοντας. This sign, then, marks misplaced texts and variants. We have no reason to think that the function would be different in the usage by Origen (or by Pamphilus? or Eusebius?) or by those who translated the copies of the Hexapla into Syriac and, in fact, the function of marking misplaced texts and variants nicely captures all of the diverse uses described by Field. Usually in textual transmission, when a word or words are inserted into the textual tradition, they are inserted at different places, so that this phenomenon also includes many of the omissions. All of the four places in the book of Ecclesiastes in Syh can be explained by the fact that the antisigma sign marks misplaced texts or variants. The evidence speaks for itself:

10  Wevers (p. 100) not recognising the sign as antisigma, says it is ninety degrees to the right. The earliest form of the sign is a backwards lunate sigma. Since the form in Syh is based on the later minuscule sigma, whether the sigma is rotated clockwise or counterclockwise, the result would be the same. Nonetheless, the name of the sign is an “anti” sigma. Neither Ziegler nor Wevers recognised the form as an antisigma.

56

The Textual Witnesses

110ab ὃς λαλήσει καὶ ἐρεῖ ᾽Ιδὲ τοῦτο καινόν ἐστιν, ἤδη γέγονεν ἐν τοῖς αἰῶσιν ————— 110a ἐστίν = 𝔐 (‫ ])הוא‬+  αυτο ↙ Syh (tr post ἤδη (10b)); + αυτο O–253 Sc 547c; + αυτος 613; > 299 705 752

Note that the position of αὐτό, unlike that of the O group, is after ἤδη in v. 10b 215ef

ἐγὼ τότε περισσὸν ἐλάλησα ἐν καρδίᾳ μου, ὅτι καί γε τοῦτο ματαιότης. ————— 215f ὅτι—ματαιότης] διοτι αφρων εκ περισσευματος λαλει 534: homoiar (16a); + διοτι (> Aeth; + ο B-613 998 d SaI) αφρων εκ (+ του 613) περισσευματος (+ καρδια Geo) λαλει (+ (sub  Syh) οτι (και ιδου pro οτι Syh) και γε τουτο ματαιοτης O Syh) B-613 998 O d 336´ 542 766B Aeth Geo Arab SaI Syh (sed hab Hilem 269,271s) = Pesh, contra 𝔐; pr (sub ÷ 788 (οὐ κεῖται παρ’ ἑβραίοις 788mg); + και 299) διοτι (οτι Dioncom 221 GregNy 364,19; > Dam; + ο 390-574-601-cII GregNy 361,18ap 364,19ap) αφρων (ανος 547txt) εκ περισσευματος λαλει (+ οτι αφρων εκ περισσευματος λεγει post fin 443) rel (HiLXX 269,285-289 GregNy 361,18 364,19 Metlem et com II.6,6s,102s II.7,7s Ol Proc CatP 22,154 Armte et ap) = Ra: ex Matth 1234 (par Luc 645)

Here a comment based on either Matthew 12:34 or Luke 6:45 was inserted into the text at different places. The comment was triggered by the statement in 215e. 217a

καὶ ἐμίσησα σὺν τὴν ζωήν, ————— 217a ἐμίσησα] -σησε k; + εγω (sub  Syh) L 798-cII 547 Ol Syh: ex (18a)

In many places in Ecclesiastes the Hebrew text has a 1 c. s. pronoun in addition to the 1 c. s. finite verb. This is always represented by ἐγώ. Here some scribe had inserted an ἐγώ into the tradition where the Hebrew had no 1 c. s. pronoun and it was appropriately marked in Syh as a misplaced text. 119c

καὶ περιπάτει ἐν ὁδοῖς καρδίας σου ————— c 119c καρδίας σου 357 542 DamL Hilem et com 348,156 351,68 = Vulg] καρδια σου 788; ἀμώμων ἡ καρδία σου 539; (+ σου Geo SaI II) ἄμωμος B-68´ 998 336´

The Ancient Translations

57

Antioch 1485 (ἀμώμως) Did 333,4s 335,26s PsChr Geo Fa1 2 SaI II; ἄμωP μοις 766; > σου 338*; + ἄμεμπτος DamM ; + ἄμωμα 613; + ἀμώμως 254´; c P + ( Syh) ἄμωμος rel (788 Anton 960 1057 1208 Dam–L M Max II 968 Met 334,2 Ol Amb Exh virg 10,69 Spec 473,15 Hi Pach 144,13 Aeth Arab Arm CPA Syh Ald Sixt)

In an earlier publication I demonstrated that the various variants ἄμωμος constitute an addition to the textual tradition (P. J. Gentry, Hexaplaric Materials in Ecclesiastes and the Rôle of the Syro-Hexapla, Aramaic Studies 1 (2003): 20-22). Among Greek witnesses only 357 and 542 attest the absence of this word. Yet 357 is closer to the earliest form of the text than any other witness and 542 happens to be a descendant of the Vorlage of Syh. The context in Ecclesiastes inspired an addition from formulaic language in the Psalter. It is not surprising, then, that Syh marks the word with an antisigma as a misplaced text. 117a

2.2.4 Asteriskoi in V

καί 1°—γνῶσιν] sub ※ V 788 Syh; > 253 L (–106 125) 130 d–254´ k 68 248´txt (= Compl; καὶ ἔδωκα—τοῦ γνῶναι 248´mg) 296´ 311 338 443 547 705 Clem II 37 Dion 212 Ol Geo ↓ : homoiot;

Here the text received by Origen was defective due to homoioteleuton and he repaired it (Commentary on Matthew XV 14) supplying the missing text from the other versions and marking it with an asteriskos. 86b

ὅτι γνῶσις τοῦ ἀνθρώπου πολλὴ ἐπ᾽ αὐτόν] sub ※ V

All MSS. and daughter versions transmit 86b: Septuaginta et Theodotion: quia scientia hominis multa super eum Hi (316,88-90). Klostermann p. 60 states, “Videntur ex Theodotion sumpte esse.” This is, however, not the case; rather, Theodotion has taken over v. 6b from the LXX, and one may observe that this is frequently his approach in revising the text. The asteriskos in V is incorrect.

2.2.5 Asteriskoi in 788

214d ὅτι συνάντημα ἓν συναντήσεται τοῖς πᾶσιν αὐτοῖς. ————— 214d ἕν] sub ※ σ΄ 788; > 998 357 Fa1

MSS. 357 and 998 (along with the Fayyumic) represent extremely early testimony to the Egyptian Group in which the ἕν was omitted. Since the text received by Origen is for the most part derived from the Egyptian Group, he inserted the text from Symmachus. Other parts of the textual tradition preserve the ἕν from the

58

The Textual Witnesses

original translation. Among all our witnesses, only 788 preserves the testimony that this was added by Origen from Symmachus. 215c

καί γε ἐμοὶ συναντήσεταί μοι, ————— 215c μοι] μου 339; + ※ θ΄ 788; > Sc O-411 L C´’ 754 338 547 613 698 705 766B 795 GregNy 361,17 364,13 Metlem et com II.6,4,35,95 OlΖ Hi Geo (sed hab Syh) = Ald, contra 𝔐

Once again, the text received by Origen did not have the pronoun μοι so he inserted it from Theodotion. No doubt it was omitted in his Vorlage as a stylistic correction, since it is redundant. Other parts of the textual tradition preserved this pronoun from the translator who assiduously followed the Hebrew parent text in his translation. 121c

ἕως ὅτου μὴ ἔλθωσιν ἡμέραι τῆς κακίας ————— 121c ὅτου B-68´ 998 C´ d 443] > 155 Didcom 338,22 339,5 Geo: cf 126a; ου rel (sub ※ 788; 539 CyrHier 897 Dam Didlem et com 338,5 338,21 339,6 Met X.5,7 Ol)

An early part of the textual tradition omits ὅτου or οὗ (> 155 Didcom 338,22 339,5 Geo). MS. 788 preserves an asteriskos showing that Origen inserted text from one of the Three. The O group dominated the rest of the textual tradition apart from B-68´ 998 C´ d 443.

2.2.6  Obeloi in 788

215ef ἐγὼ τότε περισσὸν ἐλάλησα ἐν καρδίᾳ μου, ὅτι καί γε τοῦτο ματαιότης. ————— 215f ὅτι—ματαιότης] διοτι αφρων εκ περισσευματος λαλει 534: homoiar (16a); + διοτι (> Aeth; + ο B-613 998 d SaI) αφρων εκ (+ του 613) περισσευματος (+ καρδια Geo) λαλει (+ (sub  Syh) οτι (και ιδου pro οτι Syh) και γε τουτο ματαιοτης O Syh) B-613 998 O d 336´ 542 766B Aeth Geo Arab SaI Syh (sed hab Hilem 269,271s) = Pesh, contra 𝔐; pr (sub ÷ 788 (οὐ κεῖται παρ’ ἑβραίοις 788mg); + και 299) διοτι (οτι Dioncom 221 GregNy 364,19; > Dam; + ο 390-574-601-cII GregNy 361,18ap 364,19ap) αφρων (ανος 547txt) εκ περισσευματος λαλει (+ οτι αφρων εκ περισσευματος λεγει post fin 443) rel (HiLXX 269,285-289 GregNy 361,18 364,19 Metlem et com II.6,6s,102s II.7,7s Ol Proc CatP 22,154 Armte et ap) = Ra: ex Matth 1234 (par Luc 645)

The Ancient Translations

59

2:15f was already discussed earlier. Here, apparently MS. 788 marks with an obelos words that are better marked with an antisigma in Syh. The misplaced text is inserted at different places in the textual tradition. Yet there is another possibility to consider. Epiphanius represents the lemniskos as an obelos in his portrayal of the sign. It may be possible that in the Greek tradition, there is some confusion between obelos and lemniskos. Only an exhaustive analysis of all instances can clarify this problem. 1019c καὶ τοῦ ἀργυρίου ἐπακούσεται σὺν τὰ πάντα. ————— 1019c ἐπακούσεται] pr ταπεινωσει B-S*(‑πιν-)-68´ 998 254´ 336´ 645 Did 311,25 312,19 Arab (sed hab Ald): duplex lectio, cf 𝔐; υπακουσεται O-411 106 563571 d 248´ 336´ 338 359 543 549 613 766 788 (mendose obelus ante σὺν τὰ πάντα) Dam Met IX.17,3 OlΔΙΚΜ Aeth Arab (ταπεινωσει και υπακουσεται)

Although 788 does not read ταπεινώσει ἐπακούσεται as in the Egyptian Group, it represents a faint memory of a textual tradition that had marked ὑπακούσεται as being in the Septuagint but not in the Hebrew text since the Egyptian Group has a duplex lectio. The memory is faint since the scribe put the obelos before σὺν τὰ πάντα instead of ὑπακούσεται. On the face of it, however, this case seems almost identical to cases of misplaced texts listed by Field and exemplified in the case just discussed in 2:15f. It could be, then, that 788 has employed an obelos like symbol for a lemniskos. Syh = 542

2.3  The Character of the Syro-Hexapla Translation

Among all the minuscules, 542 is the most closely related to Syh: 15a 58a 714c 722a 810c 812b 911f 101b

ἀνατέλλει] ανετειλεν 542 Syh ↓ ἐστι] pr αυτος 542 Geo Syh ἐποίησεν] pr τουτο 542 Metlem et com VI.17,2,4 CPA Syh σε] σου 125´; > 542 Didlem et com 222,20 223,13 223,14 Syh πόλει] ποιησει 542 Syhmg μακρότητος] ματαιοτητος 542 Syhtxt καί γε οὐ 3°] ουδε 125 Cass Co 13,18 Hi 328,242 Syh μεγάλης = Peshmss] μικρας 542 Hi Syh ↓ = 𝔐

68b 91c

διὰ τί 542 543 549 Syh = Pesh] διοτι rel (539 Arm) ὡς B-S*-68´’ 998 C´-260 d 336´ 443 547 645 Metlem et com VIII.1,16,95 PsChr Syn 548 Aeth Arm Fa1 SaI 2] οτι ως 339 542 698 Syh ↓ ; οτι rel (Ol Pros Dem 12 Voc 1,24 Hi) ↓

60 91d 115d

The Textual Witnesses

ἐν χειρί Syhmg (s nom) = 𝔐] ενωπιον O–V 443 542 Syhtxt Metlem VIII.1,17,96 ↓ ὅσα] ος O 542 766 Met IX.22,5 Hi Syh = 𝔐

et com

Syh preserves readings uniquely with 766 three times. This is not surprising since 766 belongs to earlier ancestors of both Syh and 542. According to the stemma, 766A B are six and seven generations from the autograph, respectively, while Syh is 20 and MS. 542 is 34 generations from the autograph. Like Syh, both 542 and 766 are distantly related to the O group. 58b 63b 810a

τοῦ ἀγροῦ εἰργασμένου] τη χωρα ειργασμενη 766 Syh πλῆθος] πλειους 766A Hi Syh εἰς τάφους] εις ταφον 357 766A Geo SaI II 2 Syh

417c 417c

θυσία] θυσιαν 998 O–637 766 Syhmg; θυσιας S k 534´ σου] σοι 299; > S O 766 Hi Geo Syhmg = 𝔐 Vulg;

3.  The Christian Palestinian Aramaic Translation (CPA) 3.1  The Witnesses/Editions of the Christian Palestinian Aramaic Translation Christa Müller-Kessler und Michael Sokoloff, A Corpus of Christian Palestianian Aramaic: Volume I, The Christian Palestinian Aramaic Old Testament and Apocrypha Version from the Early Period (Gronigen: Styx Publications 1997). F. Schulthess, Christlich-palästinische Fragmente aus der Omayyaden-Moschee zu Damaskus, Abhandlungen der Königlichen Gesellschaft der Wissenschaften zu Göttingen, VIII,3 (Berlin 1905).

The Christian Palestinian Aramaic Translation is transmitted in the following fragments as listed by Müller-Kessler/Sokoloff. Damascus Frag. II, fol. 12/11r 7:9b ἐν κόλπῳ—12c; 14a ἐν ἡμέρᾳ ἀγαθωσύνης ζῆθι; 14b-14e Damascus Frag. II, fol. 12/11v 7:15a; 16a δίκαιος πολύ—16b καὶ μή; 16b περισσά—21c ὅπως μή Damascus Frag. II, fol. 13r 7:22a πλειστάκις—22b καθόδους; 22c ὅπως—23a σοφίᾳ Damascus Frag. VI, r 11:2b ὅτι οὐ γινώσκεις—8b αὐτοῖς Damascus Frag. VI, v 11:8c τὰς ἡμέρας—9b καί 1°; 9b σε 1°—10a ἀπό 1°; 10a σου 1°—10c καί 3°; 10c ματαιότης—12:2b ἀστέρες

The Ancient Translations

61

3.2  The Textual Character of the Christian Palestinian Aramaic Translation Earlier editions of the Göttingen Septuaginta referred to this daughter version as the Syro-Palestinian version (Syp). Scholars now prefer to use the term Christian Palestinian Aramaic according to a better classification of the language in question. Some twenty variants are recorded in the apparatus: six of these are singular variants and six are shared, sometimes uniquely, with Syh. The evidence is far too fragmentary to draw conclusions. Nonethless, CPA is a daughter version of the LXX.

4.  The Coptic Translations (Co) All the Coptic texts for the edition were collated by Janet A. Timbie (Adjunct Associate Professor, The Catholic University of America).

4.1 The Sahidic Translation (Sa) 4.1.1  The Witnesses/Editions of the Sahidic Translation

The Sahidic translation is almost fully preserved in two MSS.: SaI

Rom, Bibl. Vat., Borgia copto 109 (24). X. Cent. Contents: 11-93 103-1214. Edition: A. Ciasca, Sacrorum Bibliorum fragmenta copto-sahidica Musei Borgiani II (Romae 1889) 195-214.

SaII

Ann Arbor, Univ. of Michigan Library (Special Collections Library), Mich. MS. 166. VI.-VII. Cent. Contents: 12*-1214*. Edition: L. A. Shier, Old Testament Texts on Vellum: Coptic Texts in the University of Michigan Collection, Ed. W. H. Worrell. University of Michigan Studies, Humanistic Ser. 46 (Ann Arbor 1942) 68-125.

Further additional fragments of Sa published since the appearance of SaI and SaII were also collated for the edition as follows: Sa1

Vienna, Nat. Bibl., (Papyrus Erzherzog Rainer) 2 = K 9176. ?. Cent. Contents: 89-92. Edition: W. Till, Koptische Pergamente theologischen Inhalts I. Mitteilungen aus der Papyrussammlung der Nationalbibliothek in Wien 2 (Vienna 1934) 11-12, Nr. 7.

Sa2

London, British Library, Or. 5984. VI.-VII. Cent. Contents: 66-10 72-8 28-29 81-5 8-17 93-6. Edition: H. Thompson, The Coptic (Sahidic) Version of Certain Books of the Old Testament from a Papyrus in the British Museum (Oxford 1908).

62 Sa3

The Textual Witnesses

Louvain, Universitätsbibliothek, Copt. Lov. 9. III.-IV. Cent. Contents: 612 71-2 Edition: L. Th. Lefort, Les Manuscrits Coptes de l’Université de Louvain (Louvain 1940) 59-65. 7-11 15-18 24-26.

Sa4

K 9879. ? Cent. Contents: 714d-25. Edition: C. Wessely, Griechische und Koptische Texte theologischen Inhalts I. Studien zur Palaeographie und Papyruskunde IX (Leipzig 1909) 78-79, Nr. 27ab.

Sa5

Manchester, John Rylands Library, (Coptic Biblical MSS.) Suppl. 2. V.-VI. Cent. Contents: 17-16. Edition: W. C. Till, Coptic Biblical Fragments in the John Rylands Library, Bulletin of the John Rylands Library 34 (1952) 432-458.

Sa6

Cairo, Coptic Museum of Cairo, Nr. 12761 = MSS. Section 6635. IV.-V. Cent. Contents: 510-66 610-71 74-9 712-15. Edition: N. Bosson, Un Palimpseste du Musée Copte du Caire, Le Muséon 104 (1991) 5-37.

Sa7

Berlin, Staatlichen Museen, P. Berol. 20987. IV.-V. Cent. Contents: 77-18. Unedited. Collated by J. Timbie.

Sa8

London, Private Collection. X. Cent. Contents: 21-6 28-11. Fragmentary. Unedited. Collated by J. Timbie. Text = SaI.

4.1.2  The Textual Character of the Sahidic Translation

Every deviation from the Greek Text in the Coptic versions was noted and initially recorded. Many of these variants, however, are not in the apparatus because they are due to the process of translation, or are mistakes made by the translators or later copyists, or do not support variants in Greek manuscripts. Examples: 714a 718b 718b 718b 1010c

ζῆθι] ζωτι 998; wIne Fa2 SaI II 6 = ζητει (sed hab Fa1) καί γε] om γε PsChr Fa1 SaI ἀπὸ τούτου] post σου tr Fa2 SaI 3; > Fa1 πάντα] + αγαθα SaI καὶ δυνάμεις δυναμώσει] et dabit virtutem ei in virtute SaI; in virtute dabit virtutem ei Fa1 2

The texts of the Sahidic Translation, whether complete or fragmentary, represent different translations made at different times and exhibit different approaches to the task of translation. Generalisations are always problematic, but in general, one can say that with the progress of time, the translations are less formal/ literal and more functional/dynamic in approach to translation.

The Ancient Translations

SaI

X. Cent.

SaII

VI. Cent.

Sa1

Date Unknown

Sa2

VII. Cent.

Sa3

III.-IV. Cent.

Sa4

Date Unknown

Sa5

V.-VI. Cent.

Sa6

IV.-V. Cent.

Sa7

IV.-V. Cent.

Sa8

X. Cent.

63

Usually, SaI and SaII are found together as witnesses of the form of the text attested early in Egypt by B and 998: σύν] συμπαντα 998 411 130 cII d 443 534´ 728; + παντα B-68 V 336 571c Did 82,20 SaI II 610b ὅ] οτι C 998* d k 339 766 Hi SaI II 2 6 77b τὴν καρδίαν εὐτονίας] τ. κ. ευγενειας B-68´ (‑νιας 534) 998 C (-νιας) C´–299 563 571* (‑νιας 798) Antioch 1560 Fa1 SaI II 2 6 714a ἐν ἡμέρᾳ ἀγαθωσύνης] trah ad (13c) 998 Ol Fa1 SaI II 714c τούτῳ] τουτο B-68’ O 522-540-571* 357 336´ 339 OlΓ PsChr Arm Fa1 2 SaI II 4 6 (absc 998) 1013b om στόματος 2° B-68´ 998 L cII 357 336´ 443 Did Ol Fa1 2 SaI II 1019b εὐφραίνει] του ευφρανθηναι B-68´ 998 C´-260 d PsChrlem Fa1 SaI II Hi ↓ 119c καρδίας σου] (+ σου SaI II) ἄμωμος B-68´ 998 336´ Antioch 1485 (ἀμώμως) Did 333,4s 335,26s PsChr Fa1 2 SaI II 310a

Although SaI is dated later than SaII, it frequently attests to the early textual transmission in Egypt when SaII does not. Examples: 26b

215f

319a 320c

ξύλα] > B-534 998 Met II.1,6 Amb Isa 4,25 Arm Fa1 2 SaI

ὅτι—ματαιότης] + διοτι (> Aeth; + ο B-613 998 d SaI) αφρων εκ περισσευματος λαλει (+ (sub  Syh) οτι (και ιδου pro οτι Syh) και γε τουτο ματαιοτης O Syh) B-613 998 O d 336´ 542 766B Aeth Arm SaI Syh ὅτι 1°] οὐ B-68´ 998 637 296´ 311 706 SaI WOVMi c L OlΝ Hi Fa1 2 ἐπιστρέφει] ‑στρεψει B-68´’ 637 125*(c pr m) cII 542 DamV I Sa Syh = Ald

64 416c 611a 714c 723a 88c 113a

The Textual Witnesses

ἐν] επ B-68´’ C 998 L–(125) 336´ 443 542 766 PsChr Fa1 SaI c πολλοί] > 998 C´–299 571 443 Fa1 (absc Fa2) SaI 6 σύμφωνον] ‑φων[.. 998; ‑φωνως B-68´ (‑ος 68) 357(‑ος) Did 212,21 PsChr Arm Fa1 SaI 6 πάντα / ταῦτα] tr 998 L 299 338 645 766A OlΑ Arm Fa1 2 SaI ἐξουσιάζων 2°] εξουσια B-68´ 998 357 Arm Fa2 3 SaI 2 = Ra πληρωθῶσιν] πλησθωσι(ν) B-68´’ V 998 L d (πληθ. 754) 336´ 443 PsChr Fa1 SaI: cf 18d

In fewer cases, however, SaII instead of SaI supports the witnesses of the Egyptian form of the text: 711a 912b

κληροδοσίας] absc 998 (...]ας); κληρονομιας B-68´ O d 336´ 539lem 766 PsChr Arm Fa1 2 SaII 3 κακῷ] καλω B-68 SaII: cf 417d

Limited evidence also connects SaI and SaII to the textual tradition of the Catena groups, usually when the witness of these groups is divided: 416a 58b 101a

129b

περασμός] περισπασμος 147-299-503-560-159 357 339 543 645 698 766A Fa2 SaI II OlΑ* εἰργασμένου] + και ισθι πιστος εν (επι 339) παντι· εστι(ν) βραχυ (βαρυ 336´) απο του ηρπασμενου (ηρπαγμενου 336´) cII 336´ 339 SaI II θανατοῦσαι] θανουσαι 411 390-415-425-504-522-540-601-732 357 161248c 645 Anast 641 Constit II 17,4 OlΙΚ omne lat. (praeter Hi) Aeth SaII = Compl σὺν τὸν ἄνθρωπον] συνετον ανθρωπον 998: cf 310a; συν τους ανθρωπους Fa2; συν τον λαον O–637-411 Hi Arm SaI (absc SaII) = Compl Ra 𝔐 ↓

4.2  The Fayyumic Translation (Fa) 4.2.1  The Witnesses/Editions of the Fayyumic Translation Fa1

Hamburg, Papyrus Bil. 1. 300. Edition: B. J. Diebner u. R. Kasser, Hamburger Papyrus Bil. 1. Die alttestamentlichen Texte des Papyrus Bilinguis 1 der Staatsund Universitätsbibliothek Hamburg. Genf 1989. The text is complete.

Fa2

Ann Arbor, Univ. of Michigan Library, P. Mich.inv. 3520. IV. Cent. Edition: H.-M. Schenke u. R. Kasser, Papyrus Michigan 3520 und 6868(a). Ecclesiastes, Erster Johannesbrief und Zweiter Petrusbrief im fayumischen Dialekt. Berlin: W. de Gruyter, 2003. The text is complete.

The Ancient Translations

Fa3

65

Ann Arbor, Univ. of Michigan Library, P. Mich.inv. 6868. VI.-VII. Cent. Contents: 719-22 26-28 83-913. Edition: H.-M. Schenke u. R. Kasser, Papyrus Michigan 3520 und 6868(a). Ecclesiastes, Erster Johannesbrief und Zweiter Petrusbrief im fayumischen Dialekt. Berlin: W. de Gruyter, 2003.

4.2.2  The Textual Character of the Fayyumic Translation

The texts of the Fayyumic translation, whether complete or fragmentary, represent different translations made at different times and exhibit different approaches to the task of translation. Usually, in the Coptic daughter versions, the Fayyumic translation is dependent on the Sahidic, and it should be noted, that as a general rule, both are far less precise than the Bohairic in the rendering of καί. Fa1

300

Fa2

IV. Cent.

Fa3

VI.-VII. Cent.

Cases exist where Fa1 (or Fa1 2) and 998 share unique variants: 213b 215a 321b 113b

ὑπέρ 2°] παρα 998 Fa1 ἐν 1°] τη 998 Fa1; + τη 574 705 752 SaI εἰς ἄνω] εις ουρανον 998 Fa1 2 καὶ ἐάν 2°] ἤ 998 Fa1 2 SaI

This is not surprising since Fa1 and 998 are contained in the same manuscript. Not infrequently, Fa1 2 (26b, 320c, 711a, 714c, 723a, 1013b, 119c supra), Fa2 3 (88c) or just Fa1 (416c, 611a, 77b, 714a, 714c, 1019b, 113a supra) join Sa in attesting the early textual transmission in Egypt along with B 998 and congeners. Fa3 is freer and later and hence does not always join Fa1 2 in attesting the Egyptian Text. It is also fragmentary. Further examples where Fa1 2 or just Fa1 support the Egyptian Text, sometimes independently of Sa are as follows: 28d om καί 6° B-534 998 d Fa1 SaI Ol–ΑΓ, contra 𝔐 28e τοῦ ἀνθρώπου] των ανθρωπων V-411 798c cII d–357 766B SaI, contra 𝔐; ανθρωπων B 998 357 296´ 311 336´ 706 795 Arm Fa1 311b σύν 2°] συμπαντα B-68´’ 411 130 571c cII d 443 Ath II 573 Dion Fa1 SaI 41d om καί 3° B-68´ 998 357 PsChr Fa1 2 SaI II, contra 𝔐 53c ὅτι οὐκ] ου γαρ Arm SaI; om ὅτι B-68´ 998 Fa1 PsChr = Ald: cf (7c) 515c τίς] > B-68 998 357 Fa1 2 Sa6 Amb Nab 6,28 (sed hab SaI II) 63c ἀγαθωσύνης] + αυτου 998* O-411 Fa1 66c τὰ πάντα] > B-68´ 992 998 336´ Fa1

66

The Textual Witnesses

ζωῆς] + αυτου 295 534´ Fa1 2: cf 23g κληροδοσίας] absc 998 (…]ας); κληρονομιας B-68´ O d 336´ 539lem 766 PsChr Arm Fa1 2 SaII 3 89b ποίημα] pr το B-68´ 998 V d 260 PsChr Fa1 3 SaI II 2 = Ald 815b ἅς] α V; ων C 252 543 549: ex 517g; ως 795; οσας B-68´ 998 161mg-248mg 443 534 = Ra; οσα 357 336´ 101a ἐλαίου] ελαιον B-S*-68 645 Fa1 (absc 998) 101b ὀλίγον σοφίας] ο (> 998 252mg) λογος B* 998 252mg SaII (absc Fa1) 1013b om στόματος 2° B-68´ 998 L cII 357 336´ 443 Did Ol Fa1 2 SaI II 1015a κοπώσει] κακωσει B-68´ 998 (‑σι) cII d 155 161mg-248mg 336´ 443 542 766 Didlem et com 307,8 307,22 307,25 (-σι) Metlem et com IX.14,4,65 OlΑΒΓΖΗ PsChr Arm Fa1 SaI Hilem et com 339,227 339,232 (affliget) Syh = Vulg 1019b οἶνος] + και ελαιον (ελεον S 254 336´) B-S-68´’ 998 O-411 L C´(ελαιος 130*)260 d 336´ 359 443 547 645 795 PsChrlem Arm Fa1 2 SaI = Pesh: cf Ps 48 117b ἥλιον] + ευθη 795; + ου ει (οὗ εἶ) B*-534 998 Fa1; + ου η V 119d καί 3°] + μη B-S*-68´ 998 C´–571* d 336´ Aethte (sed hab Aethap) Fa1 2 SaI (absc SaII) CPA 612b 711a

Finally, the Egyptian Text was corrected on the basis of hexaplaric sources early on and the Coptic versions attest to this. These cases will be discussed under the analysis of the Greek witnesses B-S-998 and congeners. παρ᾽ αὐτῆς] εχοντι αυτην O–V 766 Hi Fa1 ↓ : cf 512b δικαίῳ] δικαιοσυνη O 357 545c(-καιω-) 698 OlΕ Ambrst Rom 12,19 (sed hab Ambrst Eph 4,26) Hi Arm Fa1 2 SaI II 6 ↓ 722b om (22b) comma Fa1 722b καρδίαν] καρδια B-68 998* SaI (sed hab Fa2); η καρδια 125´ 725c ἀσεβοῦς ἀφροσύνην] impietatum stulti Hi Syh SaI 3 Fa1 = Vulg 𝔐 99b πάσας τὰς ἡμέρας ματαιότητός σου] πασαι ημεραι ημεραι (> 534 Fa1 3) ατμου mg σου B*-534 998 Fa1 3 ↓ ; > A Bc-68’ 637 797txt cII–260 d 338 547 OlΑΓΔΖ Κ Aethap(A2) (sed hab A2 rasuram; sed hab Did 277,1-2 Hi Arm SaII Syh et = Ald) = Sixt 510b 715b

5.  The Ethiopic Translation (Aeth) The Ethiopic Text was collated for the edition by Dr. Jan Dochhorn (Associate Professor of New Testament in the Department of Theology and Religion, Durham University).

5.1  The Witnesses/Editions of the Ethiopic Translation

The Ethiopic Text of the Book of Ecclesiastes. Edited by Samuel A. B. Mercer: (London, 1931).

The Ancient Translations

67

Literature Michael A. Knibb, Translating the Bible: The Ethiopic Version of the Old Testament (Oxford, 1999). Oscar Löfgren, Review of S. A. B. Mercer, The Ethiopic Text of the Book of Ecclesiastes, Monde Orientale 26-27 (1932, 1933) 334-345.

5.2  The Textual Character of the Ethiopic Translation

The edition of Mercer is based on a MS. from the early XVth century. To this MS. he collated some twenty-two other MSS. and then listed variants from them in the apparatus. Critics have not found his edition very satisfactory. Specialists in the ancient Ethiopic version of the Bible describe three stages in the production and textual transmission of the Ethiopic version of the Old Testament: (1) an original translation based upon the Greek version done in the V-VI centuries, (2) a revision of the original translation from Syriac based Arabic versions in the mediaeval period, and (3) a correction of the translation based on the Hebrew made in the XV-XVI centuries. Dochhorn collated the edition of Mercer for this edition. A manuscript in the apparatus of Mercer’s edition designated A2 frequently supports the LXX against his lemma text. A2, therefore, is an important witness for the text history of the Septuagint. The following sigla are employed in our Apparatus: Aethte AethA2 Aethap Aethap(A2)

= Text and Manuscripts in the Apparatus supporting the Critical Text = MS. A2 = Many MSS. in Mercer’s Apparatus excluding MS. A2 = MS. A2 plus many MSS. in Mercer’s Apparatus

We are not in a position at the present time to discern the different stages of the Ethiopic version of Ecclesiastes from the edition of Mercer. Nonetheless, the collation reveals that the Ethiopic version of Ecclesiastes was made from an early catena tradition that derived from the hexaplaric group and further back, from the Egyptian form of the text. 212d 217a 223b 612b 86a 101a

5.2.1  Agreements with the Catena Tradition

σὺν τὰ ὅσα] συμπαντα οσα 106-130-261 411 299-cII 254´-342 545 547 Dion OlΑΓ Aeth σύν] συμπασαν 411 798-cII 260c 336´ 795 Aeth SaI αὐτοῦ 2°] αυτω C´–299 609 Aeth SaI = Ald ἀριθμόν] αριθμων 411 261-545 159-503*-797 248* 252 645 OlΕ Aeth; καὶ κρίσις] κρισεως C´ Aeth Fa2 = Ald: cf (5b); θανατοῦσαι] θανουσαι 411 390-415-425-504-522-540-601-732 357 161-248c 645 Anast 641 Constit II 17,4 OlΙΚ omne lat. (praeter Hi) Aeth SaII = Compl

68

The Textual Witnesses

1019c

18b 23d 28d 215f

31b 322d 417b 64a 66c 721a 84b 914a 1011b

τοῦ ἀργυρίου] τω αργυριω C´ L 161-248c 443 613txt Max II 793 Hi Aeth SaI = Ald

5.2.2  Agreements with the O Group

ἀπὸ ἀκροάσεως] του ακουσαι (ακουειν Chr XV 684) O–253 Chr XV 684 PsVig Trin 1,7 Ruf Ex 11,6 Aeth ↓ ἐπ᾽] εν O–V-411 cII 336´ 534 Hi Aeth Arm = 𝔐 Vulg ἐντρυφήματα] εντρυφη μετα B*-S* 296 706 795 Aeth ὅτι—ματαιότης] + διοτι (> Aeth; + ο B-613 998 d SaI) αφρων εκ (+ του 613) περισσευματος λαλει (+ (sub  Syh) οτι (και ιδου pro οτι Syh) και γε τουτο ματαιοτης O Syh) B-613 998 O d 336´ 542 766B Aeth Arm SaI Syh (sed hab Hilem 269,271s) = Pesh, contra 𝔐 οὐρανόν] ηλιον Sc O-411 C´’–(390) 574 Dam Ollem et com An Jb 509 BrevGoth 218 Pel 1 Cor 7,31 SedScot Misc 64,2 Aeth Syh = Ald, contra 𝔐: cf 113c ἐν ᾧ] εως V; quod Hi Aeth ↓ ἐγγύς] εγγισον Sc O L–(125) 336´ 338 542 766 Met IV.4,4 OlΝ PsChr Isid Jud 30,6 (approxima) Hilem et com 290,248 291,254 (appropinqua) Aeth Syh ↓ txt πορεύεται] πορευσεται O–637 C´–390 425 Hiap Aeth Fa1 SaI Syh = Gra mg πορεύεται] πορευσεται S O C´–390 425 798 248 260* 338 613 645 OlΙ* Aeth Fa1 SaI (absc 992) = Ald Sixt ἀσεβεῖς] > Sc A C O k 248´ 252 296´ 311 336´ 339 542 543 549 706 766 Anton 1064 Dam Did 222,19 omne lat. Aeth Syh = Ra 𝔐 Vulg ποιήσεις] εποιησας O-411 L(–125) cII 296´ 311 338 547 645 706 Met VII.8,5s Ol Aeth Arm καί 2°] καν O–V Aeth περισσεία] οφελος O–V Aethte (sed hab Aethap(A2)) ↓

5.2.3  Agreements with Early Witnesses of the Egyptian Form of the Text

211a ἐν 1°] επι 998 139-147-159-503-560 766B Aeth, contra 𝔐 216c ταῖς ἡμέραις ταῖς ἐρχομέναις] ημεραι ερχονται 998 Aeth 57b ἐπί] εν 998 766 Aeth Syh Ath I 672 = Pesh, contra 𝔐 M PM 71b om αυτου B-S*-68´ 998 C´ 336´645 DamM M An Jb 3 Aeth

6.  The Old Georgian Translation (Geo) 6.1  The Witnesses/Editions of the Old Georgian Translation The following sources were selected and collated by Natia Dundua: Athos, ᾽Ιβήρων georg. 1; Written in 978. “Oshki Bible” (Siglum O). Jerusalem, Jer. 113; XIII.-XIV. Cent. (Siglum J).

The Ancient Translations

69

Georgian National Center of Manuscripts, A 51; XVII.-XVIII. Cent. “Mxceta Bible” (Sigel S). Georgian National Center of Manuscripts, S 1349; XVIII. Cent. Georgian National Center of Manuscripts, A 1418; XVIII.-XIX. Cent. The earliest source for the Old Georgian translation of Ecclesiastes is the Oshki Bible (978). Three of our sources, O (Oshki), J (Jerusalem), and S (Mxceta), represent one recension. The scribe of J checked his work against O. The scribe of S mostly follows O and J, but sometimes corrects his text according to the Armenian Bible (i.e. Voskan Bible, Amsterdam, 1666), which in turn was revised on the basis of the Latin Vulgate. Where S is not corrected in this way, it can be used as a control, because we have exact information that its Georgian Vorlage was older than that of both O and J. The editor of the Mxceta Bible, Sulkhan Saba Orbeliani, carefully followed his sources and did not correct anything (e.g. grammatically) himself. Thus O, J, and S undoubtedly represent the first, original, translation. The precise date of this translation is unknown, but J. Gippert has dated fragments of 1 Esdras and Isaiah paleographically to the fifth-sixth century CE (See Jost Gippert ed. (with Zurab Sarjveladze and Lamara Kajaia) The Old Georgian Palimpsest Codex Vindobonensis georgicus 2, Vol I. Brepols, 2007). On the problems of representing Greek by translation into Old Georgian one might consult the following: M. Van Esbroeck, Les Versions orientales de la Bible. Une Orientation Bibliographique, in J. Krašovec (ed), Interpretation of the Bible. Section Two: Philological and Literary Sciences. The Slovenian Academy of Arts and Sciences. Sheffield, 1998, 465-480; J. Krašovec (ed), Interpretation of the Bible. International Organization for the Study of the Old Testament. Congress Volume. Ljubljana 2007, Leiden, 2010; A. Kharanauli. Einführung in die georgische Psalterübersetzung, in Anneli Aejmelaeus and Udo Quast (eds.), Der Septuaginta-Psalter und seine Tochterübersetzungen, Symposium in Göttingen 1997. MSU 24, Göttingen 2000, 248-308; idem, Das Chanmeti-Fragment aus Jeremia—Fragen seiner Entstehung und seiner Übersetzungstechnik. Oriens Christianus, 85 (2001), 204-236; idem, Ein ChanmetiFragment der georgischen Übersetzung von Esdra I (Fragen der Authenzität, Vorlage und Übersetzungstechnik), Le Muséon, 116 (2003), 181-216. Nonetheless, the Old Georgian translation of Ecclesiastes is a literal translation and faithfully represents all the peculiarities of the Greek syntax and word-formation. The translation is also characterised by inner harmonisations of parallel texts in Ecclesiastes.

70

The Textual Witnesses

In the apparatus of the edition we give the reading of only the O, J and S manuscripts and mark it as Geo. No other sources are cited (including the Printed Bible of Prince Bakar, Moscow, 1743) because the text of the translation was later revised on the basis of the Slavic Bible. The manuscripts of the Georgian National Center of Manuscripts, A 1418 and S 1349, also show revision on the basis of the Slavic Bible and so were not used.

6.2  The Textual Character of the Old Georgian Translation

I have collaborated with Natia Dundua in the following analysis and description of the Old Georgian translation of Ecclesiastes. The collation of Geo Eccl against the Greek produces a list of approximately 183 variant readings. As a matter of first principle, a comparison was also made with Arm Eccl and the Old Georgian supports significant variants against Arm in many places and in general does not transmit the unique readings of Arm. This establishes that Geo Eccl was translated from a Greek source(s). From this group of variants 42 were selected as demonstrating genetically significant textual variation. Some of the variants shared with Greek witnesses are intrusions from parallel passages, a feature that is characteristic of Geo. What is most remarkable are the number of variants shared almost uniquely with early members of the Egyptian Group such as 998, B, and S. We have 10 instances of Geo agreeing with 998 B; 8 with 998 B-S; 1 with B-S; 3 with 998 alone; 3 with S alone; and 1 with B alone. Thirteen of these are virtually unique agreements. Some 18 of these variants are supported by various Coptic witnesses. Another 8 variants not supported by 998, B or S or any combination thereof, are supported by one or more Coptic witnesses, some uniquely. Examples are given below: 26b

ξύλα] > B-534 998 Met II.1,6 Amb Isa 4,25 Geo Fa1 2 SaI (sed hab Did 37,15 Ol Syh = Gra 𝔐)

215e

τότε] > B-S*-534´ 998 C´–571 443 547txt 645 705 Dionlem 220 GregNy 361,17 OllemΑΓΕΖΙΚΜΝ Olcom (sed hab Met II.6,5) Hilem 269,271 Geo Arm SaI (sed hab HiLXX 269,285-289 Syh), contra 𝔐

215f

ὅτι—ματαιότης] + διοτι (> Aeth; + ο B-613 998 d SaI) αφρων εκ (+ του 613) περισσευματος (+ καρδια Geo) λαλει (+ (sub  Syh) οτι (και ιδου pro οτι Syh) και γε τουτο ματαιοτης O Syh) B-613 998 O d 336´ 542 766B Aeth Geo SaI Syh (sed hab Hilem 269,271s) = Pesh, contra 𝔐; pr (sub ÷ 788 (οὐ κεῖται παρ’ ἑβραίοις 788mg); + και 299) διοτι (οτι Dioncom 221 GregNy 364,19; > Dam; +

c

The Ancient Translations

71

ο 390-574-601-cII GregNy 361,18ap 364,19ap) αφρων (ανος 547txt) εκ περισσευματος λαλει (+ οτι αφρων εκ περισσευματος λεγει post fin 443) rel (HiLXX 269,285-289 GregNy 361,18 364,19 Metlem et com II.6,6s,102s II.7,7s Ol Proc CatP 22,154 Armte et ap) = Ra: ex Matth 1234 (par Luc 645) 57d om ὅτι B-68´ 998 PsChr Geo (sed hab Hi Syh = Ald Sixt 𝔐) 515c

τίς] > B-68 998 357 Amb Nab 6,28 Geo Fa1 2 Sa6 (sed hab Dam Did 160,4 Metlem et com V.3,10,90 Ol Arm SaI II Syh = Ald Sixt Ra)

71a

ἔλαιον] ελεον B-S-68 A O 797-cII–260 k 296´ 311 336´ 339 698 706 795 Dam–V Didlem et com 196,11 196,15 196,20 196,25 PsChrcom Geo Armte Sa3 (sed hab Anton 1129 Ol Fa1 2 Hi Syh = Sixt)

85b

κρίσεως] γνωσεως 998 357 Geo

101b

ὀλίγον] ολιγω 68 (sed hab Ald); ο (> 998 252mg) λογος B* 998 252mg Geo SaII (absc Fa1)

119c

καρδίας σου 357 542 DamL Hilem et com 348,156 351,68 = Vulg] (+ σου Geo SaI II) ἄμωμος B-68´ 998 336´ Antioch 1485 (ἀμώμως) Did 333,4s 335,26s c P PsChr Geo Fa1 2 SaI II; + ( Syh) ἄμωμος rel (Anton 960 1057 1208 Dam–L M Max II 968 Met X.5,2 Ol Amb Exh virg 10,69 Spec 473,15 Hi Pach 144,13 Aeth Arm CPA Syh Ald Sixt)

EOVMi

c

The Vorlage of Geo is clearly a derivative of the early Egyptian text-type. In addition, note that there are 15 agreements with 357, one of perhaps two manuscripts that are at the head of the stemma; all other witnesses at the head of the stemma are hypothetical ancestors. This is another witness to the early date of the Vorlage behind Geo. Similarly, Geo Eccl has 28 agreements with 336´, a manuscript pair that is probably a derivative of the early Egyptian Group: 110a

ἰδέ] > 336´ Geo Fa1 2

The following chart shows agreements of Geo plus one, two, three, and four text groups. The vertical column on the left side indicates an agreement between Geo plus the group listed. Then one can consider whether this agreement is unique (+ 0 Groups) or attested by other Groups (+ 1, 2, 3, or 4 Groups). Some examples follow.

72

The Textual Witnesses

Geo + ? + 0 Groups B-S 998 Co 111b, 114ab, 26b, 53c, 57d, 515c, 85b, 812c, 101b, 119c, O 212b, 63b, L

53d,

1 Groups 2 Groups 28d, 215e, 215f, 1013b, 224b(2), 43c, 517a, 71b, 815a, 214a, 517a, 815a, 1212a 317b,

113c, 215c, 117a, 317a, 71a, 215f, 224b, 53d, 113c, 215c, 117a, 310a, 97c, 1013b, 113c, 224b, 53d, 215c, 117c,

C´ C´’ C’ cII

58b,

214a, 1212a

d

124d

28d, 317b, 43c, 215f,

k

3 Groups 117c, 310a, 317a, 71a,

215e, 71b,

1013b,

43c,

224b,

4 Groups 116b, 417d, 56b, 1019b, 116b, 417d, 56b, 1019b, 116b, 417d, 56b, 1019b, 116b, 1019b, 417d,

310a, 317a, 71a, 97c, 117a, 117c, 116b, 417d, 310a, 317a, 97c, 56b, 1019b, 117a, 117c, 71a, 56b, 97c,

Agreements of Geo with one, two, three, and four text groups produces the following totals: O 18; L 12; C´ 7; C´’ 3; C ’ 1; cII 7; d 14; k 7. As cII is cited with other C groups in C´’ or C ’, the number of agreements with cII is 11. The cII group is probably earlier than C´ and also developed from the early Egyptian Group in part. Geo Eccl is not a hexaplaric text, nor is it a strong Lucianic representative: agreements with O, L, and C´’ derive from the influence of the early Egyptian Text on these groups. The only contrary examples are 117a and 63b. 58b

M

εἰργασμένου] ειργασαμενου 548; ειργασμενος 336´ DamM ; > Geo; + και ισθι πιστος εν (επι 339) παντι· εστι(ν) βραχυ (βαρυ 336´) απο του ηρπασμενου (ηρπαγμενου 336´) cII 336´ 339 Geo Arm SaI II

1013b om στόματος 2° B-68´ 998 L cII 357 336´ 443 Did Ol Geo Armte Fa1 2 SaI II (sed hab Anton 800 Dam Met IX.13,3 Max II 981 = Sixt) c

c

113c

οὐρανόν] ηλιον Sc O L C´–571 798 161mg-248mg-252 339 443 543 547 549 Antioch 1469 GregNy 301,4 Ol Proc 11,104 (sed hab Met I.14,5,56 GregNy 299,24 300,3 300,13 La94 95 Aethap Geo (no parallel in 31b) Arm Fa1 SaI 5 Syh; absc Fa2) Hi = Pesh Tar Vulg, contra 𝔐: cf 31b

117a

καί 1°—γνῶσιν] sub ※ V 788 Syh; > 253 L (–106 125) 130 d–254´ k 68 248´txt (= Compl; καὶ ἔδωκα—τοῦ γνῶναι 248´mg) 296´ 311 338 443 547 705 Clem II 37 Dion 212 Ol Geo ↓ : homoiot

The Ancient Translations

73

417d

κακόν (κακων 534)] καλον S 253-475-411 L(–125) C´’ d–357 155 248´ 252txt 296´ 311 339 443 539 543 547 548 549 645 698 706 795 Geo (no parallel 912b) Arab Arm SaI II Met IV.4,5 (sed hab Aug Spe 8 SedScot Misc 13,17,1 Hi Fa1 (absc Fa2) Syh = Compl Gra. 𝔐 Vulg): cf 912b

56b

ὅτι σύν] οτι συ B-S-534 998 637-411 L 139c-425-563-571-797c-798 d k 252 311 336´ 338 339 443 542 543 547 549 698 766B 795 Met IV.8,6 Geo Arm Syh (sed lem et com hab OlΑΓΝ Ol–Β = Gra Ra 𝔐)

117c

ὅτι] pr εγω B-S-534´ 998 411 C´’ d k 336´ 645 705 OlΒΔΕΗΘΙΚΜΝ Geo SaI = Ald;

63b

ἐτῶν] pr των C L 563c(vid)-571 534´ OlΑ; ζωης O–V-411 Geo: cf 519a

7.  The Armenian Translation (Arm) 7.1  The Witnesses/Editions of the Armenian Translation

The following sources were selected and collated by Claude Cox (Adjunct Professor, McMaster Divinity College): H. Zohrapian (ed.). Astuatsashunch‘ Matean Hin ew Nor Ktakaranats‘ (Scriptures of the Old and New Testaments). Venice, 1805. Reprint: The Zohrab Bible. A Facsimile Reproduction of the 1805 Venetian Edition with an Introduction by Claude Cox. Classical Armenian Text Reprint Series; Delmar, NY: 1984. Reprint: Tigran A. Aivazian (ed.), Classical Armenian Bible. London, 2008. Jost Gippert, Wolfgang Schulze, Zaza Aleksidze, and Jean-Pierre Mahé (eds.). The Caucasian Albanian Palimpsests of Mt. Sinai. Volume III: The Armenian Layer. Jost Gippert (ed). Monumenta Paleographica Medii Aevi. Series Ibero-Caucasica. Turnhout, Brepols, 2010. Inhalt: 113-23; 212-226; 310-320; 45-57; 716-88; 125-1214. Jerusalem, Armenian Patriarchate, 1925; Written in 1269. Yerevan, Matenadaran, 1500; Written before 1282. London, British Library, Oriental 8833; XV Cent. Sinai, St. Catherine’s Monastery, Cod. Geo. N 13 and N 55; VIII Cent.

7.2  The Textual Character of the Armenian Translation Literature M. Abeghyan. Vark‘ Mashtots‘ (Life of Mashtots). Yerevan: Haypethrat, 1941. Claude Cox. Armenian Ecclesiastes: Arm 1 and Arm 2. Revue des Études Armeniennes 34 (2012).

74

The Textual Witnesses

The analysis was written by Claude Cox to whom I express my grateful thanks: The historical tradition (Koriwn’s The Life of Mashtots‘) asserts that the Armenian translation of the Bible was undertaken immediately after the invention of the alphabet in Samosata, Syria, ca. 406. It is still a largely unresolved question to what extent translators may have employed the Peshitta as a parent text, rather than Greek witnesses, or alongside it. The first translation was revised or checked (“established”) after the Council of Ephesus (431) when MSS. were brought back to Armenia. An earlier and a later translation type have been identified in the Gospels, Chronicles, Daniel, and now in Ecclesiastes. The textual tradition of individual books may not adhere to the two-stage generalization and the true story of the work of translation as a whole appears to be much more complex than we might conclude on the basis of Koriwn. There is no critical text for most books of the Bible in classical Armenian, so the Göttingen editions have used the old edition of Zohrapian, prepared in 1805. It is a diplomatic edition, based upon one MS., Venice 1508, dated 1319. In an apparatus Zohrapian cites variant readings from other MSS. available to him in Venice by imprecise designations like “some (witnesses attest).” The merit of Zohrapian’s edition is that he faithfully reproduced the text of the base MS. and the readings in his apparatus are well-chosen variants. Though sometimes disparaged as preserving a late, common (vulgar) text, generalizations here too deserve to be questioned. In Eccl Zohrapian’s base MS. presents well the early Armenian translation, usually designated Arm 1. Further, his extensive apparatus provides an ample portion of the revision of Arm 1, designated Arm 2. It is the earlier translation (Arm 1) that has been collated for the Göttingen edition. The parent text of Arm 1 in Eccl was a MS. closely related to the C text tradition, more particularly to cII. Its place among the Catena group witnesses is clear from the following shared variant readings: 116e 41g 58b 725c 815d

καί 2°—γνῶσιν post (17a) γνῶσιν tr C´(–797) Arm καί 6°] ∩ (2a) καί 411 C´’–260 299 547 Arm εἰργασμένου] + και ισθι πιστος εν παντι· εστι(ν) βραχυ απο του ηρπασμενου (c var) cII 336´ 339 Arm ἀφροσύνην] ευφροσυνην C´’–157 299 563 571 609 797 357 Armte συμπροσέσται] συμπορευεται 147-159

The collation of Arm Eccl against the Greek produces a list of some 108 variant readings. Agreements of Arm with one, two, and three text groups produces the following totals: O 17; L 20; C 1; C´ 12; C´’ 5; cII 29; d 18; k 15. Since cII is cited with other C groups in C´’, the number of agreements with cII is 34. Arm Eccl is not a hexaplaric text, nor is it a strong Lucianic representative: agreements with O

The Ancient Translations

75

and L derive from the textual character of the C group, which has been influenced by those types of text. At 3:19 and 12:9 Arm has unique agreements with the O group, but these appear to be coincidental in nature. Arm is most closely related to MS. B among the majuscules and shows frequent agreement with several of the unclassified MSS., especially 336´ 547 443 and 542. The Armenian is more closely related to group cII witnesses than to any individual unclassified MS. Among agreements with cII are the following. Note also 5:8, cited above. 23g

214c 310a 417b 1212b

ἀριθμόν] εν αριθμω Sc O-411 cII 547 613sup lin Arm

καί γε] > 411 cII Arm σύν] συμπαντα 998 411 130 cII d 443 534´ 728 Arm ἐγγύς] + γινου cII 795 Armte οὐκ ἔστιν] pr οτι 411 L cII 443 613 795 Arm

Arm 1 is not a literalistic translation. This makes it difficult to collate and of less text critical usefulness than its revision, Arm 2. The importance of Arm Eccl lies in its early date: the Greek MSS. of group cII, for example, are medieval or later. Agreements with cII mean that its type of text was already current in the early fifth century. Only MSS. B S and the fragmentary papyri 818 870 and 998 are earlier. The daughter versions rarely preserve original readings otherwise lost. Nevertheless, there are occasions when a version attests notable readings. Such is true of πονηρούς in place of πολλούς at 7:29: the former is otherwise known only from patristic quotations. Its preservation in Arm indicates that it was once to be found in a Greek MS., namely the parent text of Arm 1. The text of Zohrapian’s Edition, V(enice) 1508, represents Arm 1 for the Göttingen Edition of the Greek. The fragmentary Mt. Sinai palimpsest edited by Gippert, which he dates to the eighth-ninth century, preserves the same type of text as V1508 and contains no readings that suggest a different parent text; it does confirm the reliability of the later MS., V1508. Manuscripts Jerusalem 1925 and Yerevan Matenadaran 1500, so important for preserving an early type of text in some other books such as Job, attest Arm 2 for Eccl. Manuscript London/British Library Oriental 8833 (17th century), an Arm 1 witness, was consulted from time to time to determine that readings in V1508 are not preserved uniquely, but are likely typical. Though the text collated is that of V1508, close attention was also paid to the extensive quotation of Arm 2 in Zohrapian’s apparatus. The comparison of Arm 2 with Arm 1 proves decisively useful for understanding the translation technique of Arm 1. Both J1925 and M1500 were consulted for their witness to Arm 2 where its text was required beyond what Zohrapian has provided in his apparatus.

76

The Textual Witnesses

In the case of a version, it cannot be concluded that, if its reading is not cited in the apparatus, the version must attest the lemma. Issues of translation make such precision impossible. In the collation of Arm Eccl this applies, for example, to the vexatious question of whether the parent text read σύν where it has been employed to represent the Hebrew direct object marker. The problem is made more acute by the corruption of σὺν πάντα into σύμπαντα in some witnesses: Arm 1 does not distinguish the former from the latter, appearing to ignore σύν in the former case. Twice Arm 1 treats this use of σύν as the preposition “with” (7:26; 11:7). The otherwise determined textual affiliations of Arm 1 sometimes allow one to see where Arm likely belongs, but one cannot be sure. In such instances Arm has not been cited in the apparatus but does not necessarily support the lemma.

8.  The Arabic Translation (Arab) 8.1  The Witnesses/Editions of the Arabic Translation

The following sources were selected and collated by Dr. K. Martin Heide (Privatdozent, Center for Near- and Middle East Studies, Philipps-Universität Marburg):11 Brian Walton. Biblia Sacra Polyglotta. London, 1657. Reprint: Akademische Druck- u. Verlagsanstalt, Graz, Austria.

Literature Georg Graf, Geschichte der christlichen arabischen Literatur. Studi e Testi 118. (Città del Vaticano: Biblioteca Apostolica Vaticana, 1944), 1:127-131. The Arabic Text that was printed in the Paris Polyglot and again reprinted in the London Polyglot is taken from Paris, Bibl. Nat., Ar. 1 (Written 1584-1585).

8.2  The Textual Character of the Arabic Translation

Some 44 instances were selected demonstrating more or less significant variants: 14 showed agreement uniquely with one group, 6 with two groups, 15 with three groups, and 9 with more than three groups. Using E as a siglum for the “Egyptian Group” (998 B-S-68´’ Co) the results of analysis are as follows: Agreements with one group: 5 with E, 2 with O, 3 with L, 3 with 766, and 1 with 248´. 11  According to Graf, one manuscript is based on the Septuagint (Tübingen, Universitätsbibliothek, Ma VI 225 (194). XVI. Cent.). A cursory collation of this manuscript revealed a textual tradition contaminated by the Latin Vulgate of Jerome, so the Tübingen MS was not used for the edition.

The Ancient Translations

77

Agreements with two groups: 2 with E O, 1 with O L, 1 with O C, 1 with L C, and 1 with C 248´. Agreements with three groups: 1 with E O d, 3 with E C d, 2 with E L cII, and 1 each with combinations E d cII, E d 336´, O L C, O L cII, O d 248´, L C d, L cII 248´, L k 248´, cII k 248´. These data are not as confusing as they may at first seem. The Lucianic recension is based on the Hexaplaric recension. The d group is a development from the Egyptian Text and the Catena Groups and 248´ are scholarly traditions based on the earlier textual tradition in E, O, and L. The Arabic Translation is relatively late (Saadia Gaon is tenth century) and is based on a scholarly textual tradition which ultimately has its roots in the Egyptian Group, and later developments of it in the Hexaplaric, Lucianic, and Catena groups. This explains scattered agreements of Arab with E, O, L, C, and 248´. E 28d

8.2.1  Agreement with One Group τοῦ ἀνθρώπου] των ανθρωπων V-475-411 798c cII d–357 766B SaI, contra 𝔐; ανθρωπων B 998 357 296´ 311 336´ 706 795 Geo Arab Fa1

28d

ἐντρυφήματα] εντροφηματα 475; εντρυφη μετα B*-S* 296 706 795 Aeth Arab

313a

ὅς] ὅ 357 766A GregNy 441,17ap; > S* Arab

c

317b om σύν 1° 311 Arab SaI 612b

ζωῆς] + αυτου 295 534´ Arab Fa1 2: cf 23g

L 318b

δεῖξαι] + αυτους L 795 Arab (vel + αυτοις): ex (18c)

61b

ἄνθρωπον] ουρανον 130 Arab SaI 6 (absc Fa2; sed hab Fa1)

1214a

σὺν πᾶν τὸ ποίημα S A 336 698 PsChr 67,8 = Gra Ra] παν το π. L Eus Ps 80 428 532 Arab (vel συμπαν το π.) SaI Syh;

O 415a

σὺν (ουν 637) πάντας τοὺς ζῶντας V-637 336 Aeth Arab = Gra Ra] συμπαντας (-τα 998 357) τους ζωντας rel

78 715b

248´ 26a 766 318a

The Textual Witnesses

δικαίῳ] δικαιοσυνη O 357 545c(-καιω-) 698 OlΕ Ambrst Rom 12,19 (sed hab Ambrst Eph 4,26) Hi Is 2,52,34 8,19,44 17,35,32 Pel 1,39 2,3 Hi Arab Fa1 2 SaI II 6 ↓

ποτίσαι] ποιησαι 161mg-248mg Didlem 37,15 Dion Ol–ΒΓΗΝ Arab Armap

ἐκεῖ] absc 818 998; pr και Aethte (sed hab AethA2); και B-68´ 411 cII 357 OlΔΙΚΜ (sed hab Fa1 = Ald); > 766 Met III.20,4,58 Arab;

56b

ὅτι σύν] οτι συ B-S-534 998 637-411 L 139c-425-563-571-797c-798 d k 252 311 336´ 338 339 443 542 543 547 549 698 766B 795 Met IV.8,6 Geo Arm Syh (sed lem et com hab OlΑΓΝ Ol–Β = Gra Ra 𝔐); και συν Aeth; sed Hi ↓ ; tu vero Vulg; συν Ald; συ 571c SaI; om συν OlΒcom; > 68 766A Arab Fa1

912d

ὡς αὐτά] ωσαυτως 155 (ως αυτος) 298 766 DamKVM 330,286 Aeth Arab = Pesh

EO 815a

129b

OL 102a

OC 311a

LC 119e

PM c

L

OlΒ Hilem et com 324,118

8.2.2  Agreement with Two Groups σὺν τὴν εὐφροσύνην] συν τη ευφροσυνη 261 645 OlΒΗ; om σύν O–V 545c 311 336´ 534 766B Geo Arab Fa1 2 3 SaI 2 ἔτι Hi (adhuc) = Gra Ra] οτι B-68´ 998 637 254´ 336´ 542 766B Arab Arm SaI Syh = Pesh; και O–637 248´ 338 443 Met X.10,3 Geo; και οτε 296; και οτι rel (Aeth) = Ald: cf 728a

εἰς δεξιόν] εις δεξιαν O–V L(–125) 797 k 260 Did Ps 309,27 (sed hab Didlem 291,7) Aeth; εκ δεξιων 766 (-ιον 766B) Did Ps.CatA 904 Arab

αὐτοῦ] absc 998; αυτ 540 563; αυτων Sc O–V-411 147-159-299-503-560 443 543 549 766A Bas III 1605 Cyr X 712 Dam GregNy 438,2ap 440,2ap Ol–ΑΕ PsChr Arab SaI Syhmg ↓

ἐν κρίσει] εις κρισιν L C´ 161-248c 359 Antioch 1485 Anton 960 1057 1208 P Dam–M Didcom 337,1.4.7 (sed hab Didlem 336,24) Max II 968 Met X.5,4 OlΑ Amb Exh virg 10,69 Spec 474,2 Hilem et com 349,158 351,70 (sed hab Hi Pach 144,13) Arab Fa1 2 SaI II CPA Syh = Ald

The Ancient Translations

79

C 248´ 101a θανατοῦσαι] θανατωσαι 754 766A; θανουσαι 411 390-415-425-504-522-540601-732 357 161-248c 645 Anast 641 Constit II 17,4 OlΙΚ omne lat. (praeter Hi) Aeth Arab SaII = Compl;

EOd 215f

OLC 43c

E L cII 44c

8.2.3  Agreement with Three Groups (Examples Only) ὅτι—ματαιότης] διοτι αφρων εκ περισσευματος λαλει 534: homoiar (16a); + διοτι (> Aeth; + ο B-613 998 d SaI) αφρων εκ (+ του 613) περισσευματος (+ καρδια Geo) λαλει (+ (sub  Syh) οτι (και ιδου pro οτι Syh) και γε τουτο ματαιοτης O Syh) B-613 998 O d 336´ 542 766B Aeth Geo Arab SaI Syh (sed hab Hilem 269,271s) = Pesh, contra 𝔐

εἶδε(ν) (ειδι 336)] οιδε(ν) V L C´’–609 (425mg) k (46s) 342-357 155 252 311 338 E L a OVMi P c 339 443 539 543 549 602 DamV RM L (sed hab DamV M H A) Ollem–ΓΕΖ Arab = M Ald; ιδεν S 998 698 DamKM T Met III.26,9,66 OllemΕ et comΕΖ

ἀνδρός] pr και 549; pr του O 601 Fa2 SaI; (+ τω 613) ανδρι B-68’ 411 L–125´ 299-cII d k (46s) 155 (ανδρη) 252 296´ 311 336´ 339 443 542 543 547 549 645 698 706 766B 795 Didcom 117,3 Ol–Μ Amb Sat 2,30txt Ruf Om S Bas 4 Arab Arm (absc 998 Didlem 116,23; inc C; sed hab Fa1 2 SaI Syh = Ald Compl Sixt)

O L cII PM 84b ποιήσεις] ποιεις B-68´ 998 390 d 443 698 766 795 Anton 1000 Dam–KM (sed hab Did 235,24) Hi Syh = Compl (non Ald); εποιησας O-411 L(–125) cII 296´ 311 338 547 645 706 Met VII.8,5s Ol Aeth Arab Armte

9.  The Old Church Slavonic Translation 9.1  The Witnesses/Editions of the Old Church Slavonic Translation Lyubov Osinkina, The Textual History of Ecclesiastes in Church Slavonic, Diss. Oxford, 2008. 528 pages. Lyubov Osinkina, Ecclesiastes in the Church Slavonic Textual Tradition, https://dspace.gla. ac.uk/bitstream/1905/28/1/OsinkinaEdit.pdf, Accessed 26 June 2012.

9.2  The Textual Character of the Old Church Slavonic Translation

According to Osinkina, the text of Ecclesiastes is preserved in some 33 semiuncial Cyrillic manuscripts, all of East Slavonic provenance, and all on paper except for one written on parchment. All except four (later) manuscripts come from the XV and XVI Centuries. Osinkina also analysed the catena on Ecclesi-

80

The Textual Witnesses

astes in Church Slavonic, citations of Ecclesiastes in Church Slavonic texts and Ecclesiastes in Croat Glagolitic breviaries. Focusing analysis upon the manuscripts, Osinkina produced a stemma codicum and a semi-critical edition. She concluded that all the manuscripts come from a single archetype. As no critical edition of the Greek text of Ecclesiastes was available to Osinkina, she compared the results of her study with the critical apparatus of the HolmesParsons Edition. She concluded as follows: The number of coincidences between the variants readings in the Church Slavonic text with the readings in Greek manuscripts is not large and is distributed more or less evenly. The coincidences with codex Vaticanus are six, with codex Alexandrinus five, with MS. 23 also five, with MS. 106 four, with MS. 254 also four, with MSS. 157, 253, 298, 299 two. However, no clear pattern emerges from these correspondences (p. 51).

A B 23 (V) = O 253 = O 106 = L 254 = d 157 = cI

5 6 5 2 4 4 2

MS. 298 belongs to C while MS. 299 is an inconsistent member of C. The phrasing used by Osinkina is not absolutely clear to me. If, the “coincidences” entailed significant variations from the text of this edition, the evidence would show that the character of the text is late and mixed. Her data, however, are not further specified. Due to the fact that this daughter version is so late in origin and the evidence is mixed, it was not collated for the edition.

III.  The Indirect Greek Tradition (Patristic Citations) J. Allenbach, et al., eds. Biblia Patristica: Index des Citations et Allusions Bibliques dans la Littérature Patristique, Paris: Éditions du Centre National de la Recherche Scientifique. 1 (1975): Des origines à Clément d’Alexandrie et Tertullien. 2 (1977): Le troisième siècle (Origène excepté). 3 (1980): Origène. 4 (1987): Eusèbe de Césarée, Cyrille de Jérusalem, Épiphane de Salamine. 5 (1991): Basil de Césarée, Grégoire de Nazianze, Grégoire de Nysse, Amphiloque d’Iconium 6 (1995): Hilaire de Poitiers, Ambroise de Milan, Ambrosiaster 7 (2000): Didyme d’Alexandrie.

The Indirect Greek Tradition (Patristic Citations)

81

1.  Commentaries on Ecclesiastes by Greek Church Fathers (chronological) Or

1.1  Origen (Or)

= S. Leanza, L’esegesi di Origene al libro dell’Ecclesiaste, Reggio Calabria, 1975, 10-20; S. Leanza, Ancora sull’esegesi origeniana dell’Ecclesiaste, FS Ardizzoni, 1978, 491-506.

Three manuscripts contain scholia from Origen’s Commentary on Ecclesiastes: 299, 645, und 733. The text of these three manuscripts were collated for the edition. Therefore, the siglum “Or” without further addition signifies only citations in these scholia. Only one of the scholia contains a citation from Ecclesiastes: Eccl 5:1 in Fol 49v-51v of the Procopius Catena of 645 (= 645cat). See under “2. Citations From the Greek Church Fathers” for further citations from Origen. Dion

1.2  Dionysius of Alexandria (Dion)

= Dionysius Alexandrinus (L. Feltoe, Λείψανα. The Letters and Other Remains of Dionysius of Alexandria, Cambridge 1904; W. A. Bienert, Neue Fragmente des Dionysius und des Petrus von Alexandrien aus Cod. Vatop. 236, Klèronomia 5 (1973) 310.

The Letters of Dionysius contain approximately 25 citations from Ecclesiastes comprising texts from about 44 verses. The siglum “Dion” is found a dozen times in the apparatus. The closest congeners appear to be GregNy, Ol, and ancestors of the Catena Group. Most of the variants are grammatical or stylistic—indeed, almost all concern the article: 23c

καρδία] pr η O–253 L 299-cII 357 311 338c 645 698 766 Dion GregNy 312,7 OlΑΙΚΜ

1.3  Gregory Thaumaturgos (GregTh) John Jarick, Gregory Thaumaturgos’ Paraphrase of Ecclesiastes, SCS 29; Atlanta: Scholars Press, 1990.

Gregory Thaumaturgos was a disciple of Origen, so his Paraphrase of Ecclesiastes is one of our earliest sources. Nonetheless, as we have no critical edition and the character of the text is a paraphrase, it is difficult to use as a witness. He is cited in the edition only at 413b and 14a using the siglum GregTh.

82

The Textual Witnesses

1.4  Evagrius of Pontus (Evag)

Evag-E = Géhin, Paul. Évagre le Pontique: Scholies a l’Ecclésiaste. Sources Chrétiennes 397. Paris, 1993. Evag-P = Géhin, Paul. Évagre le Pontique: Scholies aux Proverbes. Sources Chrétiennes 340. Paris, 1987.

Two instances of Evag may be found in the apparatus, in both cases on the left side of the square bracket in 314d. The only other direct citation, Evag-E 68, cites 725a exactly as the text of the edition. This evidence is insufficient for conclusions although Géhin supplies the text of A as lemma since he argues that the citations of Evagrius are closest to A. 314d

αὐτοῦ 1° B-68*-534 Evag-P 285 OlM = Gra Ra 𝔐 ↓ ] absc 998; αυτων rel

Did

= Didymus Alexandrinus, Kommentar zum Ecclesiastes, (Tura-Papyrus) Teil I. 1 (Kap. 1,1-2,14) Ed. G. Binder u. L. Liesenborghs; Teil II (Kap. 3-4,12) Ed. M. Gronewald; Teil III (Kap. 5-6) Ed. J. Kramer; Teil IV (Kap. 7,1-8,8) Ed. J. Kramer u. B. Krebber; Teil V (Kap. 9,8-10,20) Ed. M. Gronewald; Teil VI (Kap. 11-12,6) Ed. G. Binder u. L. Liesenborghs: PTA 25, 22, 13, 16, 24, 9. Cited by papyrus folio and line number.

1.5  Didymus the Blind (Did)

Didymus the Blind (313-398) was Dean of the Theological School in Alexandria. The Tura Papyri (VI-VII Cent.) preserve his commentaries on Ecclesiastes, Zechariah, Genesis 1-17 and Psalms 20-46. As one might expect, this patristic writer is an important witness to the early Egyptian Text. Examples: 114a

σὺν πάντα B-S 998 V Did 25,9 Armap Gra Ra] παντα DamK Fa1 SaI 5 Syh; συμπαντα rel (Dam Met I.16,2 GregNy 303,3te Ol)

310a

σύν] συμ 106; συμπαντα 998 411 130 cII d 443 534´ 728 Geo Arab; παντα 161mg-248mg Antioch 1713 (sed hab Compl); + παντα B-68 V 336 571c Did 82,20 SaI II

1.6  Gregory of Nyssa (GregNy)

GregNy = Gregorius Nyssenus V (W. Jaeger, Gregorii Nysseni Opera, Leiden 1958 ff.)

In Ecclesiasten Homiliae by Gregory of Nyssa covers only the biblical text of 11313. GregNy agrees frequently with Ol and shows how early many scribal errors of an inner-Greek nature as well as grammatical and stylistic improvements oc-

The Indirect Greek Tradition (Patristic Citations)

83

curred in the textual tradition. These errors and improvements are often shared with O, L, and C. Examples: 21a

καρδίᾳ] pr τη O–V 139mg-147-159-503-560-798 GregNy 309,19 Co

22a

εἶπα] ειπον O 147-159-299-425c2 68*(vid) 547 698 (‑πων) cIIcat Dion 214 GregNy 310,16te Ol Dam

23c

καρδία] pr η O–253 L 299-cII 357 311 338c 645 698 766 Dion GregNy 312,7 OlΑΙΚΜ

215c

μοι] + ※ θ΄ 788; > Sc O-411 L C´’ 754 338 547 613 698 705 766B 795 Hi Geo (sed hab Syh) GregNy 361,17 364,13 Metlem et com II.6,4,35,95 OlΖ = Ald, contra 𝔐

1.7  Pseudo-Chrysostom (PsChr)

PsChr = Pseudo-Chrysostomus, Commentarius in Ecclesiasten: CCSG 4

PsChr is an early patristic witness. It seems to have affinities with the early Egyptian Group: 32b φυτεῦσαι] φυτευειν 998 357 PsChr 416c ἐν] επ B-68´’ C 998 L–(125) 336´ 443 542 766 PsChr Fa1 SaI 57d om ὅτι B-68´ 998 PsChr Geo (sed hab Hi Syh = Ald Sixt 𝔐) 714c τούτῳ] τουτο B-68’ O 357 336´ 339 PsChr Arm Fa1 2 SaI II 4 6 (absc 998) 119c καρδίας σου] (+ σου SaI II) ἄμωμος B-68´ 998 336´ Did 333,4s 335,26s PsChr Fa1 2 SaI II

Following publication of the edition, I discovered that Guillaume Bady demonstrated that the edition of Pseudo-Chrysostom’s Commentary on Ecclesiastes edited by Alexander Leanza for CCSG could be genuinely from John Chrysostom.12 In his review of my Göttingen Edition, Bady published readings from the manuscript not noted in the apparatus of Leanza’s edition that may alter the textual analysis given here.13

12  Guillaume Bady, “Questions sur l’Authenticité du Commentaire Pseudo-Chrysostomien sur l’Ecclésiaste,” Studia Ephemeridis Augustianianum 93, Institutum Patristicum Augustianianum, Rome (2005): 463-475. 13  Guillaume Bady, “Parution de l’édition critique de l’Ecclésiaste de la Septante à Göttingen,” https://biblindex.hypotheses.org/2479, accessed September 17, 2021.

84

The Textual Witnesses

1.8  Olympiodorus of Alexandria (Ol)

Commentarii in Ecclesiasten: PG 93, 477-628 = Fronton du Duc (1624) e codd. Paris gr. 174, 175. New Edition: T. Boli, Olympiodor, Diakon von Alexandreia, Kommentar zum Ekklesiastes: Eine kritische Edition. Diss. Universität Heidelberg, 2004. Robert Sinkewicz, Manuscript Listings for the Authors of the Patristic and Byzantine Periods. Greek Index Project Series 4, Pontifical Institute of Mediaeval Studies, Toronto 1992.

Early on, in research on the patristic witnesses to Ecclesiastes, Ziegler (and I in turn) discovered that the edition in Migne, PG 93, is based on the two worst manuscripts among all known sources of this patristic text. Ziegler began, and I continued to locate and collate additional manuscripts of this early sixth century work which is more complete than either Did or GregNy, and only matched in completeness by the later work of Met. Some manuscripts were collated by the Septuaginta-Unternehmen. About half-way through this work, I learned of the new critical edition of Boli from Professor Dieter Hagedorn. The book by Sinkewicz gives a list of some 21 MSS. of the Ecclesiastes Commentary of Olympiodorus. Boli’s edition of the Ecclesiastes Commentary of Olympiodorus is based on the following MSS., which were numbered by her using the letters of the Greek alphabet (Rahlfs numbers where available are given in parentheses): Α Β Γ Δ

Venice, Bibl. Marc., Gr. 22; XII.-XIII. Cent. The end from 122 is missing. (733) Paris, Bibl. Nat., Gr. 174; X.-XI. Cent. (573) Paris, Bibl. Nat., Gr. 153; XI.-XII. Cent. (562) Florence, Bibl. Laur., Plut. X 29; XIII. Cent. (386). 62e οὐκ-612e ὀπίσω is missing. Ε Genoa, Durazzo—Giustiniani A.I.10; IX.-X. Cent. Ζ Venice, Bibl. Marc., Gr. 23; X.-XI. Cent. (734) At the end of the MS., Fol. 101b ff., there is a Florilegium. It contains citations from profane authors, from Church Fathers und a number of passages from the Bible, mostly out of Proverbs. Ecclesiastes appears three times: 76ab on Fol. 121a, 81bc on Fol. 116b, 1014 on Fol. 116b. The siglum employed for these three is OlΖflor. Η Paris, Bibl. Nat., Gr. 175; XIV. Cent. Θ Rom, Bibl. Vat., Ottob. Gr. 75; XVI.-XVII. Cent. Ι Athos, Λαύρα 278 (Γ 38); XIV. Cent. s Κ Athos, ᾽Ιβήρων 614 (4734); XIII. Cent. 91-93b πᾶσιν = OlΚ

According to Boli, MS. Θ is a copy of Β and therefore is not cited here. The following late MSS. were not collated, because no significant textual value or variants can be expected from them:

The Indirect Greek Tradition (Patristic Citations)

85

Athens, M. Loberdou, gr. 21; XIX. Cent. (342?) Athos, Iberon, gr. 697; 1533. Berlin, Staatsbibl., Phillipps 1411; XVI. Cent. Paris, Bibl. Nat., Suppl. gr. 1275; XV. Cent. (see Boli, p. XI) Rom, Bibl. Vallicelliana, gr. 51; XVII. Cent. We confirmed what Boli suspected, that the following manuscript was burned and lost in the fire of 1671: Escorial, Real Bibl., gr. Θ.IV.24. The following manuscripts were unavailable (impossible to obtain): Kharkov, Bibl. Kharkov. U., gr. 2, XII. Cent. Meteora, M. Metam., gr. 593, XIV. Cent. St. Petersburg, GPB Acad.Sp. 337, ?. Cent. Two manuscripts are listed by Sinkewicz but the description of contents he provides is incorrect: 299 342

Rom, Bibl. Vat., Vat. gr. 1694; Written in 1202. (Cat.). Contents: Kap. 1-7. Athens, Historical and Ethnological Society Museum, 200; XIII. Cent. Missing 729 πολλούς—end. (Cat.).

The Catena excerpts of MS. 342 are described fully by Labate. These excerpts come from Didymus, Gregory Agrigentius (= Metrophanes), Gregory of Nyssa and Olympiodorus. See A. Labate, ‘Nuove catene esegetiche sull’Ecclesiaste’ in ΑΝΤΙΔΩΡΟΝ: Hommage à Maurits Geerard Bd. I, Wetteren 1984, pp. 252-257. MS. Sinai gr. 311 is a copy of 299 and is also described in full by Labate. Both manuscripts are identical and have a catena partly from Olympiodorus and partly from Polychronius. Thus, this is a mixed catena. See A. Labate, ‘Nuove catene esegetiche sull’Ecclesiaste’ in ΑΝΤΙΔΩΡΟΝ: Hommage à Maurits Geerard Bd. I, Wetteren 1984, pp. 257-261. Manuscripts which Boli did not acquire but we did and have collated: Μ Ν

Moscow, State Hist. Museum (GIM), Syn. Bibl., Gr. 15; 1023. Rom, Coll. gr. 7; XV. Cent. Contents: 15-25 211-57 Cat. Olymp. in Eccl. 58-fin libri alia Cat.

Boli provides a stemma and a critical text. Nevertheless, numerous mistakes were found in her collation of the manuscripts. Normally we follow the text of Boli

86

The Textual Witnesses

and use the siglum Ol in the edition for the biblical text of the Commentary of Olympiodorus and also for citations of the biblical text within the Commentary when they exist: Ol = Ollem + (Olcom) When the text chosen by Boli is based on incorrect collations, we cite the manuscripts as superscripts. We do not usually cite Ν in the edition, because this manuscript is late and offers nothing of significant textual value. By contrast, Μ is one of the earliest witnesses to Ol. When the witness of Μ is contrary to the text of Boli and offers significant variants, we cite it as OlΜ along with any other congeners. The rest of the witnesses would then be Ol–Μ. Occasionally it is necessary to cite MSS. from Boli’s apparatus. This is usually indicated by Olap as opposed to Olte, although in special situations the manuscripts are cited individually, only when the reading is a variant to the lemma of her edition. The biblical text of the Procopius Edition (Catena in Ecclesiasten, CCSG 4) is also the biblical text of the Commentary of Olympiodorus. Nonetheless 12 is not in Olympiodorus’ Commentary. MS. 733 has this verse in the margin and it stems from the other scribe who copied the Procopius Catena in the margin of MS. 733 which bears the text of the Olympiodorus Commentary. Ol represents an extremely early strand of the Catena tradition deriving from and related to the Egyptian Text and d Group. Examples: 28d om καί 6° B-534 998 d Ol Geo Fa1 SaI (sed hab Hi Arm Syh Ruf Cant 94,25 = 𝔐) c 215e om τότε B-S*-534´ 998 C´–571 443 547txt 645 705 Dionlem 220 GregNy 361,17 Hilem 269,271 Geo Arm SaI (sed hab HiLXX 269,285-289 Syh) Ol, contra 𝔐

Met

1.9  Metrophanes of Smyrna (Met)

= Metrophanes Smyrnensis (Gerard H. Ettlinger u. Jacques Noret, PseudoGregorii Agrigentini seu Pseudo-Gregorii Nysseni Commentarius in Ecclesiasten, Turnhout, 2007 [CCSG 56] Georgian Translation is XIII Cent.

Peter Van Deun, La Chasse aux Trésors, La Découverte de Plusieurs Œuvres Inconnus de Métrophane de Smyrne (IXe-Xe Siècle): Byzantion 78 (2008) 346-367.

This commentary, formerly attributed to Gregorius Agrigentinus, has been recently shown to be authored by Metrophanes of Smyrna who flourished in the latter half of the ninth century and beginning of the tenth century. The critical edition by Gerard H. Ettlinger and Jacques Noret is excellent and no variants from their apparatus need to be cited.

The Indirect Greek Tradition (Patristic Citations)

87

Met is from a part of the textual tradition similar to that of Did. Examples: 321a 411a

πνεῦμα 1° B-S*-68´ 998 L 571c d 260* 443 Didlem et com 103,22 104,14 Metlem et com III.23,2,45 PsChr Fa1 = 𝔐] pr τό rel (Metcom III.23,56 X.8,101) δύο B-68´’ C 998 d–357 338 Metlem et com IV.1,5,63 Syh = Ra] pr οι rel (539 Didlem 124,18 127,2 Didcom 127,3.6.12): ex (9a) (12b)

2.  The Greek Church Fathers (alphabetical) Here only brief information is supplied. Further information on Primary Editions and Secondary Literature are given in CC by M. Geerard, Clavis Patrum Graecorum I-V + Supplementum, Turnhout: Brepols 1983, 1974, 2003, 1980, 1987, 1998, 2018 (= Clavis I, Clavis II, Clavis III,—Clavis-Supp). Ammon

= The Letter of Ammon and Pachomian Monasticism: PTS 27

Anast

= Anastasius Sinaita: PG 89

Antioch

= Antiochus Monachus: PG 89, 1419-1856

Anton

= Antonius Melissa, Sententiae sive Loci communes: PG 136, 765-1244

AntonM

= Antonius Magnus Abbas, Epistulae 7: PG 40,977-1000

Ath

= Athanasius I-IV: PG 25-28

Bas = Basil Magnus Caesariensis I-IV: PG 29-32 Creat = Homiliae de creatione hominis [CPG 3215-16] SC 160 Ep = Epistulae (Y. Courtonne, Sainte Basile. Lettres I, II, III. Paris 1957, 1961, 1966) Spir = De spiritu sancto [CPG 2900] SC 17 Chr = Chrysostomus I-XVIII: PG 47-64 Jb.Cat = Die älteren griechischen Katenen zum Buch Hiob, PTS 40, 48, 53, 59. Cited by Chapter and Number. Clem

= Clemens Alexandrinus I-III: GCS 12, 15, 17

Clim

= Iohannes Climacus, Scala Paradisi: PG 88,631-1164

Constit

= Constitutiones Apostolorum: SC 320, 329, 336

88

The Textual Witnesses

Cos

= Cosmas Indicopleustes, Topographia Christiana: PG 88; SC 141, 159, 197

Cyr

= Cyrillus Alexandrinus I-X: PG 68-77

CyrHier

= Cyrillus Hierosolymitanus: SC 126

Dam

= Iohannes Damascenus, Sacra Parallela (O. Wahl, Der Proverbien- und Kohelet-Text der Sacra Parallela, Forschung zur Bibel 51, Würzburg, 1985) Nativ M = Sermo in nativitatem beatae Mariae virginis: PTS 29

Did Gen Hiob

= Didymus Alexandrinus = Commentarii in Genesim: SC 233, 244. Cited by section number. = Kommentar zu Hiob (Tura-Papyrus) Teil I (Kap. 1-4) Ed. A. Henrichs; Teil II (Kap. 5,1-6,29) Ed. A. Henrichs; Teil III (Kap. 7,20c-11) Ed. U. Hagedorn, D. Hagedorn, u. L. Koenen; Teil IV.1 (Kap. 12,1-16,8a) Ed. U. Hagedorn, D. Hagedorn, u. L. Koenen, PTA 1, 2, 3, 33.1. Cited by papyrus folio and line number. Ps = Psalmenkommentar (Tura-Papyrus) Teil I (Ps 20-21) Ed. L. Doutreleau, A. Gesché, u. M. Gronewald; Teil II (Ps 22-26,10) Ed. M. Gronewald; Teil III; (Ps 29-34) Ed. M. Gronewald; Teil IV (Ps 35-39) Ed. M. Gronewald; Teil V (Ps 40-44,4) Ed. M. Gronewald, PTA 7, 4, 8, 6, 12. Cited by papyrus folio and line number. Ps.CatA = Fragmenta e catenis in Psalmos: E. Mühlenberg, Psalmenkommentare aus der Katenenüberlieferung, I PTS 15, II PTS 16, Berlin-New York, 1974, 1977). Cited by scholion number. Trin = De Trinitate: PG 39 Za = Kommentar zu Sacharja (Tura-Papyrus) (SC 83-85; Ed. L. Doutreleau, (1962). Cited by papyrus folio and line number.

Epiph

= Epiphanius I-III: GCS 25, 31, 37

Eus Gen Ps

= Eusebius Caesariensis IV VIII 1,2: GCS 14, 43,1.2 = Generalis elementaria introductio [CPG 3475; PG 22] = Commentarii in Psalmos: PG 23; 24,9-76; 30,81-104 (See also C. Curti, Eusebiana I: Commentarii in Psalmos, Università di Catania, 1989).

Eustat

= Eustathius Antiochenus Clavis II.3367 (M. Spanneut, Recherches sur les écrits d’Eustathe d’Antioche avec une édition nouvelle des fragments dogmatiques et exégétiques, Lille, 1948) = PG 18, 613-704



The Indirect Greek Tradition (Patristic Citations)

GregAc

89

= Gregorius Acindynus: CCSG 31

GregNa = Gregorius Nazianzenus I-IV: PG 35-38 Briefe = Epistulae: GCS 53, Ed. P. Gallay Disc = Orationes xlv: SC 247, 250, 284, 318 [CPG 3010/27] GregNy Virg

= Gregorius Nyssenus I-X (W. Jaeger, Gregorii Nysseni Opera, Leiden 1958 ff.) = De virginitate [CPG 3165] SC 119

GregTh

= Gregorius Thaumaturgos, Metaphrasis in Ecclesiasten (J. Jarick, Gregory Thaumaturgos’ Paraphrase of Ecclesiastes, Atlanta 1990)

Herm

= Pastor Hermae (Papiasfragmente, Hirt des Hermas, Ed. Ulrich H. J. Körtner und Martin Leutzsch, Schriften des Urchristentums 3)

Hipp

= Hippolytus: GCS 1 [Cf. SC 14]

Is

= Isaias Gazaeus: PG 40, 1105-1206

Isid

= Isidorus Pelusiota: PG 78

Max

= Maximus Confessor: PG 90-91

Mel = Meletius Antiochenus Hom = Homilia in Prov. 13:22 apud Epiphanium Haeres. 73, 29-33 (GCS 37,303-308) Nil Ol

= Nilus Ancyranus (PG 79)

Hiob

= Olympiodorus, Commentarii in Ecclesiasten (Boli) = Kommentar zu Hiob: PTS 24

Cant Her

= Origenes I-IV, V, VIII, IX, X, XII, XIII: GCS 2, 3, 6, 10, 22, 33, 49 (35), 40, 41.1, 41.2, 41.3, N.F. 19; SC 132, 136, 147, 150 = GCS 2, 3; P. Glaue, Ein Bruchstück des Origenes über Genesis 1,28 [P. Bibl. univ. Giss. 17]: Mitteilungen aus der Papyrussamlung der Giessener Universitätsbibliothek 11 (Schriften der Hessischen Hochschulen), Universität Giessen 1 (1928). = Libri X in Canticum cantorum: SC 375, 376 = Dialogue with Heraclides: SC 67

Or

90

The Textual Witnesses

Jb.Cat = Die älteren griechischen Katenen zum Buch Hiob, Ed. U. Hagedorn und D. Hagedorn, PTS 40, 48, 53, 59. Cited by Chapter and Number. John = Origenes IV = SC 120, 157, 22, 290, 385. Phil = Philocalia, 1-27: SC 302, 226 PetA

= Petrus Alexandrinus (PG 18, 467-522)

PhilC

= Philo Carpasianus (PG 40, 28-153)

Proc mg CatP Gen Cant Isa Schol

= = = = = = =

PsCaes

= Pseudo-Caesarius: PG 38

Procopius Gazaeus Eccl 12 on margin of 733 by the scribe of CatP Catena in Ecclesiasten (CCSG 4) Commentarii in Genesim (PG 87) Commentarii in Cantica Canticorum (PG 87) Commentarii in Isaiam (PG 87) Scholia von Procopius Gazaeus: CCSG 4, Supplementum

PsGregNy = Auctorum incertorum (H. Hörner, Gregorii Nysseni Opera Supplementum, Leiden, 1972) Ser

= Serapion Thmuitanus (PG 40)

Sev

= Severianus Gabalensis (PG 56, 429-516)

Soz

= Sozomenus: GCS 50

Syn

= Synopsis Scripturae Sacrae, Liber Ecclesiastes: PG 28, 347-350

Theog

= Theognosti Thesaurus: CCSG 5

Tht

= Theodoretus Cyrensis I-V: PG 80-84

Vic-P

= Victorinus Poetovionensis: Hicom 290,218ss

IV.  Printed Editions (chronological) Of special importance are the three earliest printed editions: the Aldina (Ald), Complutensis (Compl) and Sixtina (Sixt). These were collated in full for the current edition. The guidelines for the citation of these three printed editions are

Printed Editions (chronological)

91

detailed below. Of the three, the Sixtina was reprinted multiple times; for this edition only the original printing was collated and used.

1.  Aldina (Ald) Ald was edited in the year 1518 and intended to present a critical text based upon a collation of different manuscripts. For Ecclesiastes, however, the text is based for the most part on MS. 68. Therefore, Ald is only cited if its text deviates from 68.

2.  Complutensis (Compl) Séamus O’Connell, From Most Ancient Sources, The Nature and Text Critical Use of the Greek Old Testament Text of the Complutensian Polyglot Bible, Academic Press: Fribourg, Vandenhoeck & Ruprecht: Göttingen, 2006.

MS. 248, which was loaned to Cardinal Ximenes from the Vatican (see Rahlfs, Verzeichnis, p. 250), forms the base of Compl. The editor of Ecclesiastes, however, did not simply reprint 248. Corrections were made largely from (a) manuscript(s) in the C´ Group. Therefore, Compl is only cited if its text deviates from 248. O’Connell demonstrates that the editor of Compl made judgements based upon the marginal notes in 248. Nonetheless, O’Connell’s analysis needs to be checked (p. 141). For example, the καί following περιλαβεῖν in 3:5b is only omitted by 248* (in the entire textual tradition). So here Compl adds it on the basis of 248c. It may be true that Compl was guided by the Hebrew, but O’Connell missed the corrrector’s note in his collation and analysis.

3.  Sixtina (Sixt) Of all the three early printed editions, Sixt shows in Ecclesiastes the greatest dependency upon a single manuscript. It presents as closely as possible the text of Codex B. When B is corrected, then it follows either B* or Bc. Therefore Sixt is noted in the apparatus only if its text deviates from B and the text of B has no corrector.

4.  Grabe (Gra) Septuaginta Interpretum tomus ultimus, ed. J. E. Grabe, Oxonii, 1709.

Grabe’s edition presents an eclectic text using Cod. A as a foundation. This edition was not collated for the apparatus; merely a few conjectures of Grabe were noted.

92

The Textual Witnesses

5.  Holmes-Parsons (HP) Vetus Testamentum Graecum cum variis lectionibus, ed. R. Holmes et J. Parsons V, Oxonii, 1823.

Volume V of this massive work was published in 1823 with Holmes as editor. It offered a complete collation of all available manuscripts, all ancient daughter translations as well as the patristic sources. The printed text to which this comprehensive material was collated by a multitude of scholars was that of the Sixtina. The value of this work is that a few manuscripts, which are now lost, were collated for the current edition from H.-P. A few readings of MS. 298, a manuscript originally conserved in the Crimea but is now missing, were noted in the apparatus. H.-P. had used MS. 159 but I was able to obtain photographs of this MS. from Moscow with the help of Yuri Zozulya. This MS., however, was in Dresden in World War II and was partially damaged during the bombing of Dresden. We have used the collation from H.-P. for the damaged portions. See Peter J. Gentry and Felix Albrecht, The Amazing History of MS. Rahlfs 159—Insights from Editing LXX Ecclesiastes, Journal of Septuagint and Cognate Studies, 44 (2011): 31-50.

6.  Rahlfs (Ra) This “Handausgabe” contains the entire Greek Old Testament. The text is based essentially on a comparison of the three main uncials A B und S. Rahlfs employed the Syro-Hexapla as his witness to O and also regarded the Bible text of Jerome’s Commentary on Ecclesiastes as essentially equivalent pure and simple to the Old Latin. There are a number of places where Rahlfs erroneously chose a reading against the entire Greek manuscript tradition on the basis of Jerome’s text because he was unaware that Jerome had corrected the Old Latin spontaneously towards the Hebrew.

B.  The Text History: Text Groups and Textual Relations of the Greek Witnesses (A.I), the Ancient Translations (A.II) and the Indirect Tradition (A.III) I.  The Hexaplaric Recension 1.  The Witnesses of the Hexaplaric Recension O = V-253-475-637 Hi Syh (Sc 411) Among notes left behind by Joseph Ziegler in his work on Ecclesiastes was a detailed analysis of the O Group. Nonetheless, after I had established the critical text and had already begun to write the Einleitung, another manuscript belonging to this extremely important text group was acquired and collated (Ra 475, XIII-XIV Cent.). This was a valuable find. Thus the entire study had to be completely reworked just before publication. The main members of the O (Origenic) group are V-253-475-637. The group is not always unified in attesting the reading of the hexaplaric text (see discussion infra). Nonetheless, variants from O or a part thereof occur in approximately 320 instances (against the critical text). Hi is clearly related to the O Group. There are 60 agreements with O, 18 of which are singular agreements. The main problem is the extent to which Hi really represents the Old Latin or the extent instead to which Jerome freely and spontaneously corrected his Old Latin source text as he produced the lemma for his Commentary. The Syro-Hexapla (Syh) also transmits a hexaplaric text, but is a distant member of the group, probably because this text derives from the Tetrapla, a later development and derivative of the Hexapla. Syh agrees with at least one member of O in 107 instances, and in 61 of these instances, O is unified. See above the section on Syh for the characterisation of this witness. Sc agrees with a member of the O Group in approximately 50 cases and against the O Group in 11; approximately eight of these are singular agreements. Clearly, Sc was using a hexaplaric text to correct S.

2.  The Critical (Aristarchean) Signs Employed by Origen The critical (Aristarchean) signs employed by Origen in the Hexapla or more accurately in the Tetrapla, have been best preserved by the Syro-Hexapla, in spite of the fact that the text of Syh is not a close member of the O group. Apart from Syh, the critical signs are preserved in Greek witnesses by only two manuscripts:

94

The Text History

V and 788. V has only asteriskoi and only twice. MS. 788 has asteriskoi in three instances–naming the source in two–and obeloi in two instances. By contrast, Syh has asteriskoi in ten instances and in seven of these names the source. Twice words are marked with an obelos, and in no less than four instances by a sign we are a calling an antisigma or lemniskos. All instances of the critical signs in all witnesses are discussed fully in the section on the Syro-Hexapla (cf. II.2).

3.  The Character of the Hexaplaric Recension The members of the O Group often diverge in their witness. First we may consider instances where the group is unified: 23d 413c 417a 511a

516b 65b

67b 711b 716b 93e 113b 116c

κρατῆσαι] κρατειν O-Sc (-τιν) προσέχειν] φυλαξασθαι O-411 ἐν ᾧ ἐὰν πορεύῃ] εν τω πορευεσθαι O-Sc-411 τοῦ δούλου (servi Amb Nab 6,29 Syhtxt) = hā‛ebed] του δουλευοντος O Hi (operanti) (σʹ θʹ ὁμοίως τοῖς οʹ Syhmg = τοῦ δουλεύοντος) = 𝔐 (hā‛ōbēd) Vulg θυμῷ] παροξυσμω O Hi Dam (sed hab iracundia Amb): ex Deut 2928(27) Ier 39(32)37 ἀνάπαυσις (-σεις, -σιν) τούτῳ (τουτο) ὑπὲρ τοῦτον ] + ουδε (και ουκ d–357) επειραθη διαφορας ετερου (εταιρου 754) πραγματος προς ετερον O d–357 547mg = σʹ πληρωθήσεται] εμπλησθησεται O-411 539; πλησθησεται 390 613 Dam OlΑ θεωροῦσι(ν)] ορωσι(ν) O-411 613 Didcom 208,17 (sed hab Didlem et com 208,2 208,15, 208,17) ἐκπλαγῇς] εκκλινης O (‑κλεινης V) Antioch 1576 SaI 3 (sed hab Fa2) νεκρούς] τεθνεωτας O-411: cf 95b οἱ νεκροί] οι τεθνεωτες O–V-411 ἐκχέουσιν] εκχεει (χεει V) O–V-411 στοιχήσει] ευθετει O–475 Hi; ευθυ 475; ευθετησει L = αʹ

See also 36a 314a 318a 322c 44c 48d 58a 512b 68a 72a 72a 722b 723b 726d 91f 94b 94b 95a 98a 98b 104a. Some of these variants entail correction of the text towards alignment with the Hebrew using readings from the Jewish revisions of the LXX, but others are simply grammatical and stylistic improvements to the text. The following chart gives an approximate idea of the instances where the witness to the hexaplaric reading is carried by only three or even by just two members of the group:

The Hexaplaric Recension

O–V (253-475-637)

O–235 (V-475-637) O–475 (V-253-637) O–637 (V-253-475) O–V 253 (475-637) O–V 475 (253-637) O–V 637 (253-475) O–253 475 (V-637) O–475 637 (V-253) O–253 637 (V-475)

68×

95

114a 115a 21a 23d 23d 27b 210a 220b 226b 36b 38a 38b 311a 311c 317d 44c 45a 45b 412a 413c 415c 417d 53b 56a 57a 510a 510b 510b 517e 62c 63b 63b 610c 611b 714a 718b 726a 82a 82b 85a 811c 812b 812b 815a 816a 816d 91a 91d 94a 95b 910d 912c 912d 914b 918b 101a 102a 102b 103a 108a 1010b 1011b 1012a 1014c 1020c 113a 116b 118a 18d 110a 116a 23b 212d 221a 224d 310b 48c 416a 57d 711b 713b 95d 105b 1010a 122a 127a 1213c 212c 67a 91f 1019c 116c 110a 111c 214a 222c 224b 311a 319a 44c 417c 51d 57d 510c 518a 65b 715a 810b 811d 815e 97a 99b 914c 1020b 121c 122c 125e 125f 129b 129b 1210c 1211b 117d 118c 120c 213e 319d 320c 416a 56a 510c 65b 67a 817e 72b 81b 1110a 723a 512b 64a 72a 721c 121a 121d 1213a 17c 28d 213a 817e

In the following instances the hexaplaric reading is represented only by 253-475637 ( = O–V): 27b 210a 36b 38a 38b 45a 412a 413c 415c 53b 56a 57a 510a 510b 62c 63b 63b 718b 726a

οἰκογενεῖς / ἐγένοντό μοι] tr O–V ᾔτησαν] επεθυμησαν O–V 539 (sed hab Hi) = σʹ Vulg φυλάξαι … ἐκβαλεῖν] tr O–V φιλῆσαι … μισῆσαι] tr O–V DamR OlΑ πολέμου … εἰρήνης] tr O–V 613 GregNy 429,1ap A-SS Helia περιέλαβε(ν)] περιλαμβανει O–V ἐπικραταιωθῇ] υπερισχυσει O–V = σʹ ἔτι] πολιν 637; + πολην 253: cf 1015b νεανίσκου] παιδος O–V = σʹ μὴ χρονίσῃς] μη βραδυνης O–V λόγων πολλῶν] + πολλα ρηματα 253 157c; + πολλα αρνηματα 475-637 ἐν χώρᾳ] εν πολει O–V Syhmg = σʹ ἐν πλήθει ἀγαθωσύνης] εν τω πληθυνθηναι (πληθηναι 253) την αγαθωσυνην (-θο- O–V 613) O–V 613 τῷ παρ᾽ αὐτῆς] τω εχοντι αυτην O–V 766 Hi Fa1 = σʹ fin] + και ουκ αναλισκει εις εαυτον O–V = σʹ ὅ τι ἔσονται] αι εσομεναι (ἔσονται 411) O–V-411 ἐτῶν] ζωης O–V-411 (pr της 475) Geo μὴ ανῇς] μη αφης O–V Did 217,19 Hi (dimittas) = αʹ σʹ ὑπὲρ θάνατον] υπερ τον θανατον O–V SaI Sa3 Anast 684 Constit I 8,20

96 82ab

811c 91d 94a

95b 912c 101a 103a 1011b 1020c

The Text History

στόμα βασιλέως φύλαξον καὶ περὶ λόγου ὅρκου θεοῦ μὴ σπουδάσῃς] εγω παραινω ρησιν βασιλεως φυλασσειν και παραβηναι ορκον θεου μη σπευσης (mend πιστευσης 475-637) O–V: cf Hi = σʹ ἐπληροφορήθη] ετολμησεν O–V = αʹ ἐν χειρί] ενωπιον O–V 443 542 Syhtxt Metlem et com VIII.1,17,96 ὅτι τίς ὃς κοινωνεῖ πρὸς πάντας τοὺς ζῶντας] οτι (> 253) τις γαρ εις (> 475) αει (> 475; α 637) διατελεσει ζων O–V; + τις γαρ εις αει διατελει ζων (τις—αει sup ras) 798 οἱ νεκροί] οι τεθνεωτες O–V-411 θηρευόμενα] συλλαμβανομενα (συνλαμ. 637 359) O–V-411 359 Hi = σʹ μυῖαι θανατοῦσαι σαπριοῦσιν σκευασίαν ἐλαίου ἡδύσματος] θανατος μυιων (μυιων θανατος 475) σηψει ελαιον μυρεψου ευωδες O–V = σʹ ὅταν ἄφρων πορεύηται] οταν πορευηται αφρων O–V-411 περισσεία] οφελος O–V Aethte (sed hab Aethap(A2)) = σʹ σὺν τὴν φωνήν = Ra] την φωνην S* O–V 645 Didlem2 315,9 Hilem et com 343,351 343,357

Instances of O–235, O–475, and O–637, all together (55), are less than the number of instances of just O–V (68). Some examples: 23b καρδία] pr η O–253 L 299-cII 357 alii 105b om ὃ B-S-68´ C O–253-411 L C´’–260 357 alii 212c 116c

ἐπελεύσεται 998] απελευσεται Sc; εισελευσεται O–475 = αʹ στοιχήσει] ευθετει O–475 Hi = αʹ

222c 122c 125e

ᾧ] ως O–637 C pau alii ἐπιστρέψωσιν τὰ νέφη] επιστρεψη τὰ νέφη V; επιστρεψει τὰ νέφη 253-475 ὅτι ἐπορεύθη] εαν πορευθη O–637

Finally, in nine places the hexaplaric reading is represented by only two witnesses. Examples: 17c 512b 72a 72b 721c 723a 81b 817e

εἰς τόπον « ‫ל־מקֹום‬ ְ ‫]א‬ ֶ προς τοπον V-475 = ad locum Hi Vulg αὐτοῦ 2°] αυτω B-68 O–475 637 539 alii πορευθῆναι] pr του 475-637 539 = 𝔐 εἰς οἶκον « ‫ל־ּבית‬ ֵ ‫]א‬ ֶ προς οικον 253-475; = ad domum La Vulg τοῦ δούλου σου = 𝔐] τους (του V) δουλους σου V-253 OlΔΙΚΜ ἐν τῇ σοφίᾳ] om τῇ B-68 O–253 475 766B (sed hab Ald) πρόσωπον αὐτοῦ = 𝔐] προ προσωπου αυτου 253-475 ὅσα] διοπερ O–V 475 = σʹ

The Hexaplaric Recension

1110a 121d

97

θυμόν] παροργισμον 253-475 Antioch 1505 φθάσωσιν ἔτη] φθασει ἔτη 253; φθαση ἔτη V-411

It is extremely seldom that the Hebrew text plays no role in the reading or witness of O and O–V. The text in the LXX normally follows 𝔐 precisely. In O–V, however, the text is sometimes transposed, see 27b 36b 8a 8b 103a. In variants involving single words, renderings by σʹ and αʹ were adopted: O σʹ 65b αʹ 116c; O–V σʹ 210a 412a 15c 510b 62c 82ab 101a 11b; O–V αʹ 811c; O–V αʹ σʹ 718b. Also, when no attribution is given in the textual tradition, one may still correctly identify the source: 53b 〈αʹ 〉 μὴ βραδύνῃς « ‫ ַאל ְּת ַא ֵחר‬cf. αʹ βραδύνειν ‫ אחר‬pi. Ps 126(127),2 Hab 2,3 see Reider-Turner p. 44. 56a 〈αʹ〉 πολλὰ ῥήματα « ‫ּד ָב ִרים ַה ְר ֵּבה‬.ְ αʹ always renders ‫ ָּד ָבר‬with ῥῆμα, see Turner p. 210. 912c 〈σ ʹ〉 συλλαμβανόμενα « ‫א ֻחזֹות‬, ֲ cf. Cant 215 πιάσατε « ‫]א ֱחזּו‬ ֽ ֶ σʹ συλλάβετε.

The following readings in O–V are also contrary to 𝔐: 413c πολιν-πολην (unintelligible) 57a εν πολει (πόλις is never used for ‫)מ ִדינָ ה‬ ְ 63b ζωης 91d ενωπιον.

4.  Singular Readings in the Four MSS. V 253 475 637 V 16c κυκλοῖ 1°] κυκλω V 116e πολλά] + καιγε V 21b ἰδέ] ιδον εγω V 23b οἶνον] + νεον V: cf οἶνος νέος (‫“ ָע ִסיס‬sweet grape juice”) Sir 9,10; 7 times in NT. 25b ξύλον πᾶν καρποῦ] ξυλον παν καρποφορον V = lignum omne fructiferum Hi Sa 28a om καί γε V Chr IV 599 Pot Ep Ath, Ep subst 20 Hi Aeth Geo Fa1 SaI = Vulg, contra 𝔐 220b ἐμόχθησα] μοχθω V: ex 218b 222c μοχθεῖ] εμοχθησεν V: ex 218b 32b πεφυτευμένον] πεφυτευμενα V 315a γενόμενον] γεγενημενον V 322d ἐν ᾧ] εως V 41f ἰσχύς] + αυτοις V 411b θερμανθῇ] θερμη V 57c μὴ θαυμάσῃς] pr και V 519a τὰς ἡμέρας = 𝔐] εν ταις ημεραις 475; τα ετη V 73b ἐν κακίᾳ] εν καρδια V SaI II 2 718c τὸν θεόν = 𝔐] τον κυριον V

98

The Text History

811d αὐτοῖς] αυτον V Geo 814d ἀσεβεῖς = 𝔐] ασεβειαι V 817a ποιήματα = 𝔐] πεποιημενα V 91c οἱ δίκαιοι] om οἱ V k 311 766A 910a τοῦ ποιῆσαι] om του V 112a μερίδα] + και γε V 115c τὰ ποιήματα = Pesh] το ποιημα V = 𝔐 123c ἤργησαν] ηρπαγησαν V 124a κλείσουσιν] κλεινουσιν V 1212b οὐκ ἔστιν] pr και V Dam 2° Geo; + σοι V

253 21c τοῦτο = 𝔐] om 253 214d om αὐτοῖς 253 43c σὺν τὸ ποίημα = 𝔐] ουπω το π. 253: ex 4,3b 59b αὐτῶν B-S* alii (αυτω Vc-411 = Gra. = ‫ ;לא‬αυτου A Sc alii) γένημα « ‫בּואה‬ ָ ‫]לֹא ְת‬ αυτου ου γεννημα 253; non fruetur eis Hi; fructus non capiet ex eis Vulg 78a ἀρχὴν αὐτοῦ = 𝔐] om αυτου 253 719b τοὺς ὄντας ἐν τῇ πόλει = 𝔐] τους εν τη πολει οντας 253 82b σπουδάσῃς] σπευσης 253 = σʹ 812e πρόσωπον αὐτοῦ = 𝔐] om αυτου 253 816c ἐπὶ τῆς γῆς] επι την γην 253 93d ἐν ζωῇ αὐτῶν] εν τη ζωη αυτων 253 105a εἶδον = 𝔐] om 253

Significant are the three word variants in 253: 1110a θυμόν « ‫]ּכ ַעס‬ ַ παροργισμον 253-475; iram Hi Vulg. παροργισμός is derived from αʹ or σʹ; cf. Ps 935(1014) 30(31)10 θυμός « ‫]ּכ ַעס‬ ַ αʹ σʹ παροργισμός Ezek 2028 LXX vacat ‫ ַּכ ַעס‬θʹ ※ θυμός σʹ παροργισμός. 123c ἤργησαν ‫]ּב ְטלּו‬ ָ αλγησουσιν 253-475; cessabunt Hi; otiosae erunt Vulg

Here we have a copyist’s error; read ἀργήσουσιν. The correct spelling is found in Hipp I/1 224 and Met ἀργήσουσιν. ἀργεῖν for Aramaic ‫ בטל‬occurs only twice in the LXX: II Esdr 424 ἤργησαν (κατηργησε 107) and ἀργοῦν (αργων B´ A 71-74106-107-120-134-370 58). 123c ὠλιγώθησαν ‫]מ ֵעטּו‬ ִ ηργησαν 253-475; imminutae sunt Hi; in minuto numero Vulg. ηργησαν is an erroneous repetition of the foregoing ἤργησαν, see above.

The Hexaplaric Recension

99

A different situation is the rendering of a phrase in individual MSS. As an example (πορεύειν) εἰς οἶκον « ‫ל־ּבית‬ ֵ ‫ ֶא‬may be selected: 4,17a 7,2a 7,2b 12,5e

εἰς οἶκον] in domum Hi Vulg εἰς οἶκον] προς οικον O; ad domum Hi Vulg εἰς οἶκον] προς οικον 253-475; ad domum Hi Vulg εἰς οἶκον] προς οικον O 766; in domum Hi Vulg

475 18d ἀπὸ ἀκροάσεως] om ἀπό 475 27b ἐγένοντο] εγενετο 475 216c ταῖς ἡμέραις ταῖς ἐρχομέναις] pr εν 475 Syh 216e καὶ πῶς ἀποθανεῖται] om καὶ πῶς 475 224d ἀπὸ χειρός] εκ χειρος 475 313a ὃς φάγεται καὶ πίεται] φάγεται et πίεται tr 475 41a ἐγώ] αν 475 53d σὺ οὖν ὅσα] συ οταν 475 58a ἐπὶ παντί] οτι εν παντι 475 517d ἐν παντὶ μόχθῳ αὐτοῦ] om αυτου 475 517e ἐὰν μοχθῇ] εμοχθησεν 475 519a τὰς ἡμέρας] εν ταις ημεραις 475 63b πλῆθος] πληθυνθη 475 | 67b πληρωθήσεται] + εις το στομα 475 71a ὑπὲρ ἔλαιον ἀγαθόν] om αγαθον 475 72b ἢ ὅτι πορευθῆναι] υπερ του πορευθηναι 475 714b καὶ ἰδὲ ἐν ἡμέρᾳ κακίας] ιδε post κακιας tr 475 722b πολλάς] πολλακις 475 729b εὐθῆ] pr δικαιον 475 83a πορεύσῃ] > 475 815c εἰ μὴ τοῦ φαγεῖν καὶ τοῦ πιεῖν καὶ τοῦ εὐφρανθῆναι] om τοῦ 3° 475 91d τοῦ θεοῦ] αυτου 475 92f ὡς ὁ ἀγαθός, ὣς ὁ ἁμαρτάνων] om ὥς 475 92g ὣς ὁ ὀμνύων] καθως ὁ ὀμνύων 475 101a μυῖαι θανατοῦσαι] μυιων θανατος 475 103c ἃ λογιεῖται] διαλογιειται 475 1010d σοφία] pr η 475 112b om ἐπὶ τὴν γῆν 475 119e καὶ γνῶθι] om και 475 1110b παράγαγε πονηρίαν ἀπὸ σαρκός σου] om ἀπό 475 122a ἕως οὗ μὴ σκοτισθῇ] om οὗ 475 125c παχυνθῇ] πληθυνθειη 475

100 1211c 1213a

The Text History

οἳ παρὰ τῶν συνθεμάτων ἐδόθησαν] om οἵ 475 ἀκούεται 998 O–475 637 359 Hi] ἄκου 637; ακουετε 475 Syh

637 110c γενομένοις] επερχομενοις 637 Did Hiob 57,12 113b ἐν τῇ σοφίᾳ] om ἐν 637 GregNy 299,23ap OlΝ 23b om καί 637 GregNy 312,7te 24b ᾠκοδόμησα] οικοδομησαι 637 (ωκ.) 23g ἡμερῶν = 𝔐] om 637 752 219b ἐξουσιάσεται] εξουσιασει 637txt 219c ᾧ 1°] ο 637 314c ἐπ᾽ αὐτῷ (-των) = 𝔐] εν αυτοις 637; επ αυτοις O–637 cII 613sup lin Fa1 322a ἀγαθόν = 𝔐] pr το 637 44a τὸν μόχθον] om τόν 637 155 = 𝔐 414b ἐγεννήθη] εγενετο 637 51b μὴ ταχυνάτω] του μη ταχυναι 637 51d ἐν τῷ οὐρανῷ] εν τοις ουρανοις 637 57b ἐπὶ τῷ πράγματι] επι το πραγμα 637 59c καί γε = 𝔐] + και 637 518d ἐν μόχθῳ αὐτοῦ] om αὐτοῦ 637, contra 𝔐 97c fin] + εν ευφροσυνη 637: ex (7a) 99f ᾧ σὺ μοχθεῖς = 𝔐] ο εμοχθης 637 914b ἐπ᾽ αὐτήν] επ αυτη 637

All variants have no significant textual value or worth. The agreement with 𝔐 in 4,4a and 7,2a is accidental. La = 𝔐 In the following instances there are readings in La which go with 𝔐 and also agree with a few Greek witnesses, particularly those from O (V-253-475-637). 3,19a ‫ִּכי ִמ ְק ֶרה‬ συνάντημα A C alii, οὐ συνάντημα B-68´ 998 alii, ὡς συνάντημα S O–637-411 C d alii, quia eventus La, idcirco unus interitus est Vulg, ὅτι συνάντημα 336´ Hi Fa1 Syh DamR Met III.20,6 PsChr 5,11a ‫ְש ֹנַ ת ָהע ֵֹבד‬ ὕπνος τοῦ δούλου (‫)ה ֶע ֶבד‬, ָ somnus servi Amb Nab 6,29, ὕπνος τοῦ δουλεύοντος O Syh, somnus operanti Hi = 𝔐 Vulg One expects the genitive operantis, which is attested by Vulg ΠHm(vid).

The Hexaplaric Recension

101

6,12c 8,13b ‫ַּכ ֵּצל‬ ἐν σκιᾷ quasi umbram Hi 6,12c, velut umbra Vulg 6,12c quasi umbra Hi Vulg 8,13b ὡς σκιάν O-411 L Hilem et com 300,3 300,95 (sed hab in umbra Amb Mort 2,4 SaI 3 6 Syh) 6,12c; ὡς ἐν σκιᾷ S* 645 766 Hilem 319,184 (sed hab in umbra HiLXX 319,195) = 𝔐 8,13b 7,8a ‫ַא ֲח ִרית ָּד ָבר‬ ἐσχάτη λόγων novissiumum sermonis La, finis orationis Vulg ἐσχάτη λόγου O Syh alii 10,17b ‫ָּב ֵעת‬ πρὸς καιρόν in tempore La Vulg Hi (Pach) Aug (civ 17,20) ἐν καιρῷ O-411 La × 𝔐 In the following instances La attests readings which go against 𝔐 but agree with a few Greek witnesses, particularly those from O (V-253-475-637). 2,5b ‫ל־ּפ ִרי‬ ֶ ‫ֵעץ ָּכ‬ ξύλον πᾶν καρποῦ ξύλον πᾶν καρποφόρον V lignum omne fructiferum La plantantur arbores, non omnes fructiferae, ut in Latinis codicibus habemus Hi 264,99s. Cf. Ps 148,9 ‫ל־ּפ ִרי‬ ֶ ‫ ֵעץ ָּכ‬ξύλα καρποφόρα PsG ligna fructifera PsH lignum fructiferum 5,15c ‫יַ ֲעמֹל‬ μοχθεῖ laboravit La Vulg ἐμόχθησεν O-411 Hi Metcom V.3,91: 6,1b ‫ל־ה ָא ָדם‬ ָ ‫ַע‬ ἐπὶ τὸν ἄνθρωπον apud homines La Vulg ἐπὶ τοὺς ἀνθρώπους O

102

The Text History

7,2a 2b 12,5e ‫ל־ּבית‬ ֵ ‫ֶא‬ εἰς οἶκον in domum 12,5e La Vulg ad domum 7,2a 2b La Vulg πρὸς οἶκον 7,2a O, 7,2b 253, 12,5e O 766 In sum, when one subtracts from the witness of the O group hexaplaric additions, corrections toward the Hebrew using readings from the Jewish revisions of the LXX, and grammatical and stylistic improvements of the text, one is close to the earliest form of the text. The witness of the O group is extremely important. All unique readings from the members of the O group were reported to demonstrate a significant truth in textual criticism: every manuscript has unique readings— most of them have no textual value. Careful scrutiny does reveal a few important readings from time to time.

II.  The Egyptian Text-Type of Codices B-S and Their Derivatives 1.  The Witnesses of the Egyptian Text-Type (E) B-S-998 68-534-613 Co (= Fa, Sa) Congeners of B Closely Related:

68´’ 68´ 68’ 534´

= 68-534-613 [602 = Copy of 613] = 68-534 = 68-613 = 534-613

Distantly Related:

336´ (= 336-728) 766 (= 766A 766B)

The major uncials are B, S, and A. MS. C belongs to the branch of the textual tradition where we find A, V belongs to the O Group, and 3010 is fragmentary and cannot be accurately classified in Ecclesiastes although it is Lucianic elsewhere.1

1  Felix Albrecht and Chiara Faraggiana di Sarzana, “A Carbonized Septuagint Palimpsest of the Libri sapientiales in Biblical Majuscule, Codex Taurinensis, Biblioteca Nazionale Universitaria,

The Egyptian Text-Type of Codices B-S and Their Derivatives

103

The following chart shows the number of grammatical-stylistic improvements, omissions, additions, and transpositions for B, S, A, and combinations of these majuscules. B S A BSA BS BA

gram.-stil. 42 31 76 6 17 5

om 16 29

add 41 24

11

8

tr 2 7

Counting all deviations from the text of the edition (except such variants as cut across family lines, e.g. ιδον for εἶδον) B-S agree in 53 instances, and A agrees with B-S in 12, with B in 4, and with S in 34. The genealogical stemma suggests that A is a branch from an ancestor common to it and S, but that S is closer to the main stem of the tree than A. Thus S is in an intermediate position between B and A. Nevertheless, A has numerous improvements to the text peculiar to it and also contamination or influences from other types of text that we do not see in B and S. It is clear from these data that B and S belong to the same part of the text tradition whereas A belongs to a different branch of the tradition. The data also show that B-S (and congeners) represent a recension; after the O group it is the earliest attested recension, since 998 dates this type of text to at least 300 CE.

2.  The Provenance of the Egyptian Text The Egyptian provenance of 998 is clear and certain. First, the manuscript has both Coptic and Greek texts of Ecclesiastes. Second, the close relations between 998 and Fa, Sa on the one hand and B-S on the other clearly identify this group of manuscripts as the early Egyptian form or type of the text.

3.  Brief Characterisation of the Egyptian Text The Egyptian text-type is characterised by careful editing to improve the text. We can give good examples of improvements to the text. Since the Greek translator was so devoted to formal equivalence, frequently the resultant translation contravened good grammar and style in Greek. Examples:

C.V. 25 (Rahlfs-Ms. 3010): its Text and Context,” paper presented at ‘New Light on Old Manuscripts: Recent Advances in Palimpsest Studies,’ 20-22 April, 2018, Vienna.

104 113e

The Text History

τοῖς υἱοῖς τοῦ ἀνθρώπου] τοις υιοις των ανθρωπων B-S

The Greek translator renders ‫ לבני האדם‬according to the number of the nouns in Hebrew whereas B-S provide an improvement whereby the plural sense of ‫ אדם‬is matched in number with ‫ בני‬according to good style in Greek. 916c

λόγοι αὐτοῦ] pr οι B-68´’-998 (vid) alii

The Greek translator rendered ‫ דבריו‬by λόγοι αὐτοῦ because he follows the articulation in the Hebrew text even if it contravenes the syntax in Greek. Thus it was natural for the Greek scribes who produced the early Egyptian Text to insert the article before λόγοι especially when the noun is determined by the possessive pronoun. 71b

καὶ ἡμέρα τοῦ θανάτου ὑπὲρ ἡμέραν γεννήσεως αὐτοῦ] om αὐτοῦ S*-68´-998 alii

Although the Hebrew text has the pronoun and therefore the Greek translator represented it in his text, a text without the pronoun is smoother in Greek. In addition to the fact that the early Egyptian Group possesses many changes, corrections, and improvements belonging to the category of inner Greek variants, it also contains correction towards to the Hebrew text achieved on the basis of Origen’s Hexapla. Examples are 29c 218b 310b 99d 1019b of which 99d is the most remarkable: 99d

πάσας τὰς ἡμέρας ματαιότητός σου] πασαι ημεραι ημεραι (> 534 Fa1 3) ατμου σου B*-534 998 Fa1 3 = αʹ

Let there be no mistake, then, or idolising of our most ancient manuscripts. The early Egyptian Text is a carefully edited and revised text, intending to improve the grammar and style in Greek as well as bring it into closer alignment with the Hebrew text. The early Egyptian Text derived from copies that came from Origen and in turn, many later text types in other parts of the early Christian world derive from the early Egyptian Text. It is possible that this Egyptian text-type is the so-called Hesychian Recension of the trifaria varietas mentioned by Jerome (Biblia Sacra Vulgata I:546, Praefatio ad Paralipomena).

4.  More Distant Relatives 336´ (= 336-728) 766 (= 766A 766B)

The A-Text

105

The manuscript pairs 336´ and 766 represent minor recensions in their own right although they have their roots in the Egyptian Group. The pair 336´ agrees with the Egyptian Group in approximately 92 instances and deviates from it in 133. In the case of 766, there are 50 agreements with B and roughly 171 disagreements. The following examples show that the genealogical roots of the manuscript pairs 336´ and 766 are in the Egyptian text-type: 18c

οὐκ ἐμπλησθήσεται] ου πλησθησεται B-534´-998 d 336´ alii

215f

ὅτι—ματαιότης] + διοτι (+ ο B-613 998 d SaI) αφρων εκ (+ του 613) περισσευματος (+ καρδια Geo) λαλει B-613 998 336´ 766B Geo alii

These manuscript pairs also have influences from other groups and many independent improvements to the text. Examples: 15b

ἕλκει] + εκει 336´

111b

γενομένοις] γενησομενοις O C´’–157 299 797 336´ 542 613 752 766 Hi alii

111c

ἔσται] εστιν A C O–637 L 609 k 443 548 766 Arm Fa1 2 SaI OlΔΚ

For further discussion of recensional activity in B, see Peter J. Gentry, Special Problems in the Septuagint Text History of Ecclesiastes, XIII Congress of the International Organization for Septuagint and Cognate Studies, Ljubljana, 2007, ed. Melvin K. Peters (Society of Biblical Literature Septuagint and Cognate Studies, no. 55. Atlanta, Georgia: Scholars Press, 2008), 137-157.

III.  The A-Text 1.  The Witnesses of the A-Text A C 155 252 296´ 549

2.  Brief Characterisation of the A-Text In the description of the Egyptian Group, Codex Alexandrinus was briefly characterised. This manuscript is about a century later than B and S and shows further editing and improvements to the text as well as influences of other parts of the text tradition.

106

The Text History

A chart shows agreements and disagreements between A and each of the four manuscripts (a manuscript pair in the case of 296´) that have their roots in A as ancestor. A=C A = 155 A = 252 A = 296´ A = 549

44 27 49 49 50

A ≠ omnes A≠C A ≠ 155 A ≠ 252 A ≠ 296´ A ≠ 549

45 43 55 43 45 44

These data help to identify manuscripts in the codices mixti category as loosely belonging to the part of the textual tradition represented by A. Like A, MSS. C 155 252 and 549 are all descendents of a hypothetical ancestor descended from one branch, whereas MSS. 296 and 548 are descended from a hypothetical ancestor through a closely related branch of the tree. An example may help to appreciate the import of these observations: 210b

οὐκ ἀφεῖλον] ουχ (ουκ 543 549) υφειλον A L(–125) 68’ 248´ 252 296´ 311 339 543 549 698 706 788 (sed hab Dionlem 216 Dioncom 217 Met II.2,2 Theog VI 151) = Gra Ra

Although the reasons why Grabe and Rahlfs chose ουχ υφειλον as the lemma in their editions are not known, at first glance the number of manuscripts supporting this reading is impressive. But when we realise that 252 and 549 are related to A, while 248´ and 296´ are more distant cousins of A, the evidence is not so weighty. It could even be the case that A was influenced by the Lucianic recension here.

IV.  The Lucianic Recension Literature Matthew M. Dickie, An Analysis of the Lucianic Recension of the Greek Ecclesiastes, Ph.D. diss., The Southern Baptist Theological Seminary, Louisville, KY, 2013. Joseph Ziegler, “Hat Lukian den griechischen Sirach rezensiert?” Biblica 40 (1959): 210-29.

Although Matthew Dickie worked under my supervision, he did the research and most of the writing for this section.

The Lucianic Recension

107

1.  The Provenance of the Lucianic Text The Lucianic recension is the work of Lucian (c. 240-312), Presbyter of Antioch, who although born at Samosata is not to be confused with Lucian of Samosata (c. 125-180), the famous rhetorician and satirist. Importantly, for his work, he had access to the Hexapla of Origen.

2.  The Witnesses of the Lucianic Recension L = 106-125-130-261-545 (443 Chr Antioch Tht)

3.  The Peculiar Character of the Lucianic Recension First, a summary of the character of the Lucianic recension is given, followed by a more detailed analysis. The chief witnesses of the Lucianic recension in Ecclesiastes are MSS. 106-125130-261-545. These same manuscripts—excluding 545—comprise a subgroup (= l) of the Lucianic recension in Sirach. It may be, then, that the Lucianic recension in Ecclesiastes is poorly attested in comparison with books like Kingdoms, although preserved nonetheless. MS. 261 better preserves the characteristics of the recension than the other witnesses, e.g.: 511a τοῦ δούλου] τω δουλω 261 Antioch 1805 Chr II 585 III 44 Tht IV 680te

This example also shows the affinities L has with the Antiochian fathers, especially Chrysostom and Antiochus Monachus. 27c ποιμνίου Ol Syh] -νιων V L 443 Chr IV 599 IX 683 Metcom II.1,53 Geo Fa1 717a πολύ] πολλα L 728 Antioch 1553

The Antiochian fathers likely preserve Lucianic readings against L, especially when they preserve grammatical/stylistic changes in agreement with O: 18d ἀπὸ ἀκροάσεως] του ακουσαι (ακουειν Chr XV 684) O–253 Chr XV 684 Trin 1,7 Ruf Ex 11,6 Aeth

LatPsVig

Influence from O is a well known characteristic of the Lucianic recension. In Ecclesiastes L preserves unique readings with O: 77b τὴν καρδίαν εὐτονίας] την ευτονιαν της καρδιας O-411 L–(125) 443 547 766 712c τὸν παρ᾽ αὐτῆς] τον εχοντα αυτην Sc O-411 L–(125) 602 613 Hi ↓).

108

The Text History

As in Sirach, L sometimes preserves a slight variation of an O reading: 89c ὃ πεποίηται] το (> L(–125)) πεποιημενον O-411 L(–125) 542 766 Chr IX 763 Metlem et com VII.11,8,58

Hence, L improves the text of O. L is chiefly distinguished from other witnesses by grammatical and stylistic improvements. These improvements include (1) doublets, (2) addition of articles, especially to nouns modified by possessive genitives and to substantival participles, (3) addition of pronouns, and (4) addition of conjunctions as the examples indicate respectively: 1011a ἐν οὐ ψιθυρισμῷ] + απουσης επωδης L ↓ App II: σʹ [αʹ 248´] ἀπούσης ἐπῳδῆς 248´ 252 318a καρδίᾳ] pr τη 540 L 41e παρακαλῶν] pr ο L C´’ k (46s) 161-248c 338 411 547 728 Antioch 1560 Chr VI 606 Dam Didcom 107,25 Met III.26,4 Ol PsChr = Ald Compl: cf (1g) 1020a ἐν συνειδήσει] pr συ C 798 L–261 125II-542 248´ 252 296´ 311 338 339 543 549 602613 698 706 795 OlΑΓΔΕΙΚΜ 22a τῷ Bas III 961] + δε L

Thus, the additions in L contribute mostly to grammatical clarity. Other grammatical and stylistic improvements include (1) employment of simplex forms for compound verbs and vice versa, (2) changes in case and/or number, and (3) lexemic changes which are synonyms, illustrated respectively: 64a πορεύεται] διαπορευεται L 298 321a υἱῶν] υιω L–130 125e αἰῶνος] αιωνων L 48b οὐκ ἔστιν 2°] ουκ υπαρχει L–(125´) 728b ἄνθρωπον] ανδρα L 443 Anast 628 Didcom 231,2 = Vulg

As in Sirach, transpositions are rare 49b αὐτοῖς] et μισθός tr L–(125) 130 Antioch 1676 OlΝ

The Lucianic Recension

109

Omissions are not a strong characteristic of the Lucianic recension. Omissions of multiple words in L are mostly errors. Some omissions defy classification, because it is unclear as to how the omission is an improvement. Still, some omissions are grammatical improvements: 215c ἐμοὶ συναντήσεταί μοι 998 Dion Ol–Ζ Syh = Compl 𝔐] om μοι Sc O-411 L C´’ 754 260 338 547 602 613 698 705 766B 795 GregNy 361,17 364,13 Metlem et com II.6,4,35,95 OlΖ Hi = Ald 321b εἰς ἄνω] om εις L C´–571c 248´ 443

In addition to these characteristics, there are many substitutions of various words and changes in morphology of verbs (i.e. tense, voice, number, person) which contribute to a stylistic text. Atticisms are not well preserved in L, but consider: 1012b καταποντίσουσιν S A O-411 252 260 359 443 543 549 645 728 766 Antioch 1541 Didcom 79,5] -ποντιοῦσιν B C 998 L C´’ d k 68 542 155 248´ 296´ 311 336 338 339 534 PM a 547 602-613 698 706 795 Anton 800 Dam–VRM H T Didlem 305,2 306,2 Ps.CatA 554 Met IX.12,3 Ol

4.  Detailed Analysis of Witnesses Attesting the Lucianic Recension The Greek minuscules 106-125-130-261-545 form the main Lucianic group. Examples include: 22a τῷ Bas III 961] + δε L 215b τοῦ ἄφρονος] τω αφρονι L 216a ὅτι GregNy 361,19 364,20] και L 223b θυμοῦ—αὐτοῦ 2°] περισπασμου (-μον 545*) και θυμου L 226b ἔδωκεν 1°] > L–(125) 48b οὐκ ἔστιν 2°] ουκ υπαρχει L–(125´) 411b καί 2° = 𝔐] + γε L–(125) 59b γένημα] + και τις ηγαπησε δωρα εν πληθει ουκ ελευσεται L–(125) 252mg ↓ 612de om ὅτι 2°—ἔσται L 714c σύμφωνον] > L 101b ὀλίγον σοφίας] ολιγη σοφια L–(125) 1010d τοῦ ἀνδρείου] τω ανδρειω L 1011 ψιθυρισμῷ] + απουσης επωδης L ↓ 116c στοιχήσει] ευθετησει L ↓ 121e μοι] μου L 125e αἰῶνος] αιωνων L

110

The Text History

129b σὺν τὸν ἄνθρωπον] + και ενωτισατο και ηρευνησε και κατεσκευασε παραβολας L↓ 1210a πολλά] + δε L 1212b σαρκός] pr της L contra 𝔐

Brief Characterisation of Individual Witnesses MS. 106

Manuscript 106 preserves few unique variants. They include: 1:3 τίς] τι 106 5:15 om γάρ 106 7:4 εὐφροσύνης] αφροσ. 106* 7:19 σοφῷ] θεω 106txt 7:29 σὺν τόν] συνετον 106 9:11 ἀπάντημα] υπαντημα 106 MS. 106 occasionally agrees with other witnesses, like B-68´’ and 998, which are known for grammatical and stylistic improvements: 220b ἐπί] εν B-S-68´’ 998 O 299-571c-797-cII 106 d 336´ 411443 547 645 705 766B 795 alii

Such variants are not necessarily Lucianic. When most manuscripts of L preserve a variant together, MS. 106 reads against the group only 8 times (3 omissions and 5 substitutions). Problems affecting legibility of the manuscript account for most deviations from L. Deviations occur mostly with manuscript 130 (6 times)— most of which are in agreement with the critical text (CT) or may be classified as orthographical. Hence, MS. 106 is firmly fixed as a member of L.

MS. 125

Manuscript 125 is poorly preserved compared to the rest of L. The manuscript also has many scribal errors. Unique readings include: Inscriptio: εκκλησιαστης βιβλιον εικοστον εβδομον 125 1:3 μοχθεῖ] μοχθοι 125 2:7 πολλή] πολλων 125 3:17 om ἐγώ 125 = Vulg 4:7 ἐπέστρεψα] -ψον 125 6:7 μόχθος] μοσχος 125 9:11 καί γε οὐ 2°] ουδε 125 LatCass Co 13,18 Hilem et com 325,148 328,242 9:11 καί γε οὐ 3°] ουδε 125 LatCass Co 13,18 Hi 328,242 Syh 10:15 κοπώσει] κοπιασει 125

The Lucianic Recension

111

12:1 ἔλθωσιν] εισελθωσιν 125 12:10 γεγραμμένον] -μενα 125 Manuscript 125 sometimes preserves variants derived from variants in L. Many of these variants are orthographical in nature and involve possible itacisms. Examples include: 8:3 θελήσῃ] θελη L–125 k 338 766 Anton 1000 DamR; θελει 125 9:10 εὕρῃ] εχη L–125 545; εχει 125-545 Careless errors in 125 abound. A few examples will suffice: 1:8 οὐκ ἐμπλησθήσεται] ουκ εμπλησθησονται 125 547 (with singular ὀφθαλμός) 2:14 συναντήσεται] -σεως 125 2:17 σὺν τὴν ζωήν] συν (+ εν 125) τη ζωη V 797 125´ 7:26 om καί 3° (26c)—ἐν αὐτῇ (26f) 125) The number of variants which 125 preserves against L, when the group also preserves a variant, is 10. A close reading of 125 shows that the manuscript does preserve Lucianic readings even amidst places where L preserves a confirmed Lucianic variant. The variants at 2:7c are an excellent example: 27c καί γε κτῆσις (κτισεις 125) βουκολίου (-λιων L d Chr IV 599 IX 683) καὶ ποιμνίου (‑νιων V L 443 Chr IV 599 IX 683) πολλὴ (πολλων 125) ἐγένετό (-νοντο 998*(c pr m) 125´ 443 alii) μοι Manuscript 125 preserves variants of readings found in manuscript 261 (e.g., 5:17 ᾧ] ον 125; ων 261). For the relationship between 125 and 261, see 125´ below.

MS. 130

Peculiar variants in manuscript 130 include the following: 5:5 στόμα] σωμα 130 6:1 ἄνθρωπον] ουρανον 130 Arab SaI 6 (absc Fa2; sed hab Fa1) 8:2 φύλαξον] ‑λαζον 130 8:14 αὐτούς 1°] αυτα 130 Manuscript 130 is a reliable witness to the Lucianic recension, because it avoids many errors, especially omissions, found in L: 1:13 ὅτι—(14b) ἥλιον] > L–130: homoiot 2:14 om καὶ ἔγνων 425 L–130: haplogr

112

The Text History

3:19 τούτου 1°] ∩ τούτου 2° L–130 797 7:23 εἶπα] ειτα 390 L–106 130 This is the case even when an error affects a large part of the tradition (e.g., 1:17a –(106 125) 130 d–254´ k 68 248´txt 296´ (17a) καί 1°—γνῶσιν] sub ※ V 788 Syh; > 253 L 311 338 443 547 705 Clem II 37 Dion 212 Ol Geo ↓ : homoiot). Similarly, in one place 130 resists conformity to L when the reading is hexaplaric (10:6 ὕψεσι] υψει L–130 Fa1 ↓ [σʹ κείμενον ἄφρονα ἐν ὕψει μεγάλῳ 252 Hi]). Sometimes 130 does not preserve unique Lucianic readings: 4:9 αὐτοῖς] et μισθός tr L–(125) 130 Antioch 1676 OlΝ 4:10 αὐτῷ (CT: αὐτῷ τῷ ἑνί)] et τῷ tr L–(125) 130) Manuscript 130 rarely agrees with O against the main group L (1:16 μου [CT: μου τῷ c λέγειν] + εν O–253 C´ 130 336 443 OlΗ SaI Syh). Although 130 tends to differ from L mainly by resisting copyists’ errors, the manuscript does preserve possible Lucianic readings:

3:19 υἱῶν] pr των 998 130 C´–798 698 Dam–M = Ald P 5:7 φυλάξαι] φυλαξει Sc O–637 L–(125) 130 161c-248c 296´ 311 706 795 Dam–M R = Compl; φυλαξεται 411 130 C´’–299 d–357 645 = Ald) P

MS. 261

Unlike 130, manuscript 261 deviates from the main group L with many grammatical and stylistic improvements. 261 agrees with Chr more often than the other manuscripts of L. Concern for grammar and style naturally places 261 in agreement with stylized texts (e.g., 3:11 ἐποίησεν] pr ὅσα 998 261). Unique readings include: 1:2 ματαιότης ματαιοτήτων 1°] + τα παντα 261*: cf sq | εἶπεν—ματαιοτήτων] omnia vanitas dixit Ecclesiastes Amb Fuga 165,11 Or V 92,13 1:9 τό 3°] τοτε 261 2:3 κατεσκεψάμην] + τε τουτο 261 2:3 ἴδω] ειδω (οιδω 261) 637 299 261 443* 698 Dionlem 215 Ol–ΔΙΚΜΝ 2:10 om (10d) 261 3:10 εἶδον σύν] ειδοσιν 261 3:10 τοῦ ἀνθρώπου] ανθρωπου 261* 3:14 αὐτῷ B-68´ 357 = Gra Ra 𝔐 ↓ ] αυτης 411; αυτοις O cII 613sup lin Fa1; αυτον 261; αυτων rel 3:14 αὐτοῦ 2°] αυτων 261 M a 4:2 ὅσοι OlΑΓΕ] ως οι 261 DamRM L 4:15 ὑπό] επι 261

The Lucianic Recension

113

5:3 εὔξῃ] εξη 261 6:11 ματαιότητα] ‑τηται 261 P 7:9 σπεύσῃς] σπευδε 252: cf 51a; σπευσει 261 DamM 7:26 αὐτῆς 2°] αυτου 261 7:26 αὐτῆς 3°] αυτου 261 OlΜ 9:5 ἐπελήσθη] επελυσθην 261* 10:10 ἐκπέσῃ] εκπεσοι 261 Variations from L, when the group preserves a variant, are common in 261. Some may be itacisms/otacisms, though in many cases 261 preserves a syntactically possible reading. Examples include: 4:3 ὅς] οστις L 147-159-425mg-503-560-571 (ωστις 261) 613 4:10 πέσωσιν] πεση 411 L(–125) (-σει 261) C´–299 (-σει 609) 443 4:10 ἐγερεῖ] εγειρει S A 998 (vid) 475 L–(125) 261 C´–299 390-260 254´ 248 296 311 339 PM 547 645 706 728 766 Antioch 1676 DamM PsChr (inc C; sed hab Anton 1108 Dionlem 227 Met IV.1,3) Syh (sed hab Aeth = Gra.) = Ald; εγειρη 261 390-cII–260 561 698 (εγηρη) 4:16 αὐτῷ] αυτων L–(125) 261 443 766A; αυτον 261 8:13 ἡμέρας OllemΑΓ et comΑ] (εν 261*) ημερα L k 260*(vid) 766 Agreement with Chr is more prevalent in 261 than other manuscripts of L. When L is divided, 261 agrees with Chr in support of both a variant or the critical text. Examples include: 4:11 om καί γε 261 Chr IX 385 5:11 τοῦ δούλου] τω δουλω 261 Antioch 1805 Chr II 585 III 44 Tht IV 680te = Pesh Manuscript 261 rarely preserves unique agreements with O (e.g., 2:11 ᾧ] ο O–V 475 261-545 = Ald), except mostly in preservation of a rare but grammatically acceptable variant of ἀγαθωσύνη: 5:17 ἀγαθωσύνην] -θοσύνην O–V 261 299-560 357 46s-631* 296´ 339 534 543 549 645 795 OlΕΗ = Compl 6:3 ἀγαθωσύνης] ‑θοσυνης O–V 261 299-425 357 631 534 543 645 Did 171,19 OlΕΗ 6:6 ἀγαθωσύνην] ‑θοσυνην O–V 261 299-609 254-357 248* 252 296´ 543 645 Did 175,26 OlΕ 7:14 ἀγαθωσύνης] -θοσυνης O–V 261 299-390-425c k–46 357 161c-248* 534 543 548 645 Didlem et com 212,21 212,25 213,1 OlΓΕ 9:18 ἀγαθωσύνην] αγαθοσυνην O–V 560-601 261 357 248* 534 543 548 645 Didcom 290,16 (absc Didlem 290,10) OlΕΗ = Compl

114

The Text History

Agreement with 125 is common (see 125´ below). Agreement with 545 is fairly common. Manuscript 261 reads with 545 in fifteen places (6 omissions and 9 substitutions), not including 7 places where one other member of L is present. Examples include: 2:11 ᾧ] ο O–V 475 261-545 = Ald 5:2 ὅτι] ο 261-545 5:3 ἐάν B-S-68´ O–V-411 C´ 539 = Ald Ra] > 261-545 248´ 260 336´ 542 766 OlΗΝ An Scrip 1,22 Fulg Ep 1,11 Spec 556,10 Hi Geo = Vulg; αν rel (Met IV.6,2 Ol–ΗΝ PsChr) 7:18 τούτου] τουτω 545c; του 261-545* 7:26 τήν] τη 261-545 8:8 καί 1°] ∩ 2° 261-545 9:3 περιφέρεια] περιφερει 261-545 9:5 ἐπελήσθη] επελυσθην 261*; επελυσθη 261c 545 9:6 τῷ] τοπω 261-545 443 9:9 ἔλαιον] ελεον V 261-545 390-295 336´ 359 Didcom 273,19 (absc Didlem) OlΗ 10:17 γῆ] > 261(|)-545 Sometimes 261 may preserve a fragmentary portion of the Lucianic recension. For example, the variant at 10:5 (ᾖν] ον 261) supposes an antecedent πονηρός instead of πονηρία in the critical text. It is possible that the variant is Lucianic and that later copyists corrected πονηρός to πονηρία. Compare 6:1 πονηρία ἣν] πονηρον ον 125´.

MS. 545

Although 545 preserves many unique variants, they often seem syntactically unlikely and awkward. Unique readings in 545 include: 1:6 τό 2°] > 545 2:3 ἑλκύσει] ελκυση 125´ 311 443; ελκυσω 545 2:4 ἐμεγάλυνα] -λυνας 545* 3:8 om τοῦ 2° 545 3:9 μοχθεῖ] μοχθοι 545 5:1 ταχυνάτω] inc C OlΔ; καταχυνατω 545*; κατασχυνατω 545c 5:17 ὅ 1°] ον 545 7:10 τούτου] τουτο 545 (not a customary reading) 8:6 πολλή] πολλυ 545*; πολλυς 545c 9:10 ἄν] > 545 Hi 10:15 ὅς] ως 545 12:1 τῆς] σης 545* 12:6 ὅτου] οπου 545

The Lucianic Recension

115

Manuscript 545 contains many errors, as evinced by its high number of corrections (545c). Many of the errors seem careless. Examples include: 2:5 ἐποίησά μοι] εποιησαμην 475 534 OlΝ; εποιησαμεν 545 3:10 σὺν τὸν περισπασμόν] συμπερισπασμον 545 5:2 ὅτι (CT: ὅτι παραγίνεται)] ο 261-545 7:2 εἰς καρδίαν] εν καρδια (καρδιαν 545) A 539 545 alii 9:14 ἔλθῃ] ελθης 545 [with nominative βασιλεύς] 10:10 τοῦ ἀνδρείου] των ανδρειων 545c; των ανδρειω 545*; τω ανδρειω L Manuscript 545 often agrees with 261 (see 261 above). One variation of 261 (125´) is peculiar (6:1 πονηρία ἣν] πονηρον ον 125´ | ἥν] ον 545txt). Variations from L are similar to general errors where carelessness seems prevalent, though 10:16 may be Lucianic. Examples are as follows: 9:10 εὕρῃ] εχη L–125 545; εχει 125-545 10:10 τοῦ ἀνδρείου] των ανδρειων 545c; των ανδρειω 545*; τω ανδρειω L 10:16 ἐν πρωΐᾳ] το πρωι L–(125) 545 443c; τω πρωι 443* 545 10:18 δόκωσις] δοκησις L–545 254 k alii; δοκησεις 545 12:14 ἐάν 1°] + τι 545*; + τε S V-411 L–545* C´–601 alii

MS. Pair 125´

The two manuscripts 125 and 261 (125´) agree 60 times. Most of the agreements are not peculiar. Nonetheless, some of them evince the Lucianic recension. Unique examples include: 3:22 γένηται] γενοιτο 125´ 4:8 ὀφθαλμός] -μοι 125´ Geo (οφθαλμ 261) = 𝔐K Vulg 5:17 τοῦ 3°] > 125´ 6:1 πονηρία ἣν] πονηρον ον 125´ 6:7 οὐ] > 125´ 6:8 om τῷ σοφῷ 125´ 7:22 σε] σου 125´ 7:22 καρδίαν] η καρδια 125´ mg 7:28 ἄνθρωπον / ἕνα] tr 125´ OlΑΒΓΖ Η 9:9 μετά] επι 125´ 9:12 om τά 125´ Dam 10:4 μὴ ἀφῇς] μη αφεις 125´ 11:10 ἀπὸ καρδίας] εν καρδια 125´ 12:2 ἐπιστρέψωσιν] αποστρεψωσι 125´ 12:12 οὐκ ἔστιν] ουκ ετι 125´

116

The Text History

Agreements with O are rare (6:5 τούτῳ] τουτο S* O-411 125´ alii; 12:7 ἐπιστρέψῃ 1°] ‑στρεψει A C O–253 125´ alii). In one place 125´ agrees with O but not L (4:4 αὐτό] ο O–V 125´ 443c Dam; > V L–125´ d–357 alii). Other deviations from L include: 9:10 σὺ πορεύῃ] συμπορευση 125´ | πορεύῃ] πορευση O–V L–125´ 298 613 Dam–KRL OlΒΗ EOVMi 10:18 στάξει] στεναξει A L–125´ 336´ 443 795 SaI; στεναζει 125´ DamV

a

Agreements with B and its congeners are not surprising. But agreements with 998 are peculiar, especially 2:7 and 4:8. Examples include: 2:7 ἐγένετο] -νοντο 998*(c pr m) 299 125´ 254 (‑νοτο) 443 645 698 728 4:8 στερίσκω] + εγω 998 125´ 6:10 τοῦ 1°] absc 818; > B-68´ C 998 125´ 147-503-560-cII d k 443 543 766 PsChr 7:22 καρδίαν] καρδια B-68 998* alii; η καρδια 125´

5.  The Antiochian Fathers The Antiochian fathers did not comment systematically on Ecclesiastes. Rather they cited individual passages from the book in various homilies and tracts. Chrysostom (Chr) is generally considered the most important witness of the Antiochian fathers because of his abundant citations.2 In Ecclesiastes, however, there are more citations from Antiochus Monachus. Citations must be evaluated with great care because of three pitfalls: 1) the citation may be a free rendering, and therefore disqualified from further evaluation and use, 2) the author may cite the same passage twice with each one differing from the other in some way (e.g., 4:12 ταχέως] ταχυ Chr IX 385; > Chr XV 720), and 3) the citation may preserve a variant which is a copyist’s error.

John Chrysostom (died 407)

Ziegler identifies three criteria which must be present in order to determine with relative certainty whether Chr preserves Lucianic readings or not: 1) Chr agrees with L and O, 2) Chr shares a grammatically or stylistically motivated variant with L, and 3) Chr agrees with the corrector of Sinaiticus (Sc), a witness to O, without support from L.3 In addition to these criteria, Ziegler mentions two other relationships which may signal a connection to the Lucianic Recension: 1) Chr 2  Ziegler, “Hat Lukian den griechischen Sirach rezensiert?” 213. 3 In his article, Ziegler uses Swete’s abbreviation (Sca) for the corrector of Sinaiticus (Ziegler, “Hat Lukian den griechischen Sirach rezensiert?” 217). For Ecclesiastes, this criterion will be broadened to include all of the O group.

The Lucianic Recension

117

agrees with O and Old Latin witnesses (La) against L, and 2) Chr agrees with L and La to preserve a proto-Lucianic reading. If Chr stands in no relation to L, O, or La, then Chr may still preserve the Lucianic Recension, especially if it can be determined that L has failed to do so. In such cases, however, certainty is unobtainable.4 In the following analysis, each citation has been placed in one of four groups: 1) Chr agrees with L against the Critical Text (CT), 2) Chr agrees with CT against L, 3) Chr disagrees with L and CT, and 4) Chr agrees with a fragmented group in CT and against CT.5 These four groups provide a vantage point from which to apply Ziegler’s criteria of analysis. (i) Chr Agrees with Unified L against CT Chr reads with L against CT ten times. Most of these variants are Lucianic—stylistically and grammatically motivated. The places of variation with incorporated notes are as follows: List 1: Chr = L (≠ CT) 1:5 om καί 1° C 411 L 390-601-798c-cII 342 547 645 Hi Aeth Arm Fa1 2 Met I.6,2 Ol Chr IX 600 Cos 2,34 4,11 9,12 10,33 Dam Max I 1140 Proc 9,48 Sev 453 Chry V 746 PsMar Scr 5 Rav 1,9 (sed hab La160 Amb Ps 118 12,22 Aug Gen ad litt 1,10) = Pesh Vulg, contra 𝔐 2:7 βουκολίου] -λιων V L d Chr IV 599 IX 683 (cf XVI 89 584) Metcom II.1,53 Geo Fa1 2:7 ποιμνίου Ol Syh] -νιων V L 443 Chr IV 599 IX 683 (cf XVI 89 584) Metcom II.1,53 Geo Fa1. MS. 443 is closely related to L. 4:1 παρακαλῶν 1°] pr ο L C´’ k (46s) 161-248c 338 411 547 728 Antioch 1560 Chr VI 606 Dam Didcom 107,25 Met III.26,4 Ol PsChr = Ald Compl: cf (1g) contra 𝔐. The participle was construed as a nominal. 7:2 ἢ ὅτι πορευθῆναι] pr ε S*(c pr m); παρα το (> V) πορευθηναι O(–475 637) Sc; υπερ του πορευθηναι 475; om ὅτι 411 L(–125) C´’ 357-754 k 338 339 443 539 543 W P 547 549 766A Antioch 1709 Chr III 155 156 VI 487 XIII 334 Cyr IV 364 DamVV M 4  Ziegler, “Hat Lukian den griechischen Sirach rezensiert?” 213-218. 5  This categorization is similar to chapter three in Brock’s analysis of L in 1 Samuel (Sebastian P. Brock, The Recensions of the Septuagint Version of I Samuel [Torino: Silvio Zamorani, 1996], 181-223).

118

The Text History

Metlem et com VI.4,2,20 OlΑΒΓΖΗ PsChr Tht II 844 An Scrip 1,19 Aug C D 17,20 Spec 528,7 Arm Fa1 2 SaI La160 Syh = 𝔐. Chr cites the text twice once including πορευθῆναι and once omitting it. Support exists in other Antiochian fathers besides Chr for both variants. Omission of a verb in a comparative clause is not unusual. The original Lucianic Recension probably preserved the infinitive, because 1) the O group, though it preserves a different variant from L, has the infinitive and 2) omission of the infinitive has no Greek manuscript support beyond patristic sources. 7:6 γέλως Did 203,1] pr ο A C 475 L(–125) d k 248´ 252 260 296´ 311 338 339 443 543 549 613 698 706 795 788 Ammon Antioch 1724 Bas III 961 Chr II 1055 Dam Metlem et com VI.8,4,29 VI.9,21 Max II 996 OlΔΕΙΚΜ Fa1 SaI 6. Noteworthy are the agreements with 443 and with the citations from Antioch and Max in this addition of the article to a noun followed by an attributive genitive. 8:1 πρόσωπον] pr το S L 260 443 795 Chr III 238 IX 508. The noun is followed by a possessive genitive. 8:9 ὃ πεποίηται] το (> L(–125)) πεποιημενον O-411 L(–125) 542 766 Chr IX 763 Metlem et com VII.11,8,58. The change from a relative clause to an attributive participle is striking and thoroughly Hellenistic.6 12:13a ἀκούεται 998 O–475 637 359 Hi] ἄκου 637; ακουετε 475 Syh; ἄκουε rel (Anton 781 973 Chr XVIII 509 Dam Eus Ps 41 GregNa I 780 1245 Disc 26,14 Max I 933 II 861 Metlem et com X.14,3,19 Ol Hi Pach 147,23 Aeth Arm [absc Fa1] Fa2 SaI II) = Pesh 12:13b ὁ ἄνθρωπος B-S-68´ A 998 253 d (357s) 252 443 542 543 549 698 766B OlΙ*= Ra] om ὁ C O–253-411 L C´’ k 248´ 296´ 311 336´ 338 339 359 547 613 645 706 766A 795 3011 Anton 781 973 Chr II 703 III 232 XV 248 XVIII 512 Jb.Cat Β 33 Dam Max I 933 II 861 Metlem et com X.14,4,20 I.4,120 Ol–Ι* PsChrcom 67,8 SaI The two changes from singular to plural in 2:7 establish a peculiar relationship. The list shows typical Lucianic improvements: addition of the article to words modified by an attributive or possessive genitive (7:6 and 8:1 respectively), addition of the article to a substantival participle (4:1), and a correction from impossible Greek (7:2).7 Similarly, influence from O is attested (8:9 and 12:13b).

6  BDF §412. 7 However, ὅτι plus infinitive may follow a verb of speaking in the Ptolemaic papyri (Mayser §50.B.I.b.2.γ).

The Lucianic Recension

119

The variant at 8:9 differs from O in style and not grammar.8 In 12:13b, the omission of the article may be due to influence from O, which affected a large part of the manuscript tradition.9 At 7:2, agreement with La160 does not suggest that the reading is proto-Lucianic, because every Latin witness, including the Vulgate, preserves the omission.10 Hence, constraint of the Latin language is doubtless operative. When Chr agrees with L against CT, the variants fall largely within Ziegler’s first tier of proofs for the Lucianic Recension: stylistic improvements and agreements with O. In addition to these criteria, agreements with other Antiochian fathers are noteworthy. (ii) Chr Agrees with CT against L There are only two citations where Chr agrees with CT against L. These citations are as follows: List 2: Chr = CT (≠ L) 5:11 ὕπνος] pr ο V-411 L(–125) 139-147-159-425mg-503-560-798-cII–260 298 311 M 534 547c pr m 613 645 Anton 1065 Antioch 1805 Dam–M Didcom 153,6 Nil 468 Ol PsChr Tht IV 680 (sed hab Chr II 585 III 44 Met V.2,2) 12:13 φύλασσε Chr II 703 III 232 XV 248 XVIII 509 Dam Did Ps.CatA 315 Max I 933 II 861 Metlem et com X.4,4,19 I.4,120 Ol] φυλαξαι C´–390 415 425* 504 540* 571c 732 797 L 298 534-613txt Both of these citations are examples of places where Chr has failed to preserve the recension.11 5:11 is noteworthy, because Chr reads against texts which preserve good Greek grammar, three of which are Antiochian (Anton, Antioch, Tht).12 (iii) Chr Disagrees with CT and L

8  In comparison with classical Greek, arthrous attributive adjectives and participles after anarthrous nouns are more common in the Greek of the Ptolemaic papyri (Mayser §64). 9  Omission of the article before nouns modified by πᾶς is not improper. The omission may have arisen due to the fact that ἄνθρωπος is a predicate noun (BDF §273). 10  Agreement with 𝔐 is probably coincidental even in the Vulgate. 11  The change of tense at 12:13 is typical of Lucian, though confusion of -ε with -αι is possible. 12  Nouns followed by an attributive genitive often have the article, though Chr uses a dative of possession (BDF §259; for the subtle distinction between a genitive and dative of possession, see BDF §189).

120

The Text History

Chr reads against CT and L forty-two times, a considerable number of variations. The origin of these variants is difficult to determine. Many of them are stylistically motivated. However, it is difficult to pronounce them as Lucianic, given that some of them may be free citations. In many cases, certainty is unattainable. There are eleven double citations, many of which agree with CT. The following lists group the citations according to Ziegler’s criteria. Extended texts from CT have been incorporated where necessary. List 3: Chr = O 1:8 ἀπὸ ἀκροάσεως (CT: καὶ οὐ πληρωθήσεται οὖς ἀπὸ ἀκροάσεως)] του ακουσαι (ακουειν Chr XV 684) O–253 Chr XV 684 PsVig Trin 1,7 Ruf Ex 11,6 (sed hab Am Mort 7,28 Tob 13,44 Eus-G Ser 9,13 Ruf Om S Bas 6 Hi = Vulg) Aeth ↓. The variant, a grammatical correction away from the Hebrew text, is attested in 252 (II App: οἱ οʹ τοῦ ἀκοῦσαι). 2:8 οἰνοχόας Did 40,25 41,18 GregNy 347,6 Met II.1,12 Ol Syh (CT: οἰνοχόον καὶ οἰνοχόας)] οινοχοουσας O d–357 336´ Chr IV 599 IX 683 XVII 123 879 Arm Fa1 SaI. The noun οἰνοχόη ‘pitcher’ was changed to a feminine participle of the verb οἰνοχοέω.13 2:14 CT: τοῦ σοφοῦ οἱ ὀφθαλμοὶ αὐτοῦ ἐν κεφαλῇ αὐτοῦ) om αὐτοῦ 1° 870 (sed hab 998) O–637-411 C ‘–299 563 571-425-601 338 534´ 539 543 766B Anast 673 Ath IV 1512 Bas I 384 II 57 537 Chr IX 345 Didlem 48,19 Didcom 48,22 Did Ps 29,13 64,14 105,6 262,17 301,23 Ps.CatA 669a 1222 Dionlem 219 GregNy 357,1te 358,11 360,17 Metlem et com II.5,2,40 Ol Proc Isa 2028 PsGregNy 67,10 Hi Geo Arm Fa1 SaI (sed hab Syh) = Vulg, contra 𝔐. This is a grammatical correction. 7:4 καὶ καρδία = 𝔐] οἱ γʹ ※ καὶ ↙ καρδία δέ Syh ↓ ; και καρδια δε 252 ↓ ; καρδια PM δε O-411 cII 534´ Antioch 1709 Chr II 1055 DamKVRM T Met VI.7,3 SaI (sed hab Fa1); om καί 601-609 106 542 645. The variant is stylistically motivated. In the list above, three variants are concerned with clarity, because they are grammatical (1:8 and 2:14) and lexical (2:8) improvements to CT. Substitution of a different conjunction in 7:4 is a matter of style. These variants may be hexaplaric corrections in Chr or variants which preserve the Lucianic Recension via O.14 13  For a discussion of the difficulties of these words, see Peter J. Gentry, “The Relationship of Aquila and Theodotion to the Old Greek of Ecclesiastes in the Marginal Notes of the SyroHexapla,” Aramaic Studies 2.1 (2004): 70-71. 14  Hexaplaric corrections also occur in Chr elsewhere (see Sebastian P. Brock, The Recensions of the Septuagint Version of I Samuel [Torino: Silvio Zamorani, 1996], 189).

The Lucianic Recension

121

Places where Chr disagrees with O and preserves stylistic and grammatical improvements are listed below with incorporated notes: List 4: Stylistic and Grammatical Improvements 2:4 οἴκους] οικιας Chr XVI 584 XVII 123 879 (sed hab Chr IV 599 IX 683) Tht I 1883. The variant is a synonym. 2:8 om καί γε V Aeth Hi Fa1 SaI Chr IV 599 Pot Ep Ath, Ep subst 20 = Vulg contra 𝔐 2:14 κεφαλῇ] pr τη 797 337 342-754 336´ 338 788 Ath IV 1512 Bas II 57 Chr IX 345 Didcom 48,22 GregNy 357,1ap PsGregNy 67,10ap (sed hab Anast 673 Bas I 384 Didlem 48,19 Did Ps 29,13 64,14 105,6 262,17 301,23 Ps.CatA 669a 1222 Dionlem 219 GregNy 357,1te 358,11 360,17 Metlem et com II.5,2,40 Ol PsGregNy 67) Co. The noun is modified by a possessive genitive. 3:7 σιγᾶν] σιγησαι Chr IX 432. The tense of the infinitive was changed to match the preceding ones. 3:7 λαλεῖν] λαλησαι 357 (-εισαι) Chr IX 433. The tense of the infinitive was changed to match the preceding ones. 4:1 αὐτοῖς 1°] αυτους 637 161c-248mg Antioch 1560 Chr VI 606 OlΒΗ: cf (1g). The pronoun was understood as an object of the participle παρακαλῶν. 5:2 περισπασμοῦ Gra Ra = sollicitudinis Hi (cf 2:26d 4:16a) = 𝔐] absc Fa1; πειρασμων 139-147-159-503-522-540-560-798 k 336(-μον) 548*vid(-μον) Chr XVIII 601 Ol Hiob 78 Syh; πειρασμος 766; πειρασμου rel (Antioch 1688 Anton 921 Dam Max II 997 Ollem et com Fa2 SaI). The noun may be a synonym. But its use in Chr is likely a preservation of an early Greek corruption and not attributable to Lucian. 5:3 om τοῦ C 571* Chr IX 248 447 DamR Eus Ps 628 885 Met IV.6,2 5:11 γλυκύς] ηδυς Chr II 585 III 44. The variant is a synonym. The noun γλυκύς can be metaphorical for something pleasant. 5:11 εἰ 1°] ειτε Anton 1065 DamAA; αν τε Chr II 585 III 44. The conditional conjuncttion was replaced with ἄν τε twice (see below). A similar variant can be found elsewhere in L (12:14). 5:11 ὀλίγον] et πολύ tr Chr II 585

122

The Text History

5:11 καὶ εἰ Antioch 1805 Dam–AVAVE Ol] ειτε και Anton 1065; αν τε Chr II 585 III 5:11 φάγεται] φαγη Chr II 585 III 44 Didcom 153,2 7:2 πορευθῆναι 1°] εισελθειν Bas III 257 Chr XIV 131 (post πένθους tr) XVI 574. The variant is synonymous with the edition, which is followed by the preposition εἰς. 7:2 οἶκον 1°] οικιαν Chr XIV 131 XVI 574 (sed hab Chr III 155 156 XI 446 XIII 334) 7:2 οἶκον 2°] οικιαν Chr XIV 131 XVI 574s (sed hab Chr III 155 156 VI 487 XI 446 XIII 334) 7:5 σοφοῦ] σοφων Chr II 1055 Syh: ex (4a) 7:6 τῶν 1° = Ald (CT: φωνὴ τῶν ἀκανθῶν)] > B 998 d 68 443 534 Chr II 1055 Dam–MM Max II 996 Fa1 2 SaI. Omission of the article in attributive genitives is normal after anarthrous nouns. 7:29 ἐποίησεν] post θεός tr Bas III 1045 Chr IX 555 Did Ps.CatA 963 Fa1 SaI 2 (sed hab Fa2) 7:29 om σύν (HT: 545 411 357 )‫את‬c 311 336´ Bas III 1045 Chr IX 555 CyrHier 384 409 Dam Did Ps 221,5 259,23 Ps.CatA 963 Olcom 65,23 Cass Co 7,4,3 Aeth SaI (sed hab Fa2) Hi Syh 8:1 φωτιεῖ] φωτιζει Chr III 238 Hi Is 15,24,43 Zach 3,12 (sed hab Hilem et com 313,1 314,22) Syh 9:16 σοφία 2°] pr η Anton 889 Antioch 1717 Chr IX 224 XVII 136 Dam (sed hab Epiph III 170 Ol PsChr). The noun is modified by an attributive genitive. 9:16 λόγοι] pr οι B 998 (vid) 571c d 68 534 602-613 Anton 889 Chr IX 224 Dam Mel Hom 303. The noun is followed by a possessive genitive. 9:16 εἰσιν ἀκουόμενοι] εισακουομενοι B 998 (vid) 68 534 (ισα-) Chr IX 224. The variant may not be an error. The preceding clause has a participle in predicate position with no finite verb. The variant brings symmetry. Note B and 998. 12:13 τέλος] περας 637 766 Chr XVIII 509 Metlem et com X.14,3,18 PsChrcom 67,6 ↓ (II App: σʹ θʹ ‫ ܣܟܐ‬Syh [= πέρας])

The Lucianic Recension

123

List 4 contains many variants which appear Lucianic. As Ziegler noted for Sirach, synonyms in Chr may evince the Lucianic Recension.15 In Ecclesiastes, Chr preserves synonyms at 2:4, 5:11, 7:2 (three variants), and 12:13. Similarly, nouns followed by a possessive or attributive genitive receive the article at 2:14 and 9:16 (twice). Changes in verbs, both modal and in tense, are typical of the Lucianic Recension (5:11, 3:7 [twice], and 8:1). Two uses of the particle ἄν at 5:11 are typical of the L group. These variants may preserve the Lucianic Recension even though the L group has failed to do so. It seems best, methodologically, to classify differing citations of the same text as free citations, especially where there is little support from Greek manuscripts or where one citation agrees with CT. Citations which may be free are as follows: List 5: Possible Free Citations in Chr 2:8 μοι 1° Chr IX 683] με 125 548*vid; > 338 Chr IV 599 4:3 om τοὺς δύο Chr IV 599 (sed hab Chr XVII 549). 4:12 ταχέως] ταχυ B 998 V d–357 68 155 336´ 443 534 Chr IX 385 PsChr: cf 8:11b; > 357 Antioch 1732 (lib) (sed hab Antioch 1676) Bas I 761 (lib) Chr XV 720 LatA-SS Helia 235 ter Amb Ep 81,8 Fuga 54 Luc 7,208 Chrom s 4,1 Fil 153,5 Hi Ep 76,1 Marc 3 Matth 3 Ps hom 59 Isid Ex 52,3 PsHi Brev 149 PsMel P 10,2,74 V 11,38,1 Pac Ep 3,27,1 PaulN Ep 23,30 Ruf Ex 9,3 (sed hab cito Hilemap 286,121 Hicom 287,152 bis Philipp 38,36; sed hab facile Chrom s 4,1 Hilemtxt 286,121 SedScot Misc 13,26,36; sed hab (dif)ficile Bed Aed 19 Tab 2,13 PsFel 1 GregM Iob 33,10 MartI 1 PaulAq For 181,2 PsSalo 53 bis Ta Ecl 8 bis? Sent 2,27 = Vulg). The omission is likely the result of free citation. The preference for ταχύ is a matter of style, given that the two stylized texts of B and 998 preserve the reading. It also may be a free citation in Chr. 4:12 ἀπορραγήσεται Chr IX 385 Dam Met IV.1,8,87] διαρραγησεται Bas I 761 Chr XV 720. The variant is a synonym but is likely a free citation. 7:2a ἀγαθόν 1° Anast 593 Antioch 1709 Bas III 257 Chr III 155 156 VI 487 XI 446 XIII 334 XVI 574 Cyr IV 364 Dam Did 197,18 Dion Isid 1241 Max II 881 Metlem et com VI.4,2,20 Ol PsChr Tht II 844] κρειττον Chr XIV 131. The variant is clearly a free citation. 7:2 ἢ ὅτι πορευθῆναι B-S* A C 998 254-342 68 248´ 252 296´ 311 336´ 534 542 548 613 645 698 706 766B 795 OlΔΕΙΚΜ] om ὅτι πορευθῆναι Anast 593 Chr XI 446 15  Ziegler, “Hat Lukian den griechischen Sirach rezensiert?” 214.

124

The Text History

XIV 131 XVI 574s Dam–VVWMP Dion Isid 1241 Max II 881 PsChrcom 74,51 Amb Ep add 14 Fuga 3 An Fris 402 Jb 3 Aug Spe 8 GregM Dial 4,4 Job 4 Hi Ez 8,27 Lucul 820 PaulN Ep 25 PsGregM Conc 14 PsIsid Test 18 Ta Ecl pr 10 Hite = Vulg. Omission of the infinitive is due likely to free citation (see List 1). 7:2c πότου] γελωτος Chr III 155 XI 446 XIII 334 XIV 131 XVI 574s Isid 1241 PsChrcom 74,51 Tht II 844 (sed hab Chr III 156 VI 487). The variant is not a synonym, though it may refer to an occasion of laughter. 8:1 αὐτοῦ 1°] > A Chr III 238 (sed hab Chr IX 508) Two of the citations appear Lucianic (4:12 and 7:2c). The fact that they are classified as free citations does not absolutely condemn them as worthless. List five exists for methodological clarity, given that double citations, especially when one citation agrees with CT, show that Chr may be aware of what the correct reading should be. It is noteworthy that Chr never preserves a variant in agreement with the entire L group while also citing CT elsewhere (see List 1). The following list contains places where citations cannot be classified according to Ziegler’s criteria or whose character is undecided: List 6: Miscellaneous 1:9 καί 1°] > 752 La160 Aethte SaI 1 Chr III 86 (sed hab Did Hiob 57,7 Eus VIII 2,22 GregNy 294,19 Max I 613 II 1412 Or I 282,6 Aug C D 12,14 Cass Co 8,21 PsMar Scr 3 (autem) Ruf Or princ 1,4,5 3,5,3 Hi SaII 5) = Vulg 2:7 πολλή] πολλων 125; πολλης 261-545; > S* 998 543 Chr IV 599 9:16 ἐξουδενωμένη] pr ουκ 357; εξου[ 998; εξουθενημενη C´–390 425 Anton 889 Mi Chr IX 224 XVII 136 DamAnt. Epiph III 170 Met IX.1,9s Mel Hom 303 (sed hab WEOV Mi AMi Antioch 1717 Dam–V V Ant. Ol PsChr); εξουδενημενη 253; εξουδενουμενη Mi A 357 DamV = Ald; εξουθενουμενη 390 DamAnt. Did Trin 816; εξουδενομενη WEOV (-θ- 425) 425 DamV . The variant is an allomorph of the word in CT.16 10:4 μεγάλας] -λης 125 698; μγλ 261; multa Ruf Om S Bas 1724 Or princ 245,22 250,30 (sed hab magna Amb Ps 37,35,2 Ruf Cant 211,19 Hi 334,56); > Chr XV 790 Proc Isa 2172 Only 1:9 is peculiar. Agreement with La160 may suggest a proto-Lucianic reading. The character of the variant, an omitted conjunction, does not permit certainty. 16  LSJ, s.v. “ἐξουθενέω.”

The Lucianic Recension

125

(iv) Chr Agrees with a Divided L Group When L is divided (i.e. at least one manuscript of the L group fails to read along with the majority of its manuscripts, excluding manuscripts that fail to preserve text), the attestation of Chr is as follows: List 7: Chr = L Minus at Least One Manuscript 1:8 ουκ εμπλησθησεται S A C O-411 C´’–147 159 503 560 L–125 k 68 125II 248´ 252 260 296´ 311 338 339 443 542 543 698 705 766 770 795 Dam–MM Metlem et com I.11,3,16 GregNy V 294,11 Max II 853 Ol = Compl Ra] οὐ πλησθήσεται B-998 d 336´ 534 602 613 645 Bas III 293 IV 1192 Chr XV 684 DamMM; ουκ εμπλησθησονται 125 547 2:8 μοι 1° Chr IX 683] με 125 548*vid; > 338 Chr IV 599 2:8 οἰνοχόον Ol Syh = 𝔐 Gra] οινοχοους A Sc O-411 574-798c-cII L–125 d–357 k(–631) 260 296´ 311 336´ 443 547 613 645 698 705 706 752c 766B 795 Chr IV 599 IX 683 XVII 123 879 Didlem 40,25 41,18 Didcom 42,3 GregNy 347,6 Met II.1,12 PsAug SermCai 94,2 Ruf Cant 94,25 Arm Fa1 SaI La94 95 Hi; οινοχοω 125 4:1 γινομένας (γειν. Sc V) B-Sc O-411vid 157-299-563-571*-609-797-cII L–125 130 d k (46s) 125II 155 248´ 252 260 296´ 311 338 339 542 543 547 549 698 728 766 Chr VI 606 Dam Met III.26,3 Ol–Α = Compl] γενομενας S* A 998 C´–157 299 563 571* 609 797 125-130 68 336´ 443 534´ 602 645 706 795 Antioch 1560 = Ald 4:11 om καί γε 261 Chr IX 385 5:11 τοῦ δούλου] τω δουλω 261 Antioch 1805 Chr II 585 III 44 Tht IV 680te = Pesh 7:5 τό] του 998 V 390-601 261 248´ 336´ 543 645 698 OlΓΗ (sed hab Antioch 1681 Chr II 1055 Did 202,1 Ol–Γ) The evidence shows that when L is divided Chr agrees most often with MS. 261. This fact is true for cases where Chr agrees with CT (2:8, 4:1) and for cases where there is no agreement with CT (2:8, 4:11, 5:11). The one exception is 7:5. The variant at 5:11 has the support of three Antiochian fathers, MS. 261, and the Peshitta—a peculiar collection of witnesses!17 17  Brock gives three possible explanations for agreements between the Peshitta and L: 1) dependence of the Peshitta on L, 2) revision of the Peshitta based on L, and 3) Lucian or a later L-revisor made sporadic use of the Peshitta. In the first explanation, the Peshitta would be a

126

The Text History

Conclusion18 The evidence shows that Chr preserves Lucianic readings. The forty-two places where Chr reads against CT and L probably includes Lucianic variants which L has failed to preserve. The impression from the evidence is that L does not preserve the Lucianic Recension as well as Chr does. This fact is not surprising.

Antiochus (i.e. Church Father)

Antiochus preserves some 150 variants. It preserves many unique readings, but absence of Greek witnesses makes it likely that Antiochus preserves free citations in such cases. Examples include: 2:26 αὐτοῦ] του ανθρωπου Antioch 1472 VA 4:9 αὐτοῖς] + ο Anton 1108 Antioch 1676 Dam–IvA M 4:9 ἀγαθός] > 342-754 Antioch 1676 DamM VA 4:9 ἐν μόχθῳ αὐτῶν] > Antioch 1676 DamA 11:8 μνησθήσεται] μνησθητε Antioch 1609 OlΕ When Antioch agrees with L most of the readings are accompanied by other text groups. Examples, including important unique agreements with L, include: 1:13 οὐρανόν] ηλιον Sc O L C´–571 798 161mg-248mg-252 339 443 543 547 549 Antioch 1469 alii contra 𝔐: cf 31b 4:9 αὐτοῖς] et μισθός tr L–(125) 130 Antioch 1676 OlΝ 5:4 σε 2°] σοι 534; > L–106 (125) 798-cII–260 k 155 338 443 795 Antioch 1760 alii 5:5 χειρῶν σου] pr των V-411 L(–125) C´’–157 260 d–254´ 68´ 155 248´ 296´ 311 336´ 443 542 547 549 706 Antioch 1688 Ol–ΓΕΖ PsChr 5:11 ὕπνος] pr ο V-411 L(–125) 139-147-159-425mg-503-560-798-cII–260 298 311 M 534 547c pr m 613 645 Anton 1065 Antioch 1805 Dam–M Didcom 153,6 Nil 468 Ol PsChr Tht IV 680 (sed hab Chr II 585 III 44 Met V.2,2) 7:2 ἢ ὅτι πορευθῆναι] om ὅτι 411 L(–125) C´’ 357-754 k 338 339 443 539 543 547 W P 549 766A Antioch 1709 Chr III 155 156 VI 487 XIII 334 Cyr IV 364 DamVV M c

c

witness to proto-Lucianic readings. Brock believes that the evidence in his corpus, 1 Samuel, supports the third explanation as most likely, though not conclusively (Brock, The Recensions of the Septuagint Version of I Samuel, 205-206). 18  Places where Chr agrees with both the Edition and a unified L group have been excluded from analysis, though there is overlap between these variants and the ones discussed in List 4 and List 5. The places of variation are 1:2 (ματαιότης 2°—ματαιότης 3°), 1:9 (τί τό 1°), 2:4 (οἴκους), 2:6 (ποτίσαι), 2:8 (καί 2°), 2:11 (τά), 3:2 (τοῦ 2°), 4:3 (τοὺς δύο), 4:12 (ἀπορραγήσεται), 7:2 (ἀγαθόν 1°), 7:2 (οἶκον 1°), 7:2 (οἶκον 2°), 7:2 (πότου), 7:3 (ἀγαθόν), 7:3 (καρδία), and 8:1 (αὐτοῦ 1°).

The Lucianic Recension

127

Metlem et com VI.4,2,20 OlΑΒΓΖΗ PsChr Tht II 844 An Scrip 1,19 Aug C D 17,20 Spec 528,7 Arm Fa1 2 SaI La160 Syh = 𝔐 7:17 πολύ] πολλα L 728 Antioch 1553 9:8 ἱμάτιά] pr τα C S O-411 L cII d k 296´ 311 339mg 359 542 543 613 645 698 706 766 795 Antioch 1713 9:11 χάρις] pr η L C´ 543 549 613 766A Antioch 1740 DamR Syn 349 = Ald 10:8 ἐν αὐτῷ] εις αυτον B-68´998 637-411 L C´–609-371 d k Antioch 1588 alii 10:8 ἐν αὐτῷ] post ἐμπεσεῖται tr O–V-411 L 443 Antioch 1588 11:9 ἐν κρίσει] εις κρισιν L C´ 161-248c 359 Antioch 1485 Anton 960 1057 1208 alii In the list above, Lucianic variants may be found in 1:13, 4:9, 5:11, 7:17, 10:8, and 11:9. Antioch rarely reads with the Critical Text against L (7 times). In such cases, MS. 443 most often reads with the Critical Text and Antioch (4 times). Agreements with single manuscripts of L are rare. The most prominent one, which is Lucianic, is 5:11 (τοῦ δούλου] τω δουλω 261 Antioch 1805 Chr II 585 III 44 Tht IV 680te = Pesh). There are some agreements with O and preservation of hexaplaric readings. The list below shows that the variants are significant: 4:1 δάκρυον] δακρυα Hi Antioch 1560 = Vulg ↓ II App: σʹ καὶ ἰδοὺ δάκρυα 248´ 5:12 αὐτοῦ 2°] αυτω B-68 O–637 539 543 Antioch 1468 alii 7:4 καὶ καρδία = 𝔐] οἱ γʹ ※ καὶ ↙ καρδία δέ Syh ↓ ; και καρδια δε 252 ↓ ; καρδια PM δε O-411 cII 534´ Antioch 1709 Chr II 1055 DamKVRM T Met VI.7,3 SaI (sed hab Fa1) II App οἱ γʹ καὶ (sub ※ Syh) καρδία δέ 252txt Syhtxt 7:16 ἐκπλαγῇς] εκκλινης O (‑κλεινης V) Antioch 1576 SaI 3 (sed hab Fa2) 9:8 κεφαλήν A B C 637 (pr την) k 130 68 252 296´ (pr την 548) 311 336 543 549 613 766A (pr την) 795 = Ra] κεφαλης S O–637-411 C´’–609 (pr τη 522*; c pr m) d (pr της 254´) 155 248´ 338 339 359 534 542 547 645 698c 728 Antioch 1713 alii 10:18 om ἡ 2° O–V 254´ 338 Antioch 1548 Dam GregNy 110te PsChr 11:10 θυμόν] παροργισμον 253 Antioch 1505 From analysis of the list above, it is possible to conclude that Antiochus has preserved the Lucianic Recension where the L Group has failed to do so, but this is not certain. Similarly, stylistic and grammatical improvements abound in Antiochus, but they cannot be declared Lucianic with certainty. Many of the variants are unique readings or are accompanied by witnesses like B-68´ 998, especially in the last half of the book. As a witness, Antiochus defies classification because of its many unique readings, conflicting citations of the same texts, and what appear to be free citations. Nonetheless, Antiochus preserves Lucianic variants, especially in its agreements with L and O.

128

The Text History

V.  The Catena-Text Critical Editions A. Labate, Catena Hauniensis in Ecclesiasten, CCSG 24, Leuven, 1992. S. Leanza, Procopii Gazaei Catena in Ecclesiasten necnon Pseudochrysostomi Commentarius in eundem Ecclesiasten, CCSG 4, Leuven, 1978. S. Leanza, Procopii Gazaei Catena in Ecclesiasten, Supplementum, Un nuovo testimone della catena sull’Ecclesiaste di Procopio di Gaza, il cod. Vindob. theol. gr. 147, CCSG 4 Suppl., Leuven, 1983. S. Lucà. Anonymus in Ecclesiasten Commentarius qui dicitur Catena Trium Patrum, CCSG 11, Leuven, 1983.

Literature FS Ardizzoni = E. Livrea und G. A. Privitera, Studi in onore di Anthos Ardizzoni, Roma 1978. M. Faulhaber, Hohelied-, Proverbien- und Prediger-Catenen. In Theologische Studien der LeoGesellschaft ed. Ehrhard u. Schindler, 4, Wien, 1902. G. Karo u. Io. Lietzmann, Catenarum graecarum catalogus, Nachrichten der K. Gesellschaft der Wissenschaften zu Göttingen, Philol.-hist. Klasse, 1902, 1-66, 299-350, 559-620. Also distributed / published separately with page enumeration 1-180 in addition to the original pagination. A. Labate, Nuovi codici della catena sull’Ecclesiaste di Policronio, Augustinianum 18 (1978), 551-553. A. Labate, L’esegesi di Evagrio al libro dell’Ecclesiaste, FS Ardizzoni, 483-490. A. Labate, La catena sull’Ecclesiaste del cod. Barb. gr. 388, Augustinianum 19 (1979), 333-339. A. Labate, Le catene sull’Ecclesiaste del cod. Collegio gr. 16. In Κανίσκιν, Studi in onore di Giuseppe Spadaro, hrsg. Anna Di Benedetto Zimbone und Francesca Rizzo Nervo, Rubbettino, 2002 (Medioevo Romanzo e Orientale, 12), 183-194. A. Labate, L’apporto della catena Hauniense sull’Ecclesiaste per il testo delle versioni greche di Simmaco e della LXX, Rivista Biblica 35 (1987), 57-61. A. Labate, Sulla catena all’Ecclesiaste di Policronio, Studia Patristica 18,2 (1989), 21-35. S. Leanza, Le catene esegetiche sull’Ecclesiaste, Augustinianum 17 (1977), 545-552. S. Leanza, La catena all’Ecclesiaste di Procopio di Gaza del Cod. Marc. gr. 22 (ff. 67v-83r). In Studia codicologica, hrsg. K. Treu (Texte und Untersuchungen zur Geschichte der altchristlichen Literatur, 124), 279-289. S. Leanza, Ancora sull’esegesi origeniana dell’Ecclesiaste, FS Ardizzoni, 491-506. S. Leanza, Il commentario sull’Ecclesiaste di Dionigi Alessandrino: Scritti in onore di Salvatore Pugliatti, V, Milano, 1978, 397-429. S. Leanza, L’Esegesi di Origene al Libro dell’Ecclesiaste, Reggio Calabria, 1975, 10-20. S. Leanza, Sul Commentario all’Ecclesiaste di Didimo Alessandrino, Studia Patristica 17,1 (1982), 300-316.

The Catena-Text

129

S. Lucà, L’esegesi di Nilo di Ancira sul libro dell’Ecclesiaste, Sileno 3 (1977), 13-39. S. Lucà, La catena dei 3 Padri sull’Ecclesiaste, FS Ardizzoni, 555-582. S. Lucà, Gli scolii sull’Ecclesiaste del Valliceliano gr. E 21, Augustinianum 19 (1979), 287-296. A. Rahlfs, Verzeichnis der griechischen Handschriften des Alten Testaments, Nachrichten der K. Gesellschaft der Wissenschaften zu Göttingen, Philol.-hist. Klasse, 1914, Beiheft (= MSU 2), Berlin, 1914.

1.  The Various Catenae of Ecclesiastes and the Catena-Groups Catena MSS. are MSS. which usually present the biblical text and, either around the biblical text or following it, comments appropriate to that text excerpted from different patristic writers, chained or linked together. Hence the name catena. More than one type of catena manuscript tradition exists for Ecclesiastes; indeed, the catena traditions for Ecclesiastes are complex and rich. Beginning from information in Rahlfs’ Verzeichnis where manuscripts are either classified as “Cat.” or not, the following chart shows the catena type of each MS. using the research of Faulhaber, Karo/Lietzmann, and more recent studies of the catena traditions in Ecclesiastes. Another chart follows reversing the orientation and listing catena manuscripts according to the text groups to which they attest or witness. Legend for Columns in Chart A = Bible Text B = Collated by Septuaginta-Unternehmen C = Classified by Faulhaber or Karo/Lietzmann or the Later Research D = Dated before 1500 E = Dated after 1500 Y = Yes N = No The Catena Types 1 = The “3 Father”-Catena (VII. Cent.) 2 = The Procopius-Catena (VI. Cent.) 3 = The Polychronius-Catena (VI. Cent.) 4 = The Olympiodorus-Catena (VI. Cent.) 5 = The Hauniensis-Catena (Mixed Catena Type) 6 = The Sinaitica-Catena (Mixed Catena Type) 7 = The Barberini-Catena (MS. 645—Mixed Catena Type)

130

The Text History

Catena Manuscripts Listed by the Type of Catena 139 147 149 159 260 295 297 299 336 338 342 348 352 353 371 386 390 411 415 436 437 464 471 47520 485 486 503 504 522 539 540 560

Y Y Y Y Y Y N Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y

A

B Y Y Y Y Y CCSG11 Jaeger Y Y Y Y N N N Y Y Y Y Y N N N CCSG2419 Y N21 N Y Y Y Y Y Y

C 3 3 5 N 5 1 Gr-Nyss 6 3 3 1+5

D X/XI XII XI XII X/XI XV XII 1202 XIV XIV XIII

1 4 = OL Δ 3 3b

XIII/XIV 1075

5 O 3a 3 3 3 3 3a 3 3

XV/XVI

XIII/XIV XIII/XIV XII/XIII XIII XI XIII XIII

E

XVI 1540 1540 XVI XVI 1556 XVI 1550 1549 XVI XVI

19  Copy of 260. 20  This MS. contains an unidentified Commentary on Ecclesiastes, but is certainly not a Catena MS. 21  Sister MS. of 464; related to MS. 563.

131

The Catena-Text

561 562 563 571 573 574 632 645 720 732 733 733II 734 752 754 755 798 3007 3011 3012 Ber. Phill. gr. 1422 Ber. Phill. gr. 1484 Bibl. Casanat., 203 Bibl. Vallic. 51 Mos. Syn. B. gr. 15 Ottobon. gr. 75 Ottobon. gr. 112 Vat. gr. 1521 Yale U. Lib. g. 273

Y Y Y Y Y Y Y Y Y Y N23 Y Y Y Y Y Y Y Y Y Y Y Y Y Υ Y Y Y Y

Y Bo. / Zi. Y Y Boli (Y: 1-2) N Y N22 Y CCSG 4 Y Y Y Y N Y N Y N N N N Y Υ Boli N N N

1 4 = OL Γ 3a 3a 4 = OL Β 3 3 4a? 3 3 2 4 = OL Α 4 = OL Ζ 3 2 ? 2+5 624 ? 4+?

XIII XI/XII XII XI XIII X/XI XII XII X XI XIV XIV XV

1 4 = OL 4 = OL Μ 1023 4 = OL Θ 1 3 3

XVI XVI XVI 1573

XVI XVI 1510 XVI/XVII XVII XVI XVI XVII XVI XVII 1583

Boxes are placed around manuscripts incorrectly classified as catena manuscripts (e.g. manuscripts of Olympiodorus’ Commentary on Ecclesiastes) and around manuscripts whose date is late and hence are not used. A few manuscripts whose date is after 1500 are used because the textual witness is needed for a particular type of catena.

22  Daughter of MS. 732. 23  The lemma is the Bible text of Olympiodorus Commentary = 733II. 24  Descendant of MS. 299.

132

The Text History

Catena Manuscripts Listed by Text Groups 1. O

= V-253-475-637 + 411

Jerusalem, Patr.-Bibl., Τάφου 370 XVI Moscow, State Hist. Museum, Syn. Bibl., Gr. 355 XIII/XIV 2. C

411 475

3b +Com

= 139-147-159-299-390-415-503-504-522-540-560-563-571-574-732-798

Milan, Bibl. Ambr., A 148 inf. Venice, Bibl. Marc., Gr. 21 Genoa, Missione urb., 2 Moscow, RGADA, Fonds 1607, Inv. 1.22 Oxford, Bodl. Libr., Laud. gr. 30 (A) Rome, Bibl. Vat., Vat. gr. 1694 Paris, Bibl. Nat., Gr. 154 Oxford, Bodl. Libr., Auct. E. 2. 16 Cambridge, Trinity Coll., B. 7. 3 Paris, Bibl. Nat., Coisl. 194 Paris, Bibl. Nat., Gr. 151 Rome, Coll. gr. 16 Leiden, Univ.-Bibl.; Vulc. 50 Oxford, Bodl. Libr., Auct. E. 2. 17 und 18 Paris, Bibl. Nat., Gr. 172 Paris, Bibl. Nat., Gr. 176

X/XI X/XI 1075 XII XII 1202 XII XII/XIII XIII XIII XIII XIV XV/XVI XVI XVI XVI

139 732 390 159 147 299 563 503 522 540 560 798 415 504 571 574

3 3 3 3 3 6 3a 3 3 3 3 2+5 ? 3 3a 3

MS. 299 deviates frequently from the C Group; it is a mixed catena. 3. cII

= 295-371-561-752(ab 312) (The “Three-Father”-Catena = CCSG 11)

Paris, Bibl. Nat., Gr. 152 Rome, Bibl. Vat., Ottob. gr. 56 Rome, Bibl. Casanat., 203 Rome, Bibl. Ottobon. gr. 112 Escorial, Real Bibl., R-I-3 Vienna, Nat. Bibl., Theol. gr. 115

XIII XV XVI XVI XVI XVI

561 295 636 — 371 752

4. 260-149-471 Cat. Hauniensis = CCSG 24 Copenhagen, Kgl. Bibl., Gamle Kgl. Saml. 6 X/XI Vienna, Nat. Bibl., Theol. gr. 11 XI Moscow, State Hist. Museum, Syn. Bibl., Gr. 147 XIII/XIV

260 149 471

5 5 5

133

The Catena-Text

5. No Biblical Text—Procopius Cat. = CCSG 4 Venice, Bibl. Marc., Gr. 22 6. d

= 254-342-357-754

XII 254´

Vienna, Nat. Bibl., Theol. gr. 147 Athens, Hist. and Ethn. Society Mus., 200

733

2

754 342

2 1+5

Ζ Μ Β Γ Α Δ

4 4 4 4 4 4

Ε Κ Ι Η

4 4 4 4 4

= 254-754 XI XIII

7. Olympiodorus Commentary (These are not catena MSS.) Venice, Bibl. Marc., Gr. 23 Moscow, Syn. B. gr. 15 Paris, Bibl. Nat., Gr. 174 Paris, Bibl. Nat., Gr. 153 Venice, Bibl. Marc., Gr. 22 Florence, Bibl. Laur., Plut. X 29

X 1023 XI XI/XII XII XIII/XIV

734 — 573 562 733II 386

Also Olympiodorus Commentary MSS. Genoa, Durazzo - Giustiniani A.I.10 Athos, ᾽Ιβήρων 614 (4734) Athos, Λαύρα 278 (Γ 38) Paris, Bibl. Nat., Gr. 175 Rome, Vallic. gr. 67 (E 21)

IX / X XIII XIV XIV XIV

3011

Precise description of the work of Olympiodorus is difficult: should it be classified as a catena or as a commentary? His work lies, in fact, on the boundary of both categories. Certainly he copied materials from all earlier commentaries known to him. Nonetheless, the earlier materials are forged together into a commentary. Here his work is classified as a commentary, but this problem explains that some manuscripts are listed in Rahlfs’ Verzeichnis under the rubric “Cat.” Olympiodorus Manuscripts Too Late For Use Rome, Coll. gr. 7 Rome, Bibl. Vallicelliana, gr. 51 Rome, Bibl. Vat., Ottob. gr. 75

XV XVI XVII

3012 Θ

4 4 4

134

The Text History

8. Codices mixti (336´ = 336-728) Paris, Bibl. Nat., Coisl. 193 Rome, Bibl. Vat., Barber. gr. 388 Athos, ᾽Ιβήρων 555 Athos, ᾽Ιβήρων 676

XI XIII XIV XIV

539 645 336 338

3a 4a? 3 2

Discussion with Detlef Fraenkel, May 30, 2008 Detlef analysed MS. 539 and believes this manuscript represents the earliest form of the catena tradition which served as the basis for both the Procopius Catena and the Polychronius Catena. 9. Gregory of Nyssa—Homiliae in Eccl (See Edition by Jaeger) Rome, Bibl. Vat., Vat. gr. 1802

XII

297

10. Copies, Derivatives, and Manuscripts Too Late For Use Vienna, Nat. Bibl., Theol. gr. 11 Moscow, Syn. Bibl., Gr. 147 Sinai, Cod. gr. 311 Berlin, Staatsbibl., Phill. 1411 Berlin, Staatsbibl., Phill. 1412 Modena, Bibl. Est., Gr. 155 Munich, Staatsbibl., Gr. 131 Madrid, Bibl. Nac. 4749 Turin, Bibl. Naz. B. II. 8 New Haven, Yale University Libr. gr. 273 Basel, Univ.-Bibl., A. VII. 6 Madrid, Bibl. Nac. 4781 Munich, Staatsbibl., Gr. 292 und 294 Rome, Bibl. Angel., Gr. 113 Vienna, Nat. Bibl., Theol. gr. 199 Rome, Vat. gr. 1521 Rome, Bibl. Casanat., 203 Rome, Bibl. Ottobon. gr. 112 Berlin, Staatsbibl., Gr. 1422 Berlin, Staatsbibl., Gr. 1484 Rome, Bibl. Ottobon. gr. 75

XI XIII/XIV 1510 1540 1540 1550 1549 1556 1573 1583 XVI XVI XVI XVI XVI XVI XVI XVI XVI/XVII XVII XVII

149 471 3007 352 353 464 485 436 720 — 348 437 486 632 755 — 636 — — — —

5 6

3a 3 3 3 3 3

4

The Catena-Text

135

2.  The Catena Groups The term Catena Group is used in the Göttingen Septuaginta for the text type of the Bible text of the catena MSS. In any particular catena manuscript, the text type of the Bible text and the textual tradition of the catena itself may not match. Normally, however, when the Bible text of the catena belongs to the same text type, the textual tradition of the catena will also be the same. C´’ C = 139-147-159-299-390-415-503-504-522-540-560-563-571-574-732-798 cI = 157-425-601-609-797 157´ = 157-797 cII = 260-295(ab 312)-371-561-752(ab 312) C´ = C + cI C´’ = C + cI + cII

The siglum cII is used for sub-types of catena traditions in Ecclesiastes that are related to the main C Group, although they have a different place in the textual tradition: MS. 260 is the witness for the Catena Hauniensis (so named because the MS. is in Copenhagen) and 295-371-561 are used as the witnesses for the socalled Catena Trium Patrum (the Three-Father Catena). cI = 157-425-601-609-797 These five MSS. 157 425 601 609 797 form a Group that is designated as cI; they are, to be sure, not catena manuscripts, but they are affiliated with C. No doubt they originated by someone copying only the Bible text of a catena manuscript and omitting the catena part. Examples: 3:19d οὕτως] ουτω cI-260 OlΝ 9:12a καί γε] + τουτο C-cI (=157-425-601-609-797) = C´ 10:14b γενόμενον] γενησομενον C-cI = C´ C

= 139-147-159-299-390-415-503-504-522-540-560-563-571-574-732-798

C designates the main group of the catena manuscripts. These sixteen manuscripts all transmit the Polychronius Catena. The siglum C refers to the text type of the Bible text in the Catena-MSS.

136

The Text History

Examples: txt

c

1:6a ἐκεῖ πορεύεται] εκπορευεται C´(–299B 798 ) 543 549 698 Cos 5,33 = Ald mg 2:3b ἑλκύσει] εκλυσει C –139 299 798*-cI–425 797 GregNy 311,16ap = Ald c 2:10b οὐκ ἀφεῖλον] ουκ απειχον (απει 601) C´–299 798 c 2:10f μερίς μου] μερισμος C´–299 798 = Ald c c 2:10g om ἀπό C´–571 798 2:12d σὺν τὰ ὅσα] παντα οσα C´–299 = Ald 2:20a ἀποτάξασθαι] αναταξ. 411c C´–299 574 = Ald 2:21c om ἐν αὐτῷ C´–299 Hi = Ald c 3:14d ἀφελεῖν] αφαιρειν (αφερειν 609 Ald) C´–563 571 = Ald 3:20b ἀπό] εκ C´’ 339 543 = Ald 3:20b τὸν χοῦν] om τόν 411 C´Did 103,2(vid) = Ald 3:21b om αὐτό C–299 563 571-425-601 311 542 PsChr Arm Fa2 = Ald 4:1g καί 6°] ∩ (2a) καί 411 C´’–260 299 547 = Ald 5:4b τὸ εὔξασθαί σε καί] ευξαμενον 411 C´ Chrys IX 597 = Ald 5:18c ἀπ᾽ αὐτοῦ] απ αυτων C´–601 Hi Armte = Ald Vulg 7:26b θηρεύματα] θηραμα 411 C–299 563 571*-159-425-601 260c = Ald 8:15d  συμπροσέσται] συμπορευσεται 411 C–147 159 563 571*-425-601 613 = Ald ↓ ; συμπορευεται 147-159 Arm 8:16d om ὅτι C´ = Ald 8:17d τοῦ ζητῆσαι] του ευρειν C´–609 = Ald: ex (17h) 9:1e εἰδώς] post ἄνθρωπος tr S C´–(157) 299 (503) 645 = Ald 9:9b ματαιότητός 1°] νεοτητος C–(298): ex Prov 518b 9:15b αὐτός] αυτοις C´OlΗ = Ald 10:14b γενόμενον] γενησομενον C´-298: cf 19a 12:8c μνησθήσεται] ‑σονται C´ 125 357 766

Noteworthy is the fact that MS. 299 deviates frequently from the main Group. Cases also can be found where the main Group is divided over a particular reading. Examples:

2:9a 2:26e 3:17a 4:14a 5:19a

καί γε] γε V 139mg-147-159-299-503-560-563-571-798 cI–425 601 547* Fa1; γεγονε 390-415-425-504-522-540-574-601-732 = Ald προσέθηκα] + σοφιαν C´–147 159 503 560 = Ald προσθεῖναι] πορευθηναι 390-415-504-540-574-732-425-601 καρδίᾳ μου] pr τη 637-475 390-415-425-504-540-601-732 OlΒΗΝ Co = Ald ἐξελεύσεται] απελευσεται 390-415-425-504-540-601-732 ὅτι οὐ πολλὰ μνησθήσεται] om οὐ C´–157 299 563 571 609 797

cII

= 260-295(ab 312)-371-561

2:1c

137

The d-Group

The three MSS. 295 371 561 form a group, the co-called Three-Father Catena, that is designated as cII. MS. 260 is our only source for the Catena Hauniensis (so named because the MS. is in Copenhagen). There are more agreements between cII and 260 in the apparatus than disagreements, so 260 is for this reason included in the cII Group.

3.  Text-History of the Catena-Groups The following genealogical tree of the catena traditions in Eccl was produced by Faulhaber in 1902:

3 Väter-Catene

Polychronius-Catene

Prokop-Catene

Olympiodor-Catene

Hs. 645

Faulhaber explains: Die » 3 Väter «-Catene bildet eine eigene Überlieferungsschicht. Aus der ProkopCatene als der fruchtbaren Muttercatene ist decrescendo der Polychroniustypus und crescendo der Olympiodortypus und indirect, nämlich durch die OlympiodorCatene, die Rension des Barber. III. 107 [= 645] hervorgegangen (p. 166).

VI. The d-Group d 254´

= 254-342(fin 729)-357(fin 1211c οἳ παρά)-754 = 254-754

These four MSS. 254 342 357 754 form a group, that is designated as d.

138

The Text History

Examples of Agreement in Significant Variation from Original Text: 216c 320b 513b 514b 76c 713b 715a 101b 1014d 114a 1110a 128a

ταῖς ἡμέραις] om ταῖς d ἀπὸ τοῦ χοός] υπο του χοος d χειρί] pr τη 475 d ἥκει = 𝔐] ην 475 d init] pr ὅτι d κοσμῆσαι] επικ. d Didcom Hilem et com ματαιότητός μου = 𝔐] νεοτητος μου d ἀφροσύνης = 𝔐] αφρονος d ὅ τι] διοτι d ἄνεμον] ανεμους d Met IX.22,2 Arm ἀπόστησον] αποστρεψον d εἶπεν ὁ ᾽Εκκλησιαστής, / τὰ πάντα ματαιότης = 𝔐] tr d

Among the variants 7:15a can be highlighted: against 𝔐 d had ἐν ἡμέραις νεότητός μου in dependence upon 11:9b and 12:1b ἐν ἡμέραις νεότητός σου. MS. 357 In approximately 52 instances, MS. 357 deviates from the d-Group. In many of these cases of deviation, 357 is less recensional than the other witnesses of the dGroup. In fact, MS. 357 is one of our best single witnesses to an early independent branch of the genealogical tree. MS. 357 also has many unique readings. As usual, these are of little value, but at 119c it attests the original text with 542. Examples: 28e 218b

οἰνοχόον] οινοχοους A Sc O-411 L–125 574-798c-cII-752c k (–631) d–357 alii μοχθῶ] κοπιω B-68´ 998 O d–357 336´ ↓

48d 713c 119c

πλούτου] δωρων 698 (δωρον); δολου 357 ὃν ἂν ὁ θεός] ο ανθρωπος 357  ἄμωμος ↙ Syh] om 357 542 Hilem et com 348,156 351,68 = Ge Ra Vulg

VII. The k-Group

k 46´

= 46-337-631 = 46-631

The Manuscript Pair 248´

139

These three MSS. form a group that is designated by the siglum k. Examples: 49b

411a 416c 53c 55b 83b 83c

ἐν μόχθῳ αὐτῶν] υπερ τον ενα k: ex (9a)

αὐτοῖς] pr εν 46´ 337 om 46´ 337 θέλημα] φρονημα 46´ 337 ἄγνοια] αγνοημα 601 k Dam (3b) post (3c) tr 46´ 337 ὅτι] το k

This peculiar recension is a derivative of the early Egyptian Group. Examples: 113c 117b 117d 23e 219c

γενομένων] γινομενων B-534´ 411 L cII d k 336´542 645 698 Hi Syh παραφοράς] παραβολην k Fa1 2 SaI ὅτι] pr εγω B-S-534´ 998 C´’ d k 336´ 411 645 705 Geo Ol τοῦ ἀνθρώπου] των ανθρωπων B-S-68´’ 998 O–V 253-411 C´’ d k alii ᾧ 2°] absc 998; > B*-S* 411 L C´’–260 797 k 155 336´

VIII.  The Manuscript Pair 248´ 248´ = 161-248 For Canticum, Reinhart Ceulemans believes that 161 is a copy of 248. His reasons are as follows: (1) 161 and 248 are extremely similar, which shows that they are related; (2) 161 has all of the information from 248, plus additional errors (e.g. sigla that are given incorrectly in 161, but correctly in 248; omissions of information that is provided correctly in 248); (3) 248 on the other hand has no unique readings (i.e. errors that are not also present in 161). This indicates that 161 is a copy of 248. The page layout confirms this: in 161 space was reserved for the hexaplaric and other readings, which in 248 were added in the margins and in between the lines of the main text. In 161, everything is moreover written by one and the same hand, whereas in 248 the hexaplaric and other readings were added later on (either by the original hand or by another). The evidence for Ecclesiastes is more complicated and inconclusive. There are errors in 248 not in 161. There are hexaplaric readings in 161 not in 248. MS. 248 seems to have positioned the indices for the marginal readings better than 161. There are places where 248 agrees with the corrector of 161. Without exhaustive analysis, it appears prudent to leave the readings for both manuscripts in the apparatus. Nonetheless, the manuscript pair is designated 248´ so that the number 248 is more visible in the apparatus in connection with its relation to Compl.

140

The Text History

IX.  The Codices Mixti (al) MS. 411 MS. 411 is a late manuscript where the scribe used a manuscript from the hexaplaric group to correct his Vorlage. This resulted in a mixed textual tradition and hence the manuscript is classified as a codex mixtus. A chart summarises the relations of 411 to other witnesses: 411

=

(Nearly) Unified Text Group

411

=

O 184 (21 Singular Agreements) L 152 C´ 147 cII 193 d 120 k 133

411

≠

(Nearly) Unified Text Group 70 (36 Singular)

411

= =

O (When Group Not Unified) V 20 253 7 475 10 637 20

Although 411 agrees with cII more frequently than with O, 411 shares 21 variants with O alone, and only one with cII alone.

X.  Individual Manuscripts—Unclassified Some individual manuscripts are so fragmentary, or the extant and legible portion of the text is so scant that they cannot be properly classified or related to the textual tradition. The evidence is inadequate to permit any meaningful placement genealogically in the history of the textual tradition, even as codices mixti. Here we would list the fragmentary papyri (818 870 969 992) and the burned palimpsest majuscule 3010.25 25  Based on analysis of the entire manuscript, F. Albrecht believes the text of 3010 is Lucianic. See F. Albrecht, “Palimpsesthandschriften der griechischen Weisheitsbücher in Majuskelschrift,” in E. Bons, et al., eds., Die Septuaginta. Themen, Manuskripte, Wirkungen (WUNT 444),

The Later Greek Translators/Revisors

141

XI.  The Later Greek Translators/Revisors Literature A.M. Ceriani, ed., Codex Syro-Hexaplaris Ambrosianus photolithographice editus (Monumenta sacra et profana, 7; Milan: Typis et impensis Bibliothecae Ambrosianae, 1874). Peter J. Gentry, “Hexaplaric Materials in Ecclesiastes and the Rôle of the Syro-Hexapla,” Aramaic Studies 1 (2003), 5-28. Peter J. Gentry, “Propaedeutic to a Lexicon of the Three: The Priority of a New Critical Edition of Hexaplaric Fragments,” Aramaic Studies 2.2 (2004), 145-74. Peter J. Gentry, “The Relationship of Aquila and Theodotion to the Old Greek of Ecclesiastes in the Marginal Notes of the Syro-Hexapla,” Aramaic Studies 2.1 (2004): 63-84. Peter J. Gentry, “‘The Role of the “Three” in the Text History of the Septuagint’: II. Aspects of Interdependence of the Old Greek and the Three in Ecclesiastes,” Aramaic Studies 4.2 (2006): 153-192. K. Hyvärinen, “Die Übersetzung von Aquila” (ConBOT, 10; Lund, 1977). J. Jarick, ‘Aquila’s Koheleth’, Textus 15 (1990): 131-139. P.S. Marshall, “A Critical Edition of the Hexaplaric Fragments of Ecclesiastes” (Ph.D. diss. The Southern Baptist Theological Seminary, 2007). J.D. Meade and P.J. Gentry, “Evaluating Evaluations: The Commentary of BHQ and the Problem of ‫הֹוללֹות‬ ֵ in Ecclesiastes 1:17,” in Gillian Bonney and Rafael Vicent (eds.), SOPHIA— PAIDEIA: SAPIENZA E EDUCAZIONE (Sir 1,27): Miscellanea di Studi offerti in onore del prof. Don Mario Cimosa, Nuova Biblioteca di Scienze Religiose 34, Rome: Libreria Ateneo Salesiano, 2012), 197-217. Yun Yeong Yi, “Translation Technique of the Greek Ecclesiastes” (Ph.D. diss. The Southern Baptist Theological Seminary, 2005).

1.  Introductory Observations The Greek translation of Qohelet incorporated into the body of texts loosely termed “the Septuagint” is characterised by extreme formal equivalence. It is debated, in fact, as to whether the translator was Aquila or not.26 As one might presume, scholiasts who selected readings in variation from the LXX supplied not only a fair number from Symmachus, whose renderings are more oriented to ­ übingen: Mohr-Siebeck, 2020, 501-514, esp. 505-507 and F. Albrecht and C. Faraggiana di SarT zana, “A Carbonized Septuagint Palimpsest of the Libri sapientiales in Biblical Majuscule, Codex Taurinensis, Biblioteca Nazionale Universitaria, C.V.25 (Rahlfs-Ms. 3010): Its Text and Context,” in: C. Rapp, et al., eds., New Light on Old Manuscripts. The Sinai Palimpsests and Other Advances in Palimpsest Studies (Veröffentlichungen zur Byzanzforschung), Vienna: Verlag der Österreichischen Akademie der Wissenschaften, 2021. 26  Cf. J. Jarick, ‘Aquila’s Koheleth’, Textus 15 (1990): 131-139 and Peter J. Gentry, “The Relationship of Aquila and Theodotion to the Old Greek of Ecclesiastes in the Marginal Notes of the Syro-Hexapla,” Aramaic Studies 2.1 (2004): 63-84.

142

The Text History

the demands of the target language, but also a number attributed to Aquila and Theodotion, whose translations are just about as wooden. Note that almost half of the marginal notes concerning Theodotion simply affirm that his text is the same as the text of the Fifth Column of the Hexapla. Work towards collecting, collating, and analysing afresh the materials for the Second Apparatus of the Göttingen Ecclesiastes and preparation of the edition for the Hexapla Project/Institute have been conducted as a co-ordinated effort. Alongside collations made by the Septuaginta-Unternehmen, Phillip Marshall completed in 2007 at The Southern Baptist Theological Seminary a doctoral dissertation under my supervision that constituted a major (albeit preliminary) step towards the edition for the Hexapla Project.27 Marshall’s analysis and critical edition is important due to his fresh study of materials in the Commentary on Ecclesiastes by Jerome and in margins of the Syro-Hexapla. Nonetheless, research has progressed since his work. When the edition in preparation for the Göttingen Septuaginta was ready for publication in 2013, an important manuscript was discovered by Reinhart Ceulemans, namely 788, which preserved as much in terms of text as all of the other sources together (approximately 213 [frequently longer] readings). Thanks to the kindness of Ceulemans, these materials are included in the Göttingen Edition, but analysis of them must await his own publications. The following description of hexaplaric materials, including that of the Syro-Hexapla, is based on my own research.

2.  The Witnesses of the Later Greek Translators The hexaplaric materials are derived from three types of sources: (1) marginal notes in manuscripts transmitting the biblical text of Ecclesiastes, (2) citations from Church Fathers whose texts are available either in surviving commentaries and exegetical works or in various catena traditions that excerpted these commentaries and exegetical works, and (3) marginal notes of the Syro-Hexapla, a translation of the Tetrapla, a recension derived from the Fifth Column of the Hexapla with translations of many readings from the Three in the margins.

2.1  Marginal Notes in Greek Bible Manuscripts

The following manuscripts are the main sources in Greek for readings of the Three Jewish Revisors and associated hexaplaric notes. The arrangement follows the Rahlfs Numbers and provides library numbers as given in the Einleitung of the Göttingen Edition:

27  Phillip S. Marshall, “A Critical Edition of the Hexaplaric Fragments of Ecclesiastes,” (Ph.D. diss. The Southern Baptist Theological Seminary, 2007).

The Later Greek Translators/Revisors

143

161

Moscow, Russian State Archive of Ancient Documents, Fonds 1607, Inv. 1.7 (earlier Dresden, Sächsische Landesbibliothek, A. 170); XIV. Cent. 248 Rome, Bibl. Vat., Vat. gr. 346; XIII. Cent. 252 Florence, Bibl. Laur., Plut. VIII 27; X. Cent. 336 Athos, ᾽Ιβήρων 555; XIV. Cent. 411 Jerusalem, Patr.-Bibl., Τάφου 370; XVI. Cent. 539 Paris, Bibl. Nat., Coisl. 193; XI. Cent. Excerpts from Ecclesiastes with hexaplaric readings and catena excerpts associated with the respective textual excerpts. 788 Tirnavos, Δημοτικὴ Βιβλιοθήκη 25; X. Cent. (Cat.).

Some characterisation and quantification of these sources is both necessary and useful. Most of the marginal readings are derived from 161, 248, 252, 539, and also 788. The number of readings that are preserved in each source is approximated in a chart: Source 161 248 252 336 411 539 788

Number of Readings 318 328 208 1 3 85 212

MSS. 161 and 248 preserve approximately 318 and 328 readings respectively. These two manuscripts are closely related in their textual tradition. This is indicated not only by congruence in both the number and the text of the marginal notes, but is obvious from the system used in both manuscripts to index the marginal readings. Exactly the same indices and signs are used in both 161 and 248, although the format and pagination of the manuscripts are not the same. 248 is approximately a century earlier and has more notes than 161. Since there is a tendency both in manuscripts with marginal notes and in catena manuscripts for the number of the readings to decrease as copying continues through the centuries, it appears that 248 may be an earlier witness to this tradition. MS. 248 has approximately ten more readings than 161. Nonetheless, no major analysis has been conducted to this point that could prove the exact nature of the relationship between these two manuscripts. It is doubtful that 161 is simply a direct copy of 248 since the placing and the spacing in the indices is more accurate in 161. MS. 252 preserves about 208 readings of the Three. Some 147 of these are in combination with 161 and 248 and a further five with 248 where 161 has no read-

144

The Text History

ing. In approximately 24 instances, MS. 252 is the sole witness of the reading and in approximately 16 further instances, it is joined by 539, Hi or Syh, i.e. to the exclusion of 161 and 248. In the remaining 60 instances, 252 preserves a longer reading of which other witnesses, such as 161 and 248 for example, preserve only part. 252

Combination + 539 + Syh + 539 Syh + Hi + Hi Syh

Instances 1 9 3 2 1

These rough statistics are sufficient to show that in terms of genealogical relationship or textual tradition, 161 and 248 come from a common parent and that this parent and the parent of 252 probably have a common parent in turn:

MS. 252 is tenth century while MSS. 161 and 248 are Fourteenth and Thirteenth Centuries respectively. Thus 252, in spite of copyist’s errors, goes back independently to a parent related to the parent of 161 and 248. As noted previously, there is a tendency to lose readings in the transmission through the centuries. MS. 539 is an odd manuscript. It contains only excerpts of the biblical text of Ecclesiastes, followed by hexaplaric readings without attribution and these in turn are followed by scholia and excerpts from commentaries of the Church Fathers. A chart indicates how the 85 readings preserved in 539 are correlated with other sources: 539

Combination + 161 248 etc. + 252 + 252 Syh + Syh + O Group = Sole Witness

Instances 75 1 3 4 1 1

The Later Greek Translators/Revisors

145

Although a majority of readings preserved by 539 are also in common with 161 and 248, some are in common only with 252 and a few only in common with Syh. Once 539 is the sole witness. To include 539, perhaps the stemma could now be revised as follows:

In approximately 10 instances, 539 preserves only part of the text transmitted by 161, 248 and other sources. MS. 539 may in fact represent the earliest form of the complicated and various catena traditions that have survived for Ecclesiastes.28 A long reading of twenty-one words attributed to Symmachus is preserved solely by MS. 336 amidst scholia of Evagrius of Pontus on Ecclesiastes. Paul Géhin described the note in his published research on the scholia of Evagrius, but did not include the actual text of the reading from Symmachus in his publications: ce manuscrit offre un premier choix de scholies aux folios 246-249 et un second choix aux folios 259v-261. Les deux séries, complémentaires, transmettent d’une façon assez négligée un total de 17 scholies (pour certaines d’une façon incomplète). Dans la première série une leçon hexaplaire portant Eccl. 5,19 est intercalée entre les scholies 44 et 45.29

MS. 411 is sixteenth century and manuscripts after 1500 are not normally used in Göttingen Editions. Nonetheless, this manuscript has readings in common with the O Group, witnessed by only four sources. It is thus an important witness in spite of its late provenance. Felix Albrecht, a resident scholar of the SeptuagintaUnternehmen discovered one hexaplaric reading in the margins and two embedded in the biblical text, also witnessed by the members of the O Group. Some five or so hexaplaric readings are actually embedded in the text of manuscripts in the O Group.

28  Suggested by Detlef Fraenkel in conversation at the Septuaginta-Unternehmen in Göttingen, June 2008. 29  Paul Géhin, Évagre le Pontique: Scholies à l’Ecclésiaste, Sources Chrétiennes 397 (Paris: Les Éditions de Cerf, 1993), 29-30.

146

The Text History

2.2  Patristic Sources

Of all the commentaries on Ecclesiastes by patristic writers only four survived in Greek and one in Latin.30 Three of these contain readings of the Three and along with scholia from a fourth source constitute the evidence found in commentaries and exegetical works. The sixth century Commentary on Ecclesiastes by Olympiodorus preserves 6 readings.31 A commentary attributed to Gregory Agrigentinus, but now recognised as written by Metaphranes of Smyrna (IXX century) preserves 3 readings.32 One reading of Symmachus is preserved in surviving scholia of the fourth century father Evagrius of Pontus.33 Of these ten readings, only three in Olympiodorus are unattested by other sources. In Latin we have preserved a complete commentary on Ecclesiastes by Jerome.34 Jerome explicitly states in his Commentaries on the Psalms and in the Letter to Titus that he went to Caesarea and used Origen’s Hexapla there.35 He made extensive notes from the Hexapla which he used both for his Commentary on Ecclesiastes and for the later translation in the Vulgate. Perhaps he even had copies of hexaplaric texts in his library in Bethlehem. His access to and direct use of Origen’s Hexapla is evident in the Commentary on Ecclesiastes for he frequently cites one or more of the Jewish recensions in the course of his exegesis and explanation of the text and preserves for us some 65 readings in this way. A brief chart shows that the readings preserved by Jerome are independent of the Greek manuscript tradition. More than half of the readings are of course also preserved in Greek in the manuscripts just described. In some 12 cases, however, Jerome preserves a full text and the Greek manuscripts offer only a fragment of the same text. In 4 cases, the only other witness is the Syro-Hexapla. In roughly 17 cases, Jerome is the sole source, but in 9 of these he actually provides the text in Greek and not in Latin.36 Thus few readings remain that require retroversion 30  There is also an unidentified commentary on Ecclesiastes in Ra 475—unedited and unpublished. 31  Theodora Boli, “Olympiodor, Diakon von Alexandria, Kommentar zum Ekklesiastes: Eine kritische Edition,” (Ph.D. diss. Heidelberg, 2004). 32  Gerard H. Ettlinger and Jacques Noret, eds., Pseudo-Gregorii Agrigentini seu PseudoGregorii Nysseni Commentarius in Ecclesiasten (CCSG 56; Turnhout: Brepols, 2007). For the attribution to Metaphranes of Smyrna, see Peter Van Deun, “La Chasse aux Trésors: La Découvert de Plusieurs Œuvres Inconnues de Métrophane de Smyrne (IXe-Xe Siècle,” Byzantion 78 (2008): 346-367. I am indebted to Reinhart Ceulemans for drawing my attention to this. 33  Paul Géhin, Évagre le Pontique: Scholies à l’Ecclésiaste (Sources Chrétiennes 397; Paris: Les Éditions du Cerf, 1993). 34 M. Adriaen, ed., S. Hieronymi Presbyteri Commentarius in Ecclesiasten (CCSL 72; Brepols: Turnhout, 1959). 35 Jerome, Commentariolus in Psalmos I,4 and Epist. ad Tit. ad II,9 (PL 26, 734). 36  Jerome actually gives the reading in Greek in a total of 16 instances. In seven cases the reading is also attested by other sources, only in 9 of them is Jerome is the only source for the reading.

The Later Greek Translators/Revisors

147

from Latin. This is not as easy as creating retroversions from the Syro-Hexapla given Jerome’s method and style of translation. Hi

Description of Witness part also in Greek MSS. + Syh = Sole Witness in Greek = Sole Witness in Latin

Instances 12 4 9 8

2.3  Catena Traditions

Another kind of patristic source are the catena manuscripts. Excerpts from the commentaries of the Church Fathers relating to the same stretch of biblical text were chained or linked together. These comments were either placed around the biblical text or between passages of the biblical text in a linear layout. For Ecclesiastes the catena traditions are complex and only a brief account of them can be given here. There are essentially four or five separate traditions: (1) an anonymous work commonly called the Catena Trium Patrum,37 (2) the Catena Hauniensis,38 (3) Olympiodorus Catena,39 (4) the Procopius Catena,40 and (5) the Polychronius Catena. There are also manuscripts which show a mixture of these basic types. Fortunately, these have been analysed extensively by Labate, Leanza, and Lucá. The only major type not yet available in a critical edition is the Polychronius Catena and this has now been studied by Felix Albrecht at the Septuaginta-Unternehmen. Together, we have checked the following manuscripts as representative of the Polychronius Catena: 139

Milan, Bibl. Ambr., A 148 inf.; X.-XI. Cent. (Cat.).

159

Moscow, Russian State Archive of Ancient Documents, Fonds 1607, Inv. 1.22 (earlier Dresden, Sächsische Landesbibliothek, A. 107); XII. Cent. (Cat.).

522

Cambridge, Trinity Coll., B. 7. 3; XIII. Cent. Eccl. c. Catena Excerpts.

560

Paris, Bibl. Nat., Gr. 151; XIII. Cent. (Cat.).

37  Santo Lucà, ed., Anonymus in Ecclesiasten Commentarius, qui dicitur Catena Trium Patrum Corpus (CCSG 11; Turnhout: Brepols, 1983). 38  Antonio Labate, ed., Catena Hauniensis in Ecclesiasten (CCSG 24; Turnhout: Brepols, 1992. 39  A. Labate, “Nuove catene esegetiche sull’Ecclesiaste,” in Ἀντίδωρον. Hulde an Dr. Maurits Geerard bij de voltooiing van de Clavis Patrum Graecorum, I (Wetteren, 1984), 241-263. 40  Sandro Leanza, ed., Procopii Gazaei Catena in Ecclesiasten, necnon Pseudochrysostomi Commentarius in Eundem Ecclesiasten (CCSG 4; Turnhout: Brepols, 1978).

148

The Text History

The number of readings preserved in each of these major types of catena is tabulated as follows: Catena Trium Patrum Catena Hauniensis Procopius Catena Polychronius Catena

2 14 0 2

The Olympiodorus Catena offers nothing that is not already in the extant Commentary. The instances in the Polychronius Catena are also from the Commentary of Olympiodorus. Nonetheless, all the sources have been painstakingly searched for hexaplaric readings. Date of MS. 159 Peter J. Gentry and Felix Albrecht, The Amazing History of MS Rahlfs 159—Insights from Editing LXX Ecclesiastes, Journal of Septuagint and Cognate Studies, 44 (2011): 31-50.

For the discovery and analysis of MS. 159, see the article by Albrecht and Gentry. Now that the character and history of MS. 159 is properly understood, the date given for it in Rahlfs’ Verzeichnis must be re-evaluated. The date given in Rahlfs’ Verzeichnis of X./XI. is derived from the catalogue of manuscripts in the Dresden Library produced by Schnorr von Carolsfeld.41 A reference to Matthaei shows influence from Professor Matthaei in this matter. Nonetheless, iron gall ink was not generally used until the beginning of the twelfth century and Oskar Lehmann, an expert in ligatures, dates the manuscript to XII/XIII on the basis of certain ligatures.42 The catalogue of manuscripts of the Synodal Library gives the twelfth century as the date for the parent from which 159 was cut and extracted. This is much more realistic and ought to be followed.43 This re-evaluation, however, does not detract from the importance of its witness since its textual tradition is not identical to that of MS. 139. 41  Franz Schnorr von Carolsfeld, Katalog der Handschriften der Königliche öffentliche Bibliothek zu Dresden. 1 Band: Enthaltend die Abteilungen Mscr. Dresd. A-D und F-H (Leipzig, 1882), 66. 42 Oskar Lehmann, Die Tachygraphischen Abkürzungen der griechischen Handschriften (Leipzig: Teubner, 1880), 73, 82, 95. 43  Archimandrite (Filantropov, Vasilii), Vladimir, Sistematicheskoe opisanie rukopisei Moskovskoi Sinodal‘noi (Patriarshei) biblioteki. Pt. 1: Rukopisi grecheskie. (Moscow, 1894), 43.

The Later Greek Translators/Revisors

149

Examples of Collation of 159 Two or three examples will illustrate how the manuscript can be read and also its importance. For each example, first the text of Ecclesiastes is given according to the edition followed by the First Apparatus where relevant. Then the folio numbers in MS. 159 for the passage are provided (i.e. the folios bearing the regular script and the folios bearing the mirror script) and the reading in MS. 159 is discussed. 1:16 1:16

ἐλάλησα ἐγὼ ἐν καρδίᾳ μου τῷ λέγειν ᾽Εγὼ ἰδοὺ ἐμεγαλύνθην καὶ προσέθηκα σοφίαν ἐπὶ πᾶσιν, οἳ ἐγένοντο ἔμπροσθέν μου ἐν Ιερουσαλημ,

om μου 2° 147-159-503-560 Folios in 159: 2V and 6R 6R = Regular Script 2V = Mirror Script To read 1:16, and especially the words ἔμπροσθέν ἐν Ιερουσαλημ at the end in 1:16d, both mirror and regular script must be taken together letter by letter. MS. 159, which Ziegler assigned to the cI group on the basis of the collations in Holmes-Parsons and Klostermann, reads with 147 503 and 560 of the main C group in omitting the pronoun after ἔμπροσθέν. Ziegler’s cI group consisted of manuscripts which were not catena manuscripts, but had the same Bible text as the manuscripts in the main catena group (C). Since 159 is a catena manuscript, it belongs in the main catena group (C). 2:09 2:09

καὶ ἐμεγαλύνθην καὶ προσέθηκα παρὰ πάντας τοὺς γενομένους ἔμπροσθέν μου ἐν Ιερουσαλημ· καί γε σοφία μου ἐστάθη μοι.

150

The Text History

Folios in 159: 4R and 6V 4R = Regular Script 6V = Mirror Script To read 2:09a-c, once again both mirror and regular script must be read together. Note in the word ἐμεγαλύνθην, that the letters μεγαλ are clear in the mirror script, while the last part of the word is best attested in the regular script. The initial epsilon is partly represented in both regular and mirror scripts. Once again, 159 joins 147 503 and 560 against the remaining 16 manuscripts of both C and cI groups in not adding σοφίαν after προσέθηκα. 3:21 3:21

καὶ τίς οἶδεν τὸ πνεῦμα υἱῶν τοῦ ἀνθρώπου

εἰ ἀναβαίνει αὐτὸ εἰς ἄνω, καὶ τὸ πνεῦμα τοῦ κτήνους εἰ καταβαίνει αὐτὸ κάτω εἰς γῆν;

τοῦ ἀνθρώπου] om τοῦ 637 139-147-503-560 357 = Ald; ανθρωπων 159 LatAug Anap 4,23,37 Epte 238,15 Trinap 14,22 Hilem te 280,282 Hicomm ap 282,321 Arm Fa SaI (sed hab sing Dam Didlem et comm 103,22 104,7 104,14 Metlem et comm III.23,2,45,56 Ol PsChr LatAug Ante 4,23,37 C D 13,24 Epap 238,15 Trinte 14,22 Hilem ap 280,282 Hicomm te 282,321 Hi Did 54); domini Lic 2,6; Adam Vulg Folios in 159: 9V and 10R

09V = Regular Script 10R = Mirror Script MS. 159 reads ανθρωπων instead of ἀνθρώπου. This reading is singular and unique. While it is certainly a copyist’s error, it is interesting for the history of interpretation as a large number of patristic citations in Latin also have this reading. In addition, neither Matthaei nor Klostermann correctly read the manuscript at this point.

2.4  The Syro-Hexapla

Last but by no means least is the Syro-Hexapla. Our sole source for the Syro-Hexapla is the eighth century Codex Ambrosianus in Milan, available in the facsimile by

The Later Greek Translators/Revisors

151

Ceriani.44 Elsewhere in description of this source I have noted that it offers many different kinds of materials, all relevant for hexaplaric studies.45 First is the text itself which includes words marked by an asteriskos in ten instances and names the source in eight of them. Only one other witness (i.e. Codex Venetus) preserves any asteriskoi for Ecclesiastes and does so twice, erroneously in both instances. Twice words are marked in the text of Syh by an obelos, and four times by an antisigma or lemniskos. Second, there are marginal notes offering at least six different types of information: (1) a few notes offer textual variants, (2) many marginal notes offer readings from the later Jewish Revisers Aquila, Symmachus, and Theodotion, (3) fifteen notes offer words in Greek to clarify evidence when retroversions may be uncertain, (4) seven notes offer explanatory glosses—these are always distinguished from hexaplaric notes by being circled, (5) longer scholia are found in the top and bottom margins, with appropriate indices to connect them to the text, and (6) eight of the longer scholia contain seven citations from the Ecclesiastes Commentary of Olympiodorus and one from Evagrius. Syh preserves a total of 250 marginal readings. The relationship of the marginal notes in Syh to the other sources is of great interest and is charted as follows (the count is only proximate but gives data from which conclusions can be drawn): Combination 161 248 252 539 Syh 161 248 252 Syh 161 248 252 Syh TOTAL V Syh Syh = Sole Witness

Instances 39 16 14 8 77 1 103

44  A. M. Ceriani, ed., Codex Syro-Hexaplaris Ambrosianus photolithographice editus (Monumenta sacra et profana, 7; Milan: Typis et impensis Bibliothecae Ambrosianae, 1874). Serious study must not rely on the edition of H. Middeldorpf, Codex Syriaco-Hexaplaris, Pars I (Berlin, 1835) pace W. Strothmann, Konkordanz des Syrischen Koheletbuches nach der Peshitta und der Syrohexapla (Göttinger Orientforschungen, I: Reihe: Syriaca, Band 4; Wiesbaden: Harrassowitz, 1973), p. iv. 45  See the following studies of the Syro-Hexapla (Codex Ambrosianus) in Ecclesiastes: Peter J. Gentry, “Hexaplaric Materials in Ecclesiastes and the Rôle of the Syro-Hexapla,” Aramaic Studies 1 (2003), 5-28, idem, “The Relationship of Aquila and Theodotion to the Old Greek of Ecclesiastes in the Marginal Notes of the Syro-Hexapla,” Aramaic Studies 2.1 (2004): 63-84, idem, “Propaedeutic to a Lexicon of the Three: The Priority of a New Critical Edition of Hexaplaric Fragments,” Aramaic Studies 2.2 (2004), 145-74, idem, “‘The Role of the “Three” in the Text History of the Septuagint’: II. Aspects of Interdependence of the Old Greek and the Three in Ecclesiastes,” Aramaic Studies 4.2 (2006): 153-192 and Phillip S. Marshall, “A Critical Edition of the Hexaplaric Fragments of Ecclesiastes,” (Ph.D. diss. The Southern Baptist Theological Seminary, 2007).

152

The Text History

In some 23 instances, Syh preserves a longer reading and other witnesses only part, and in perhaps 35 instances, Syh preserves part of a longer reading preserved in other witnesses. These data corroborate the proposal I have advanced elsewhere that the marginal notes come from a separate source than the text, a source similar to the Catena Text Traditions. Approximately 60% of the notes it transmits are also found in the archetype of the Greek witnesses while 40% are independent of this tradition. The fact that longer notes in the bottom and top margins cite from Evagrius and Olympiodorus where they have readings of the Three also demonstrates that those who produced Syh gathered materials from several sources.

3.  Comparison of Recent Research and Sources Used by Field The sources scoured for the new Göttingen Edition of Ecclesiastes as well as for the edition in the Hexapla Project may now be compared with the sources used in 1875 by Field for his Origenis Hexaplorum quae supersunt. First, Field lists the sources given in Montfaucon: Codex Regius, num. 1890 Codex Regius, num. 19902 Codex Regius, num. 2435 Codex Regius, num. 2436 Hieronymi Commentarius in Ecclesiasten Notes from Drusius and “Roman” Edition

= Ra 560 = Ra 562 = Ra 561 = Ra 563

MSS. 560 and 563 belong to the Polychronius Catena, Ra 561 to the Catena Trium Patrum, and Ra 562 is a manuscript of Olympiodorus’ Commentary on Ecclesiastes. Montfaucon’s sources for marginal notes in Greek manuscripts were slim. Second, to Montfaucon’s list of sources Field adds MSS. 161, 248, and 252 as collated and published in the edition of Holmes and Parsons. The Auctarium of Field also gives “Mat” as another manuscript source. His source for “Mat” was the collation made by G. L. Spohn of marginal readings in 161 in 1785. Spohn employed the siglum “Mat” because the manuscript was in the possession of Professor Matthaei in Leipzig. In 1788 Matthaei sold this manuscript to the Library in Dresden and was also the one who collated it along with 159 and 160 for the edition of Holmes and Parsons. Thus Field mistakenly listed the evidence for MS. 161 twice. Nonetheless, the evidence given by Field was a major step forward. To the list of sources used by Field we have added MSS. 336, 411, 539 and 788 and a fresh search of the Commentaries and catenae. Analysis of the Commentary by Olympiodorus rests upon more than a dozen manuscripts and of the Polychronius Catena on sixteen manuscripts, some 3 or 4 centuries older than the ones used previously. The fresh collations made by the Septuaginta-

The Later Greek Translators/Revisors

153

Unternehmen of 161, 248, and 252 have also revealed mistakes in the collations made for Holmes-Parsons. Finally, Field used the edition of Middeldorpf for Syh as the facsimile by Ceriani did not appear until one year before publication of his monumental work. The analysis of Syh by myself and Marshall is based upon a better source, therefore, than the evidence given for Syh in Field. Although the body of hexaplaric materials for Ecclesiastes has grown both quantitatively (more than doubled since Field) and qualitatively as a result of extensive research, and indeed is fairly rich in relation to what remains for some books, we still have a limited amount of evidence for the work of the Three in Ecclesiastes.

4.  On the Textual Character of the Three Later Translations For Qohelet we have (in very approximate numbers) about 130 readings attested for Theodotion, 215 for Aquila, and 380 for Symmachus. This is precisely what we would expect, since the original translation belongs to the kaige group of texts, but probably was not made by Theodotion himself.46 Since the translation is most like Theodotion, there would be few readings from Theodotion different in a significant way, more from Aquila, and many more from Symmachus since his effort was to produce a translation more sensitive to the demands of the target language. In addition, approximately 50 readings are attributed to Aquila and Theodotion together, but only two to Theodotion and Symmachus and none to Aquila and Symmachus together. An example is given from each of the Jewish revisers. Theodotion. In 5:10c OG renders ‫ם־ראּית ֵעינָ יו‬ ְ ‫( ִּכי ִא‬K = ‫ר ִאּיַ ת‬,ְ Q = ‫ ְראּות‬or ‫ְ)ראֹות‬ by ὅτι ἀρχὴ τοῦ ὁρᾶν ὀφθαλμοῖς αὐτοῦ; (construing ‫ אם‬as ‫אם‬, ֵ cf Ez 21:26). A marginal reading attributed to θ΄ in Syh reads ‫ܕܠܡܚܙܐ ܵܒܥܝܢܐ ܕܝܠܗ‬ ܼ ‫ = ܐܐܠ ܐܢ‬εἰ μὴ τοῦ ὁρᾶν ὀφθαλμοῖς αὐτοῦ. What we see here is that Theodotion corrects the original translation to what is now the normal reading of the text. In fact, almost half of the notes attributed to Theodotion read θʹ ὁμοίως τοῖς οʹ, telling us that Theodotion has the same text as that in the Fifth Column of the Hexapla.47 This also confirms that the OG belongs to the kaige group in character.48

46  Yun Yeong Yi, “Translation Technique of the Greek Ecclesiastes” (Ph.D. diss. The Southern Baptist Theological Seminary, 2005), 217-218. 47  See Peter J. Gentry, “Hexapla Materials in Ecclesiastes and the Rôle of the Syro-Hexapla,” Aramaic Studies 1 (2003): 5-28. 48 See Peter J. Gentry, “Pre-Hexaplaric Translations, Hexapla, post-Hexaplaric Translations,” in Textual History of the Bible: The Hebrew Bible. Volume 1A Overview Articles, eds. Armin Lange and Emanuel Tov (Leiden: Brill, 2016), 211-235.

154

The Text History

Aquila. As early as 1871 Heinrich Grätz proposed that the original translation of Qohelet was by Aquila. Most notable in the translation is the rendering of ‫ את‬+ direct object by σύν + accusative in Greek (e.g. 2.17, 3.10, 11, 17bis and passim).49 How can we explain, then, the fact that there are even more readings attested for Aquila than Theodotion in the case of this book? Some have proposed that Aquila produced two versions. One was the original translation and the other a later revision—the source of the marginal notes attributed to Aquila. Others, like Barthélemy denied the marginal notes attributed to Aquila as coming from him.50 These proposals are not satisfactory since the character of the marginal notes better matches the system and translation technique of Aquila as Hyvärinen and Jarick have shown.51 According to MSS. 161 248 and 788 Aquila renders περασμός in 4:8c instead by τέλος. This is characteristic since Aquila has a system whereby forms of the same root in Greek match forms of the same root in Hebrew, regardless of the part of speech. Symmachus. In 1:9a for ‫ מה־ׁשהיה הוא ׁשיהיה‬OG has τί τὸ γεγονός; αὐτὸ τὸ γενησόμενον which Symmachus revises to τὸ προόν, αὐτό ἐστι τὸ ἐσόμενον. One can immediately see the improvement according to the criteria of later Hellenistic Greek. Many of the readings for Symmachus are preserved only in Latin by Jerome who, as we know, had a preference for the text of Symmachus and made use of it as a tool in dealing with difficult passages in Hebrew and for his rendering in the Vulgate.

5.  Text Critical Value for the Hebrew Bible An example of the importance of the hexaplaric materials for Ecclesiastes is illustrated by the crux criticorum at 1:17. Until the discovery of MS. 788, all of our Greek witnesses attested παραβολάς where 𝔐 has ‫הֹוללֹות‬. ֵ Already in 1937 Gordis proposed that the original translation had παραφοράς, now supported by 788sup lin and also adopted in the critical edition of the Göttingen Septuaginta as the original text. Yet the problem in the text in the Greek textual tradition is also matched by a problem in the Hebrew tradition. The editor of BHQ proposed ‫הֹוללּות‬ ֵ as the preferred text in Hebrew, but John Meade and I have argued that proper analysis of both Greek and Hebrew traditions supports MT.52 49 Ibid. 50 Barthélemy, *Devanciers, 31-2. 51  K. Hyvärinen, “Die Übersetzung von Aquila” (ConBOT, 10; Lund, 1977), 88-99 and J. Jarick, J. Jarick, ‘Aquila’s Koheleth’, Textus 15 (1990): 131-139. 52  For full discussion, see J.D. Meade and P.J. Gentry, “Evaluating Evaluations: The Commentary of BHQ and the Problem of ‫הֹוללֹות‬ ֵ in Ecclesiastes 1:17.”

C.  Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes I.  General Assessments of Translation Technique 1. Overview Rahlfs’ programmatic statement, made in his pilot project on Genesis in 1926, applies for the reconstruction of the earliest accessible stage of the text of Ecclesiastes: “In der Herstellung des Textes bin ich nicht einer einzelnen Hs. gefolgt, sondern habe jedesmal diejenige Lesart aufgenommen, welche mir nach dem Gesamtstande der handschriftlichen Überlieferung und unter Vergleichung des hebräischen Textes als die beste erschien.”1 Indeed, this is true for all critical editions since no one manuscript consistently preserves the best form of the text; every manuscript has scribal mistakes and does not perfectly transmit the original text. Textual critics must follow an eclectic method in attempting to establish the earliest attainable form of the text. The difference between creating critical editions of classical texts and of the text of the Septuagint is primarily twofold: (1) the text of the Septuagint is a translation, and (2) the text of the Septuagint is a canonical and sacred text for both Jews and Christians. The fact that the text is a translation establishes ipso facto two different approaches to the study of the text: (1) one approach aims to determine the meaning intended by the translator, and (2) another aims to determine the meaning as understood by the various communities reading and using this translation throughout the entire history of textual transmission. The canonical character of the text affects both approaches to the study of the text. For both translators and tradents, the text being copied was the Word of God. The Greek Text of Ecclesiastes originated as a translation of a Hebrew original at a particular place and time. The act of translation establishes the notion of an “original text” and validates the genealogical method of textual criticism, whose basic principles were forged by Paul Maas in Textkritik (1950). The first step is to eliminate common mistakes made by copyists and scribes in transcribing texts, i.e. unconscious and unintentional errors as opposed to intentional changes in the text. Frequent errors of sight and sound are listed in an organised manner in the “Grammatica und Orthographica” section of the edition. Such errors were not collected or recorded systematically for the Book of Ecclesiastes; collation of the manuscripts began in 1929. An exhaustive and

1  Alfred Rahlfs, Genesis, Gesellschaft der Wissenschaften, Göttingen; Stuttgart: Privilegierte Württembergishce Bibelanstalt, 1926, 34.

156

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

model treatment of common errors is given by Udo Quast in his edition of Ruth. One may also consult “A Note on Scribal Error” by John W. Wevers.2 The second step is to observe significant variants. Agreement between manuscripts in the case of significant variants normally determines genealogical relationship. Thus, the textual critic commences with long lists of agreements in significant variants among manuscripts. One begins with deviation from the original text that is obvious and establishes families or groups on this level. Later on, a better understanding of the genealogical relationships and history of the transmission of the text may be used for more difficult problems where deciding the original text among variants is not so simple. Significant variants usually entail changes made intentionally by the scribes, although unintentional mistakes such as omission of larger sections due to parablepsis can sometimes be used to establish family relationship. The translator of Ecclesiastes was committed to an approach to translation based on extreme formal equivalence between source and target languages in the corpus of translations we call the Septuagint. For example, according to the rules of grammar in Hebrew, only the last member of a bound phrase is articulated if the entire noun phrase is definite. Greek grammar and style, however, normally articulates both nomen regens and nomen rectum or neither in noun phrases. The Greek translator of Ecclesiastes, however, follows the rule for his Hebrew parent text. So, the Greek copyists who had no access or grasp of the parent text of the translator made many improvements in grammar and style. Sometimes a scribe struggled to make sense of a rendering so formal that the Greek was almost unintelligible and employed the context of the Greek translation to smooth things out. These “improvements” are what we call “inner Greek” scribal errors. In addition, copyists changed the text intentionally due to exegetical, historical, or theological reasons. Changes to the text also arose because copyists whose native speech was Byzantine Greek conformed the Hellenistic Greek of the translation of the text to grammatical or lexical choices natural to a later period of Greek. In the history of the transmission of the text, normally copyists had no access to the original parent text from which Ecclesiastes was translated. Scholars who did, such as Origen, and Jerome, affected the subsequent textual transmission. Other tradents who did not know Hebrew nonetheless employed the translations of the Jewish revisers, Theodotion, Aquila, and Symmachus, whose readings were accessible through the Hexapla, to correct the text towards the Hebrew. To reach the earliest form of the text accessible on the basis of the extant evidence entails, as Rahlfs noted, principles of textual criticism, and knowledge of the translation technique of the Greek translator.

2  John W. Wevers, “A Note on Scribal Error,” Canadian Journal of Linguistics 17 (1972): 185-190.

General Assessments of Translation Technique

157

2.  The Hebrew Vorlage of the Translator and Translation Technique Literature A. Schenker, et al., eds. Biblia Hebraica Quinta. Fascicle 18: General Introduction and Megilloth, Stuttgart: Deutsche Bibelgesellschaft, 2004. Peter J. Gentry, Ecclesiastes: The New English Translation of the Septuagint, edited by A. Pietersma and B. G. Wright. Oxford: Oxford University Press, 2007. Yun Yeong Yi, Translation Technique of the Greek Ecclesiastes. Diss. The Southern Baptist Theological Seminary, 2007.

2.1  The Masoretic Text (𝔐/MT) The Hebrew Vorlage used by the Greek translator was almost identical to our current Masoretic Text. This is based not only on my own thorough study, but is also the conclusion of the dissertation by Yi. Unfortunately for Y. A. P. Goldman, editor of Qoheleth for Biblia Hebraica Quinta, no critical edition of the Septuagint was available in 2004. Nonetheless, in addition to Rahlfs’ Handausgabe, Goldman was given access to the Kollationshefte for Ecclesiastes in Göttingen prior to 1998 and he also had unpublished materials on Ecclesiastes for Critique Textuelle de l’ancien Testament by D. Barthélemy. In the apparatus constructed by Goldman for Ecclesiastes he prefers a different Hebrew text as ‘original’ in 59 instances. Forty-seven of these entail an appeal to G = the Septuagint (including five G*). He employs the siglum G* to indicate that there is variation in the textual tradition of the Greek Ecclesiastes at this point, but the asteriskos indicates the original reading of G in his judgement. Of these 59 cases, only about ten represent instances where a different consonantal text is involved. Some of these differences entail confusion of beth and kaph, daleth and resh, waw and yodh. Conversely, the siglum G*, indicating Goldman’s preference for the original text of the LXX is used in 48 instances: 21 of these agree with the edition here and 27 do not. What these data show is that Goldman unnecessarily exaggerates the differences between the Vorlage of the LXX and the Masoretic Text. 2.2  The Dead Sea Scrolls (𝔔) Literature É. Puech, Un nouveau fragment du manuscritb de l’Ecclésiaste (4QQohéletb ou 4Q110). Revue de Qumrân 76 (2000) 617-621. E. Ulrich, Ed. Qumran Cave 4: XI. Psalms to Chronicles. Discoveries in the Judaean Desert XVI. Oxford: Clarendon Press, 2000.

158

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

The available texts are extremely fragmentary. Contents: 4QQoha = 4Q109 Col. I: Fragment 1 i Col. I: Fragment 2 Col. II: Fragmente 1 ii, 3-6 i Col. III: Fragmente 1 iii, 6 ii, 7

Qoh 5:13-17 Qoh 6:1? Qoh 6:3-8, 12-7:6 Qoh 7:7-10, 19-20

4QQohb = 4Q110, 4Q468-1 Fragmente 1-3

Qoh 1:8-16

No conclusions can be drawn from the apparatus e silentio. The evidence is cited using the siglum 𝔔 on left and right side of the square bracket for all relevant variants. 115 514 515 63 64 64 66 68 74 75 75 76 77 719

ἐπικοσμηθῆναι] ‫?𝔔 לגבור ;𝔐 לגתקן‬ καθώς 𝔐] ‫𝔔 כיא‬ καί γε 𝔐] absc Fa1; om καί SaI 𝔔 ὑπὲρ αὐτὸν τὸ ἔκτρωμα 𝔐] ὑπὲρ αὐτόν post ἔκτρωμα tr Co 𝔔 ἦλθεν 𝔐] ‫𝔔 בה‬ πορεύεται = ‫𝔔 הלך ]𝔐 ילך‬sup καὶ εἰ = ‫𝔔 ואם לוא ]𝔊 𝔐 וְ ִאּלּו‬ ὅ τι = (ὅτι) 𝔖] ‫( 𝔔 כמה‬sed hab ‫)𝔐 כי מה‬ ἐν οἴκῳ = 𝔐] ‫𝔔 בית‬ ἐπιτίμησιν = 𝔐] ‫𝔔 גערות‬ ὑπὲρ ἄνδρα ἀκούοντα = ‫𝔔 טוב ל[שמע ;*𝔔 טוב [למע ]𝔐 מאיׁש ׁשמע‬c: ex 75a καί γε = 𝔐mss] ‫𝔐 גם‬mss 𝔔 απολλυει = 𝔐] ‫𝔔 ויע]וה‬ βοηθήσει= 𝔔] confortabit Hilem et com 309,305te 309,322ap = 𝔐

In the apparatus of BHQ, Goldman prefers the reading of 𝔔 to 𝔐 only for ‫ הלך‬in 64. This would go against the Vorlage of the LXX.

2.3  The Syriac Peshitta (Pesh/𝔖) Literature Alexander A. di Lella, J. A. Emerton und D. J. Lane, eds. The Old Testament in Syriac according to the Peshitṭa Version. Part II. Fasc. 5. Proverbs, Wisdom of Solomon, Ecclesiastes, Song of Songs. Leiden: E. J. Brill, 1979.

General Assessments of Translation Technique

159

G.W. Anderson, Canonical and Non-Canonical: Cambridge History of the Bible edited by P.R. Ackroyd and C.F. Evans, Vol. 1. Cambridge, 1970, 113-159. Beck, Edmund, ed. Des Heiligen Ephraem des Syrers Sermones, 1. Corpus scriptorum christianorum orientalium, 305 : Scriptores Syri, 130. Louvain: Secr. du CorpusSCO, 1970. Deppe, Klaus, ed. Kohelet in der syrischen Dichtung: 3 Gedichte über des Kohelet-Buch von Afrēm, Jakob von Sarug und Johannes von Mossul. Wiesbaden: Harrassowitz, 1975. G. Janichs, Animadversiones criticae in versionem syriacum Peschitthonianam librorum Koheleth et Ruth, Leipzig, 1869. Kamenetzky, A.S. Die P’shita of Kohelet textkritisch und in ihrem Verhältnis zu dem massoretischen Text, der Septuaginta und den andern alten griechischen Versionen. ZAW 24, 1904, 181-239. Lane, D.J. “Lilies that Fester…” The Peshitta Text of Qoheleth. Vetus Testamentum 29, 1979, 481-490. Schoors, Antoon. The Peshitta of Koheleth and Its Relation to the Septuagint, in After Chalcedon. Studies in Theology and Church History edited by C. Laga et al. Uitgeverij: Peeters, 1985, 347-357. Weitzman, Michael. Peshitta, Septuagint and Targum, in VI Symposium Syriacum 1992: University of Cambridge, Faculty of Divinity, 30 August-2 September, 1992 edited by René Lavenant, Rome: Pontifical Biblical Institute, 1994, 51-84.

The collation and analysis here is largely the work of John D. Meade to whom I express my deepest gratitude. I remain responsible for what is written here. In the past, the Syriac Peshitta has been called a “daughter version” of the Septuagint (so Anderson). Agreements between the Peshitta and the Septuagint were sometimes understood in terms of dependent relationship on the part of the Peshitta. Recent research, particularly by Jerome Lund and Michael Weitzman, has moved the consensus of scholarship to a different position: as a general rule the Peshitta is independent of the Septuagint. Clear criteria were established by Weitzman to demonstrate that agreement does not always indicate dependence. Other factors such as polygenesis in the process of translation, common access to interpretive traditions of Second Temple Judaism, and common Vorlage, can account for agreements besides direct dependence. Only if factors such as these are eliminated as possible reasons for agreement between Peshitta and LXX can dependence on LXX be clearly demonstrated. Careful analysis indicates that the Peshitta of Qohelet is dependent upon the Greek Ecclesiastes only in a sporadic and non-systematic manner. Such clear cases of dependence, then, would make the Peshitta an early witness to the text history of the Greek Ecclesiastes. 2:7 𝔐 𝔊 𝔖

2.3.1  Examples of Polygenesis [...] ‫יתי‬ ִ ִ‫ָקנ‬ ἐκτησάμην] + μοι 766B Fa SaI ‫ܩܢܝܬ ܠܝ‬

160

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

Peshitta 2:5, 6, 8 et passim has ‫ ܠܝ‬following 1cs finite verbs, which continue the Hebrew construction begun in 2:5. Thus the Peshitta translator arrived at his rendering of 2:7 from the immediate context. The addition of μοι attested in the Greek textual tradition by 766B (as well as in Fa SaI) are also cases where scribes have harmonised the reading in 2:7 according to the context. Since both the Peshitta translator and the Greek scribe would have had independent access to the context, it is not necessary to posit dependence of Peshitta on the Greek tradition. 2:11 𝔐 𝔊 𝔖

‫ּוב ָע ָמל‬ ֶ καὶ ἐν (+ παντι 411 GregNy 352,5) μόχθῳ ‫ܘܒܟܠ ܥܡܐܠ‬

Peshitta harmonises ‫ ֶב ָע ָמל‬to the fuller phrase ‫ ܒܟܠ ܥܡܐܠ‬in 2:11, which also occurs in 1:3 and 5:17. The reading in 2:11 attested by 411 and GregNy 352,5 has probably been harmonised to that in 1:3 or 5:17. Thus polygenesis accounts for the agreement between this strand of the Greek tradition and translator of the Peshitta. 8:9b 𝔐 𝔊 𝔖

‫ת־ל ִּבי‬ ִ ‫וְ נָ תֹון ֶא‬ ἔδωκα τὴν καρδίαν ‫ܘܝܗܒܬ ܠܒܝ‬

8:9b is an example of polygenesis due to the linguistic constraints set by Greek and Syriac. Hebrew uses infinitive absolutes in sequence with finite verbs; however, this option is not available in Greek and Syriac. Therefore, although 𝔊 and 𝔖 agree against 𝔐, probably each translator chose the past tense finite verb independently in order to render the Hebrew Vorlage. 10:6a 𝔐 𝔊 𝔖

‫נִ ַּתן ַה ֶּס ֶכל‬ ἐδόθη ὁ ἄφρων (= αʹ σʹ Hi Vulg Tar) ‫ܝܬܒ ܣܟܐܠ‬

10:6a provides a good example of polygenesis. The consonantal text ‫ סכל‬is vocalised ‫( ֶס ֶכל‬folly) in 𝔐 here and vocalised ‫( ָס ָכל‬fool) at 2:19, 7:17, 10:3(bis), and 10:14. 𝔊, Peshitta along with αʹ σʹ Hi Vulg Tar read ‘fool’ in all instances. Assimilation of 10:6a to the other occurrences of ‘fool’ in Ecclesiastes could occur independently in the versions. Common exegetical tradition may also account for this agreement between 𝔊 and 𝔖, but polygenesis or translation technique is the best explanation.

General Assessments of Translation Technique

2:12d 𝔐 𝔐mss 𝔊 𝔖

161

2.3.2  Examples of Common Vorlage ‫ָעׂשּוהּו‬ ‫ עׂשהו‬Kenn 4 17 18 30 77 99 107 109 111 117 118 125 128 129 151 157 158 166 167 170 172 181 188 192 196 213 218 224 240 244 245 260 384 651 680 ἐποίησεν αὐτήν (= Hi Vulg) ‫ܥܒܕܗ‬

Although one could posit polygenesis as the reason for agreement between 𝔊 and 𝔖 in 2:12d, the evidence of Hi and Vulg combined with attestation of the singular verb in the mediaeval Hebrew MSS. provides credibility for positing a common Vorlage in this case. 7:25d 𝔐 𝔐mss 𝔊 𝔖

‫ֹהולֹלות‬ ֵ ‫וְ ַה ִּס ְכלּות‬ ‫ והוללות‬Kenn 1 17 76 99 117 125 136 145 151 180 188 201 212 223 224 384 674; ‫ והללות‬18 147 καὶ σκληρίαν καὶ περιφοράν ‫ܘܣܟܠܘܬܐ ܘܡܬܥܒܪܢܘܬܐ‬

The explanation at 2:12d will suffice probably for positing a common Vorlage, to which 𝔊 and 𝔖 had access independently, since many mediaeval Hebrew MSS. share this same reading. 3:18d 𝔐 𝔊 𝔖

2.3.3  Examples of Common Exegetical Tradition ‫וְ ִל ְראֹות‬ καὶ τοῦ δεῖξαι (= σ´ Hi Vulg) ‫ܘܠܡܚܘܝܘ‬

𝔐 is vocalised as a Qal Inf. Cstr. from ‫ ראה‬+ ‫ל‬. Both 𝔊 and 𝔖, together with Symmachus and Jerome, presumably read the text as a Hiphil Infinitive Construct ‫ל ְראֹות‬.ַ Rarely, the preformative he of the Hiphil Stem is syncopated or elided in biblical Hebrew (see Gesenius § 53q). The Greek translator, however, may have vocalised this form as Hiphil with elided he as is standard in Qumran and Mishnaic Hebrew (Qimron § 310.145; Segal, § 43). 8:1c 𝔐 𝔊 𝔖

‫וְ עֹז ָּפנָ יו יְ ֻׁשּנֶ א‬ καί ἀναιδὴς προσώπῳ αὐτοῦ μισηθήσεται ̈ ‫ܘܕܚܨܝܦܢ‬ ̈ ‫ܐܦܘܗܝ ܢܣܬܢܐ‬

162

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

Perhaps this is the clearest example of common exegetical tradition in Ecclesiastes. Although no other version has the reading “to hate”, there may be haggadic influence from BTaan 7b, “Do not read yeshuneh [changed] but yesuneh [hated]”, which demonstrates the presence of a common reading of this verse to which both the Peshitta and Septuagint attest.3 4:15c 𝔐 𝔊 𝔖

2.3.4  Dependence of the Syriac Version on the Septuagint ‫ַהּיֶ ֶלד ַה ֵּׁשנִ י‬ τοῦ νεανίσκου τοῦ δευτέρου ‫ܥܠܝܡܐ ܬܢܝܢܐ‬

On the surface both versions appear to have an independent translation of the Hebrew text, but it is difficult to reach such a conclusion when the whole context is consulted. In 4:13a, 𝔐 has ‫ֹטוב יֶ ֶלד‬, while 𝔊 has ἀγαθὸς παῖς and 𝔖 has ‫ܛܒ ܛܠܝܐ‬. In 4:15c, 𝔊 changes the equivalent for the same Hebrew word and 𝔖 also changes its reading. It is possible that that each did this independently, but it seems more probable that 𝔖 is aware and makes use of the subtleties of the Greek translation here. 7:2d 𝔐 𝔊 𝔖

[…] ‫יִ ֵּתן‬ δώσει ἀγαθόν ‫ܝܗܒ ܛܒܬܐ‬

7:2 has ἀγαθόν as object for the verb “give” whereas Ra had no object in his edition. Antoon Schoors concluded the object was added later in the textual transmission of 𝔖 as the text was brought into closer alignment with the LXX although there is no MSS. evidence to support this conclusion.4 If the object is in OG, then the translators of 𝔖 would have had access to this reading at the time of translation. Why should we posit an intermediate step in the textual transmission of 𝔖 for which there is no manuscript evidence? 10:16a 𝔐 𝔊 𝔖

‫י־לְך ֶא ֶרץ‬ ָ ‫ִא‬ οὐαί σοι, πόλις ‫ܘܝ ܠܟܝ ܡܕܝܢܬܐ‬

3  Antoon Schoors, “The Peshitta of Kohelet,” 350 n. 21. J. Rabbinowitz, Ta‘anith in The Babylonian Talmud: Seder Mo‘ed, ed, I. Epstein, (London: Soncino, 1938), 29. 4  Ibid., 352.

General Assessments of Translation Technique

163

This example appears to have escaped the notice of Goldman and Schoors, but is relevant since 𝔊 has departed from 𝔐 and 𝔖 has followed 𝔊 in this instance. 𝔐 uses ‫ עיר‬in 10:15, but then employs ‫ ארץ‬is used in v. 16 and v. 17. 𝔊 has πόλις in v. 15 and v. 16 but γῆ in v. 17. 𝔖 appears to depend on 𝔊 for its reading in 16a, but in v. 17 it continues to read ‫ ܡܕܝܢܬܐ‬for ‫ארץ‬. 2:15f (= 16a 𝔖) 𝔐 […] 𝔊 διότι ἄφρων ἐκ περισσεύματος λαλεῖ 𝔖 ‫ܡܛܠ ܕܣܟܐܠ ܡܢ ܝܬܝܪ ܡܡܠܠ‬

Variation in placement of this text in the Greek textual transmission as well as translation technique demonstrate almost certainly that it is a gloss derived from the NT Gospels, although retained by Ra.5 Nonetheless, the question remains open as to whether this text was original in 𝔖 or added later due to revision towards 𝔊.6 The earliest witnesses in 𝔊, namely B and 998, still post date 𝔖 by about 100 years.7 The form of the text in the Peshitta does not match the form of the text that we have preserved in the Old Syriac Version of the Sinaiticus: ‫ ܡܢ ܬܘܬ̈ܪܐ ܗܘ ܓܝܪ ܕܠܒܐ ܡܦܩ ܦܘܡܐ‬.‫ܣ‬. Curetonianus has a similar text form as that in Sinaiticus.8 Therefore the reading must have come from 𝔊 at an earlier date. The early evidence of the Greek insertion lends plausibility that the 𝔖 translator may have had access to a form of this text c. 200 CE.

2.4  The Cairo Genizah Fragment (𝔎) 2.4.1  The Witnesses/Editions of the Cairo Genizah Fragment

Cambridge University Library T-S Misc.28.74 (herein 𝔎) = FGP C127731, C127732

Literature D. S. Blondheim, Échos du Judéo-Hellénisme, REJ (1924): 1-14. Reprint: Les Parlers judéoromans et la Vetus Latina (Paris: Édouard Champion, 1925. Peter J. Gentry, The Greek Genizah Fragment of Ecclesiastes and its Relation to Earlier Greek Versions: Biblical Greek in Context edited by J. K. Aitken and T. V. Evans, Biblical Tools and Studies, Leuven: Peeters, 2015, 91-116.

5  Schoors also reads this evidence in this way. Schoors, “Peshitta of Kohelet,” 351. 6  Ibid., 351. Schoors considers this an example of later revision. 7  The text in these witnesses occurs in the following form: διότι ὁ ἄφρων ἐκ περισσεύματος λαλεῖ. 8  George Anton Kiraz, Comparative Edition of the Syriac Gospels: Aligning the Sinaiticus, Curetonianus, Peshitta and Harklean Versions, Vol. 1 Matthew, (New York: Brill, 1996), 167.

164

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

N. Fernández Marcos, Some Thoughts on the Later Judaeo-Greek Biblical Tradition, Bulletin of Judaeo-Greek Studies 2 (1988): 14-15. N. R. M. de Lange, Two Genizah Fragments in Hebrew and Greek. In J. A. Emerton and S. C. Reif (eds.), Interpreting the Hebrew Bible: Essays in Honour of E. I. J. Rosenthal, Cambridge: Cambridge University Press, 1982, 64-75. N. R. M. de Lange, A Greek Translation of Kohelet (Ecclesiastes). In N. de Lange, Greek Jewish Texts from the Cairo Genizah, Texte und Studien zum antiken Judentum, 51; Tübingen: Mohr-Siebeck, 1996, 71-78, 354-357. N. R. M. de Lange, Japheth in the Tents of Shem: Greek Bible Translations in Byzantine Judaism. Texts and Studies in Medieval and Early Modern Judaism, 30; Tübingen: Mohr-Siebeck, 2015.

2.4.2  On the Textual Character of the Cairo Genizah Fragment

The text was first published by Blondheim in 1924 and later re-edited twice with Greek transcription, analysis and comments by Prof. de Lange in 1982 and 1996. Photographs of the fragment are available on the Cairo Genizah website of Cambridge University. My own analysis is the most recent and thorough. The fragment contains a translation of Qoheleth into Greek, but represented by transliteration in Hebrew script according to the system of Tiberian Hebrew. The translation is made from the Hebrew text into Later Byzantine or Mediaeval Greek—influenced by the Old Greek. Relevant connections between the Cairo Genizah translation and the textual tradition of the LXX are few and tenuous and may be briefly commented upon as follows. In a few places the text of 𝔎 is identical to unique readings in 998 and Fa. 𝔎 reads παρά with 998 Fa instead of ὑπέρ in 2:13b and omits ἕν with 998 359 Fa in 2:14d. Should we align 𝔎 with the Egyptian Text on the basis of such scant evidence? It is after all, from the Genizah in Cairo, Egypt. We might expect an Egyptian text-type. But 𝔎 has τότε in 2:15e against its omission in B S* 998 and also reads in 2:16c against the Egyptian Text τῶν ἡμέρων τῶν ἐρχαμένων, a reading presumably equivalent to the variant ταῖς ἡμέραις ταῖς ἐρχομέναις in the LXX text tradition. Since 𝔎 was rendering the Hebrew slavishly, the inclusion of τότε is not surprising. These last two variants may be corrections in the Egyptian Text of a stylistic nature not in the common text inherited by 𝔎. In 2:19a 𝔎 has ἔστιν instead of LXX ἔσται. This agrees with 609 k 357 543 Fa in the LXX text tradition. Twice 𝔎 renders ‫יהיה‬, in 2:18c he employs ἔσται in a relative sentence and in 2:19a he uses ἔστιν in an indirect question. 𝔎 also reads ἐξισιάσει with 637txt in 2:19b instead of LXX ἐξουσιάζεται. Nothing can be proven from these data. In 2:15d LXX employs ἵνα τί for ‫ למה‬in Hebrew whereas 𝔎 renders by διὰ τί. In an earlier study I analysed all renderings of ‫ למה‬in the Three (Peter J. Gentry, “The Relationship of Aquila and Theodotion to the Old Greek of Ecclesiastes

General Assessments of Translation Technique

165

in the Marginal Notes of the Syro-Hexapla,” 80-81). The data show that while Aquila uses either ἵνα τί or εἰς τί for ‫למה‬, Theodotion prefers ἵνα τί but does use εἰς τί on occasion if the sources can be trusted. Not much evidence exists for the rendering of ‫ למה‬by Symmachus, but in Isaiah 58:3 διὰ τί is attributed to him. The σʹ reading here for Eccl 2:15 is εἰς τί. There are OG translators, however, who use διὰ τί for ‫( למה‬e.g. Exod 1:13, 5:22, Ps 41:9, Job 3:11, 7:20, 10:19, 13:24), so no clear connection to Symmachus is possible. For 2:16 de Lange notes that μνημόσυνον for ‫ זכרון‬instead of μνήνη as in LXX is one of two readings attributed to Symmachus. The evidence in App II for 2:16 includes a reading attributed to Symmachus which may be either ἀνάμνησις or μνημόσυνον. One cannot be sure since we are dealing with a retroversion from the Syro-Hexapla. Nevertheless, a glance at Hatch and Redpath’s Concordance to the Septuagint shows that the equivalence μνημόσυνον for ‫ זִ ָּכרֹון‬is already common in LXX Exodus (Exod 12:14, 13:9, 17:14, 28:12(bis), 28:23(29), 30:16, 36:14(39:7), so that we cannot use this to connect 𝔎 to Symmachus. In 2:14b 𝔎 employs πορεύομαι for ‫ הלך‬in Hebrew. De Lange notes that “this is the Greek verb normally used for the root ‫ הלך‬in ‘the Three’ (Aquila, Symmachus and Theodotion).”9 But it is also the equivalent standard for ‫ הלך‬for all translators in the body of texts loosely referred to as the Septuagint. Only a glance at the fifteen columns of πορεύομαι in Hatch and Redpath’s Concordance to the Septuagint is necessary to see this. We cannot connect 𝔎 to the Three on the basis of this verb. Perhaps more striking is the case in 2:21e. For MT ‫ רעה רבה‬LXX has πονηρία μεγάλη. 𝔎, however, renders κακία πολλή which is close to κάκωσις πολλή in Symmachus. A marginal note in 161 and the reading of 547 also have πολλή. The fact that the Bible text of Jerome supports πολλή is indication of influence from Symmachus in Jerome and these two witnesses in the LXX text tradition. In sum, I find one instance of possible influence from the three in 𝔎. We can say that it is a translation made from the Hebrew text. The translator knew and was influenced by the LXX Ecclesiastes. He advanced and perfected patterns in the LXX and developed further the approach seen in the καίγε tradition. There is little observable influence from Aquila and the translation technique is not as extreme or systematic as what we see in Aquila. The Greek is popular (rather than literary), but difficult to date precisely. Since the text comes from the Cairo Genizah and is Greek represented in Hebrew script, one is inclined to say that the text is Jewish. The translation was made directly from the Hebrew text but was influenced strongly by the Septuagint. The translator not only follows the Septuagint translation in his own work, but he extends and perfects the patterns of translation technique observable in the OG. The strong influence of the LXX and negligible influence from the Three 9  N. de Lange, “A Greek Translation of Kohelet (Ecclesiastes),” 72 n 5.

166

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

is perhaps unexpected, but then, while OG Ecclesiastes is not Aquila, the approach to translation is one that is a position of the Jewish community in the matter of representing the Hebrew Scriptures in Greek.

2.5  The Latin Vulgate (𝔙) A. Gasquet, et al., eds. Biblia sacra iuxta Latinam Vulgatam versionem ad codicum fidem, iussu Pii pp. XI, cura et studio monachorum Sancti Benedicti Commissionis pontificiae a Pio pp. X institutae sodalium, Romae : Typis polyglottis Vaticanis, 1926-1995. Robert Weber, et al., eds. Biblia Sacra iuxta Vulgatem Versionem. Stuttgart: Deutsche Bibelgesellschaft, 1969, 4th ed., 1994.

For the Vulgate, the edition of the Benedictine community of San Girolamo in Rome (Gasquet, et al. Biblia Sacra) has been the principal textual base. Ecclesiastes is found in volume XI (Proverbia, Ecclesiastes, Canticum Canticorum) published in 1957. The edition of R. Weber, Biblia Sacra, was also consulted. The Latin Vulgate was translated directly from the Hebrew and is a fairly good witness to the proto-Masoretic Text. There are places where Jerome was influenced by the Septuagint, but many more where he was influenced by Symmachus (see above § 1.2 on The Old Latin Translation). The siglum for the Vulgate is placed after the equal sign since it belongs largely to the witnesses for the parent text.

3.  Translation Technique: On the Character of the Translator of Ecclesiastes The Greek translation of Ecclesiastes in the Septuagint is characterised by extreme formal equivalence, so much so that scholars have suggested that it is the work of Aquila himself, a revisor of the Septuagint who flourished circa 120 CE. While the identity of the translator is still undetermined according to the most recent research, the character of the translation reveals that in fact some patterns are identical to those considered classical Aquila and others are clearly not Aquila. The approach of the translator to his task is one where faithfulness is defined and measured by the degree of quantitative alignment between Greek translation and Hebrew original. As a result, the text is difficult to read in places and almost incomprehensible at times from the point of view of the native speaker of Hellenistic Greek who has no knowledge of or recourse to the source text in Hebrew. The attempt of the translator to align his translation with the source language and text affects both lexical and syntactic choices in Greek. Choosing the same lexeme in Greek for every occurrence of a particular lexeme in the Hebrew regardless of the contextual meaning results in a high degree of stereotyping. Few calques or isolate renderings, however, are to be found. Like Aquila, then,

General Assessments of Translation Technique

167

consistency and uniformity in Hebrew-Greek equivalents is extremely rigid. Unlike Aquila, however, we do not see a set of Hebrew-Greek equivalents that is so highly refined and specialised that distinct equivalents are employed for each Hebrew lexeme, nor does the translator attempt to represent the root system of Hebrew by forming a set of equivalents from a single Greek stem to be used for Nouns and Verbs derived from a single root. Even more than in lexical equivalents, it is in the morpho-syntactic structures that the attempt to align the Greek and the Hebrew is most apparent and also where the results are so awkward or even absurd. One obvious and outstanding example is the rendering of ‫ את‬by σύν (1:14, 2:12, 2:17, 2:18, 3:10, 3:11bis, 3:17bis, 4:1, 4:2, 4:3, 4:4bis, 4:15, 5:3, 5:6, 7:14, 7:15, 7:18, 7:26, 7:29, 8:8, 8:9, 8:15, 8:17bis, 9:1bis, 9:15, 10:19, 10:20, 11:5, 11:7, 12:9, 12:14).10 That σύν is found with the dative in 9:11 is normal (σύν πᾶσιν αὐτοῖς); that it is found, however, with the genitive (9:15) or accusative (rel) to mark the direct object is contrary to Greek grammar. This characteristic of the translator may be further illustrated. In 2:1 ‫ וראה בטוב‬is rendered by καὶ ἰδὲ ἐν ἀγαθῷ. The verb ‫ ראה‬followed by ‫ ב‬here is idiomatic, conveying the sense ‘to enjoy’ (so NRSV). The Greek translator of Ecclesiastes rigidly renders ‫ ראה‬by εἶδον and ‫ ב‬by ἐν. Contextually, “to see in good” makes no sense in Greek. Again, a construction Article + πᾶς + Noun normally contrasts the whole with the part.11 Yet in 3:17 ‫ כי־עת לכל־חפץ‬is rendered ὅτι καιρὸς τῷ παντὶ πράγματι (cf. 3:1, 4:8, 4:16). The article τῷ is employed to represent the ‫ ;ל‬this effect creates difficulties for the Greek reader and skews the meaning of Hebrew while attempting to represent it formally in all respects. Other cases are perhaps readable, but awkwardly so. Note, for example, that ‫ עד אׁשר לא־יבאו ימי הרעה‬in 12:1 is rendered by ἕως ὅτου μὴ ἔθλωσιν ἡμέραι τῆς κακίας. With the negative μή, the result is awkward, but not necessarily impossible Greek. Elsewhere the rigid approach of the translator is carried off more cleverly. Hebrew ‫( גם‬1:17, 2:1, 2:7, 2:8, 2:14, 2:15, 2:15, 2:19, 2:21, 2:23, 2:23, 2:24, 2:26, 3:11, 4:4, 4:8, 4:8, 4:8, 4:11, 4:14, 4:16, 4:16, 5:9, 5:16, 5:18, 6:5, 6:9, 7:14, 7:21, 7:22, 7:22, 8:10, 8:12, 8:14, 8:16, 9:1, 9:1, 9:6, 9:6, 9:6, 9:12, 9:13, 10:20, 12:5) or ‫ וגם‬is (1:11, 3:13, 5:15, 6:3, 6:7, 7:6, 7:18, 8:17, 9:3, 9:11, 9:11, 9:11, 10:3, 11:2) is always rendered by καίγε except in a double translation in 7:22b where the equivalent is simply καί. Nonetheless, the equivalence ‫גם‬/‫ = וגם‬καίγε, while illustrating the spirit of the Greek translator of Ecclesiastes, is unlikely to have originated with him. Here the translator extends and perpetuates

10  2:12 and 9:11 are included against Ra; see J. Ziegler, “Die Wiedergabe der nota accusativi ’et, ’aet- mit σύν,” ZAW 100 Sup (1988):222-233. This, of course, does not affect the NETS Translation. 11  F. Blass and A. Debrunner, A Greek Grammar of the New Testament and Other Early Christian Literature, 9th-10th ed., edited and translated by R. W. Funk (Chicago: University of Chicago Press, 1961), § 275.7.

168

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

a tradition. Clearly, then, the Greek translator of Ecclesiastes belongs to the socalled καίγε tradition of translators.12

II.  Basic Principles for the Constitution of the Text An apparently eclectic assortment of problems in the textual transmission of LXX Ecclesiastes is examined and evaluated here. This exercise shows the reader how to make use of the data in the apparatuses to assess the reconstruction of the earliest form of the text as proposed by the editor. It illustrates the kind of research behind the choices made by the editor and shows how issues which might be considered by some to be secondary to the constitution of the text are inextricably intertwined and of potential significance to disciplines only laterally related to the study of the Septuagint. For each example, the Hebrew text of 𝔐 is provided alongside the text of the edition. These discussions reveal competing Tendenzen: (1) genealogical method of textual criticism, (2) probabilities derived from translation technique, (3) the possibility of corrections motivated by improvement of grammar and style, and (4) corrections toward the Hebrew parent text mediated by several sources.

1.  Inner Greek Scribal Errors Each translator followed a different approach to the task of translation. Broadly speaking, all translations can be categorised on a continuum between extreme formal and quantitative correspondence from source to target language on the one hand and dynamic, functional correspondence or free translation on the other. The person who produced the translation in Greek of Ecclesiastes followed an approach of extreme formal and quantitative correspondence between Greek and Hebrew so that for almost one hundred years his work was mistaken for that of Aquila himself. As a result, there are parts of this translation that are difficult for even a native speaker of Hellenistic Greek to comprehend unless they also knew Hebrew and could consult the Hebrew source text. Since scribes who copied the text did not know Hebrew and had no recourse to the parent text, sometimes they attempted to correct or revise the text on the basis of the context. Such corrections are commonly referred to as inner Greek scribal errors.13 Consider, first, the case of 4:9-10a. 12 See Peter J. Gentry, “Pre-Hexaplaric Translations, Hexapla, post-Hexaplaric Translations,” in Textual History of the Bible: The Hebrew Bible. Volume 1A Overview Articles, eds. Armin Lange and Emanuel Tov (Leiden: Brill, 2016), 211-235. 13  On inner Greek scribal errors, see J. W. Wevers, “The Use of Versions for Text Criticism: The Septuagint,” in La Septuaginta en la Investigacion Contemporanea (V Congreso de la IOSCS), ed. Natalio, Fernándes Marcos. Madrid: CSIC, 1985, 15-24.

Basic Principles for the Constitution of the Text

169

4:9-10a

𝔐 4:9 ‫ן־ה ֶא ָחד‬ ָ ‫טֹובים ַה ְּׁשנַ יִ ם ִמ‬ ִ ‫ׁש־ל ֶהם ָׂש ָכר טֹוב ַּב ֲע ָמ ָלם‬ ָ ֶ‫ֲא ֶׁשר י‬ 4:10a ‫ת־ח ֵברֹו‬ ֲ ‫ִּכי ִאם־יִ ּפֹלּו ָה ֶא ָחד יָ ִקים ֶא‬

Edition

4:9 ἀγαθοὶ οἱ δύο ὑπὲρ τὸν ἕνα, οἷς ἔστιν αὐτοῖς μισθὸς ἀγαθὸς ἐν μόχθῳ αὐτῶν· 4:10a ὅτι ἐὰν πέσωσιν, ὁ εἷς ἐγερεῖ τὸν μέτοχον αὐτοῦ ————— A B C S (870) 998 O L C´’ d k min verss ————— 10 πέσωσιν] πεση L(–125) (-σει 261) C´–299 (‑σει 609) 411 443 La160 Hilem et com 286,116 287,140 Aeth Arm Didlem 124,12 Ps.CatA 1103 Amb Ep 81,3.6 ter Inst virg 11,74 Chrom Matth 22,3,5.6 ter PetrChr 170,5 = Ald 𝔙; absc 870 Didlem et com 126,9 126,10 | ὁ εἷς] inc C; > 534; + ο ετερος 411 L(–125) C´–299 Amb Inst virg 11,74 Chrom Matth 22,3,5 = Ald | ἐγερεῖ 870] …]ρει Didlem 124,11 (sed hab Didcom 124,14 127,13); εγειρει S A 998 (εγει̣[…) 475 L– PM (125) 261 C´–299 390-260 254´ 248 296 311 339 547 645 706 728 766 Syh Antioch 1676 DamM PsChr (inc C; sed hab Aeth Anton 1108 Dionlem 227 Metlem et com IV.1,3,45 Ol = Gra Ra) = A Ald; εγειρη 261 390-cII–260 561 698 (εγηρη); εγειρι 342-357 155 336 534 548; εγεροι DamA ; + τον ενα 357 | μέτοχον] μετεχοντα 139-147-159-503-560-798; πλησιον 336´ ————— 10 πέσωσιν] αʹ σʹ θʹ ὁμοίως τοῖς οʹ πέσωσιν Syh

The Hebrew text can be rendered as follows: 9 10a

Two are better than one when they have a good reward for their labour; for if they fall, the one can lift up his partner.

In verse 10a, the end of the protasis and the beginning of the apodosis is clearly marked in 𝔐 by the fact that the verb ‫ יִ ּפֹלּו‬is in pause. According to the critical reconstruction of the Greek text, the translation is literal and the translator has fastidiously followed the order of the words in his source text: 9

Better are two than one, because they have a good reward in their toil. 10 For if they fall, the one will raise his partner up14 14  Peter J. Gentry, “Ecclesiast,” in A New English Translation of the Septuagint edited by Albert Pietersma and Benjamin G. Wright (Oxford: Oxford University Press, 2007).

170

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

Nonetheless, for scribes copying the Greek text without any knowledge of the Hebrew, delimiting the protasis and apodosis was not so transparent. Instead of both people falling down and each having a partner to help him get up, the position of ὁ εἷς naturally suggests a singular verb so that one of them falls and the other person who remains standing is the one who raises his fallen comrade. In this case ὁ εἷς belongs to the protasis and is construed as the subject of the verb πέσωσιν and the form is corrected from third person plural to third person singular πέσῃ. Since ὁ εἷς is now no longer understood as the subject of ἐγερεῖ, the same scribes supplied ὁ ἕτερος as an explicit subject. It is now abundantly clear, contrary to the source text, that the benefit of two is that when only one falls, the other person who is still standing can then raise his fallen comrade. Following the lemma in the Göttingen Edition are three separate sections. The first is called the Kopfleiste which lists the witnesses extant for the text on that page of the edition. This is extremely useful since some manuscripts are fragmentary or have missing pages. One does not have to return to the Einleitung or Introduction to discover this for each manuscript, but can determine right away which witnesses are available for the problem in question. The second section is the First Apparatus (Apparatus I) which provides the text history of the Old Greek. The third section is known as the Second Apparatus (Apparatus II) and supplies extant readings from revisions of the Septuagint derived from Origen’s Hexapla. These later revisions have frequently corrupted and influenced the textual transmission of the Old Greek as recorded in the First Apparatus. The first lemma in Apparatus I is πέσωσιν. All variation in the manuscript tradition is provided before a vertical line (|) which separates this instance of variation from the next. Following the square bracket is the first variant πεση, a Third Person Singular Aorist Active Subjunctive. This reading is supported by 411, a manuscript which belongs to the codices mixti but occasionally supports the hexaplaric or O group. The siglum L, representing the Lucianic recension, follows O in the order in the apparatus, since Lucian based his revision on a hexaplaric text. Following the symbol L is –(125) in superscript. This means that all members of the Lucianic group except 125 support the variant πέσῃ. The number 125 is in parentheses to show that this is due to a larger omission in this manuscript. If one consults A. I. 2. one will see that MS. 125 has a number of large lacunae. Next is C´. Upper case italicised C refers to the main group of catena manuscripts and the prime indicates the addition of the first sub-group designated cI; so, C´ = C + cI. In the Ecclesiastes Edition, cI represents a group of five manuscripts which are not catena manuscripts, but whose text is derived from the biblical text of manuscripts in the tradition represented by C. Following the symbol C´ is –299 in superscript. This means all manuscripts comprising C + cI except 299 have the variant πέσῃ (i.e. twenty-one manuscripts). Here MS. 299 is not in parentheses and can be automatically assumed to support the lemma πέσωσιν unless it is listed in sup-

Basic Principles for the Constitution of the Text

171

port of another variant. Since there is only one variant to the lemma, 299 can be assumed to have the lemma (e silentio). Following the siglum C´ in parentheses is (-σει 609) indicating that one of the members of the cI group has an itacistic variant but may nonetheless be considered to support the variant πέσῃ. The same is true for MS. 261 of the L group, which also supports the variant πέσῃ. Finally, from a list of some thirty unclassified minuscules, only 443 also has πέσῃ. Following the main witnesses—the Majuscules, Papyri, and Minuscules—are listed the evidence of the Early Daughter Versions and then the Church Fathers. Greek Fathers are listed first and Latin Fathers afterwards. Normally all citations in the first five hundred years are listed for the Greek Fathers. Here we have a citation of this verse of Ecclesiastes in catena fragments of Didymus the Blind’s Commentary on the Psalms critically edited by E. Mühlenberg. This is significant since the witness of the Commentary on Ecclesiastes by Didymus was not preserved at this point. While citations from the Church Fathers may not always be the most reliable witness to a text for a variety of reasons, at the same time they do represent indirectly Greek manuscripts that are older than most of the witnesses to the text and so can be of great importance. In this case we see that the reading of the Lucianic and Catena groups goes back to an earlier point in the textual transmission than the date of the manuscripts in these groups. No less than eight citations from three Latin Fathers are listed in a concise fashion. The principle for citing the Latin Fathers differs from that for citing the Greek Fathers, as it relates to the problem of the Old Latin. Actual manuscript sources of the Old Latin for Ecclesiastes are extremely sparse and, as we have seen, the lemma of Jerome’s Commentary on Ecclesiastes is not a reliable witness to the Old Latin as Rahlfs thought. Therefore, our most important source for the Old Latin is frequently citations in the Latin Fathers. As a general rule, only quotations in the Latin Fathers which are clearly not from the Vulgate are cited in the apparatus. In this instance, since the Third Person Singular verb is obviously not a correction to the Hebrew text, the citations of the Latin Fathers along with the evidence of both the lemma and commentary of Jerome’s Commentary and La160 clearly attest the almost lost text of the Old Latin. This witness can be combined with the citation in Didymus to show how early this inner Greek scribal error entered the textual transmission of the Septuagint Ecclesiastes. The same could be said of the value of citing the Armenian and Ethiopic daughter versions—these translations were made from Greek manuscripts that are earlier than most of the minuscules cited in the apparatus. At the end an equal sign shows that the reading πέσῃ or the equivalent in Latin is also that of the Aldine and Vulgate. It is the normal practice to collate for the Göttingen Editions the three earliest and most important Printed Editions of the Septuagint, the Complutensian Polyglot (1514-17), Aldine (151819), and Sixtine (1587). In Ecclesiastes these editions are based largely on MSS.

172

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

248, 68, and B respectively. Economy is maintained in the apparatus by citing them only when the Complutensian differs from 248, when the Aldine differs from 68 and when the Sixtine differs from B. This notation in the apparatus here, is important, then, as an instance is recorded where obviously the editors of the Aldine corrected the manuscript they were using as their source on the basis of manuscripts in the catena tradition. Readers interested in the Latin tradition are saved the trouble of checking the Vulgate as the note indicates that Jerome maintained the Third Person Singular against his Hebrew text in his translation of the Vulgate. Just before the vertical line marking a new problem, we read “absc 870, Did 126,9 126,10.” MS. 870 is a IV-V century papyrus and Did refers to the Tura Papyrus preserving the Commentary on Ecclesiastes of Didymus the Blind deriving from the IV century. The siglum absc is an abbreviation for abscissus and indicates that these extremely early sources are damaged at this point and so cannot be used as witnesses in this problem. The notation absc is important to exclude evidence that may be deduced e silentio as we see in the next paragraph. Readers and users of the edition might well want to know what precisely is the support for πέσωσιν. This can be readily adduced by subtracting from the Kopfleiste manuscripts listed for variants that deviate from the lemma and this yields the following list of witnesses: A C B-S-68´’-998 O 299-cII d k 155 248´ 252 296´ 311 336´ 338 339 542 543 547 549 645 698 706 766 795 Syh According to the principles of the Göttingen Septuaginta, only Greek majuscules and minuscules can be deduced e silentio from the apparatus. In this case, one daughter version and citations in five patristic sources also support the lemma: Aeth Anton Didcom Dion Met Ol. This is indicated by the notation ‘sed hab’ in parentheses. Thus, the notation (sed hab Aeth Anton 1108 Dionlem 227 Metlem et com IV.1,3,45 Ol = Gra.) means that the Ethiopic, and citations from Anton, Dion, Met and Ol have the lemma rather than the variant. This notation also alleviates the necessity to list the witness before the square bracket. The Second Apparatus provides any extant materials from the Jewish revisors. These are normally fragments surviving from Origen’s Hexapla. The note αʹ σʹ θʹ ὁμοίως τοῖς οʹ πέσωσιν represents a retroversion from the Syriac of the SyroHexapla. The margin and text of Syh are as follows: ‫ܡܩܝܡ ܠܫܘܬܦܐ ܕܝܠܗ܂‬ ܼ ‫ܡܛܠ ܕܐܢ‬ ܼ ‫ܢܦܠܘ ݂ܢ ܚܕ‬ ܿ ‫܂ܐ܂ܣ܂ܬ܂ ܀‬Index super ‫ܢܦܠܘܢ‬ Syhmg: ‫ܢܦܠܘܢ܀‬ ‫ܒܗ‬ ܼ ‫ܒܕܡܘܬܐ‬ ܼ ݂ ܼ

Approximately 70 notes employing the notation ὁμοίως are in the margin of Syh in Ecclesiastes. I have discussed the meaning and purpose of notes of this type at

Basic Principles for the Constitution of the Text

173

length in articles published in Aramaic Studies.15 These marginal notes appear to be derived from a different manuscript than the manuscript used as the exemplar for the text of the Syro-Hexapla. In all probability they came from a manuscript early in the catena tradition where the lemma was πέσῃ and the purpose of the scholiast was to note that the three Jewish revisors had a reading identical to that of the οʹ text, i.e. the Fifth Column of Origen’s Hexapla, i.e. πέσωσιν, although this reading was not his lemma. A citation from Olympiodorus’ Commentary on Ecclesiastes forms part of the Catena at this point and the comment of Olympiodorus makes clear that πέσωσιν was the reading of the οʹ text. A solid vertical line in the apparatus separates the variation on the lemma πέσωσιν from another problem—variants on the lemma ὁ εἷς, each divided by a semicolon. First, inc C indicates that the uncial C, a V century palimpsest, is unreadable at this point. At first, collation for the Göttingen Edition was made from the published collations of Tischendorf.16 Later, a careful collation was made from examination of the manuscript in person by Felix Albrecht.17 Albrecht was able to clarify many readings that were uncertain in the edition of Tischendorf. Nonetheless, this passage remains difficult to read. If, however, new photographic techniques were applied to this manuscript, possibly the under-writing might be accurately read. MS. 534, but not its congeners 68 and 602-613 omitted ὁ εἷς. No obvious palaeographic or scribal error is readily apparent as the reason for this. Since 534 has πέσωσιν in the protasis and the itacistic spelling εγειρι in the apodosis, the omission of ὁ εἷς does not alleviate the main problem in the text. A number of witnesses have ὁ ἕτερος after ὁ εἷς—almost exactly the same witnesses that have the Third Person Singular verb as already noted. A second solid vertical line introduces another case of variation which at first seems more complex than is really the situation. The lemma of the edition has a Future Active Indicative of ἐγείρω, while a number of manuscripts attest a Present Active Indicative ἐγείρει or a Present / Aorist Active Subjunctive ἐγείρῃ; εγειρι is an itacistic spelling in which one cannot determine whether the scribe intended the Present Indicative or Present Subjunctive. Probably the Subjunctive is also an error for the Present Indicative motivated by itacism. The Future ­Indicative is almost certainly the original text. The Greek translator renders yiqtol forms 15  See Peter J. Gentry, “Hexaplaric Materials in Ecclesiastes and the Rôle of the Syro-Hexapla,” Aramaic Studies 1 (2003): 5-28, idem, “Propaedeutic to a Lexicon of the Three: The Priority of a New Critical Edition of Hexaplaric Fragments,” Aramaic Studies 2.2 (2004): 145-74, and idem, “The Relationship of Aquila and Theodotion to the Old Greek of Ecclesiastes in the Marginal Notes of the Syro-Hexapla,” Aramaic Studies 2.1 (2004): 63-84. 16  Constantin von Tischendorf, Codex Ephraemi Syri rescriptus sive fragmenta utriusque Testamenti e codice Graeco Parisiensi celeberrimo quinti ut videtur post Christum seculi (Leipzig: Bernh. Tauchnitz, 1845). 17  Felix Albrecht, “Codex Ephraemi Syri rescriptus. Neue Lesarten zum Septuagintatext des Koheletbuches,” ZAW 122 (2010): 272-279.

174

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

in Hebrew in 179 instances: in 114 he employed a Future Indicative, in 32 an Aorist Subjunctive—all but one in dependent clauses, in 12 a Present Indicative, in eight a Participle, in four a Present Subjunctive—all in dependent clauses, in four an Aorist Indicative, in four an Infinitive, and in one instance a Present Imperative.18 None of the cases of Present Indicative are found in the apodosis of conditional sentences. The persistent commitment of the translator to extreme formal and quantitative correspondence between source and target languages makes it highly probable that he employed a Future Indicative here. The fact that a Present General Condition in Greek in which one normally employs a Present Indicative (or rarely Subjunctive, cf. Job 41:18) in the apodosis rather than a Future More Vivid Condition in which one usually employs a Future Indicative in the apodosis easily explains why Greek scribes changed the Future to a Present.19 The Present General Condition better suits this context. Condition

Protasis

Apodosis

Present General Future More Vivid

ἐάν + Subjunctive ἐάν + Subjunctive

Present Indicative Future Indicative

Once more, the textual witnesses supporting the lemma may be deduced e silentio by subtracting witnesses for variants from the Kopfleiste as follows: B-68’ O-411 299-561 k 161 252 338 443 542 543 549 795 Note that in the Kopfleiste 870 is in parentheses. This IV-V century papyrus contains only excerpts from Ecclesiastes and preserves only part of the text for the page of the Göttingen Edition in question. This warns the reader not to draw conclusions e silentio. It is precisely the reason why 870 is listed before the square bracket when it attests the lemma. Since the Present Indicative is attested as early as 300 CE in 998 (Hamburger Papyrus Bil. 1), a full listing of the patristic witnesses is important to show that what is proposed as the original text is attested as early as that of the inner Greek scribal error. The equals sign followed by the sigla Gra Ra indicates that the reading chosen as original text in the Göttingen Edition was adopted as early the edition of Grabe in 1709,20 a major milestone in critical editions of the Septuagint. 18  For an exhaustive presentation of all the evidence, see Yun Yeong Yi, “Translation Technique of the Greek Ecclesiastes” (Ph.D. diss. Southern Baptist Theological Seminary, 2005), 141-159. 19  See H. W. Smyth, Greek Grammar (Cambridge, MA: Harvard University Press, 1920), §§ 2297, 2336. 20  J. E. Grabe, ed., Septuaginta Interpretum tomus IV (Oxford, 1709).

Basic Principles for the Constitution of the Text

175

To return to matter of the witness of 998, note that this is represented in the apparatus as follows: 998 (εγει̣[…). 998 is listed as a witness supporting the variant εγειρει. In parentheses, the reader is shown exactly what is extant in this manuscript: one can clearly read epsilon, gamma, epsilon and a damaged iota. Then a lacuna occurs in the papyrus. The editio princeps of the papyrus, however, offers εγε[ρει] as its text.21 For the Göttingen Edition, I have carefully checked all instances of lacunae in the manuscript against the photographs provided. There is no doubt that an iota follows the epsilon, similar to the iota in the word εγειρε at the end of the next line. The editors count agreements and disagreements with the major uncials as follows:22 Agreements with 998 B 343 S (Sc) 172 (37) A 171 C 177

Against 998 41 208 (53) 211 155

It is by no means impossible that 998 could have agreed with A and S against B here, but the editors have a predilection for restoring lacunae according to B. At any rate, some uncertainty over the reading of 998, which can be easily gathered from the apparatus, shows again the importance of the patristic testimony to indicate the early date for both the original text and the inner Greek scribal error that arose at this point. 9:6

Inner Greek Scribal Error in Best Witnesses

𝔐 9:6c

‫עֹולם‬ ָ ‫ין־ל ֶהם עֹוד ְל‬ ָ ‫וְ ֵח ֶלק ֵא‬

Edition

9:6c καὶ μερὶς οὐκ ἔστιν αὐτοῖς ἔτι εἰς αἰῶνα ————— A B C S 998 O L C´’–299 k d–342 min–359 verss ————— 6 καί 4°] pr οτι d (357 vid); + γε B-S*-68´’ 998 V-411 L C´’–260 d 155 336´ 443 645 698 766 Arm Ol ————— 21  B. J. Diebner, and R. Kasser, eds., Hamburger Papyrus Bil. 1. Die Alttestamentlichen Texte des Papyrus Bilinguis 1 der Staats- und Universitätsbibliothek Hamburg (Cahiers d’Orientalisme XVIII; Geneva: Patrick Cramer, 1989), 247. 22  Ibid., 34.

176

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

No readings are extant for 9:6 in the Second Apparatus. In the First Apparatus, there are two variants associated with the fourth instance of καί in the verse. In the d group, the καί is preceded by ὅτι. Normally the d group consists of four manuscripts (254-342-357-754), but the Kopfleiste shows that 342 does not attest this text. Consultation of the Einleitung reveals that 342 has a lacuna from πολλούς in 7:29 to the end of the text. So, in this particular problem, the siglum d only represents three manuscripts and the reading is uncertain in one of them, i.e. 357. The insertion of ὅτι is clearly secondary, supplied by scribes who wanted to clarify the relationship between 9:6ab and 9:6cd: 6 καί γε ἀγάπη αὐτῶν καί γε μῖσος αὐτῶν καί γε ζῆλος αὐτῶν ἤδη ἀπώλετο, καὶ μερὶς οὐκ ἔστιν αὐτοῖς ἔτι εἰς αἰῶνα ἐν παντὶ τῷ πεποιημένῳ ὑπὸ τὸν ἥλιον.

Separated by a semicolon is a second variant. Significant and weighty witnesses have καί γε instead of καί. These are the Egyptian Group (B-S*-68´’ 998) supported by the d Group, part of the O Group (V-411), the Lucianic and Catena groups, a dozen or so other minuscules and the Bible text of the Commentary on Ecclesiastes by Olympiodorus, one of the four or five patristic works on Ecclesiastes that have survived and dating to the early sixth century. From the criteria employed by Vinel,23 the sheer number and weight of these witnesses would suggest this reading is original. A reader may well want to know what witnesses support the lemma of the Göttingen Edition. These may be determined e silentio by subtracting support for the variant from the Kopfleiste as follows: A C Sc O–V k 125II 161 248 252 260´ 296´ 311 338 339 542 543 547 549 706 795 Hi Syh Sa2 Fa Arm = Ra What may not be determined e silentio in the system used in these critical editions is precisely what daughter versions and church fathers support the lemma. In this case the Sacra Parallela of John of Damascus and the Commentary of Metrophanes (Dam Metlem et com VIII.5,7,51 Hi Syh Sa2 Fa1 3 Aeth) are the only available patristic sources along with most of the daughter versions. The notation sed hab in parentheses in effect lists witnesses before the square bracket. The patristic sources are from the Seventh and Ninth Centuries respectively, and so earlier than all of the minuscules listed for this variant. Thus, two majuscules, a fair number of minuscules and most of the daughter versions support the lemma. In this particular problem, translation technique has a large role to play. 23  F. Vinel, L’Ecclesiaste: Traduction du texte grec de la Septante, introduction et notes (La Bible d’Alexandrie, 18; Paris: Cerf, 2002).

Basic Principles for the Constitution of the Text

177

Careful comparison of source and target texts show that the translator of the Greek Ecclesiastes was extremely literal and rendered his source text both lexically and syntactically in a highly stereotypical manner. LXX Ecclesiastes belongs to the so-called καίγε tradition and he consistently employed καίγε for ‫ גם‬or ‫וגם‬ in the Hebrew source text. The one-to-one correspondence between Hebrew text and Greek translation is perfect. Hebrew ‫( גם‬1:17, 2:1, 2:7, 2:8, 2:14, 2:15, 2:15, 2:19, 2:21, 2:23, 2:23, 2:24, 2:26, 3:11, 4:4, 4:8, 4:8, 4:8, 4:11, 4:14, 4:16, 4:16, 5:9, 5:16, 5:18, 6:5, 6:9, 7:14, 7:21, 7:22, 7:22, 8:10, 8:12, 8:14, 8:16, 9:1, 9:1, 9:6, 9:6, 9:6, 9:12, 9:13, 10:20, 12:5) or ‫ וגם‬is (1:11, 3:13, 5:15, 6:3, 6:7, 7:6, 7:18, 8:17, 9:3, 9:11, 9:11, 9:11, 10:3, 11:2) is always rendered by καίγε except possibly in a double translation in 7:22b where the equivalent is simply καί. In addition, waw in MT is rendered by καί in 356 instances and by zero in five while only three instances of asyndeton are rendered by καί.24 Thus we can be almost one hundred per cent sure that the translator rendered ‫ וְ ֵח ֶלק‬by καὶ μερίς and not by καί γε μερίς. On the other hand, Greek scribes who had no recourse to the Hebrew text would be almost certain to insert καί γε here given the context here in 9:6ab where we have καί γε three times. The scribes were also familiar with the expressions normally used by the translator as a whole. It would be so natural to insert γε here and as 998 shows, this occurred fairly early in the textual transmission and consequently dominated a large part of the textual transmission. The witness of the Hexaplaric Group is important here because it shows that the text received by Origen did not have the γε and there is no evidence that Origen removed elements in the Fifth Column to make the text match the Hebrew. Origen’s οʹ Text is also as early as the witness of 998. 9:7

Inner Greek in Other Witnesses

𝔐 9:7a

‫ֵלְך ֱאכֹל ְּב ִׂש ְמ ָחה ַל ְח ֶמך‬

Edition

9:7a Δεῦρο φάγε ἐν εὐφροσύνῃ ἄρτον σου ————— A B C S 998 O L C´’–299 d–342 k min verss ————— 7 ἄρτον B O–637-411 254´ 68 125II 534 542 DamK Met VIII.6,2 = Ra] absc 998; pr τον rel ————— 7 ἐν εὐφροσύνῃ] σʹ μετ᾽ εὐφροσύνης 161 (ind mend ad πίε (7b)) 248 (ind mend ad πίε (7b))

24  For an exhaustive presentation of all the evidence, see Yun Yeong Yi, “Translation Technique of the Greek Ecclesiastes,” 337-339.

178

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

It is interesting to compare the example in 9:6c with that of 9:7a. The critical text as reconstructed for the Göttingen Edition is δεῦρο φάγε ἐν εὐφροσύνῃ ἄρτον σου. A large number of manuscripts have an article before ἄρτον. To determine exactly which manuscripts are included in the siglum rel one must subtract the witnesses before the square bracket as follows: A C S 637 L C´’ k 357 155 161 248 252 260´ 296´ 311 336´ 338 339 359 443 534602-613 543 547 549 645 698 706 766 795 Again, the difficulty of the present system in the Göttingen apparatus is that the reader may not be able to determine e silentio citations from the church fathers included in rel. There are three: CyrHier 1104 Dam–K Ol. The abbreviation absc means abscissus; there is a lacuna in 998 so that it may not be used as a witness in this problem. This is unfortunate, since this is a papyrus from around 300 CE. Like the problem in 9:6c, here too, knowledge of translation technique offers an Archimedean point by which one may gain leverage over the problem. It is difficult to decide merely on the grounds of families of manuscripts and the age, character, and weight of the witnesses. This problem is treated by Joseph Ziegler in a magisterial study on the approach of the Greek translator of Ecclesiastes to articulation—his last work before his death in 1988.25 Nouns in the Hebrew parent text having pronominal suffixes are normally unarticulated by the translator. This contravenes the requirements of syntax in Greek but maintains formal correspondence to the code of the source text. The evidence is fully provided in the article by Ziegler. Here, Greek scribes naturally supplied the article since it is the norm in Greek when the noun is accompanied by a pronoun in the genitive. Neither the problem in 9:6c nor 9:7a is that important. In both cases the variation in the text is insignificant. In both cases a knowledge of the habits and patterns of the translator—absolutely consistent, fixed and rigid in the case of OG Ecclesiastes—provides an Archimedean point to gain leverage over deciding only on the basis of manuscript groupings. Both cases are excellent examples of the kind of improvements to the text made by Greek scribes who struggled with the woodenness of the translation and had no access to the parent text to see what the translator was attempting to achieve in his translation. What is interesting, however, is that a similar constellation of groups and individual manuscripts support the original text in one instance and the inner Greek scribal error of the text in the other. The following chart shows witnesses that support the lemma in one case and the vari25  J. Ziegler, “Der Gebrauch des Artikels in der Septuaginta des Ecclesiastes,” in Studien zur Septuaginta—Robert Hanhart zu Ehren, edited by D. Fraenkel, U. Quast and J. W. Wevers (Abhandlungen der Akademie der Wissenschaften in Göttingen, Philologisch-Historische Klasse, 190; Mitteilungen des Septuaginta-Unternehmens, 20; Göttingen: Vandenhoeck & Ruprecht, 1990), pp. 83-120.

Basic Principles for the Constitution of the Text

179

ant in the other. This is followed by a shorter list of witnesses which have the inner Greek scribal error in both cases or the original text in both instances. 9:6c Lemma Variant 9:7a Lemma Variant

A C 637 k 161 248 252 260´ 296´ 311 338 339 543 547 549 706 795 B V-411 68 534 B V-411 68 534 A C 637 k 161 248 252 260´ 296´ 311 338 339 543 547 549 706 795

Witnesses Having Inner Greek Scribal Error in Both Instances: S C´’ L 155 336´ 443 534-602-613 645 698 766 Witness Having Original Text in Both Instances: 253 125II-542 Note that the witnesses supporting the lemma in the first case support the variant in the second and vice versa. Clearly, one cannot always reconstruct the earliest form of the text merely from preferences for certain manuscripts or families of manuscripts.

2.  Variants Preserved Almost Entirely in Patristic Sources 7:20 𝔐

‫ִּכי ָא ָדם ֵאין ַצ ִּדיק ָּב ָא ֶרץ‬ ‫ֲא ֶׁשר יַ ֲע ֶׂשה־טֹוב וְ לֹא יֶ ֱח ָטא‬

Edition

7:20 ὅτι ἄνθρωπος οὐκ ἔστιν δίκαιος ἐν τῇ γῇ, ὃς ποιήσει ἀγαθὸν καὶ οὐχ ἁμαρτήσεται. ————— A B C S 998 O L C´’ d k min verss ————— 20 ἄνθρωπος] pr ο 371; post ἔστιν tr 336´ 443 Didcom 221,8 Ps.CatA 539 (sed hab Hilem 309,307 Dam Metlem et com VII.1,3,81 PsChr Hi Gal 384); post δίκαιος tr V; > Ath IV 92 Or IV 377,9: cf 3 Reg 846 | οὐκ ἔστιν] post δίκαιος tr L Ol | om δίκαιος Armte Agnellus 45 Coll Avell 97,9 ConcilCarth Reg 115 ConcilMilev 7 Lucul 810 826 PsAug Ful 197,21 202,12 PsSalo Ecl 1005 Ruf Lev 459 SedScot Eph 1005

The quantitative approach of the Greek translator provides as a literal rendering of the Hebrew text: “for as to humanity, there is not a just person in the earth who will do good and not sin.” In the Hebrew, the sentence structure entails left dislocation or y movement—an element is removed from the clause and given first

180

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

position.26 Such a literal rendering produces an awkward construction in Greek. Some scribes resolved this by moving ἄνθρωπος after the copula. Some scribes omitted ἄνθρωπος while others omitted δίκαιος. This last variation is attested only by 11 citations in 9 Latin Fathers and by the Armenian. One of the main principles in the Göttingen Editions is that citations from patristic sources are only given when they support a Greek manuscript. Here the citations from the Latin Fathers do not support any Greek witness. Nonetheless, in conjunction with the Armenian, there is the distinct possibility that a Greek witness did attest the omission just as Greek witnesses also attested other inner Greek “improvements” to an awkward sentence structure. The omission of the word “just” may also represent the Old Latin at this point.

3.  Interdependence of LXX Text-History and Text-History of the Jewish Revisors 6:5 𝔐

‫א־ר ָאה וְ לֹא יָ ָדע‬ ָ ֹ ‫ם־ׁש ֶמׁש ל‬ ֶ ַ‫ּג‬ ‫נַ ַחת ָלזֶ ה ִמּזה׃‬

Edition

6:5 καί γε ἥλιον οὐκ εἶδεν καὶ οὐκ ἔγνω, ἀνάπαυσις τούτῳ ὑπὲρ τοῦτον. ————— A B C S (818) 998 O L C´’ d k min–(155) verss ————— 5 ἀνάπαυσις] αναπαυσεις B-S-998-68 A O–637 d–357 296´ 311 547 698 706 752 795 Syh (absc 818; sed hab Hi Aeth Arm Ol Metlem et com V.5,14,100 Amb Jb 2,4,15 An Scrip 1,10 Spec 392,13 = Gra. Ra 𝔐 𝔔) = Compl; αναπαυσιν 475-637-411 C´’ k 443 766A SaI II 2 6 Fa1 2 Did 174,27 ↓ ; + και ουκ επειραθη διαφορας ετερου (εταιρου 754) πραγματος προς ετερον d–357 ↓ ; post τούτῳ tr 336´ | τοῦτον] τουτο 601 k; τουτων 261 645; τουτου 637; + ουδε επειραθη (επιραθη V) διαφορας (διαφθορας 637) ετερου (ετερον 475) πραγματος προς (> 637) ετερον O 547mg ↓ ————— ݁ 5 καὶ οὐκ ἔγνω ἀνάπαυσις τούτῳ ὑπὲρ τοῦτον] σʹ ‫ܐܬܢܣܝ ܒܫܘܚܠܦܐ ܕܣܘܥܪܢܐ ܐܚܪܢܐ‬ ‫ܘܐܠ‬ ‫ ܠܘܬ ܐܚܪܢܐ‬Syh (= καὶ οὐκ ἐπειράθη διαφορᾶς ἑτέρου πράγματος πρὸς ἕτερον ↑) | ἀνάπαυσις] αʹ σʹ θʹ ‫( ܢܝܚܐ‬i.e. αʹ θʹ ἀνάπαυσιν ↑; σʹ διαφορᾶς: omnes singulares) Syh

26  If an element is removed and not replaced by a pronoun, the fronting is called y movement. See T. Givón, Syntax, 2 vols. (Philadelphia: John Benjamins, 1984, 1990).

Basic Principles for the Constitution of the Text

181

Here in 6:5 the textual transmission of the Old Greek (OG) is corrupted by an entire line from Symmachus: καὶ οὐκ ἐπειράθη διαφορᾶς ἑτέρου πράγματος πρὸς ἕτερον. It is possible to demonstrate solely on the basis of principles of textual criticism that this line is clearly secondary in the textual tradition since it is attested by three of the four manuscripts in the d group after ἀνάπαυσις and by the O group at the end of the verse. Normally the presence of the same text in different locations is a clear sign of a later insertion to the original text. This is confirmed by the fact that the text begins by καὶ οὐκ in d–357 and by a different clause connector, οὐδέ, in the O group. Furthermore, a glance at the Hebrew text shows that the line constitutes a double translation, another clear indication of something secondary in the textual tradition. Both the lemma and margin of the Syro-Hexapla as well as the lemma of Jerome’s Commentary and the text of the Vulgate are significant for this problem: Syh ‫ܙܐ ܘܐܠ ݂ܝܕܥ ܵܢܝܚܐ܂‬ ݂ ‫ܘܐܦ ܫܡܫܐ ܐܠ ݂ܚ‬ ‫ܗܢܐ ܝܬܝܪ ܡܢ ܗܢܐ ܂‬ Syhmg: ‫ ܂ܐ܂ܣ܂ܬ܂ ܢܝܚܐ ܀‬Index super ‫ܵܢܝܚܐ‬ ܿ Syhmg: ‫ܐܬܢܣܝ‬ ‫ ܂ܣ܂ ܘܐܠ‬Index super ‫ ܗܢܐ‬1° ܿ ‫ܐܚܪܢܐ ܠܘܬ ܐܚܪܢܐ ܀‬ ‫ܒܫܘܚܠܦܐ ܕܣܘܥܪܢܐ‬ Hi Vulg

nec cognovit, requies huic magis quam illi neque cognovit distantiam boni et mali

While the text of Syh corresponds entirely to the text of 998 A B S alii, a marginal note provides a reading attributed to Aquila, Symmachus, and Theodotion for the term ‘rest’ and also a longer marginal note attributed to Symmachus. The index for the Symmachus reading although connected to ‫ ܗܢܐ‬supplies an alternative to OG καὶ οὐκ ἔγνω, ἀνάπαυσις τούτῳ ὑπὲρ τοῦτον. This text attributed to Symmachus in the Syro-Hexapla matches precisely the additions in the O and d groups. The rendering ἐπειράθη for ‫ ידע‬may at first seem strange but the translation “he experienced” is in fact contextually sensitive and is also employed by Symmachus for ‫ ידע‬in 8:5, in contrast there as in 6:5 to the equivalent γινώσκω used by OG. The rendering distantiam in Jerome’s Vulgate Translation is derived from διαφορᾶς in Symmachus, while boni et mali is a free adaptation of his own. The lemma of Jerome’s Commentary, however, in this case represents the Old Latin uncontaminated by Symmachus and not corrected spontaneously towards the Hebrew as is sometimes the case with Jerome. Another problem in the same stretch of text has to do with the word ‘rest’. There are three possibilities: (1) the Nominative Singular ἀνάπαυσις adopted as the lemma of the Göttingen Edition, (2) the form ἀναπαύσεις which could be construed either as Nominative Plural or Accusative Plural supported by the

182

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

witnesses B-S-68-998 A O–637 d–357 296´ 311 547 698 706 752 795 Syh = Compl, and (3) the Accusative Singular ἀνάπαυσιν supported by the witnesses 637411 C´’ k 260´ 443 766A Did 174,27 Fa SaI. Papyrus 818, dated to the end of the third century, is damaged at this point and cannot be used as a witness. Support for the lemma has been provided in parentheses using ‘sed hab’ as only Greek MSS. could be determined from the Kopfleiste e silentio: C L 357 125II 248´ 252 336´(post τούτῳ tr) 338 339 534´ 542 543 549 645 766B Hi Arm Ol Metlem et com V.5,14,100 Amb Jb 2,4,15 An Scrip 1,10 Spec 392,13 = Gra. Ra. Both Grabe and Rahlfs, earlier editors, have correctly seen the Nominative Singular ἀνάπαυσις as the lectio difficilior in spite of strong support in terms of character, date, and number of witnesses for the other two variants. Again, the problem is one of inner Greek scribal error. Due to itacism, the Nominative Singular and Accusative Plural are identical, and the noun is more easily construed as the object of ἔγνω than as subject of a nominal sentence in v. 5b independent of the sentence in v. 5a. Since the construction with the Nominative Singular matches that in the Hebrew, while the Accusative does not, the evidence for inner Greek scribal error is clear and persuasive. Before considering the Accusative Singular variant ἀνάπαυσιν, note that a marginal note in the Syro-Hexapla attributes the term ‘rest’ in the singular to all Three revisors. The second marginal note, however, attributes a longer reading to Symmachus for a lemma that overlaps with that for the first note. In combination with the Greek manuscripts in the textual tradition of the OG that are influenced by the Symmachus reading it is clear that Symmachus rendered ‫ נחת‬by διαφορᾶς rather than a form of the noun ἀνάπαυσις. The first note, then, can be interpreted in two ways: (1) either only Aquila and Theodotion had ‘rest’ and the attribution to Symmachus is incorrect, or (2) the purpose of the note is to indicate that the Three Revisors had a term in the singular while the lemma has a noun in the plural. In the second scenario, the exact equivalent in lexical terms was not the purpose of the scholiast, only the issue of grammatical number. Therefore, consideration of the Accusative Singular ἀνάπαυσιν attested by the Catena group and part of the O group results in suggesting that this variant is a kind of hyper-correction to the Hebrew text by Greek scribes who had no knowledge of Hebrew or direct access to the Hebrew text, but used the Three as a sort of guide for this. It may also be just a simple correction based upon better style: the singular is more contextually suitable than the plural. The first mistake involved itacism and construing the noun as object of ἔγνω and then, based on a textual tradition that had ἀναπαύσεις, the Accusative Singular ἀνάπαυσιν arose as a stylistic correction or hyper-correction to the Hebrew via the Three. What the variants in the First and Second Apparatus show, then, is that critical reconstruction of the text of the Three is based on analysis of the textual tradition of OG and vice-versa, knowledge of the text of the Three may clarify the textual tradition of the OG.

Basic Principles for the Constitution of the Text

7:12 𝔐

183

‫ִּכי ְּב ֵצל ַה ָח ְכ ָמה ְּב ֵצל ַה ָּכ ֶסף‬

Edition

7:12 ὅτι ἐν σκιᾷ αὐτῆς ἡ σοφία ὡς σκιὰ τοῦ ἀργυρίου, ————— A B C S 998 O L C´’ d k min(–155) verss ————— 12 ὅτι] pr σκεπει σοφια ως σκεπει το αργυριον 503 ↓ | ἐν] η k ; + τη 299 OlΑΓ Co | σκιᾷ] κακια 357 | αὐτῆς] αυτοις 609; αυτου OlΒΗΖ; αυτη 534 OlΔΙΚΜ | ἡ σοφία] της σοφιας Sc V; εν σοφια 797; ως σοφια 357; om η 253; > 543 | om ὡς σκιά—σοφίας 357: homoiot | ὡς σκιά] ως κακια 571*; ω σκια 998 299-571c 336: haplogr | om τοῦ B-68´ 998 d(–357) 336´ 443 PsChr SaI 6 ————— 12 ὅτι—ἀργυρίου] σʹ ὅτι (> Hi) ὡς (> 161 248 252 539) σκέπει σοφία ὁμοίως (ὡς 161 248 252 539) σκέπει (+ et Hi) τὸ ἀργύριον 161 248 252 539 (s nom) Hi Syh

An entire line corresponding to 7:12a has been preserved from the revision of Symmachus. This line has been transmitted in Greek in four manuscripts as a marginal note as well as being cited in the Commentary of Jerome and being preserved in Syriac in the margin of the Syro-Hexapla. The Latin and Syriac evidence is as follows: Hi: quomodo protegit sapientia, similiter protegit et pecunia ܿ ‫ܒܗ ܒܕܡܘܬܐ‬ ܿ ܿ ‫ܕܡܣܬܪܐ ܚܟܡܬܐ‬ Syh: ‫ܡܣܬܪ ܟܣܦܐ‬ ‫܂ܣ܂ ܡܛܠ ܕܐܝܟܢܐ‬ ܼ The textual value of the non-Greek witnesses is significant here as P. Marshall notes: “both Syh and Hi agree against the Greek MSS. in including ὡς (quomodo ܿ Hi; ‫ ܕܐܝܟܢܐ‬Syh) before σκέπει 1° and in reading ὁμοίως (similiter Hi; ‫ܒܗ ܒܕܡܘܬܐ‬ 27 Syh) before σκέπει 2°.” This reconstruction was already posited by F. Field in 1875.28 The tradition transmitted in the marginal notes of the four Greek MSS. may well represent corruption and influence from the MSS. of the OG in respect to the ὡς before the second instance of σκέπει in this verse. The text of the Vulgate is remarkably similar to the text of Symmachus: Vulgate: sicut enim protegit sapienta sic protegit pecunia 27 Phillip S. Marshall, “A Critical Edition of the Hexaplaric Fragments of Ecclesiastes” (Ph.D. diss. Southern Baptist Theological Seminary, 2007), 202. 28  F. Field, Origenis Hexaplorum quae supersunt (2 vols.; Oxford: Clarendon Press, 1875), 2:392 n. 25.

184

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

W. W. Cannon included the case of 7:12 in his list of instances where Jerome’s Latin Vulgate was influenced by Symmachus.29 Indeed, evidence abounds that demonstrates Jerome’s dependence upon readings in Symmachus to guide his translation of the Vulgate. S. D. Weeks, however, has proposed that Jerome’s rendering in the Vulgate may be due to his reading kaph in his parent Hebrew text instead of beth.30 Yet this suggestion is also a possible explanation for the rendering of Symmachus and even of that of the Old Greek for at least the second beth in the line. With this discussion of the revision of Symmachus in mind, one can clearly see that the text of LXX in only one manuscript, 503, has been corrupted by the reading of the text of Symmachus, producing in fact, a double rendering of the source text at this point. What is also noteworthy is that the variant preserved in the text of 503 corresponds to the tradition for Symmachus that is preserved in the margin of the four Greek MSS. 161, 248, 252, 539 in contrast to that preserved by Jerome and Syh. The reading of 503 is hardly significant for the constitution of the text, but reveals a genealogical relation between the tradition in the Catena Group to which 503 belongs and the textual tradition for the text of the Three represented in the four Greek MSS. Such information is of value in sorting out the text history in spite of the fact that the evidence is scant in this instance as well of being of interest to scholars seeking to understand either the catena tradition or the relationship of Jerome and the Three to the text history of OG. 2:25 𝔐

Ra

2:25

‫ּומי יָ חּוׁש חּוץ ִמ ֶּמּנִ י‬ ִ ‫אכל‬ ַ ֹ ‫ִּכי ִמי י‬

ὅτι τίς φάγεται καὶ τίς φείσεται πάρεξ αὐτοῦ;

Edition

2:25 ὅτι τίς φάγεται καὶ τίς πίεται πάρεξ αὐτοῦ; ————— A B C S 998 O L–(125) C´’ d k min verss ————— 25 πίεται] absc 998; φείσεται Hi (parcet) Syh (sed hab Dionlem 223 GregNy 364,2 370,22 Ol La94 95 PsIgn 227,3 (bibit)) = Gra. Ra. ↓ | αὐτοῦ] αυτων 766B ————— 29  W. W. Cannon, “Jerome and Symmachus: Some Points in the Vulgate Translation of Koheleth,” ZAW, N.F. 4 (1927): 191-199. 30  See Phillip S. Marshall, “A Critical Edition of the Hexaplaric Fragments of Ecclesiastes” 202.

Basic Principles for the Constitution of the Text

185

25 φάγεται] αʹ φείσεται σʹ ἀναλώσει 161mg (ind ad τίς 1°) 248mg (ind ad φάγεται) | φείσεται] αʹσʹ ὡσαύτως φείσεται θʹ πίεται Syh (ind ad φείσεται) ܿ ‫ܡܢܗ܂‬ ‫ܘܡܢܘ ܢܚܘܣ ܣܛܪ‬ ܼ ‫ܕܡܢܘ ܢܐܟܘܠ‬ ܼ ‫ܡܛܠ‬ ݁ mg Syh : ‫܂ܐ܂ܣ܂ ܗܘ ܕܐܠ ܗܟܘܬ ܢܚܘܣ܀‬Index super‫ܢܚܘܣ‬ Syhmg: ‫܂ܬ܂ ܢܫܬܐ܀‬Index super‫ܢܚܘܣ‬ Field: αʹ σʹ ὡσαύτως, φείσεται. Field: θʹ πίεται

Rahlfs (1935) reconstructs φείσεται as OG, as did Grabe before him (1709). His apparatus gives as support O La†, meaning Origen’s recension, the Old Latin and at most not more than one additional minuscule. While we do not know all the reasoning behind Rahlfs’ reconstruction, we can safely say that he chose φείσεται because the reading better corresponds to the Hebrew parent text, it better approximates Aquila (and he thought the OG Aquila or Aquilanic) and it is supported by both Jerome’s Old Latin and the Text of the Syro-Hexapla—the latter being a witness for Origen’s Text. One of the most enlightened treatments of this crux criticorum is that of Jan de Waard in 1979.31 While he is concerned with establishing the Hebrew text, he provides excellent insights and a good summary of previous solutions. For the reading φείσεται he lists Origen’s Recension, the Vetus Latina, Aquila and Symmachus as witnesses, also noting parcet in Jerome. This looks like five witnesses compared to the two listed by Rahlfs. The reading parcet is, in fact, the Old Latin of Rahlfs. So, aside from αʹ and σʹ, he has only the same two witnesses as Rahlfs. Indeed, since 1935, no further witness for φείσεται has come to light in spite of the † sign given by Rahlfs. [Recently, A. Schoors has argued for the analysis of Ellermeier that ‫ חוׁש‬means to ‘fret’ or ‘worry’ against de Waard who argued for the meaning ‘enjoy’.32 The central issue, here, however, is not how modern scholars construe the meaning of Ecclesiastes but how the Greek translator understood his source text.] Both readings must be considered paleographically: tisfeisetai versus tispietai. If one were to assume a dittography of the sigma, a spi could have been read as fi. This would have occurred in the period of the uncials. It assumes πίεται is the lectio difficilior and φείσεται arose solely as an inner-Greek 31  J. de Waard, ‘The Translator and Textual Criticism (with Particular Reference to Eccl 2,25)’, Biblica 60 (1979), pp. 509-529. 32  See Antoon Schoors, The Preacher Sought to Find Pleasing Words: A Study of the Language of Qoheleth, Part II: Vocabulary (Orientalia Lovaniensia Analecta 143; Leuven: Peeters, 2004), 384-386. On p. 65 Schoors renders 2:25 “who can eat or who can have enjoyment/worry apart from him?” and on p. 185 he translates “who can eat and who can enjoy (or fret) without him?” Thus Schoors appears to place the solution for which he argues on pp. 384-386 secondary in his renderings.

186

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

development. Yet this does not explain how πίεται dominates the text tradition, and it is a real stretch to confuse the two words on this basis alone.33 One might assume πίεται arising as an inner-Greek development as it more naturally goes with φάγεται, and φείσεται is contextually odd. So φείσεται would be the lectio difficilior. Yet πίεται may be considered just as possibly based upon the same parent text as φείσεται. De Waard is worth citing in full here: There is no serious reason whatsoever to believe that this reading goes back to a different Hebrew Vorlage yišteh. In fact, it can be explained in three different ways: (a) as a Verlegenheitslesart and an introduction of the pair ’kl—šth from verse 24b; (b) as a specific rendering of MT ya­ḥûš, taken in the generic sense of “enjoy”; (c) as the rendering of a Hebrew verb which got lost in Hebrew, but which still exists in Arabic (ḥ asa—to drink). Possibilities (a) and (b) are the more probable ones and they are not necessarily mutually exclusive, since the existence of the word pair ’kl— šth in verse 24b may have inspired the specific rendering. It should at least be noted here that possibility (b) did not get the attention it deserves. Anyway, text-critically one can retain the important conclusion that in none of these cases the reading goes back to a different Vorlage.34

If, as I have argued, OG Ecclesiastes is closer to the καίγε tradition than to Aquila, the first proposal would fit the translation technique well. I deem it likely that the OG translator had difficulty with the verb, whether read ‫ יָ חּוׁש‬or ‫יָ חּוׂש‬, and provided a contextually based rendering. It also makes sense that Theodotion retained this text in his revision. As I have maintained, the marginal notes in the Syrohexapla come from a different manuscript than the manuscript which was the Vorlage for the Text. Fiftyfive notes have ὁμοίως τοῖς οʹ, one employs ὡς instead of ὁμοίως (ὡς οἱ οʹ 3:10), and fourteen have just ὁμοίως. Only 2:25 has ὡσαύτως. Normally ὁμοίως indicates sources having a text identical to the lemma and ὁμοίως τοῖς οʹ indicates sources having a text identical to the οʹ text. The scholiast doubtless had πιέται in his text and noted in the margin that “Aquila and Symmachus in like manner had φείσεται, while Theodotion had πιέται.” He had to add φείσεται because it was not his lemma and did not add τοῖς οʹ because theirs was not equal to the οʹ text. According to the Colophon for Ecclesiastes in Codex Syro-Hexaplaris Ambrosianus, the parent text for the text of Syh was corrected by Eusebius and 33  An ingenious solution is proposed by McNeile who suggests that an original πείσεται could explain both πίεται and φείσεται and could be based upon ‫ חוש‬understood according to the meaning in Aramaic and Post Biblical Hebrew ‘feel pain’ > enjoy. See A. H. McNeile, An Introduction to Ecclesiastes (Cambridge, University Press, 1904), p. 158. Positing an intermediate step in the process, however, which has no attestation whatsoever, is just too ingenious. 34  J. de Waard, ‘The Translator and Textual Criticism’, p. 522.

Basic Principles for the Constitution of the Text

187

Pamphilus.35 This can explain why it is not the οʹ text and why it is different from the O Group. It also assumes a different meaning for ὡσαύτως from ὁμοίως: why should we assume that the two words are used with identical intent or meaning? The readings of the Three must be carefully sorted out in this passage as well. Field gives φείσεται for αʹ σʹ and πίεται for θʹ based upon the marginal note in Syh. The reading ἀναλώσει attributed to σʹ in 248 appears to be ignored. The evidence of 248 and Syh are as follows: 248

Ind ad φάγεται. Margin: αʹ φείσεται σʹ ἀναλώσει.

Syh

Ind ad φείσεται. Margin: αʹ σʹ ὡσαύτως φείσεται θʹ πίεται.

When one considers the text in Hebrew and the equivalents normally used by LXX translators, it is probable that in 248 ἀναλώσει is for φάγεται and that φείσεται is for LXX πίεται. Thus, Aquila may have read τίς φάγεται καὶ τίς φείσεται while Symmachus had τίς ἀναλώσει καὶ τίς φείσεται. We may thus assume that the note is correct in both 248 and Syh: both αʹ and σʹ are similar (ὡσαύτως): “Who will spend/eat and who will be thrifty?”, whereas OG and θʹ have “Who will eat and who will drink?” For proof that this is the correct meaning of ὡσαύτως in the marginal note of Syh, see a similar case in 3:11.36 As an aside, it should be noted that a lexicon of the Three would have to assign different meanings for φείσεται in the cases of Aquila and Symmachus (the context requires ‘spare’ for Aquila and ‘be thrifty’ for Symmachus) and that this could not be done without first reconstructing a critical text of the Three. As for the Latin reading parcet, the probability that Jerome corrected the Bible text of his Commentary on the basis of Aquila is strong. La94 95 has bibet (supported by the citation in Pseudo-Ignatius) and this is much more likely to be the Old Latin. We can now summarise as follows. The reading πίεται is based upon the same parent text as φείσεται. It has better claim to fit the translation technique of the OG since OG is not Aquila and closer to Theodotion. Jerome’s Bible text is not the Old Latin and Syh is probably not Origen’s recension here, but a correction of it based upon Aquila. Rahlfs’ supports for φείσεται are removed and we should adopt the reading of the entire Greek manuscript tradition including the Old Latin as in La94 95. Not only does Aquila help us to reconstruct the true OG, but the reconstruction of the OG helps us sort out our hexaplaric witnesses and reconstruct αʹ and σʹ for the entire verse once we understand the original lemma of the marginal note and the correct meaning of ὡσαύτως. 35  For a discussion on the colophon and its significance, see Peter J. Gentry, ‘Hexaplaric Materials in Ecclesiastes and the Rôle of the Syro-Hexapla’, Aramaic Studies 1.1 (2003), pp. 6-7. 36  3.11 employs ὡς (‫ )ܐܝܟ‬whereas 2.25 uses ὡσαύτως (‫)ܗܟܘܬ‬.

188

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

7:6 𝔐

Ra

7:6a

‫ִּכי ְכקֹול ַה ִּס ִירים ַּת ַחת ַה ִּסיר‬

ὅτι ὡς φωνὴ τῶν ἀκανθῶν ὑπὸ τὸν λέβητα,

Edition

7:6a ὡς φωνὴ τῶν ἀκανθῶν ὑπὸ τὸν λέβητα, ————— A B C S 998 O L(–125) C´’ d k min(–155) verss ————— 6 ὡς B-S-68´’ 998 C´ 357 296´ 311 338 443 645 706 795 SaI 6 Fa1 Ammon Antioch 1724 c a V Bas III 961 Dam (ωσπερ DamKVRMH TL A Max II 996) Amb Exh virg 11,76 BenA Conc 1126 Eugip Reg 28,74 Spec 557,8 (et sicut Reg Mag 179,183)] pr ὅτι (sub ※ Syh) rel (Hi Syh Arm Arab Did Metlem et com VI.8,3,28 PsChr = Ra 𝔐 𝔔 Peshmss Vulg) ↓ ————— 6 ὡς φωνὴ τῶν ἀκανθῶν ὑπὸ τὸν λέβητα] διὰ γὰρ φωνὴν ἀπαιδεύτων ἐν δεσμωτηρίῳ γίνεταί τις 161 248 | τῶν ἀκανθῶν] ἀπαιδεύτων Syh (‫)ܕܐܠ ܪܕܝܐ‬

The first problem is presented as background for issues arising later in this study. Rahlfs’ Text has ὅτι ὡς at the beginning of 7:6a. His apparatus shows only that ὅτι is omitted in B and S. No doubt his choice was based on the fact that the Greek translation is extremely literal and he could not imagine that the translator would omit a word in his parent text. Moreover, Rahlfs believed that the Bible text of Jerome’s Commentary on Ecclesiastes was a reliable source for the Old Latin and therefore one of the earliest witnesses, albeit indirect, to the LXX. Rahlfs’ choice is not sound for a number of reasons. First, as I have shown in other studies, in a few instances he based his edition solely upon Jerome’s Text going against the entire manuscript tradition in Greek.37 Good evidence is supplied in my Grinfield Lectures on the Septuagint to show that Jerome corrected the Old Latin towards his Hebrew text in an impromptu fashion as he recorded the text of his commentary and so the Bible text of the Commentary is not a reliable witness to the Septuagint.38 Instead, the citations from the Latin Fathers constitute a better witness to the Old Latin and they all have an equivalent for ὡς but not for ὅτι.

37  See Peter J. Gentry, “Propaedeutic to a Lexicon of the Three: The Priority of a New Critical Edition of Hexaplaric Fragments,” Aramaic Studies 2.2 (2004) 145-74. 38  Peter J. Gentry, “‘The Role of the “Three” in the Text History of the Septuagint’: II. Aspects of Interdependence of the Old Greek and the Three in Ecclesiastes,” Aramaic Studies 4.2 (2006): 153-192.

Basic Principles for the Constitution of the Text

189

Second, the witness of the Syro-Hexapla is important in this problem. The equivalent for ὅτι is preceded by an asteriskos and followed by a metobelos. This is a clear indication that the ὅτι was not part of the text Origen received as the Septuagint, but was added from one of the Three and appropriately marked to show that the text of the Septuagint was lacking what corresponds to this word in Origen’s Hebrew text. The Syro-Hexapla is one of the most reliable sources for preserving the diacritical marks used in the Fifth Column of the Hexapla and at the beginning of Ecclesiastes in Rahlfs’ Text he indicates the Syro-Hexapla in the apparatus as his source for the οʹ text. He ought to have taken this witness far more seriously. Third, Rahlfs did not account sufficiently for the fact that in a number of instances, the parent text of the Greek translator may have differed from 𝔐. At present this is best seen by studying the commentary of Goldman in the Biblia Hebraica Quinta.39 In this particular problem he proposes the same text as I do for the earliest form of the Old Greek. It is possible that by either haplography in the parent text or by parablepsis due to homoiarcton the Greek translator read ‫ כקול‬instead of ‫כי כקול‬. Fourthly, while the text of the O group in this case contaminated a large part of the textual tradition, B S are strongly supported now by 998, the Catena group and a good number of unclassified minuscules. And formal correspondence on the part of the translator is maintained. The decision to banish ὅτι from the critical edition should not be difficult to acknowledge. 8:8d 𝔐

Ra

8:8d

‫וְ ֵאין ִמ ְׁש ַל ַחת ַּב ִּמ ְל ָח ָמה‬

καὶ οὐκ ἔστιν ἀποστολὴ ἐν τῷ πολέμῳ,

Edition

8:8d καὶ οὐκ ἔστιν ἀποστολὴ ἐν ἡμέρᾳ πολέμου, ————— A B C S 998 O L(–125) C´’ (–299) d k min(–155) verss ————— 8 καί 2° ∩ 3° 252 | ἀποστολή] υποστολη 543 549 | ἐν ἡμέρᾳ πολέμου = 𝔖] in bello Hilem et com 316,102 317,117 = ἐν τῷ πολέμῳ Ra. 𝔐 ↓ ————— 8 καί 2°—πολέμου] σʹ οὐδε ἔστι(ν) παρατάξασθαι (-ται 161) εἰς πόλεμον 161 248 252

39  A. Schenker et al. eds., Biblia Hebraica Quinta. Fascicle 18: General Introduction and Megilloth (Stuttgart: Deutsche Bibelgesellschaft, 2004).

190

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

Rahlfs’ Handausgabe has ἐν τῷ πολέμῳ instead of ἐν ἡμέρᾳ πολέμου as in the Edition. The words ἐν τῷ πολέμῳ constitute a conjecture based, no doubt, upon in bello in the Bible text of Jerome and the fact that this was precisely equivalent to 𝔐. Rahlfs’ conviction that the OG was, in fact, the rendering of Aquila meant that he did not permit any reading that did not represent complete and precise formal equivalence to the Hebrew text. Rahlfs also believed that the Bible text of Jerome’s Commentary was an accurate and completely reliable witness to the Old Latin. The Bible text of Jerome, in fact, represents here his Hebrew text and not the Old Latin, as he sporadically and spontaneously corrected the Old Latin base for the Bible text according to the Hebrew. Furthermore, as in many similar situations, there may well also be influence from Symmachus whose translation is functional, yet accurately represents the Hebrew text. As the Second Apparatus shows, the rendering of Symmachus for 8:8d is preserved in marginal notes in three manuscripts as follows: οὐδε ἔστι(ν) παρατάξασθαι εἰς πόλεμον and at least in regard to the prepositional phrase corresponds formally to the Hebrew while ἐν ἡμέρᾳ πολέμου does not. Jerome’s rendering in the Vulgate nec sinitur quiescere ingruente bello certainly follows the interpretation of Symmachus. All manuscripts of Ecclesiastes have ἐν ἡμέρᾳ πολέμου and this should be taken as the text of OG. It may be that the OG translator had a different parent text at this point or that due to an error of sight in the process of translation he read ‫ ּב)יו(ם מלחמה‬instead of 𝔐 ‫ּב ִּמ ְל ָח ָמה‬.ַ It may just be that this is one of a number of places where the OG translator must be allowed to have a rendering that does not correspond formally to the Hebrew in a mechanical way. Elsewhere I have argued as have other scholars, that LXX Ecclesiastes is definitely not Aquila. The approach to translation fits somewhere within the καίγε tradition and is similar to the work of Theodotion. This assessment of translation technique in OG Ecclesiastes allows for the recognition that while the translator is committed to a high level of formal correspondence, there is the possibility, as in the renderings of Theodotion in Job for example, for some functional renderings and some sporadic and spontaneous departure from an extreme formal and quantitative approach to translation. The line immediately previous to v. 8d is also relevant, but contains difficult problems. The evidence for the line is supplied and a brief discussion of the problems before returning to the issues involved in 8:8d. 8:8c 𝔐

Ra

8:8c

‫וְ ֵאין ִׁש ְלטֹון ְּביֹום ַה ָּמוֶ ת‬

καὶ οὐκ ἔστιν ἐξουσία ἐν ἡμέρᾳ τοῦ θανάτου,

Basic Principles for the Constitution of the Text

191

Edition

8:8c καὶ οὐκ ἔστιν ἐξουσιάζων ἐν ἡμέρᾳ θανάτου, ————— A B C S 998 O L(–125) C´’ (–299) d k min(–155) verss ————— 8 ἐξουσιάζων 2°] pr ο OlΑΓ; εξουσια B-68´ 998 357 SaI 2 Fa2 3 Geo = Ra | ἐν ἡμέρᾳ 1°] ημερας A; ημεραις 609 | om ἐν 2° OlΒΗ | θανάτου] pr του O-411 130 542 766A SaI II 2 Fa3 (sed hab Fa2) = Ra 𝔐; + και ουκ εστιν υποστολη εν ημερα θανατου 543: dittogr | καί 2° ∩ 3° 252 ————— 8 NIL

Before considering the relevance of 8:8c for 8:8d, a couple of problems must be discussed. For the Handausgabe Rahlfs chose ἐξουσία, the reading of B, against ἐξουσιάζων, the reading of S and A. Now that all available sources before 1500 have been collated we can see that B is supported by 998 and its congeners 68 and 534 and one member of the d group (357), while the rest of the textual tradition supports S and A. No doubt B and 998 provide an extremely early witness, but we should pause before adopting their witness against the rest of the tradition. Consideration of the approach and habits of the translator gives us an Archimedean point from which we can gain leverage to move the world in this problem. All four instances of the verb ‫ ׁשלט‬in the Hebrew Qoheleth are rendered by ἐξουσιάζω (2:19, 5:18, 6:2, 8:9). The noun or adjective ‫ ִׁש ְלטֹון‬is rendered by a participle of ἐξουσιάζω in 8:4 and by the noun ἐξουσία in 8:8, at least according to Rahlfs’ Text. The adjective ‫ ַׁש ִּליט‬is rendered by a participle of ἐξουσιάζω in all three occurrences (7:19, 8:8, 10:5). It is clear that the patterns of the OG translator constitute a probability against the choice of Rahlfs in 8:8. Consideration of internal evidence also does not support the choice of Rahlfs very well. If ἐξουσία is, in fact, original, perhaps ἐξουσιάζων arose due to palaeographic factors from ΕΞΟΥΣΙΑ ΕΝ, although such an explanation is not highly convincing. Indeed, this argument could provide support for the other reading as well. If ἐξουσιάζων is original, one can easily explain ἐξουσία as an attempt to match 8c with the form and structure of 8d so that a noun, ἐξουσία, must match the noun ἀποστολή. Replacement of ἐξουσιάζων by ἐξουσία could be an inner Greek scribal error. In conclusion, the weight of external evidence, internal evidence, and the probability of how the translator would work in this instance are against Rahlfs. The variant offered by B and 998 belongs to a group of stylistic corrections to OG characteristic of this part of the textual tradition.

192

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

Second, note that the O group and 411 as well as four additional minuscules have an article before θανάτου. In addition 125II and 542 form a manuscript pair and so count as a single witness, and moreover are related to the text of Syh. This problem is also treated by Ziegler in his study on the approach of the Greek translator of Ecclesiastes to articulation. Normally, nouns in bound phrases which are not articulated in Hebrew are also not articulated in Greek. In approximately 10 instances, individual manuscripts articulate the nomen rectum. When the bound phrase is articulated in Hebrew, the nomen rectum is articulated also in Greek although in all instances but three some witnesses lack the article. In 5:10a, the article is supported by only one witness and in 8:8c, as we have seen, by nine sources counted as five separate witnesses. Rahlfs’ Text has the article before θανάτου in 8c—in his case solely on the basis of V (O = V-253-637)—and τῷ πολέμου in 8d. Thus, he relegates the article to the apparatus in 5:10a and retains it in 8:8c. Ziegler argues the article belongs in the apparatus in both instances. No doubt Rahlfs retained it in 8:8 because of formal correspondence to the Hebrew. Why does the O group have the article? Normally Origen did not correct the Fifth Column on the basis of the Hebrew, but the colophon to Syh indicates it was translated from manuscripts of the Hexapla corrected by Pamphilus and Eusebius. Thus, correction towards the Hebrew is possible in manuscripts of the O group. Conversely, possibly scribes omitted the article in 8c since the matching phrase in 8d did not have it. This last possibility would have to have occurred extremely early to influence all the text tradition except manuscripts representing Origen’s Fifth Column and so is not persuasive. The relevance of 8:8c for 8:8d may now be considered. Is it not possible that the OG translator adopted a freer approach rendering ‫ ַּב ִּמ ְל ָח ָמה‬to make it match the phrase in 8c? This is more probable than the approach of Rahlfs who has to assume that the copyists did the same thing but so early that it affected even 998 and indeed the entire surviving textual tradition. The example of 8:8d shows that even in following the entire manuscript tradition one has to balance consideration of translation technique, the work of Origen, the work of Jerome, and weigh probabilities of changes made by copyists against the probability of what came from the hand of the translator. 8:10 𝔐 ‫יתי ְר ָׁש ִעים ְק ֻב ִרים וָ ָבאּו‬ ִ ‫ּוב ֵכן ָר ִא‬ ְ ‫ן־עׂשּו‬ ָ ‫ּומ ְּמקֹום ָקדֹוׁש יְ ַה ֵּלכּו וְ יִ ְׁש ַּת ְּכחּו ָב ִעיר ֲא ֶׁשר ֵּכ‬ ִ

Ra

8:10

καὶ τότε εἶδον ἀσεβεῖς εἰς τάφους εἰσαχθέντας, καὶ ἐκ τόπου ἁγίου ἐπορεύθησαν

Basic Principles for the Constitution of the Text



193

καὶ ἐπῃνέθησαν ἐν τῇ πόλει, ὅτι οὕτως ἐποίησαν. καί γε τοῦτο ματαιότης.

Edition

8:10 καὶ τότε εἶδον ἀσεβεῖς εἰς τάφους εἰσαχθέντας, καὶ ἐκ τόπου ἁγίου, καὶ ἐπορεύθησαν καὶ ἐπῃνέθησαν ἐν τῇ πόλει, ὅτι οὕτως ἐποίησαν. καί γε τοῦτο ματαιότης. ————— A B C S 998 O L(–125) C´’ (–299) d k min(–155) verss ————— 10 εἰς—εἰσαχθέντας] sepultos et venerunt Hi = 𝔐 ↓ | εἰς τάφους] εις ταφον 357 766A Syh SaI II 2 Geo; εις τοπον 336´; εις κολασιν 159mg 411mg: ex Olcom; post εἰσαχθέντας tr 766B SaI II 2 Fa2 3 (absc Fa1) | εἰσαχθέντας] αχθεντας C 798-295-260 543 549 OlΙΚΜ Met VII.12,2; καταχθεντας 766B; εισταχθεντας 698 | om καί 2° 357 | ἐκ] απο O-411 539; οσοι 766; > OlΑ | τόπου Hi Syh SaI = 𝔐] > Fa1 2 3; του rel (539) | ἁγίου] + επορευθησαν Sc O–637 254´ 338 547 OlΑΓ; + απεστησαν 766 | om καί 3° Sc 637-411 797-cII 539 Hi Fa1 2 3 Arm Met VII.13,2 = Ra 𝔐 ————— 10 εἰσαχθέντας—τόπου ἁγίου] αʹ καὶ ἦλθον ἐκ τόπου ἁγίου 161 248 252 (s nom); σʹ οἳ (> 161 248) καὶ ὁπότε (ποτε 252 Syh) περιῆσαν (ησαν 252 Syh) ἐν τόπῳ ἁγίῳ 161 248 252 539 (s nom) Syh | καὶ ἐπορεύθησαν—πόλει] αʹ θʹ iverunt et gloriati sunt in civitati; σʹ revertebantur (ανεστρεφον 161 248; ανεστρεφοντο 252; αναστρεφομενοι 539) laudentes in civitati Syh

In 8:10b Rahlfs’ Text has καὶ ἐκ τόπου ἁγίου ἐπορεύθησαν whereas the Göttingen Edition has καὶ ἐκ τόπου ἁγίου, / καὶ ἐπορεύθησαν. The difference entails proposing that the OG translator either had a parent text ‫ והלכו‬or misread ‫ יהלכו‬as such—the problem concerns the similarity of waw and yodh in the Hebrew script of the Herodian period. The difference in line division is not relevant. Yohanan Goldman, editor of Ecclesiastes for Biblia Hebraica Quinta has καὶ ἐκ τοῦ ἁγιου καὶ ἐπορεύθησαν G* in his apparatus. This constitutes a claim for his readers that he is giving them the original LXX even though problems exist in its textual transmission. He devotes almost a page and a half to all the problems in this verse in the Textual Commentary portion of BHQ and proposes the following as the original Hebrew text, which I have set beside 𝔐 for comparison: 𝔐 ‫יתי ְר ָׁש ִעים ְק ֻב ִרים וָ ָבאּו‬ ִ ‫ּוב ֵכן ָר ִא‬ ְ [10] ‫ן־עׂשּו‬ ָ ‫ּומ ְּמקֹום ָקדֹוׁש יְ ַה ֵּלכּו וְ יִ ְׁש ַּת ְּכחּו ָב ִעיר ֲא ֶׁשר ֵּכ‬ ִ

194

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

Goldman’s Proposed Orig Text (BHQ)

‫יתי ְר ָׁש ִעים ְק ֵר ִבים יָ בֹאּו‬ ִ ‫ּוב ֵכן ָר ִא‬ ְ [10] ‫ן־עׂשּו‬ ָ ‫ּוב ִמ ְק ָּדׁש יְ ַה ֵּלכּו וְ יִ ְׁש ַּת ְּבחּו ָב ִעיר ֲא ֶׁשר ֵּכ‬ ְ Only part of the discussion by Goldman is necessary for our purposes: Nowhere does ‫ הלך‬piel mean “to go out,” a meaning often proposed here. Qoheleth probably meant “to walk in,” and the alteration of the preposition ‫ ב‬in ‫ ובמקדש‬is easily explained in view of the preceding scribal error involving ‫קרבים‬. If the wicked are not entering the sanctuary (‫)קרבים‬, but are brought to the grave (‫)קברים‬, they can no longer be brought into the sanctuary. That is why the original ‫ ובמקדש‬has been altered to ‫( וממקדש‬G), connecting this to the verb ‫ יהלכו‬which follows—and so imposing an alien meaning on this verb. In G and M, an attempt to make a clear separation between the wicked dead and the sanctuary may be observed. In the proto-M text, a cj. was read instead of the ‫ י‬in ‫( יבאו‬G); ‫( ובאו‬M). In the Vorlage of G, this same phenomenon transpired with ‫( יהלכו‬M) becoming ‫( והלכו‬G)—Rahlfs omits the cj., and, in fact, chooses the Origenian text, but the large majority of the Greek mss. attest the cj. καὶ here. The history of the text can be traced in two main stages: (1) A confusion between ‫ קרבים‬and ‫ קברים‬led to the alteration of ‫ ובמקדש‬into ‫( וממקדש‬G); (2) Interpreting that form, G then read ‫והלכו‬, leaving ‫ וממקדש‬in an erratic position (which gave rise to the insertion of ἐπορεύθησαν after καὶ ἐκ τοῦ ἁγίου by a corrector of Codex Sinaiticus). The proto-Masoretic text avoided this by reading ‫ ובאו‬instead of ‫יבאו‬, thus integrating ‫ וממקדש‬within the second phrase, and adding ‫ מקום‬for the sake of enlarging the concept of “sanctuary” (perhaps “synagogue” was meant: see Gordis, Koheleth—The Man 286).40

Whether original LXX is ἐκ τοῦ ἁγίου or ἐκ τόπου ἁγίου, his parent text almost certainly had [‫ מ]ן‬and not ‫ב‬. Aquila has ἐκ τόπου ἁγίου supported by Hi, the Peshitta and the Targum. Symmachus has ἐν τόπῳ ἁγίῳ, but this is best explained as a dynamic rendering and the Vulgate is doubtless based upon Symmachus. Support in the textual tradition for a Hebrew text with ‫ ב‬is non-existent. So, Goldman’s proposal that an original ‫ ובמקדש‬became ‫ וממקדש‬and ‫ מקום‬was added as a clarification is not persuasive. In fact, the expression ‫ מקום קדוש‬suits Qoheleth (cf. ‫ מקום‬1:5, 1:7, 3:16bis, 3:20, 6:6, 8:10, 10:4, 11:3) while ‫ מקדש‬as a way of referring to the central sanctuary is not otherwise attested. The article τοῦ is contrary to the translation technique of OG, whereas the τόπου ἁγίου fits perfectly. I acknowledge with Goldman that the Bible text of Jerome’s Commentary is no basis for proposing τόπου instead of τοῦ as original, 40  A. Schenker et al. eds., Biblia Hebraica Quinta. Fascicle 18: General Introduction and Megilloth (Stuttgart: Deutsche Bibelgesellschaft, 2004), 101*.

Basic Principles for the Constitution of the Text

195

but the witness of the Sahidic and Syro-Hexapla cannot be explained this way and remain a huge problem for Goldman’s proposal. The change from τόπου to τοῦ is easily explained as an inner Greek scribal error occurring early enough to affect our entire manuscript tradition in Greek. The parent text of OG certainly had ‫ וישתבחו‬and may have had ‫והלכו‬, but otherwise was virtually identical to 𝔐. By contrast, Goldman’s proposal involves several stages and has no evidence for ‫קרבים‬, the cornerstone of his suggestion. This example reminds us that the apparatus is not simply for those interested in the constitution of the OG. Scholars of the Hebrew text and early ancient versions are helped by references that aid the analysis of questions which interest them and have to do with the place of the LXX in the larger history of the text of the Old Testament.

4.  Jerome Correcting the Old Latin on the Basis of his Hebrew Text The impact of both Origen and Jerome on the subsequent textual transmission of the Septuagint is immense and frequently difficult to disentangle. The problem of the Old Latin in relation to LXX Ecclesiastes may be briefly summarised from my Grinfield Lectures on the Septuagint as follows: Unfortunately no complete manuscripts exist of the text of the Old Latin for Ecclesiastes. In addition to patristic citations of the Old Latin we have three sources. First is Jerome’s Commentary on Ecclesiastes which reproduces the entire text in the lemma. Rahlfs used the siglum La for this source in his Handausgabe. This contravenes the principles of the Göttingen Editions as La properly designates only manuscripts of the Old Latin. Hence the siglum Hi is used in the Edition. Second, approximately 26 verses of the Old Latin of Eccl are found alongside Job and fragments of Proverbs and Canticles in Cod. 11 of the Stiftsbibliothek in St. Gallen (eighth century).41 The siglum for this text is La160. Third, we have marginal notes made in 1561 in a Spanish Vulgate Bible from a now lost tenth century Valvanera Abbey Bible. Almost identical are marginal notes in a twelfth century Bible.42 These are indicated by the sigla La94 95. Rahlfs’ main source for the Old Latin, then, was the lemma of Jerome’s Commentary—what kind of a source, we shall soon see. Significant articles published

41  C. P. Caspari, Das Buch Hiob in Hieronymus’s Übersetzung aus der alexandrinischen Version, Christiana, 1893. 42  Collations for the Göttingen Edition are based upon a photograph of the manuscript. I discovered that María Ángeles Márquez is preparing a critical edition of these marginal notes and she graciously shared with me a preliminary version of her text. See MSS. 94 and 95 in Roger Gryson, Altlateinische Handschriften / Manuscrits Vieux Latins I (Freiburg: Herder, 1999).

196

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

by A. Vaccari43 and S. Leanza,44 in 1958 and 1987 respectively, argue that Jerome’s work on Ecclesiastes had three stages: (1) first, he revised the Old Latin on the basis of the OG, (2) second, he made a translation based directly upon the Hebrew but dependent upon the Old Latin, (3) thirdly, he made a translation based upon the Hebrew and completely independent of the Old Latin.45 According to Vaccari and Leanza, La160 is a witness to the first stage, the Bible text of the Commentary represents the second stage, and the Vulgate represents the third stage. Leanza differs from Vaccari in that while Vaccari treats the Bible text of the Commentary as the Old Latin corrected occasionally towards the Hebrew, Leanza sees it as a new translation indebted somewhat to the Old Latin.46 Jerome explicitly states in his Commentaries on the Psalms and the Letter to Titus that he went to Caesarea and used Origen’s Hexapla there.47 He made extensive notes from the Hexapla which he used both for his Commentary on Ecclesiastes and for the later translation in the Vulgate. This is of great interest for our present focus on hexaplaric sources. The following analysis is based upon my own research before discovery of the work of Leanza and Vaccari. 9:07

Jerome Corrects Old Latin on the Basis of the Hebrew Text

𝔐 9:7a

‫ֵלְך ֱאכֹל ְּב ִׂש ְמ ָחה ַל ְח ֶמָך‬

Edition

9:7a Δεῦρο φάγε ἐν εὐφροσύνῃ ἄρτον σου ————— A B C S 998 O L C´’–299 d–342 k min verss ————— 7 δεῦρο φάγε] veni manduca Amb Exh virg 11,75; vade (+ et Hilem) comede Hilem et com 324,112 326,164 (sed hab veni comede HiLXX 326,184 Hi Ep 96,1 PsAug Pal 17) = vade ergo et comede Vulg —————

43  A. Vaccari, ‘Recupero d’un lavoro critico di S. Girolamo’, in A. Vaccari, Scritti di Erudizione e di filologia, II (Storia e Letteratura, 67; Roma, 1958), pp. 83-146. 44  S. Leanza, ‘Le tre versioni geronimiane dell’Ecclesiaste’, Annali di storia dell’esegesi 4 (1987), pp. 87-108. 45  I am indebted to my colleague Professor Gregg Allison for gracious and expert help in reading the Italian. 46  S. Leanza, ‘Le tre versioni geronimiane dell’Ecclesiaste’, p. 98. 47 Jerome, Commentariolus in Psalmos I,4 and Epist. ad Tit. ad II,9. Some dismiss this testimony, but Jerome’s commentaries are full of hexaplaric readings.

Basic Principles for the Constitution of the Text

197

The biblical lemma of Jerome’s Commentary represents the Old Latin sporadically corrected by Jerome towards the Hebrew.48 He was also aided and influenced in this spontaneous revision of the Old Latin by renderings of the Jewish revision of Symmachus. Normally, then, accurate determination of the Old Latin from La160 and Jerome’s Commentary is problematic and emphasises the importance of quotations in the Latin Fathers. As a general rule, only citations in the Latin Fathers which are clearly not from the Vulgate are referenced in the apparatus. Examples drawn from 9:7 and 9:8 illustrate the problem with the Old Latin. In 9:7a ‫ ֵלְך ֱאכֹל‬is rendered by the Greek translator as δεῦρο φάγε. The rendering of ‫ הלך‬as a hortatory complement in an imperatival sequence by δεῦρο is standard in the LXX, and is also employed by the Eccl translator in 2:1, although the meaning is moving from ‘come’ in classical texts to ‘go’ as a general rule.49 Jerome modified the Old Latin veni comede to vadi et comede. This case is easy to prove since in the course of his commentary he actually cites the version of the Old Latin and specifies it as Septuagint in contrast to the lemma of his Bible text. In addition, a citation in Jerome’s Epistula 96 and Pseudo-Augustine as well as a similar rendering in Ambrose, a father well known for preserving the Old Latin, support the conclusion that veni comede is the Old Latin representing the LXX while vade et comede is a revision of the Old Latin in the Bible text of the Commentary that with the addition of ergo ended up as his translation in the Vulgate. The Edition employs parentheses with ‘sed hab’ to show Old Latin sources that support the lemma in contrast to the renderings in Jerome’s Commentary that deviate from it as well as the somewhat idiosyncratic citation in Ambrose. 9:8

Jerome Corrects Old Latin on the Basis of the Hebrew Text

𝔐 9:8b

‫אׁשָך ַאל־יֶ ְח ָ ֽסר‬ ְ ֹ ‫וְ ֶׁש ֶמן ַעל־ר‬

Edition

9:8b καὶ ἔλαιον ἐπὶ κεφαλήν σου μὴ ὑστερησάτω. ————— A B C S 998 O L C´’–299 d–342 k min verss ————— 8 ἐπὶ κεφαλήν] υπερ κεφαλην 359 CyrHier 4,8; η κεφαλη 706 | κεφαλήν A C B-68 637 (pr την) 130 k 252 296´ (pr την 548) 311 336 543 549 613 766A (pr την) 795 = Ra] absc 998 Didlem; κεφαλης S O–475 637-411 C´’–609 (pr τη 522*; c pr m) d (pr της 254´) 155 248´ 338 339 359 534 542 547 645 698c 728 Antioch 1713 (pr της) CyrHier 4,8 (pr της) 48  This analysis was made before discovery of the articles by Vaccari and Leanza. 49  See E. Eynikel and J. Lust, “The Use of δεῦρο and δεῦτε in the LXX,” Ephemerides Theologicae Lovanienses 67/1 (1991): 57-68.

198

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

Didcom 273,19s (pr της) OlΑΓΖΔΕΙΚΜ; κεφαλη 475 L–130 609 298 443 698* 766B (pr τη) 788 OlΒΗ | μὴ ὑστερησάτω] μη στερησατω 571c; non deficiat Hilem 324,115 (sed hab desit Amb Inst virg 17,110 Hicom 327,188 327,198) ↓ = Vulg ————— 8 μὴ ὑστερησάτω] σʹ οὐ μὴ διαλίπετω 161 248

Another example is the case of 9:8b. Full discussion of the problem in the text is given later on, but for now we may note that for ἐπὶ κεφαλήν the lemma of Jerome’s Commentary has de capite whereas the citations in Ambrose reading in capite almost certainly preserve the Old Latin. Again, Jerome has corrected the verb desit in his Old Latin as witnessed by citations within his own commentary and another in Ambrose to deficiat, a verb no doubt influenced by διαλίπετω in Symmachus. In a number of passages Vinel was right to reject the text of Rahlfs where he chose or conjectured a reading on the basis of the lemma of Jerome against all the witnesses in Greek. Nonetheless, her reasons for rejecting readings selected by Rahlfs are inadequate. One must, to be sure, be careful about going against all or a large number of witnesses. But one must control the history of the text and here, demonstrating that Jerome occasionally corrected the Old Latin according to his Hebrew Vorlage is what makes his testimony in these places invalid as a witness to the LXX via the Old Latin. 5:5b Jerome Corrects Old Latin on the Basis of the Hebrew Text? 5:5b MT

‫אמר ִל ְפנֵ י ַה ַּמ ְל ָאְך ִּכי ְׁשגָ גָ ה ִהיא‬ ַ ֹ ‫וְ ַאל־ּת‬

Edition/Ra καὶ μὴ εἴπῃς πρὸ προσώπου τοῦ θεοῦ ὅτι Ἄγνοιά ἐστιν, App I πρὸ προσώπου τοῦ θεοῦ] in conspectu angeli Hilem et com 292,45 294,96; ante faciem dei Spec 522,12; coram angelo Hicom 293,61 = Vulg ↓ | πρὸ προσώπου] το προσωπον S; > προ 106-261-545txt 706 Syh ‫ܘܐܠ ܬܐܡܪ ܩܕܡ ܦܪܨܘܦܐ‬ ̇ .‫ܐܝܬܝܗ‬ ‫ܕܐܠܗܐ ܕܐܠ ܝܕܥܬܐ‬ ̣ Syhmg: ‫ ܂ܐ܂ܣ܂ܬ܂ ܕܡܐܠܟܐ ܀‬Index super ‫ܕܐܠܗܐ‬

The case of 5:5b is given to illustrate that one cannot always determine whether Jerome’s modification of the Old Latin was influenced directly from the Hebrew text or from the Three. Here the Old Greek has rendered ‫ ִל ְפנֵ י ַה ַּמ ְל ָאְך‬by πρὸ προσώπου τοῦ θεοῦ. Jerome offers 5:5b three times, once in the lemma and twice in the commentary itself:

Basic Principles for the Constitution of the Text

199

et ne dixeris in conspectu angeli, quia ignorantia est Hilem 292,45 et ne dixeris coram angelo, quia ignorantia est Hicom 293,61s ne dixeris in conspectu angeli, quoniam ignoratio est Hicom 294,96 Vulg neque dicas coram angelo non est providentia (= Aug Spe 8 Co-Lat 215,2) ne dicas ante faciem dei quia ignorantia est Spec 522,11 All of the texts in both lemma and commentary of Jerome’s Commentary on Ecclesiastes correspond to MT rather than to the OG. Citations in the Speculum of Augustine and the Lateran Council have been conformed to the Vulgate. The only citation preserving the Old Latin is the Speculum of Pseudo-Augustine. The rendering in conspectu angeli in Jerome’s Commentary looks like a rapid revision of the Old Latin while the rendering coram angelo is the more functional equivalent later adopted in the Vulgate. The only information on the Three comes from a marginal note in the Syro-Hexapla where we are informed that Aquila, Symmachus, and Theoodotion all have the word for ‘angel’ or ‘messenger’, no doubt ἄγγελος. Although Jerome does mention Aquila in his commentary at this point, it is in connection with the word ignorantia. So, we have no way of proving whether the modification of the Old Latin in the lemma of his commentary is based directly on the Hebrew or influenced by the Three.

5.  Further Illustration of Text-Critical Principles in Establishing the Text 9:4

Text History of LXX Corrupted by Renderings from Symmachus

𝔐 9:4a

‫ל־ה ַחּיִ ים‬ ַ ‫י־מי ֲא ֶׁשר ֻיְב ַחר ֶאל ָּכ‬ ִ ‫( ִּכ‬Q ‫)יחבר‬

Edition

9:4a ὅτι τίς ὃς κοινωνεῖ πρὸς πάντας τοὺς ζῶντας; ————— A B C S 998 O L C´’–157 299 d–342 k min–359 verss ————— 4 τοὺς ζῶντας] + τις γαρ εις αει διατελει ζων (τις—αει sup ras) 798 ↓ ————— 4 ὅτι τίς—ζῶντας] σʹ τίς γὰρ εἰς ἀεὶ διατελέσει ζῶν 161 248 252 (s nom) 539 (s nom) Hi Syh

9:4a offers an example of an instance where the text of one of the Three Jewish Revisors has corrupted the text history of the Septuagint. One manuscript, 798, has the text τίς γὰρ εἰς ἀεὶ διατελεῖ ζῶν following ζῶντας. Fortunately, we have a reading from Symmachus for this line preserved in six separate witnesses and

200

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

the text is given in the Second Apparatus. The text in 798, therefore, constitutes a double translation of the line. A down arrow in the First Apparatus alerts the reader to this corruption of the textual transmission of the LXX from one of the Three. Since MS. 798 is in the Catena Group, it is not surprising that one can find readings from the Three influencing the text in this particular part of the text tradition. 9:4

Correction of LXX toward Hebrew Based on Symmachus

𝔐 9:4a

‫ל־ה ַחּיִ ים‬ ַ ‫י־מי ֲא ֶׁשר ֻיְב ַחר ֶאל ָּכ‬ ִ ‫( ִּכ‬Q ‫)יחבר‬

Edition

9:4a ὅτι τίς ὃς κοινωνεῖ πρὸς πάντας τοὺς ζῶντας; ————— A B C S 998 O L C´’–157 299 d–342 k min–359 verss ————— 4 κοινωνεῖ] -νησει L 296´ 311 336´ 443 602-613 698 706 795 ————— 4 ὅτι τίς—ζῶντας] σʹ τίς γὰρ εἰς ἀεὶ διατελέσει ζῶν 161 248 252 (s nom) 539 (s nom) Hi Syh

Another variant for the same text in 9:4a illlustrates another issue in the text history. There is a Ketib/Qere problem between ‫ יבחר‬and ‫ יחבר‬in the Masoretic Text. The LXX Ecclesiastes witnesses to the Qere. This is not the issue in question. Note that the L group and eleven minuscules representing no more than eight separate witnesses because of manuscript pairs have a future tense (κοινωνήσει) while the majority of the textual tradition has a present tense (κοινωνεῖ). Once more, analysis of translation technique can guide the text critic. The Greek translator renders yiqtol forms in Hebrew in 179 instances: in 114 he employed a Future Indicative, in 32 an Aorist Subjunctive—all but one in dependent clauses, in 12 a Present Indicative, in eight a Participle, in four a Present Subjunctive—all in dependent clauses, in four an Aorist Indicative, in four an Infinitive, and in one instance a Present Imperative.50 The persistent commitment of the translator to extreme formal and quantitative correspondence between source and target languages can be seen in the fact that 64% of the time he used a Future Indicative. Nonetheless, the translator is sensitive to contextual factors and there are a dozen cases of Present Indicative, eleven of which are in gnomic statements 50  For an exhaustive presentation of all the evidence, see Yun Yeong Yi, “Translation Technique of the Greek Ecclesiastes,” 141-159.

Basic Principles for the Constitution of the Text

201

as is exactly the case here. These patterns in translation technique make the Present Indicative the lectio difficilior at this point and add weight to the strong support in terms of external evidence for κοινωνεῖ. How, then, did the Future Indicative arise in the text history? I am not suggesting that Greek scribes were bilingual and had access to the parent text in Hebrew, but they may have used the Three Jewish Revisors as an aid to correct the LXX on the basis of the Hebrew, a method used by Origen himself. The L group has a number of readings like this and once again, a glance at the Second Apparatus shows that Symmachus has a Future Indicative. Readings from Aquila and Theodotion are not preserved here and it is also possible that the correction is more directly due to their rendering. 9:8

Freedom of the Translator

𝔐 9:8b

‫אׁשָך ַאל־יֶ ְח ָ ֽסר‬ ְ ֹ ‫וְ ֶׁש ֶמן ַעל־ר‬

Edition

9:8b καὶ ἔλαιον ἐπὶ κεφαλήν σου μὴ ὑστερησάτω. ————— A B C S 998 O L C´’–299 d–342 k min verss ————— 8 ἐπὶ κεφαλήν] υπερ κεφαλην 359 CyrHier 4,8; η κεφαλη 706 | κεφαλήν A B C 637 (pr την) k 130 68 252 296´ (pr την 548) 311 336 543 549 613 766A (pr την) 795 = Ra] absc 998 Didlem; κεφαλης S O–475 637-411 C´’–609 (pr τη 522*; c pr m) d (pr της 254´) 155 248´ 338 339 359 534 542 547 645 698c 728 Antioch 1713 (pr της) CyrHier 4,8 (pr της) Didcom 273,19s (pr της) OlΑΓΖΔΕΙΚΜ; κεφαλη 475 L–130 609 298 443 698* 766B (pr τη) 788 OlΒΗ | μὴ ὑστερησάτω] μη στερησατω 571c; non deficiat Hilem 324,115 (sed hab desit Amb Inst virg 17,110 Hicom 327,188 327,198) ↓ = Vulg ————— 8 μὴ ὑστερησάτω] σʹ οὐ μὴ διαλίπετω 161 248

We can now return to 9:8b and analyse in detail the main problem there. The entire verse can be literally rendered, “At all times let your garments be white and oil on your head not be lacking.” The problem concerns the case of κεφαλή as the object of the preposition ἐπί. For each variant of κεφαλή some witnesses have the article. The insertion of the article is secondary since, as we have already seen in 9:7a, the absence of the article follows the patterns of the translator whereas the addition of the article better suits syntactic requirements in Greek for nouns with a possessive pronoun. So, the addition of the article is another example of inner Greek scribal error and will not be considered further in this analysis.

202

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

The problem is difficult to decide on the basis of external evidence. Of the three main options, genitive, dative, or accusative case of κεφαλή, the dative is the least likely as it is probably due to haplography of sigma in 609 298 and 698* whose congeners have the genitive or due to loss of supralinear nu in L–130 and 766B whose congeners have the accusative. Damage to 998 and the Tura Papyrus of Didymus means their witness is not extant at this point. For the accusative we have the majuscules A B C, one member of the O group, the k group and a dozen or so unclassified minuscules. Supporting the genitive, we have S, most of the O group, the Catena and d groups, and about sixteen unclassified minuscules. There is weighty evidence for both genitive and accusative. According to the principles of syntax in Greek, the genitive is used to indicate location with verbs of rest while the accusative is used to mark extent in space or with verbs of motion. The preposition ἐπί followed by the dative seems to aid in specifying relations that could be indicated by case alone, especially in the classical period. Apart from 9:8, the Greek translator of Ecclesiastes employed ἐπί for ‫ על‬in 28 instances. A chart lists these and indicates the function. E stands for extent in space, M for motion, and Loc for locative. Variants for passages where the textual tradition is divided are also listed. 01:06 01:12 01:16 02:17 02:20 03:14 03:17 05:01 05:01 05:01 05:05 05:07 05:07 06:01 08:06 08:14 08:16 09:12 09:14 10:04 10:07 10:07

 καὶ ἐπὶ κύκλους αὐτοῦ ἐπιστρέφει τὸ πνεῦμα  ἐγενόμην βασιλεὺς ἐπὶ ᾽Ισραὴλ  καὶ προσέθηκα σοφίαν ἐπὶ πᾶσιν  ὅτι πονηρὸν ἐπ᾽ ἐμὲ τὸ ποίημα  τοῦ ἀποτάξασθαι τὴν καρδίαν μου ἐπὶ παντὶ μόχθῳ  ἐπ᾽ αὐτῷ οὐκ ἔστιν προσθεῖναι  ὅτι καιρὸς τῷ παντὶ πράγματι καὶ ἐπὶ παντὶ τῷ ποιήματι  μὴ σπεῦδε ἐπὶ στόματί σου  ὅτι ὁ θεὸς ἐν τῷ οὐρανῷ, καὶ σὺ ἐπὶ τῆς γῆς  ἐπὶ τούτῳ ἔστωσαν οἱ λόγοι σου ὀλίγοι  ἵνα μὴ ὀργισθῇ ὁ θεὸς ἐπὶ φωνῇ σου  μὴ θαυμάσῃς ἐπὶ τῷ πράγματι  καὶ ὑψηλοὶ ἐπ᾽ αὐτούς ᾽ʹΕστιν πονηρία … καὶ πολλή ἐστιν ἐπὶ τὸν ἄνθρωπον  ὅτι γνῶσις τοῦ ἀνθρώπου πολλὴ ἐπ᾽ αὐτόν  ἔστιν ματαιότης, ᾖ πεποίηται ἐπὶ τῆς γῆς  τοῦ ἰδεῖν τὸν περισπασμὸν τὸν πεποιημένον ἐπὶ τῆς γῆς  ὅταν ἐπιπέσῃ ἐπ᾽ αὐτοὺς ἄφνω  καὶ οἰκοδομήσῃ ἐπ᾽ αὐτὴν χάρακας μεγάλους  ἐὰν πνεῦμα τοῦ ἐξουσιάζοντος ἀναβῇ ἐπὶ σέ  εἶδον δούλους ἐφ᾽ ἵππους  καὶ ἄρχοντας πορευομένους ὡς δούλους ἐπὶ τῆς γῆς

A=M — D A=? D D D D G = Loc D D D A=E A=E A=E G = Loc G = Loc A=M Advers Advers A=M G = Loc

203

Basic Principles for the Constitution of the Text

11:01 11:02 11:03 11:09 12:06 12:07

 ᾽Απόστειλον τὸν ἄρτον σου ἐπὶ πρόσωπον τοῦ ὕδατος  ὅτι οὐ γινώσκεις τί ἔσται πονηρὸν ἐπὶ τὴν γῆν  ἐὰν πληρωθῶσιν τὰ νέφη ὑετοῦ, ἐπὶ τὴν γῆν ἐκχέουσιν  καὶ γνῶθι ὅτι ἐπὶ πᾶσι τούτοις ἄξει σε ὁ θεὸς ἐν κρίσει  καὶ συντριβῇ ὑδρία ἐπὶ τὴν πηγήν  καὶ ἐπιστρέψῃ ὁ χοῦς ἐπὶ τὴν γῆν, ὡς ἦν

A=M A=E A=E D A=? A=?

Passages with Variants: 1:6 2:17 3:14 3:17 5:1 5:1 5:1

5:5 5:7 5:7

6:1 8:6 8:16 9:12 9:14 10:4 10:7 11:1

κύκλους Amb Ps 118 12 Bed Rat 8 Hi Ezech 1, Zach 2, Ep 100 Isid Nat 27,21 Hi = Vulg] κυκλου OlΒΓ La94 95 La160 PsMar Scr 5; κυκλοις 390-574 ἐπ᾽ ἐμέ] εμοι 357 ↓ | ἐμέ] εμοι 336´ ἐπ᾽] απ 299 k 296´ 336 549 795; εν 673 | αὐτῷ B 357 68 534 = Gra Ra 𝔐 ↓] αυτης 411; αυτοις O cII 149 260 602 613sup lin; αυτον 261; αυτων rel = Ald Compl ἐπί] …]μ̣ 998; εν 818 248c 645 OlΕΖΙ Aeth SaI; > Cass Co 21,12,3 ἐπὶ στόματί] in ore Spec 522,8 Aeth Hi Syh = 𝔖; εν πνευματι 357: ex 79a | ML στόματί] -τος DamM T OlΗ ἐπὶ τῆς γῆς] εν τη γη Anast 684 Ath III 569 Did Ps.CatA 622a Hicom 291,10 Ild Bap 111 Ruf Ps 36,5,4 Spec 522,10 (sed hab Aug Spe 8 Hilem 291,3 = Vulg) Aeth ἐπὶ τούτῳ S* A C C´’ d–357 k 155 161txt-248txt-252 260´ 311 338 339 543 547 549* 645 698 706 OlΒΕΖΗΘΙΚ = Ald Compl Ra] επι τουτο 797 357 296´ 549c 795; δια τουτο 969 O B-Sc-68´’(-τω 534) L–(125)(-τω 545*) 125II 161mg-248mg 336´(‑τω 336) 411 443 542 766 Hilem 291,3 (propter hoc) Syh SaI Fa1 PsChr OlΑΓ Met IV.5,4 Anast 684 Or Phil 1,28 = Sixt: cf 811c; pr καί 147; absc 998 Fa2 c φωνῇ] pr τη 159 298; φωνης 299-572-371 k 261 155 260´ 336´ 534 548 DamH ; II A φωνην 130 125 542 766 ; post η ras 1 litt 248 τῷ πράγματι] plur Arm; το πραγμα 637; τω θαυματι 357 αὐτούς = Ra] αυ̣[…(spat 4 lit) 998 (sed hab Fa1); αυ 155(||); αυτοις O–253 mg L–106 (125) C´–147 (157) 159 425 503 522* 540 560 571 609 797-260 k 248 296´ 311 339 543 547 549 645 706 795 788 Hi Aeth PsChr Ollem–Μ et com = Ald Sixt Gra; αυτης S A C 411 147-159-425mg-503-540-560-571-609-797-cII–260 106 161 252 542 613 698 766 Met IV.9,5 Geo Syh, contra M; αυτησοι 253 EOVMi V A ; τὸν ἄνθρωπον] τω ανθρωπω 545c; των ανθρωπων 298 534 698 DamV τους ανθρωπους O αὐτόν] αυτω L(–125) 296΄ 766B OlΖ; αυτων 503* 609 766A ἐπὶ τῆς γῆς] ε. την γην 253; super terram Hi (in terra Vulg) c ἐπ᾽ αὐτούς] επ αυτοις C´’–571 543 766 Ol; om επ k 260* 311 338 728 ἐπ᾽ αὐτήν] εις αυτην 766B; επ αυτη 637; εν αυτη V; επ αυτης 443 ἐπὶ σέ] εν σοι cII ἵππους] ιπποις C΄–390 571* 601 DamR; ιππων 357 M ἐπὶ πρόσωπον] εις προσωπον 357 155 DamM ; επι προσωπου B(c) 157-563c H T ΓΖΜ 571*-609-797 68-534 Dam Ol PsChr

204 11:2 11:3 11:9 12:6 12:7

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes M

τὴν γῆν] της γης C΄’ L 252 260΄ 296΄ 336΄ 338c vid 602-613 766 = Ald DamM Didcom 319,22 ἐπὶ τὴν γῆν] επι της γης C´–157 563 571 609 797; in terram Amb Ep 2,4 (sed hab super terram Hi) ἐπὶ πᾶσι] εν πασι O-411 Fa Sa Didcom 337,1 Anton 960 (sed hab Anton 1057 1208) Max II 968 Hi Pach 144,13 (sed hab Amb Exh virg 10,69 Hi Spec 39) = 𝔖 ἐπὶ τὴν πηγήν] επι τη πηγη B 998(vid) d 68 534; απο της γης 336΄ ἐπὶ τὴν γῆν] εις την γην cII Ol Aug Cass Hi Ep 131,26,2 Hi Syh = 𝔖

Most cases where variation is found are not serious except for 11:2 and 3. As a general rule, the customary rule in syntax just noted is preserved by the translator. When the object is γῆ, the Genitive is used in 5:1, 8:14 and 16, and 10:7 against the Accusative in 12:6 and 7. This is odd and may require separate treatment at another time. The noun κεφαλή as the object of the preposition ἐπί is found in fourteen instances elsewhere in the LXX as follows: 49:26 [εὐλογίαι] ἔσονται ἐπὶ κεφαλὴν Ιωσηφ καὶ ἐπὶ κορυφῆς ὧν ἡγήσατο ἀδελφῶν Num 06:07 ὅτι εὐχὴ θεοῦ αὐτοῦ ἐπ᾽ αὐτῷ ἐπὶ κεφαλῆς αὐτοῦ Deut 33:16  ἔλθοισαν ἐπὶ κεφαλὴν Ιωσηφ καὶ ἐπὶ κορυφῆς δοξασθεὶς ἐν ἀδελφοῖς Judg 09:53 καὶ ἔρριψεν γυνὴ μία κλάσμα ἐπιμυλίου ἐπὶ κεφαλὴν Αβιμελεχ 2 Sam 02:25 καὶ ἔστησαν ἐπὶ κεφαλὴν βουνοῦ ἑνός 2 Sam 03:29 ἀθῷός εἰμι … ἀπὸ τῶν αἱμάτων … καταντησάτωσαν ἐπὶ κεφαλὴν Ιωαβ 4 Macc 15:20 καὶ ἐπὶ κεφαλαῖς κεφαλὰς ἀποδειροτομουμένας Ps 132:2 ὡς μύρον ἐπὶ κεφαλῆς τὸ καταβαῖνον ἐπὶ πώγωνα Prov 10:06 εὐλογία κυρίου ἐπὶ κεφαλὴν δικαίου Prov 10:22 εὐλογία κυρίου ἐπὶ κεφαλὴν δικαίου Sir 27:25 ὁ βάλλων λίθον εἰς ὕψος ἐπὶ κεφαλὴν αὐτοῦ βάλλει Sir 44:23 εὐλογίαν πάντων ἀνθρώπων … κατέπαυσεν ἐπὶ κεφαλὴν Ιακωβ Jer 38:07  εἶπεν κύριος τῷ Ιακωβ Εὐφράνθητε καὶ χρεμετίσατε ἐπὶ κεφαλὴν ἐθνῶν Lam 03:54 ὑπερεχύθη ὕδωρ ἐπὶ κεφαλὴν μου Gen

Passages with Variants: Gen 49:26  κεφαλήν] -λης A Fc 15´-17-72´-135-426 L 246 n Tht I 223; της κεφαλης 30´-343-344´ Num 06:07 κεφαλῆς] -λην 72 246 Deut 33:16 κεφαλήν] -λης 72 C-414 344mg 799; κορυφην 75´

Basic Principles for the Constitution of the Text

205

Prov 10:06 κεφαλὴν] -λης S* A Prov 10:22 κεφαλὴν] -λης S A Sir 27:25 κεφαλὴν] -λης 248 Jer 38:07 κεφαλὴν] -λης A-410; -λας V

The Oxford Lexicon by Liddell-Scott-Jones notes that with κεφαλή, the accusative is rare compared to the genitive and dative. In the LXX, however, the accusative seems to be used even in cases that might be construed as locative. The problem in 9:8 is that the verb ὑστερέω is capable of being understood as indicating either motion or rest. One may visualise “oil lacking on the head” or mentally think of “a lack of pouring oil over the head.” This verb is found in 21 other passages in the LXX, none of them terribly illuminating for this problem.51 The rigid and stereotypical approach to lexical equivalents on the part of the translator requires that he use ἐπί for ‫ על‬and ὑστερέω for ‫חסר‬. These are his fixed equivalents for these items. One area, however, where he is completely free to show his own understanding of the text is in the case of the object of ἐπί. The genitive seems to be the lectio facilior as is evident from the variants in both Eccl and elsewhere in the LXX. Rahlfs was no doubt right to choose the accusative as the best text. 9:9

Error in Hebrew or LXX Textual Transmission?

𝔐 9:9 ‫ל־יְמי ַחּיֵ י ֶה ְב ֶלָך‬ ֵ ‫ר־א ַה ְב ָּת ָּכ‬ ָ ‫ם־א ָּׁשה ֲא ֶׁש‬ ִ ‫ְר ֵאה ַחּיִ ים ִע‬ ‫ל־יְמי ֶה ְב ֶלָך‬ ֵ ‫ן־לָך ַּת ַחת ַה ֶּׁש ֶמׁש ָּכ‬ ְ ‫ֲא ֶׁשר נָ ַת‬ ‫ִּכי הּוא ֶח ְל ְקָך ַּב ַחּיִ ים‬ ‫ר־א ָּתה ָע ֵמל ַּת ַחת ַה ֶּׁש ֶמׁש‬ ַ ‫ּוב ֲע ָמ ְלָך ֲא ֶׁש‬ ַ

Edition

9:9 καὶ ἰδὲ ζωὴν μετὰ γυναικός, ἧς ἠγάπησας, πάσας τὰς ἡμέρας ζωῆς ματαιότητός σου τὰς δοθείσας σοι ὑπὸ τὸν ἥλιον, πάσας τὰς ἡμέρας ματαιότητός σου, ὅτι αὐτὸ μερίς σου ἐν τῇ ζωῇ σου καὶ ἐν τῷ μόχθῳ σου, ᾧ σὺ μοχθεῖς ὑπὸ τὸν ἥλιον. ————— A C B-S 998 O L C´’–299 d–342 k min verss ————— 51  Num 9:7, 9:13, Deut 15:8, Neh 9:21, Job 36:17, Ps 22(23):1, 38(39):4, 83(84):11, Eccl 6:2, 10:3, Cant 7:2, Sir 7:34, 10:27, 11:11, 11:12, 13:4, 26:28, 51:24, Hab 2;3, Dan LXX 4:30 (Dan Th 5:27).

206

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

9 om καί 1° Hi Fa1 3 (sed hab Aeth Geo Arm Hi Mal 2 = 𝔖) = Ra. 𝔐 | ἰδέ] ειδε 637 706 Did 274,5 | ζωήν] ζων 357 | μετά] επι 125 | ἧς ἠγάπησας] > 147-159-560; + και 296 | om ἧς 295 | om πάσας 1°—σου 1° 298 | om πάσας 1° 425 | om τάς 1° C O–637 B-68 L–130 390 357-754 248*-252 336 338 411 443 542 543 549 766 Met VIII.7,2 (absc 998; sed hab Did 277,1 Ol = Ald Sixt) | om ζωῆς k 336 Hi Hi Mal 2 | ματαιότητός 1°] νεοτητος C–(298): ex Prov 518b; νηπιοτητος 698 | σου 1° ◠ 2° 253 L 797txt 252txt 359 543 = 𝔖 | τὰς δοθείσας] της δοθεισης 336 Arm; αι εδοθησαν 475-637 338 411 542 Met VIII.7,3 | ἥλιον 1° ◠ 2° 766A | πάσας τὰς ἡμέρας ματαιότητός σου] πασαι ημεραι ημεραι (> 534 Fa1 3) ατμου σου B*-534-998 Fa1 3 ↓ ; > A 475-637 Bc-68 797txt cII–260 d k 338 547 Aethap(A2) Olap (sed hab Hi Syh SaII Aethte Arm Did 277,1-2 Olte = Ald) = Sixt | om τάς 2° C S 155 252mg 411 443 542 549 766B OlΕΙΜ | ματαιότητός 2°] pr της 411 147; νεοτητος 155: ex Prov 518b | αὐτό] αυτω 637 390; αυτη 125 766B; αυτος S | σου 3°—σου 4°] ζωης σου 797 | om τῇ C 637 645 788 | om ἐν 2° S 645 | om τῷ k | om σου 5° 998 k 336 411 Hi Fa3 (absc Fa1) Didcom 278,9 = 𝔐 | ᾧ σὺ μοχθεῖς] ο εμοχθης 637 | ᾧ] pr εν Sc; ο 147; ος 698 | σύ] σοι 563-571* 548 Didlem et com 278,6 278,9; > A 797 443 728 Fa1 Geo = 𝔙 | μοχθεῖς] -θης 296 336 359 OlΜ; -θις V 998 | ἥλιον 2° ◠ ἥλιον 911a 357 ————— 9 ἰδὲ ζωήν] σʹ ἀπόλαυσον ζωῆς 161 248 | (9b) πάσας τὰς ἡμέρας ζωῆς ματαιότητός σου] αʹ ἀτμοῦ σου (πᾶσαι αἱ ἡμέραι) 161 248 (ind ad ματαιότητός 2°) | (9d) πάσας τὰς ἡμέρας ματαιότητός σου] αʹ πᾶσαι αἱ ἡμέραι (+ ἡμέραι 248) ἀτμοῦ σου 161 248 (ind ad ματαιότητός 2°) | ματαιότητός 2°] οἱ λʹ ἀτμοῦ 252

Before considering 9:9, a brief discussion of the actual readings in MSS. 161, 248 and 252 in the Second Apparatus is necessary. The marginal note for each manuscript is as follows: 161

αʹ ἀτμοῦ σου πᾶσαι αἱ ἡμέραι ἀτμοῦ σου (Ind ad ματαιοτητος 2°)

248

αʹ ἀτμοῦ σου πᾶσαι αἱ ἡμέραι ἡμέραι ἀτμοῦ σου (Ind ad ματαιοτητος 2°)

252txt is lacking 9:9cd. This is supplied in the margin opposite 9:9b. In the margin opposite 9:9e and below the previous marginal note are the words οἱ λο ἀτμοῦ. P. Marshall, in his provisional critical edition of hexaplaric remains for Ecclesiastes, has observed rightly that the marginal notes 161 and 248 constitute in fact two distinct readings, one corresponding to 9:9b and the other to 9:9d, even though the index is tied only to ματαιότητός in 9:9d.52 His suggestion, however, that the repetition of ἡμέραι in 248mg is due to dittography is probably not right.53 In a discussion of the problem with Detlef Fraenkel he proposed more plausibly 52 Phillip S. Marshall, “A Critical Edition of the Hexaplaric Fragments of Ecclesiastes” (Ph.D. diss. Southern Baptist Theological Seminary, 2007), 265. 53  Although made in consultation with myself (as his supervisor) at the time.

Basic Principles for the Constitution of the Text

207

that the doubled ἡμέραι in 248mg, also in B* 998, and secondarily removed in 161mg and 534 Fa1.2 points to full Aquila readings for both 9:9b and d. This is how the information has been represented above in the Second Apparatus. The main issue in 9:9 may now be considered. This last problem is difficult and perplexing. There certainly is corruption in the textual transmission, but it is not certain whether the corruption occurred already in the Hebrew Vorlage of the Greek translator or in the course of textual transmission after the Greek translation was made. Y. A. P. Goldman, editor of Qoheleth for Biblia Hebraica Quinta argues for the former position as follows: ‫ ֲא ֶׁשר‬2 to ‫ ֶה ְב ֶלָך‬2 S [Syriac] attests the shortest form of the verse, without v. 9aβ. That part is attested in G, but in a dubious way: (1) τὰς δοθείσας for ‫אשר נתן‬, as indicated by McNeile (An Introduction, 150), is “foreign to the style of the translation”; (2) the translation of the last three words (‫ )כל ימי הבלך‬πάσας (τὰς) ἡμέρας ματαιότητός σου is omitted in Codex Alexandrinus and part of the medieval mss.; and they are imported (in the nominative) from the Aquila version in the text of B and 998. The omission in S and in G or their Vorlage might have resulted from an accidental parablepsis due to homoioteleuton, but on the other hand that part of the verse is rather redundant. ‫ל־יְמי ֶה ְב ֶלָך‬ ֵ ‫ ָּכ‬Codex Alexandrinus and a majority of witnesses of the Greek tradition do not have these words, which in Codex Vaticanus (first hand) and 998 seem to have been imported from the Aquila version; see McNeile, An Introduction, 150).54

In order to deal fairly with the analysis proposed by Goldman, the arguments proposed by McNeile need to be cited in full: v. 9. καὶ ἴδε MSS. S.H. Pesh | M. ‫ ראה‬Σ Hier. τὰς δοθείσας … ἀτμοῦ σου. It is probable that the two clauses (a) ‫אשר נ" ל" ת" השמש‬, and (b) ‫כל ימי הבלך‬, were absent from the pre-Aḳ iban text used by the translator. In the case of (b) the amounts to a certainty. For (i) B (alone) has πᾶσαι ἡμέραι ἡμέραι (sic) ἀτμοῦ σου, which is clearly a corruption of Aq. πᾶσαι αἱ ἡμέραι α’. σ. (ii) In CSV 147-157-159. 161-248. 296 S.H. a literal rendering has been supplied from the earlier similar clause in the verse—πάσας [τὰς] ἡμέρας [τῆς] ματαιότητός σου. (iii) The clause is omitted in A curss. rel. Pesh. Hier. Tg. and in some Heb. MSS. K. and de R. Clause (a) is found in all Greek MSS. (exc. 106-261) and in S.H. But Pesh. om. The rendering τὰς δοθείσας is foreign to the style of the translation, and may have been 54  A. Schenker et al. eds., Biblia Hebraica Quinta. Fascicle 18: General Introduction and Megilloth (Stuttgart: Deutsche Bibelgesellschaft, 2004), 104*.

208

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

supplied from Σ or Θ, as clause (b) was from Aq. It is omitted, together with the first clause, in ten MSS. K. de R. ἐν τῇ ζώῃ σου MSS. S.H. | om. pron. M. Pesh. Hier. (om. pron. with μόχθῳ also). This may, however, be a mistake of a Greek scribe, owing to the occurrence of σου with nine other words in vv. 7-9.55

Parts of this complicated problem can be resolved through exhaustive study of the translator’s technique. First, at the beginning of v. 9, Rahlfs reconstructs the earliest form of the text without the conjunction καί. This was based solely on the biblical lemma of Jerome’s Commentary which he considered a strong witness to the Old Latin. Now we can add only the Fayyumic Version to the witness of Jerome’s Commentary. Although both are extremely early witnesses, the addition of the copula conjunction is characteristic of the Fayyumic so only the lemma of the Commentary by Jerome is a serious witness. Did Jerome correct his Old Latin on the basis of the Hebrew or is his lemma really the Old Latin? From the perspective of translation technique, it is possible that the translator rendered asyndeton or zero in his parent text by καί (1:8c, 3:21a, 8:9a). Normally, however, waw is rendered by καί and zero by zero in Greek. The correspondence is 99%. Also, from the point of view of stylistic structure, we have a series of proverbs each marked at the onset by asyndeton (v. 7a, 8a, 9a, 10a) and by a copula conjunction in between, e.g. καὶ πίε v. 7b. The combination of stylistic structure and translation technique seems a strong argument, while the addition of a καί is the lectio facilior. Was this an inner Greek scribal error that doubtless occurred very early or did the parent text of the translator have the conjunction? Is the Peshitta influenced by the OG or is his conjunction from his parent text? Alternatively, it is possible that the parent text of the translator had a waw. It would be natural between the two pairs of proverbs. We could view the Peshitta as support and the testimony of Hi Mal 2 as the Old Latin with the biblical lemma of Jerome’s Commentary a correction to the Hebrew of his time. Second, McNeile followed by Goldman claims that rendering ‫ן־לָך‬ ְ ‫ ֲא ֶׁשר נָ ַת‬by τὰς δοθείσας σοι is “foreign to the style of the translation.” Yet study of translation technique shows that the Greek translator employed a participle for qatal in 18 instances56 and for yiqtol in eight,57 all in relative clauses.58 The majority of these cases involve the verb ‫היה‬, but the construction itself is well within norms for the style of the translator. This shows the importance of exhaustive analysis of 55  A. H. McNeile, An Introduction to Ecclesiastes (Cambridge, University Press, 1904), 150. 56  1:9 (bis), 1:10, 1:13, 1:14, 2:7, 2:9, 2:17, 3:15, 4:2, 4:3, 5:8, 7:19, 8:16, 8:17, 9:6, 9:9, 11:8. 57  1:9 (bis), 1:11 (bis), 2:18, 8:7, 10:14 (bis). 58  For complete discussion, see see Yun Yeong Yi, “Translation Technique of the Greek Ecclesiastes” (Ph.D. diss. Southern Baptist Theological Seminary, 2005), 151-154, 167-171.

209

Basic Principles for the Constitution of the Text

translation technique and not relying on Probeschriften for results. Goldman and McNeile, then, have no reason on the basis of translation technique to suggest that 9c is secondary and not part of the original translation. What about the arguments of Goldman and McNeile based on external evidence that 9:9c and d were not in the original Greek translation? Most of the evidence recorded for the Göttingen Edition was available to both of them. This case is a good opportunity to show how to read the apparatus of the Göttingen Edition. One must eliminate from consideration minor mistakes and the issue of articulation. This leaves the following apparatus: 9 πάσας 1°] ∩ 2° k OlΒΗ; > 425 | σου 1° ∩ 2° 253 797txt L 252txt 359 543 OlΖ = 𝔖 | ἥλιον 1° ∩ 2° 766A | πάσας τὰς ἡμέρας ματαιότητός σου C S V-411 C´(797mg)-260´ 125II 155 161-248 252mg 296´ 311 336´ 339 443 542 549 645 698 706 766B 795 Did 277,1-2 OlΕΙΜ Arm Hi SaII Syh = Ald] πασας τας ημερας ματαιοτητας σου k; πασαι ημεραι ημεραι (> 534 Fa1.2) ατμου σου B* 998 534 Fa1.2 ↓ ; > A Bc 637 cII mg d 68 338 547 602-613 OlΑΓΔΖ Κ = Sixt | σου 3°—σου 4°] ζωης σου 797 | ἥλιον 2° ∩ ἥλιον 911a 357 txt

9 πάσας 1° ∩ 2° OlΒΗ | om πάσας 1°—σου 1° 298 | σου 1° ◠ 2° 253 L 797txt 252txt 359 543 = 𝔖 | ἥλιον 1° ◠ 2° 766A | πάσας τὰς ἡμέρας ματαιότητός σου] πασαι ημεραι ημεραι (> 534 Fa1 3) ατμου σου B*-534-998 Fa1 3 ↓ ; > A 475-637 Bc-68 797txt cII–260 d k 338 547 Aethap(A2) Olap (sed hab Hi Syh SaII Aethte Arm Did 277,12 Olte = Ald) = Sixt | σου 3°—σου 4°] ζωης σου 797 | ἥλιον 2° ◠ ἥλιον 911a 357 Omissions and parablepses are noted at five different locations in the apparatus: Omitting 9b and 9c: Omitting 9c and 9d: Omitting 9d: Omitting 9def: Omitting 9f—11a:

OlΒΗ txt 253 L 797txt 252txt 359 543 OlΖ = 𝔖 mg A Bc 637 cII–260 d k 68 338 547 602-613 OlΑΓΔΖ Κ = Sixt 766A 357

Since the reading of B* 998 534 Fa1.2 is clearly a correction added from Aquila, the archetype of these witnesses also omitted 9d. Note carefully the witnesses that attest to 9:9c and d: C S V-411 C´(797mg)-260´ 125II 155 161-248 252mg 296´ 311 336´ 339 443 542 549 645 698 706 766B 795 Did 277,1-2 OlΕΙΜ Arm Hi SaII Syh = Ald McNeile’s supposition that 9c is secondary is groundless and his argument that 9d is supplied from 9b is not sound. The O Group is V-253-637 and sometimes 411. MS. 542, closely related to Syh also attests the Fifth Column of the Hexapla.

210

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

Since V-411-542 and Syh have 9c and d, this is surely the text received by Origen in his Fifth Column before filling in the minuses vis à vis the Hebrew from the Three. This is as early a witness as 998 which attests the parablepsis. The argument of Goldman and McNeile that the original text of LXX or even the Hebrew Vorlage omitted 9c and 9d is simplistic and not supported by the complex of parablepses and omissions in the textual tradition in Greek. The best explanation accounting for the data must assume a Hebrew parent text for LXX identical to 𝔐 and parablepses occurring in the textual transmission of the LXX. First, a parablepsis resulted in the omission of 9c and d. Second, a partial correction added only 9c. The precursor of 998 and B* added 9d from Aquila. mg Only one manuscript has 9c as a correction, i.e. OlΖ and this is certainly secondary. Therefore, since no serious witness has the text of 9c as a correction, one cannot explain the omission of only 9d. This is the crux criticorum of the thesis of Goldman and McNeile (i.e. that which completely crucifies it). 6:6a

Corruption of the LXX Text History from Aquila

𝔐 6:6a

‫וְ ִאּלּו ָחיָ ה ֶא ֶלף ָׁשנִ ים ַּפ ֲע ַמיִ ם‬

Edition/Ra

6:6a καὶ εἰ ἔζησεν χιλίων ἐτῶν καθόδους App I χιλίων / ἐτῶν] χιλια ετη 161* 248c (cor ex αʹ 248mg) ↓ ; πολλων ετων Didcom 172,1; tr 342 Didcom 170,1 (sed hab Didlem 175,26) | καθόδους] καθοδεις 253; παροδους 411; εν οδοις SaI; duplices Hilem 297,13 = 𝔐; duobus Hicom 297,20 = Vulg

6:6a offers an interesting instance where the textual tradition of the Old Greek has been corrupted in a minor way by a reading from Aquila. There is no doubt that the original Greek translator had the same Hebrew text as MT, which can be literally translated “and if he lived a thousand years two times.” The Greek translator has καὶ εἰ ἔζησεν χιλίων ἐτῶν καθόδους, “even if he lived recurrences of a thousand years.” Both the translation in the OG and the textual tradition are of interest. First, the curious rendering of ‫ ַּפ ַעם‬by κάθοδος appears to provide support for the claim that the Septuagint is, in fact, the work of Aquila. The dual of ‫ ַּפ ַעם‬is rendered by καθόδους here and the plural is rendered by the same in 7:22. Normally the word κάθοδος means ‘a descent, going down, way down’. Since κατά also means ‘back’ as well as ‘down’ in composition, another meaning already attested in Classical Greek is ‘a going back, return’. From this second use apparently the meaning ‘recurrence’ is derived although the earliest instances listed in Liddell-Scott-Jones are either in Ecclesiastes or attributed to Aquila. Already in 1892 Dillmann listed Exod 34:24, 1 Sam 3:10 and 1 Kgs 22:16 as attributed to Aquila and suggested

Basic Principles for the Constitution of the Text

211

that the hexaplaric reading resembling 7:22a is mistakenly attributed to Aquila and belongs instead to Symmachus.59 The Reider-Turner Index to Aquila in 1996 expands the list to nine instances and includes the instance in 7:22b as Aquila.60 The problematic double translation in 7:22 will be considered shortly, but all the renderings of ‫ ַּפ ַעם‬in both Old Greek and also later revisions / translations as well as all instances of κάθοδος should be surveyed in order to acquire an accurate picture. The constructions in which ‫ ַּפ ַעם‬is found in the Hebrew text and the ways in which these are rendered are so varied it is difficult to know how to present the evidence or summarise it. First, constructions are listed where ‫ ַּפ ַעם‬means ‘instance’ or ‘occurrence’. These are arranged by number: singular, dual, and plural plus cardinal and ordinal numerals. Then peculiar constructions followed by instances where the noun means ‘foot’ or ‘footstep’. Renderings of later revisors/ translators, where they exist, are given in parentheses. Renderings of ‫ ַּפ ַעם‬in Old Greek and Later Translations ‫ » ַּפ ַעם‬ἅπαξ Neh 13:20 ‫ » ַה ַּפ ַעם‬ἅπαξ Gen 18:32, Judg 6:39(bis, AB) ‫( » ַה ַּפ ַעם‬καὶ > A) τὸ ἅπαξ Judg 15:3(AB), (ἔτι > A) τὸ ἅπαξ Judg 16:18(AB), τὸ ἅπαξ Judg 16:28(AB) ‫ » ַּב ַּפ ַעם ַהּזאת‬τὸ ἅπαξ τοῦτο 2 Kgdms 17:7 (alia νυνὶ ταύτην) ‫ » ַּב ַּפ ַעם ַהּזאת‬zero Jer 10:18 (αʹ εν τω απαξ τουτω σʹ εν τη εποδω ταυτη 86) ‫ » ַּפ ַעם‬χρόνον Prov 7:12 ‫ » ַה ַּפ ַעם‬νῦν Gen 2:23 (σʹ θʹ ἅπαξ), Exod 10:17 ‫ » ַה ַּפ ַעם‬τὸ νῦν Exod 9:27 ({αʹ} τὸ σαμʹ ἐν καιρῷ 135 550(s nom)-cIIcat–25 414(s nom) 18(s nom) 646(s nom)) ‫ » ַה ַּפ ַעם‬νῦν ἔτι τοῦτο Gen 29:35 ‫ » ַה ַּפ ַעם‬ἀπὸ τοῦ νῦν Gen 46:30 ‫ » ַה ַּפ ַעם‬ἐν τῷ νῦν καιρῷ Gen 30:20 ‫ » ַע ָּתה ַה ַּפ ַעם‬ἐν τῷ νῦν καιρῷ Gen 29:34 ‫ » ַּב ַּפ ַעם ַהּזאת‬ἐν τῷ νῦν καιρῷ Exod 9:14 ‫ » ּגַ ם ַּב ַּפ ַעם ַהּזאת‬ἐπὶ τοῦ καιροῦ τούτου Exod 8:32(28) ‫ » ּגַ ם ַּב ַּפ ַעם ַה ִהוא‬ἐν τῷ καιρῷ τούτῳ Deut 9:19 (αʹ καθόδῳ M), 10:10, Jer 16:21 (αʹ εν τω απαξ τουτω 86 Syh (sub αʹθʹ)

59 A. Dillmann, ‘Über die griechische Übersetzung des Qoheleth’, Sitzungsberichte der Königlich preussischen Akademie der Wissenschaften (Berlin, 1892), p. 7. 60  See J. Reider, An Index To Aquila, (Supplements to Vetus Testamentum 12; Brill: Leiden, 1966), p. 122.

212

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

‫ » ַּפ ַעם ַא ַחת‬εἰς ἅπαξ Josh 10:42 (σʹ μιᾷ ὁρμῇ), Isa 66:8 ‫ » ְּב ַּפ ַעם ַא ַחת‬εἰς ἅπαξ 2 Kgdms 23:8 ‫ » ְּב ַּפ ַעם ַא ַחת‬ἐν καιρῷ ἑνί 1 Chr 11:11 (Alia εἰς ἅπαξ) ‫ » ַּפ ַעם ַא ַחת‬εὐθέως Josh 6:11 (σʹ περίοδον μίαν), πάλιν Josh 6:14 (ἅπαξ sub ※ ; σʹ μιᾷ ὁδῷ) ‫ » ַּפ ַעם ַא ַחת‬zero Josh 6:3 (ἅπαξ sub ※ ; αʹ μιᾷ ὁδῷ σʹ μιᾷ περιόδῳ Field), 1 Kgdms 26:8 ‫ » ַּפ ֲע ַמיִם‬δίς Gen 41:32, Num 20:11, 1 Kgdms 18:11 (sub ※), 3 Kgdms 11:9 ‫ » זֶ ה ַּפ ֲע ַמיִם‬δίς Gen 43:10 ‫ » ַּפ ֲע ַמיִם‬ἤδη δεύτερον τοῦτο Gen 27:36 ‫ » ַּפ ֲע ַמיִם‬δίς ἐπὶ τὸ αὐτό Nah 1:9 (θʹ δευτερον Hi Syh) ‫ » ַּפ ֲע ַמיִם‬καθόδους Eccl 6:6 ‫ » ַּפ ֲע ַמיִם‬zero 1 Kgdms 18:11 ‫ » ָׁשלׁש ְּפ ָע ִמים‬τρεῖς καιρούς Exod 23:17, 34:23 (αʹ καθόδους M), 34:24 (αʹ (absc 321) καθόδους 458(s nom) 85´-130(s nom)-344-346(s nom)), Deut 16:16 (αʹ καθόδους M), 2 Chr 8:13 ‫ » ָׁשלֹוׁש ַּפ ֲע ַמיִם‬θʹ ὁδοὺς τρεῖς Job 33:29 (σʹ δὶς τρίς) ‫ » ָׁשלׁש ְּפ ָע ִמים‬τρίς 1 Kgdms 20:41, 3 Kgdms 17:21, 4 Kgdms 13:18, 13:19, 13:25 ‫ » ָׁשלׁש ְּפ ָע ִמים‬τρισσῶς 3 Kgdms 7:4, 7:5 ‫ » ָׁשלׁש ְּפ ָע ִמים‬τρίτον Num 24:10, Judg 16:15(AB) ‫לׁשים ְּפ ָע ִמים‬ ִ ‫ּוׁש‬ ְ ‫ » ָׁשלֹוׁש‬τριάκοντα καὶ τρεῖς δίς Ezek 41:6 (σʹ αἱ διαβάσεις Syh) ‫ » ָׁשלׁש ְּפ ָע ִמים‬zero 3 Kgdms 9:25 (καθόδους sub ※ αʹ) ‫ » ַא ְר ַּבע ְּפ ָע ִמים‬zero Neh 6:4 ‫אֹו־ׁשׁש ְּפ ָע ִמים‬ ֵ ‫ » ָח ֵמׁש‬πεντάκις ἢ ἑξάκις 4 Kgdms 13:19 ‫ » ֶׁש ַבע ְּפ ָע ִמים‬ἑπτάκις Gen 33:3, Lev 4:6, 4:17, 8:11, 14:7, 14:16, 14:51, 16:14, 16:19, 25:8, Num 19:4, Josh 6:15, 4 Kgdms 4:35, 5:10, ἑπτάκι 3 Kgdms 18:43, 4 Kgdms 5:14 ‫ » ֶׁש ַבע ְּפ ָע ִמים‬ἑξάκις Josh 6:15 ‫ » ֶׁש ַבע ְּפ ָע ִמים‬zero Josh 6:4 (ἑπτάκις sub ※), Josh 6:15 ‫ » ֶע ֶׂשר ְּפ ָע ִמים‬δέκατον Num 14:22 ‫ » ֶע ֶׂשר ְּפ ָע ִמים‬zero Neh 4:6 ‫ » זֶ ה ֶע ֶׁשר ְּפ ֲע ִמים‬zero Job 19:3 (σʹ θʹ τοῦτο δέκατον) ‫ » ֵמ ָאה ְּפ ָע ִמים‬ἑκατονταπλασίονα 2 Kgdms 24:3 ‫ » ֵמ ָאה ְּפ ָע ִמים‬ἑκατονταπλασίως 1 Chr 21:3 ‫ » ֶא ֶלף ְּפ ָע ִמים‬χιλιοπλασίως Deut 1:11

Basic Principles for the Constitution of the Text

213

‫ » ְּפ ָע ִמים ַרּבֹות‬πλεονάκις Ps 106(105):43 ‫ » ְּפ ָע ִמים ַרּבֹות‬πλειστάκις Eccl 7:22a (αʹ? σʹ?) ‫ » ְּפ ָע ִמים ַרּבֹות‬καθόδους πολλάς Eccl 7:22b ‫יׁשית‬ ִ ‫ » ַּפ ַעם ֲח ִמ‬zero Neh 6:5 (Alia τὸν πέμπτον Field) ‫יעית‬ ִ ‫ » ַּב ַּפ ַעם ַה ְּׁש ִב‬τῇ περιόδῳ τῇ ἑβδόμῃ Josh 6:16 ‫ד־ּכ ֶּמה ְּפ ָע ִמים‬ ַ ‫ » ַע‬ποσάκις 3 Kgdms 22:16 (αʹ ἕως πόσων καθόδων, σʹ ποσάκις), 2 Chr 18:15 ‫ם־ּב ַפ ַעם‬ ְ ‫ » ְּכ ַפ ַע‬κατὰ τὸ εἰωθός Num 24:1 ‫ם־ּב ַפ ַעם‬ ְ ‫ » ְּכ ַפ ַע‬καθὼς ἀεί Josh 16:20A ‫ם־ּב ַפ ַעם‬ ְ ‫ » ְּכ ַפ ַע‬ὡς ἅπαξ καὶ ἅπαξ Josh 16:20B, Judg 20:30B, 20:31B, 1 Kgdms 3:10 (αʹ (ὡς) κάθοδον ἐν καθόδῳ: Field, Cod. Reg. 243 s nom), 20:25 (σʹ ὡς εἰώθει) ‫ם־ּב ַפ ַעם‬ ְ ‫ » ְּכ ַפ ַע‬καθὼς ἅπαξ καὶ ἅπαξ Judg 20:30A, 20:31A ‫ » ַּפ ַעם‬διάβημα Ps 17(16):5, 85(84):14 (σʹ κατάβασις), Ps 119(118):133, 140(139):5, Cant 7:2 ‫ף־ּפ ָע ַמי‬ ְ ‫ » ְּב ַכ‬τῷ ἴχνει τοῦ ποδός μου = πούς 4 Kgdms 19:24, (σʹ Isa 37:25) ‫ף־ּפ ָע ַמי‬ ְ ‫ » ְּב ַכ‬zero Isa 37:25 (σʹ ἐν ἴχνει ποδός μου) ‫ » ַּפ ֲע ֵמי ַד ִּלים‬πόδες πραέων = πούς Isa 26:6 (θʹ βήματα πενήτων), Ps 57(56):7, Prov 29:5 (σʹ εʹ διάβημα), Judg 5:28(B) ‫ » ַּפ ַעם‬ἴχνος Judg 5:28(A) (᾽ʹΑλλος: δισσοί) ‫ » ַּפ ַעם‬χείρ Ps 58(57):11 (πούς οἱ λ), 74(73):3 (σʹ πούς) ‫ » ַעל ַא ְר ַּבע ַּפ ֲעמ ָֹתיו‬ἐπὶ τὰ τέσσαρα κλίτη Exod 25:12 (θʹ (σʹ θʹ 344; αʹ σʹ 130-321) μέρη M 707(nom absc) 85(s nom)-130-321-344; σʹ πλευράς (-ρα 343) M 343(s nom); γωνιάσματα Fb) ‫ » ַעל ַא ְר ַּבע ַּפ ֲעמ ָֹתיו‬zero Exod 37:3 ‫ » ַא ְר ָּב ָעה ַפ ֲעמ ָֹתיו‬zero 3 Kgdms 7:30 ‫הֹולם ָּפ ַעם‬ ֶ (one hammering an anvil) » ἅμα ἐλαύνων Isa 41:7 (αʹ καθοδον) Occurrences of κάθοδος in Old Greek and Later Translations ?? » κάθοδος 1 Esdras 2:20 ‘a way down’ ‫ » ַּפ ַעם‬κάθοδος Deut 9:19 (αʹ), 1 Kgdms 3:10 (bis; αʹ), Isa 41:7 (αʹ) ‫ » ַּפ ֲע ַמיִם‬καθόδους Eccl 6:6 ‫ » ְּפ ָע ִמים‬καθόδους 3 Kgdms 9:25 (sub ※ αʹ) ‫ » ְּפ ָע ִמים‬καθόδους Exod 34:23 (αʹ), 34:24 (αʹ), Deut 16:16 (αʹ) ‫ » ְּפ ָע ִמים ַרּבֹות‬καθόδους πολλάς Eccl 7:22 ‫ד־ּכ ֶּמה ְּפ ָע ִמים‬ ַ ‫ » ַע‬αʹ ἕως πόσων καθόδων 3 Kgdms 22:16

214

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

Both Old Greek and later translators employ a wide variety of renderings ranging from highly formal to highly functional equivalents. Adverbs ending in ‑ακι(ς) and ordinal adverbs are examples of excellent functional equivalents. More literal are cases of ‫ ָׁשלׁש ְּפ ָע ִמים‬rendered by τρεῖς καιρούς—the majority of instances from Exodus, normally one of the most dynamic translations in the Pentateuch (23:17, 34:23, 34:24). The rendering ὁδοὺς τρεῖς from Theodotion Job (33:29) is similar to τρεῖς καιρούς and also not dissimilar to κάθοδος. It is true that is only attributed to Aquila elsewhere, but other renderings are also attributed to Aquila: ἐν τῷ ἅπαξ τούτῳ for ‫ ַּב ַּפ ַעם ַה ִהוא‬in Jer 10:18 and 16:21, μιᾷ ὁδῷ for ‫ ַּפ ַעם ַא ַחת‬in Josh 6:3, πλεονάκις καιροῦ for ‫ ְּפ ָע ִמים ַרּבֹות‬in Eccl 7:22a61 and ἐν καιρῷ for ‫ַה ַּפ ַעם‬ in Exod 9:27. No doubt Wevers is right to consider the attribution in Exod 9:27 false and Eccl 7:22a is a special situation to be discussed later. But μιᾷ ὁδῷ in Josh 6:3 is similar to ὁδοὺς τρεῖς in Theodotion Job 33:29. I conclude that κάθοδος is certainly a characteristic equivalent for ‫ ַּפ ַעם‬in Aquila, but is also possible for a translation in the καίγε tradition before Aquila. In addition to translation technique, however, the textual tradition in 6:6a contributes to the picture. The text of MS. 161 was corrected from χίλια ἔτη to χιλίων ἐτῶν. MS. 248 also has a correction in the text. Three tiny changes are made supra lineam: α is written above ων in the first word, η is written above ων in the second word, and an acute accent is placed above the first iota of χιλίων. In addition, the corrector wrote ακλ, the symbol for Aquila in the margin. In this way the corrector indicates not only the reading of Aquila, but intends this reading to correct the text of the first hand. There is no reason not to accept the testimony that Aquila’s text read καὶ εἰ ἔζησεν χίλια ἔτη καθόδους instead of καὶ εἰ ἔζησεν χιλίων ἐτῶν καθόδους. Thus, an Aquila reading has corrupted the textual transmission of the Old Greek. In addition, this is a clear example to show that while the Old Greek has Aquila like traits, the translator is not as rigid as Aquila in syntax. 5:10

Corruption of the LXX Text History from Symmachus

𝔐 5:10

Ra

5:10b

‫יה‬ ָ ‫ה־ּכ ְׁשרֹון ִל ְב ָע ֶל‬ ִ ‫ּומ‬ ַ ‫ם־ר ִאּיַ ת ֵע ָינֽיו׃‬ ְ ‫( ִּכי ִא‬Qr ‫ְ)ראּות‬

καὶ τί ἀνδρεία τῷ παρ᾽ αὐτῆς ὅτι ἀλλ᾽ ἢ τοῦ ὁρᾶν ὀφθαλμοῖς αὐτοῦ;

61  According to the margin of MSS. 161, 248, and 252.

Basic Principles for the Constitution of the Text

215

Edition

5:10b καὶ τί ἀνδρεία τῷ παρ᾽ αὐτῆς ὅτι ἀρχὴ τοῦ ὁρᾶν ὀφθαλμοῖς αὐτοῦ; App I τῷ] το 563c k 252 443c 543 547 645 698 706 OlΗΙ*; των 299 795 OlΑ; τον 336´ | παρ᾽ αὐτῆς] εχοντι αυτην O–V 766 Hi Fa ↓ : cf 512b | αὐτῆς OlΑΒΓΕΖΘΙΚ] αυτοις OlΗ; αυτη 342; αυτην 602-613 | ὅτι ἀρχὴ τοῦ ὁρᾶν] nisi ut videat Hi: ex ‫ם־ראֹות‬ ְ ‫( ִּכי ִא‬cf 𝔐) | om ὅτι 637 754 602 613 | ἀρχή B S* A C 998 411 L–(125) C´’–(157) k 68 125II 161 248 252 260´ 296´ 311 336´ 339 txt 443 534 542 543 547 549 698 706 795 Did 151,2 OlΑ ΒΓΕΖΗΘΙΚ Co: ex ‫( ֵאם‬cf Ez 2126)] αρχει c Αsup lin c –357 645; αρκει 545 ; αχρι Ol ; αλλ η S O d 602-613 766 Metlem et com V.1,5,47 = Ra 𝔐 ܿ ‫ܘܡܢܐ ܓܒܪܘܬܐ‬ Syh ‫ܠܗܘ ܕܠܘܬܗ܂‬ ܵ ‫ܚܙܐ ܒܥܝܢܐ ܕܝܠܗ܂‬ ܼ ‫ܡܛܠ ܕܪܝܫܐ‬ ܼ ‫ܕܠܡ‬ mg Syh : ‫ ܂ܣ܂ ܐܐܠ ܐܢ ܒܠܚܘܕ‬Index super ‫ܡܛܠ‬ ܵ ‫ܚܙܬܐ‬ ‫ܕܥܝܢܐ ܕܝܠܗ ܀‬ mg Syh : ‫ ܂ܬ܂ ܐܐܠ ܐܢ‬Index super ‫ܡܛܠ‬ ‫ܕܠܡܚܙܐ ܵܒܥܝܢܐ ܕܝܠܗ ܀‬ ܼ mg Syh : ‫ ※ ܂ܣ܂ ܡܢܐ ܓܝܪ ܢܨܚܢܐ‬Index ad ‫ܘܡܢܐ‬ ܿ ‫ܕܐܝܬܝܗ‬ ܿ ܿ ‫ܠܗ܂‬ ‫ܠܗܘ‬ ‫ܐܐܠ ܐܢ ܒܠܚܘܕ‬ ܵ ‫ܚܙܬܐ‬ ‫ܕܥܝܢܐ ܕܝܠܗ↙ ܀‬ App II καὶ τί—αὐτοῦ] (※ Syhmg) σʹ τί γὰρ ἀνδραγάθημα τῷ ἔχοντι (σχοντι 149cat-260cat-471cat) αὐτό, εἰ μὴ μόνον θεωρία ὀφθαλμῶν αὐτοῦ; 149cat-260cat-471cat Metcom Olcom Syhmg | ὅτι—αὐτοῦ] εἰ (absc 252) μὴ μόνον θεωρία ὀφθαλμῶν αὐτοῦ 161 248 252 (s nom)

The example of 5:10b is clearer than the previous one. The rendering τῷ παρ᾽ αὐτῆς for ‫יה‬ ָ ‫ ִל ְב ָע ֶל‬is a clever one, accomplishing both formal and functional equivalence at the same time. This idiomatic use of ‫ ַּב ַעל‬in Hebrew is consistently rendered by the translator the same way in five of seven instances (5:10, 5:12, 7:12, 8:8, 12:11). In 10:20, however, ‫ ַּב ַעל ַה ְּכנָ ַפיִם‬is rendered by ὁ ἔχων τὰς πτέρυγας, and in 10:11 ‫ ַּב ַעל ַה ָּלׁשֹון‬idiomatically by τῷ ἐπᾴδοντι. Several sources for at least the reading of Symmachus are extant here. The Syro-Hexapla has opposite the text two marginal readings, one attributed to Symmachus and one to Theodotion. An index also directs the reader to a longer note at the bottom of the page where an excerpt from Olympiodorus’ Commentary on Ecclesiastes is translated into Syriac and the same reading from Symmachus is given in a fuller version, marked at the beginning by an asteriskos and at the end by a metobelos. No retroversion from Syriac is necessary since the Commentary by Olympiodorus is preserved in Greek and the Symmachus readings are also preserved in the Commentary by Metrophanes and the Catena Hauniensis. From this reading it is clear that the variant ἔχοντι αὐτήν found in 253 and 637 from the O Group, MSS. 766A and 766B, as well as the Fayyumic and the lemma of Jerome’s

216

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

Commentary is derived from Symmachus. The fact that Jerome adopted a reading from Symmachus offering a slightly smoother style is no surprise. The difference between my edition and that of Rahlfs in 5:10c is also worth noting. The commentary by Goldman in the Bible Hebraica Quinta offers a concise explanation: Rahlfs’ text ὅτι ἀλλ᾽ ἤ is witnessed by a corrector of Codex Sinaiticus as well as V 253, which are hexaplaric mss. The translation of ‫ כי אם‬elsewhere in Qoh is either: εἰ μή (3:12; 8:15), or ὅτι ἐάν (4:10; 11:8) when ‫ כי‬is read as a cj. on its own followed by a conditional sentence. Thus, ὅτι ἀλλ᾽ ἤ in 5:10 is not the Old Greek, but rather a late evolution of G inspired by a revised text. G* ὅτι ἀρχή inteprets the second word of ‫ כי אם‬as ‫“ ֵאם‬point of departure” (cf. Ezek 21:26). σʹ εἰ μὴ μόνον is a reinforced translation of M.62

The fact that Rahlfs based his edition on hexaplaric manuscripts is not in principle problematic. Apart from minuses and pluses, Origen is not reported to have altered his received text. Nonetheless, the problem in 5:10b just considered shows correction and corruption of the text in the O Group. The Colophon of the Syro-Hexapla also reports that Eusebius and Pamphilus corrected this text. Rahlfs’ confidence in going against the majority of witnesses was doubtless based on the fact that ὅτι ἀλλ᾽ ἤ seemed to correspond better to MT. He could not imagine the translator deviating from his parent text. Goldman has shown that the translator had a parent text close to MT, but not identical (at least in vocalisation). His explanation removes support for ἀρχή as a free rendering of some sort. Finally, the apparatus constructed for this verse supports the thinking of Goldman. One manuscript from the O group, one from the d group as well as 613 (and its copy 602) omit the ὅτι showing that something in ὅτι ἀλλ᾽ ἤ is secondary. 5:11

Corruption of the LXX Text History from Symmachus and Theodotion

𝔐 5:11

‫אכל‬ ֵ ֹ ‫ם־ה ְר ֵּבה י‬ ַ ‫ם־מ ַעט וְ ִא‬ ְ ‫תּוקה ְׁשנַ ת ָהע ֵֹבד ִא‬ ָ ‫ְמ‬ ‫וְ ַה ָּׂש ָבע ֶל ָע ִׁשיר ֵאינֶ ּנּו ַמּנִ ַיח לֹו ִליׁשֹון׃‬

Edition/Ra 5:11

γλυκὺς ὕπνος τοῦ δούλου, εἰ ὀλίγον καὶ εἰ πολὺ φάγεται· καὶ τῷ ἐμπλησθέντι τοῦ πλουτῆσαι οὐκ ἔστιν ἀφίων αὐτὸν τοῦ ὑπνῶσαι.

62  A. Schenker et al. eds., Biblia Hebraica Quinta. Fascicle 18: General Introduction and Megilloth (Stuttgart: Deutsche Bibelgesellschaft, 2004), 84*.

Basic Principles for the Constitution of the Text

217

App I comma absc Didlem | om (11a) 425txt | τοῦ δούλου] τω δουλω 261 Antioch 1805 Chr II 585 III 44 Tht IV 680te = 𝔖; του δουλευοντος O Hi (sed hab servi Amb Nab 6,29) = 𝔐 𝔙 ↓ | καί 2°—ὑπνῶσαι] et saturitas divitis non sinit eum dormire Hi (sed hab et satiato (etsi quis satiatus fuerit Ps 1,28,4) divitiis non est qui sinat eum dormire Amb Nab 6,29 Ps 1,28,4) = 𝔙 ↓: cf 𝔐 | ἐμπλησθέντι] εν (εμ- V) πλησμονη O ↓ Syh ‫ܒܕܐ܂‬ ݂ ‫݂ܚܠܝܐ ܫܢ݂ ܬܗ‬ ݂ ‫ܕܥ‬ Syhmg: ‫ ܂ܣ܂ܬ܂ ܒܕܡܘܬ‬Index super ‫ܒܕܐ‬ ݂ ݂ ‫ܕܥ‬ ‫ܫܒܥܝܢ ܀‬ App II καί 2°—ὑπνῶσαι] σʹ ἡ δὲ πλησμονὴ τοῦ πλουσίου οὐκ ἐᾷ καθεύδειν 161 248; et saturitas divitis non sinit eum dormire Hi (saturitas autem divitis non sinit dormire eum Vulg)

The text tradition of 5:11 offers a new twist to things because it illustrates not only influence from Symmachus upon the textual transmission of the Old Greek, but also shows the difficulty at times of determining the text of the Three. There are two different problems in this verse. First, note that while the textual tradition of the Old Greek witnesses to τοῦ δούλου suggesting that the translator vocalised his parent text as the noun hā‛ebed, the O group and the lemma of Jerome’s Commentary represent a parent text that reads the participle hā‛ōbēd as in 𝔐. Again, the original Old Latin is only preserved in citations by Ambrose in De Nabuthae and Explanatio XII psalmorum. The marginal note in the Syro-Hexapla indicates that Symmachus and Theodotion have the same translation as the οʹ text. But what is the οʹ text here? The lemma of Syh reads ‫ܒܕܐ‬ ݂ This clearly represents τοῦ δούλου for at least two ݂ ‫ܕܥ‬. reasons. First, the dot below ensures reading the noun‛abdā, ‘servant’ rather than the participle ‛ābdā /‛ōbdō. Secondly, examples do exist in Syh where τοῦ δούλου and τοῦ δουλεύοντος are distinguished in Syriac. In Mal 1:6 Syh has ‫ܒܕܐ‬ ݂ ‫ ݂ܥ‬for ܿ δοῦλος but in 3:17 and 18 Syh has ‫ ܦܠܚ‬three times for the participle of δουλεύω and employs a dot above. Thus, Syh in Eccl 5:11 clearly supports τοῦ δούλου and not τοῦ δουλεύοντος in MSS. which represent the οʹ text. The marginal note states that Symmachus and Theodotion have the same as the οʹ text. Thus, according to Field in 1875 and the Editor of BHQ in 2004, Symmachus and Theodotion have τοῦ δούλου. Both mention that V and 253 read τοῦ δουλεύοντος and Goldman speaks of them as “hexaplaric mss.” We now also have 637 and these three MSS. form the O group, which in theory, apart from the minuses and pluses, should attest the OG as received by Origen. I made a special study of the marginal notes in Syh of Eccl which contain the word ὁμοίως and came to the following result, surprising as it may seem:

218

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

If the lemma of the manuscript from which the marginal notes in Syh were copied and translated differed from the οʹ text, the marginal note would read only ὁμοίως (indicating that αʹ σʹ θʹ agreed with the lemma, but not with the οʹ text). If, however, the lemma and the οʹ text were the same, the note would read ὁμοίως τοῖς οʹ to show that αʹ σʹ θʹ agreed with the οʹ text as well the lemma, or in some instances with the οʹ text against the lemma.63

The marginal note for Eccl 5:11 has ὁμοίως τοῖς οʹ because it came from a source, like the Catena group, that had τοῦ δούλου as lemma and the scholiast responsible for the note knew that σʹ θʹ had a reading the same as the οʹ text, but not the same as his lemma. The O group is a clear witness to the οʹ text, the text of Origen’s Fifth Column. The marginal note in Syh means that σʹ θʹ read τοῦ δουλεύοντος, the same as οʹ text or O group, but this was not the lemma of the scholiast. It also makes better sense that σʹ θʹ had a reading equivalent to 𝔐 and wanted to correct or revise the OG in this direction. Normally Syh agrees with the O group and Ziegler, in the Göttingen Collation Books, actually construed Syh as reading the participle, because of the marginal note. Here, however, Syh reads with the main textual tradition and as usual, the marginal note is now confusing as to the reading of σʹ θʹ. Goldman, Editor of BHQ, believes that the text of 𝔐 is secondary, influenced by theological considerations.64 This may be, but Symmachus and Theodotion do not support the OG at this point. Instead, they are responsible for a corruption in the O group in the textual transmission. This fits with similar corruption from Symmachus in the O group already seen in earlier examples. Another item of interest in this verse is the reading ἐν πλησμονῇ in the O group instead of τῷ ἐμπλησθέντι in 5:11c. This is an adaptation clearly influenced by the Symmachus reading attested by MSS. 161 and 248. The lemma of Jerome’s Commentary abandons the Old Latin preserved in citations of Ambrose in De Nabuthae and Explanatio XII psalmorum and replaces the entire line by Symmachus. 6:2

Corruption of the LXX Text History from Symmachus

𝔐 6:2c-e

‫וְ ֵאינֶ ּנּו ָח ֵסר ְלנַ ְפׁשֹו‬ ‫ר־יִת ַאּוֶ ה‬ ְ ‫ִמּכֹל ֲא ֶׁש‬ ‫ֹלהים ֶל ֱאכֹל ִמ ֶּמּנּו‬ ִ ‫יטּנּו ָ ֽה ֱא‬ ֶ ‫וְ לֹא־יַ ְׁש ִל‬

63  Peter J. Gentry, “Propaedeutic to a Lexicon of the Three: The Priority of a New Critical Edition of Hexaplaric Fragments,” Aramaic Studies 2.2 (2004): 150. 64  A. Schenker et al. eds., Biblia Hebraica Quinta. Fascicle 18: General Introduction and Megilloth (Stuttgart: Deutsche Bibelgesellschaft, 2004), pp. 85*, 37.

Basic Principles for the Constitution of the Text

219

Edition/Ra

6:2c-e καὶ οὐκ ἔστιν ὑστερῶν τῇ ψυχῇ αὐτοῦ ἀπὸ πάντων, ὧν ἐπιθυμήσει, καὶ οὐκ ἐξουσιάσει αὐτῷ ὁ θεὸς τοῦ φαγεῖν ἀπ᾽ αὐτοῦ, App I τῇ ψυχῇ] την ψυχην 337 795 | αὐτοῦ 1°] > 390 298; + και ουκ αναλισκει εις εαυτον O–V = (2c) 𝔐 ↓ ܿ Syh ‫ܕܝܠܗ܂‬ ‫ܘܐܠ ܐܝܬܘܗܝ ܕܚܣܝܪ ܠܢܦܫܐ‬ ‫ܡܢ ܟܠܗ ܿܗܘ ܡܐ ܕܢܪܓ ܂‬ ݁ ‫ܘܐܠ‬ ‫ܡܢܗ܂‬ ‫ܠܡܐܟܠ‬ ‫ܡܫܠܛ ܠܗ ܐܠܗܐ‬ ܼ ܼ ܿ Syhmg: ‫ ܂ܣ܂ ܘܐܠ ܡܦܩ ܼܗܘ‬Index super ‫ܘܐܠ‬ ‫ܡܕܡ ܥܠܘܗܝ ܡܢ ܟܠܗܝܢ ܿܗܢܝܢ ܕܐܢ‬ ܿ ‫ܢܪܓ ܂ ܘܐܠ‬ ‫ܝܗܒ ܠܗ ܫܘܠܛܢܐ‬ App II ἐξουσιάσει] σʹ δίδωσιν 161 248 Field: καὶ οὐκ ἀναλίσκει εἰς ἑαυτὸν ἀπὸ πάντων ὧν ἂν ἐπιθυμήσῃ, καὶ οὐ δίδωσιν αὐτῳ ἐξουσίαν ὁ θεὸς τοῦ φαγεῖν ἀπ᾽ αὐτοῦ.

6:2c-e is a further instance of where the οʹ text is influenced by Symmachus, but the lemma of Syh follows the main text tradition against the O group. In 253 and 637, two of the four manuscripts from the O group, the words καὶ οὐκ ἀναλίσκει εἰς ἑαυτόν are found following τῇ ψυχῇ αὐτοῦ. The marginal note in Syh shows that these words are part of a longer reading from Symmachus given there. Thus, we have another case of corruption within the textual tradition of the OG, but this time, the corruption helps to preserve the original text of Symmachus in Greek. The Aph‘el of ‫ ܢܦܩ‬can mean to spend,65 but retroversion from this verb in Syriac with such a broad range of usage might have been difficult without the preservation of the text in Greek. Here as well as in the previous case in 5:11a, the lemma of Syh also goes against the O group. 5:6

5:6

MT

Rahlfs’ Edition Influenced by Symmachus via Jerome

‫ִּכי ְבר ֹב ֲחֹלמֹות וַ ֲה ָב ִלים ְּוד ָב ִרים ַה ְר ֵּבה‬ ‫ֹלהים יְ ָ ֽרא‬ ִ ‫ת־ה ֱא‬ ָ ‫ִּכי ֶא‬

65  See J. Payne Smith, Compendius Syriac Dictionary (Oxford, Oxford University Press, 1902), s.v. and M. Sokoloff, A Syriac Lexicon (Winona Lake, IN: Eisenbrauns/Gorgias Press, 2009), s.v.

220 Ra

5:6

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

ὅτι ἐν πλήθει ἐνυπνίων καὶ ματαιότητες καὶ λόγοι πολλοί· ὅτι σὺν τὸν θεὸν φοβοῦ.

Edition 5:6 ὅτι ἐν πλήθει ἐνυπνίων καὶ ματαιοτήτων καὶ λόγων πολλῶν ὅτι σὺν τὸν θεὸν φοβοῦ. ̣ܵ ‫ܡܛܠ ܕܒܣܓܝܐܘܬܐ‬ Syh ‫ܠܡܐ܂‬ ܼ ‫ܕܚ‬ ܵ ܵ ‫ܘܕܣ̈ܪܝܩܘܬܐ‬ ‫ܘܕܡܐܠ ܣܓܝܐܬܐ܂‬ ܼ ‫ܕܚܠ܂‬ ‫ܡܛܠ ܕܐܢܬ ܡܢ‬ ܼ ‫ܐܠܗܐ‬ ܼ Syhmg: ‫ ܂ܣ܂ ܡܛܠ ܣܓܝܐܘܬܐ‬Index super ‫ܡܛܠ‬ ̣ܵ ‫ܣܪܝܩܘܬܐ‬ ‫ܠܡܐ܂‬ ܼ ܼ ‫ܕܚ‬ ܵ ܵ ‫ܘܡܐܠ ܣܓܝܐܬܐ ܀‬ App I καὶ 1°—πολλῶν] et vanitates et verba plurima Hi ↓ = ματαιότητες καὶ λόγοι πολλοί Ra.: cf plurimae vanitates et sermones in numeri Vulg | om καί 2°—πολλῶν 797: homoiot | λόγων πολλῶν tr 339; + πολλα ρηματα 253 157c; + πολλα αρνηματα 637: cf Hi | λόγων] λογω 155 App II ὅτι 1°—πολλῶν] σʹ διὰ γὰρ πλήθους ὀνείρων ματαιότητες καὶ λόγοι πολλοὶ ψεύδων 161 248

Some of the observations made concerning the character of the Bible text of Jerome’s Commentary and influence from Symmachus can be employed to improve reconstruction of the text of OG. Rahlfs’ Text is based solely upon the lemma of the Commentary by Jerome against all the witnesses known to him. Perhaps he felt that the character of the translator supported a text whose syntax in Greek matched that of 𝔐. Yet enough evidence has been assembled to show that Jerome’s lemma is not a reliable witness to the Old Latin or its parent text, the Old Greek. Here as well, the lemma of Jerome is identical to Symmachus and must represent influence from this Jewish reviser. In her edition of Ecclesiastes for La Bible d’Alexandrie, Françoise Vinel also proposes an original text based upon the Greek witnesses rather than relying solely upon the lemma of Jerome. Referring to the last half of 5:6a she states, “Rahlfs calque sur le latin l’emploi du nominatif pour les deux dernier substantifs; nous gardons la série des trois génitifs dépendant de en phḗthei (sic), conformément au text donné par BSA.”66 In her commentary she also affirms that the last two nouns are genitives dependent upon πλήθει instead of forming the subject of the nominal sentence as in MT and Symmachus. She does not, however, indicate 66  F. Vinel, L’Ecclesiaste: Traduction du texte grec de la Septante, Introduction et notes (La Bible d’Alexandrie, 18; Paris: Les Éditions du Cerf, 2002), pp. 31, 134-135.

Basic Principles for the Constitution of the Text

221

Symmachus as the source of the error and her translation actually reflects the text of Rahlfs. While the text of OG appears awkward, and difficult issues are found in the textual tradition concerning reading σύν or σύ in 5:6b, the syntax is not impossible. 5:6b can be read as a noun clause which is the subject and 5:6a is the predicate: “for in the abundance of dreams and vanities and many words it is that you must fear God.” We may not always expect that the OG translator had the same parent text as we do or construed it in the same way. 5:7-8

Is the Old Greek Equivalent to Theodotion or Symmachus?

𝔐 5:7d-8a ‫יהם׃‬ ֶ ‫ּוגְ ִבֹהים ֲע ֵל‬ ‫יִתרֹון ֶא ֶרץ ַּבּכֹל ִהיא‬ ְ ְ‫( ו‬Qr ‫)הּוא‬ ‫ֶמ ֶלְך ְל ָׂש ֶדה נֶ ֱע ָ ֽבד׃‬

Ra

5:7d-8a καὶ ὑψηλοὶ ἐπ᾽ αὐτούς. καὶ περισσεία γῆς ἐν παντί ἐστι, βασιλεὺς τοῦ ἀγροῦ εἰργασμένου.

Apparatus in Rahlfs: εν V†] επι rel. Edition

5:7d-8a καὶ ὑψηλοὶ ἐπ᾽ αὐτούς. καὶ περισσεία γῆς, ἐπὶ παντί ἐστι βασιλεὺς τοῦ ἀγροῦ εἰργασμένου. ݁ ܿ Syh ‫ܥܠܝܗ ܘܝܬܝܪܘܬܐ ܕܐܪܥܐ܂‬ ‫ܘܪܡܐ‬ ܼ ݁ ‫ܗܘ ܐܝܬܘܗܝ ܡܠܟܐ܂‬ ݂ ‫ܒܟܠ‬ ܼ ‫ܐܠܬܪܐ ܕܦܠܝܚ܂ ܀‬

݁ Syhmg: UyEloi Index super ‫ܘܪܡܐ‬ ݁ ݁ ܿ mg ‫ܘܪܡܐ‬ Syh : ‫܂ܣ܂ ܘܝܬܝܪ ܪܡܝܢ‬Index inter et super ‫ܥܠܝܗ‬ ܼ ‫ܠܥܠ ܡܢܗܘܢ ܀‬ ݁ ܿ ‫ܘܪܡܐ‬ Syhmg: ‫܂ܐ܂ܬ܂ ܒܬܪܗܘܢ‬Index inter et super ‫ܥܠܝܗ‬ ܼ Syhmg: ‫܂ܬ܂ ܒܟܠ ܐܝܬܘܗܝ‬Index super ‫ܒܟܠ‬ ܼ ‫݁ܡܠܟܐ ܕܚܩܐܠ‬ ‫ܕܦܠܝܚܐ ܀‬ N.B. Field has ‫ ܦܠܝܚܐ‬instead of ‫ܕܦܠܝܚܐ‬ App I

ἐπί A C B-S 411 L–(125) C´’–(157) d k 68 161 248 252 260´ 296´ 311 336´ 339 443 534 543 547 549 645 698 706 795 Dam GregAg 945 Ollem et com = Ald] absc

222

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

998; ἐν O 125II 542 602-613 766 Hi Syh Did 145,1 Syn 349 = Ra 𝔐 ↓ | ἐστι] pr αυτος 125II-542 Syh: cf 𝔐; pr αυτου 766; εσται 766B; + αυτος O; + και C App II καὶ 1°—αὐτούς] σʹ καὶ (> 161) ὑψηλότεροι ἐπάνω αὐτῶν 161 248 252 (s nom) | ἐπί—εἰργασμένου] σʹ ἐπὶ παντὶ αὐτός ἐστι βασιλεὺς τῇ χώρᾳ εἰργασμένῃ 161 248; οʹ (sic 252) ἐν παντί αὐτὸ βασιλεὺς τῇ χώρᾳ εἰργασμένῃ 252

Rahlfs chose ἐν παντί as original text in 5:8b. As his apparatus shows, he based his choice solely upon Codex Venetus. The dagger indicates that at the most, not more than one minuscule supports this witness. The edition of Holmes-Parsons would have been Rahlfs’ source of information on the minuscules and no other support is listed there except the citation of the Synopsis Scripturae Sacrae spuriously attributed to Athanasius. Why would Rahlfs choose the text of just one manuscript against all the other witnesses? Naturally we do not know his mind, but his choice brings the Old Greek in line with our Masoretic Text. It is hard to imagine the translator being so free with his parent text. Codex Venetus belongs to the O Group and doubtless Rahlfs knew the hexaplaric nature of this source, but unless one can prove that the text received by Origen was corrected using the Three apart from minuses and pluses, there is no good reason to reject the witness of this text. Thus far Rahlfs seems to have a leg on which to stand. Is it possible that the reading of Symmachus somehow influenced almost the entire textual tradition of the Old Greek at this point? Such a hypothesis cannot be ruled out a priori, but one must show a mechanism for it in the history of the text tradition. Research for the Göttingen Edition reveals more manuscript support for the choice made by Rahlfs: eight minuscules in addition to Codex Venetus, two patristic citations as well as the Bible text of Jerome’s Commentary and the text of the Syro-Hexapla. This evidence must be scrutinised. MSS. 253 and 637 belong with Codex Venetus to the O Group. MSS. 125II and 602 are copies of 542 and 613 respectively. MS. 542 is the closest relative of the Syro-Hexapla among all our Greek manuscripts. While the textual affiliation of 613 and 766A and 766B have not been fully worked out as yet, these manuscripts may be suspected of corrections towards the Hebrew text based upon information from the Three. While MS. 252 has the majority text as its lemma, the marginal reading shows the scribe was aware that this was not the οʹ text clearly witnessed by all members of the O Group and Syh. It is not surprising that the bible text of Jerome’s Commentary agrees with MT. If the Old Latin text that he received supported ἐπί, he quickly brought this text into alignment with his Hebrew text. What was the parent text of the translation by Symmachus? While his renderings are certainly more dynamic than either Aquila or Theodotion, he does

Basic Principles for the Constitution of the Text

223

faithfully follow his Hebrew parent and one might reasonably assume that his parent text had ‫ על כל‬instead of ‫ בכל‬as in MT. This is supported by the Targum. The Latin Vulgate is so free at this point that one cannot use it as a clear witness. It is hard to overturn the evidence of the majority of Greek witnesses. If we posit ἐπὶ παντί as original text, then the Greek translator would be literally representing his parent text even though this is not the same as MT. What Rahlfs did not realise is the extent to which the parent text of the Greek translator differs in minor ways from our present Masoretic Text. Indeed, Y. Goldman, editor of Qoheleth for the Biblia Hebraica Quinta has chosen alternative readings against MT in some 58 instances. In 47 of these he appeals to the OG. Goldman also believes that ἐπί represents the original Greek translation here and argues as follows: As noted by Barthélemy in CTAT, 5, ad loc., ‫ עבד‬nifal always refers to the working of the land in the Hebrew Bible (Deut 21:4; Ezek 36:9, 34). Thus G can be read: “And the benefit of a land, above all, is a king to a cultivated ground”; for ‫ על כל‬meaning “above all,” see Pss 95:3; 96:4; 97:9. M can be understood: “And the benefit of the land in everything is a king to a cultivated ground.” The two readings are very close. However, the text transmitted by M conveys the idea of a general well-being coming to a land with a king and cultivated ground. The text as reflected in G gives a more constrasted expression between the powers of a commercial culture, which produces administration and all kinds of social strata with their iniquities (v. 7), and what is best “above all” for a land, a society living from agriculture and having a king at its head. V insuper universae seems to align with G, but V is very free, and Jerome reads like M in his commentary.

In the apparatus of the Biblia Hebraica Quinta, the editor employs the siglum ‘ideol’ to indicate that the text represented by MT and supported by the lemma of Jerome’s Commentary, the Peshitta, and Theodotion, is ideologically motivated. There is evidence that Origen’s Hexaplaric Text was corrected, not least by the statement in the colophon of the Syro-Hexapla that affirms such to be the case, and correction from Theodotion is a mechanism which can easily explain the situation in the text history, at least far more easily than going against almost the entire text tradition. I believe Goldman is right to argue for ἐπί as original text, but either way, one of the Three, either Symmachus or Theodotion, has corrupted the textual transmission of the Old Greek. 6:8

Rahlfs’ Edition Influenced by Aquila and Theodotion

𝔐 6:8

‫ן־ה ְּכ ִסיל‬ ַ ‫ה־ּיֹותר ֶל ָח ָכם ִמ‬ ֵ ‫ִּכי ַמ‬ ‫יֹוד ַע ַל ֲהֹלְך נֶ גֶ ד ַה ַחּיִ ים‬ ֵ ‫ה־ּל ָענִ י‬ ֶ ‫ַמ‬

224 Ra

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

6:8 App

ὅτι τίς περισσεία τῷ σοφῷ ὑπὲρ τὸν ἄφρονα; διότι ὁ πένης οἶδεν πορευθῆναι κατέναντι τῆς ζωῆς. τις > BS*

NETS

For what surplus has the wise over the fool, since the needy knows to walk before life?67

Edition 6:8 ὅ τι περισσεία τῷ σοφῷ ὑπὲρ τὸν ἄφρονα; διὰ τί ὁ πένης οἶδεν πορευθῆναι κατέναντι τῆς ζωῆς; Syh ‫ ܐܬ( ܡܛܠ ※ ܕܡܢܐ ܗܝ↙ ܝܬܝܪܘܬܐ‬super ‫)ܕܡܢܐ‬ ‫ܠܚܟܝܡܐ ݁ܛܒ ܡܢ ܫܛܝܐ܂‬ ‫ܕܥ‬ ݂ ‫ܡܛܠ ܡܢܐ ܡܣܟܢܐ ݁ܝ‬ ‫ܠܡܐܙܠ ܠܘܩܒܠ ܵܚܝܐ܂‬ ݁ ‫܂ܣ܂ ܡܢܐ ܕܝܢ‬Index super ‫ܡܛܠ‬ Syhmg: ‫ܠܒܝܫܐ܂ ܀‬ Syhmg: ‫܂ܐ܂ܬ܂ ܒܕܡܘܬ ܫܒܥܝܢ܀‬Index super ‫ܡܛܠ‬ App I ὅ τι περισσεία τῷ σοφῷ] τι ουν περισσον σοφω Hi (quid enim est amplius sapienti) Olcom ↓ ; quid habet amplius sapiens Vulg | ὅ τι 992 B-S*-998 O C´ d 68 443 534 542 547 728 795 Met Fa] ητις 336; + τίς A C Sc L cII k 248´ 252 260´ 296´ 311 338 339 411 543 549 613 645 698 706 752 766 Ol Arm Syh (sub ※ αʹθʹ) = Gra. Ra ↓ | περισσεία] περισσευει O | διὰ τί 125II 542 543 549 Syh] absc 992 998; διοτι rel | διὰ τί—πορευθῆναι] quid pauperi nisi scire ut vadat Hi ↓ App II ὅ τι περισσεία] σʹ τί οὖν περισσόν 161 248

Two major problems plague the history of the OG Text in 6:8. In the first half of the verse, Rahlfs chose ὅτι τίς περισσεία τῷ σοφῷ as the rendering of the Greek translator for ‫ה־ּיֹותר ֶל ָח ָכם‬ ֵ ‫ּכי ַמ‬.ִ His apparatus indicates that this choice involved going against B and S*. These early majuscules are now supported by the witness of 998, a papyrus dating to 300, 992, a papyrus dating to the sixth century, the O group—apparently an uncontaminated witness to Origen’s text at this point, the Catena group along with the d group—another group of catena manuscripts, and a good number of generally reliable minuscules. Against this are arrayed Sc A C and a number of minuscules well connected to hexaplaric sources. Why would Rahlfs opt for a text so poorly supported in terms of textual witnesses? It is not 67  Official abbreviation for A New English Translation of the Septuagint and the Other Greek Translations Traditionally Included Under that Title, see http://ccat.sas.upenn.edu/nets

Basic Principles for the Constitution of the Text

225

difficult to understand the basis for the choice made by Rahlfs. ὅτι τίς corresponds perfectly to ‫ ִּכי ַמה‬in MT and no good reason from translation technique was available to show how and why the translator would have ignored ‫ ָמה‬here, when he never does so elsewhere apart from the problem in 6:8b. The problem is even more serious when one notes that the Syriac equivalent for τίς is explicitly marked by an asteriskos and metobelos in the Syro-Hexapla and this word is specifically attributed to Aquila and Theodotion. This is damning evidence against Rahlfs’ choice showing that τίς in the textual transmission is corruption from the Three just as surely as the lemma of Jerome is once more stolen from Symmachus in 6:8a and not a shred of it representing the Old Latin. The important and reliable witness of Syh ought to be the Hound of Heaven for the conscience of the text critic here. Yun Yeong Yi’s exhaustive analysis of translation technique in Ecclesiastes shows that the Greek translator employed a variety of approaches to rendering ‫מה‬, all of them contextually sensitive. ‫ למה‬is rendered by ἵνα μή (5:5c, 7:17b), ἵνα τί (2:15d), and μήποτε (7:16b). ‫ מה‬followed by ‫ ׁש‬is rendered by τί τό (1:9a, 1:9b, 8:7a, 10:14c) or just τό 3:15a, and by εἴ τι (6:10a) and ‫ במה ׁש‬by ἐν ᾧ ἐάν (3:22b). ‫ מה‬alone is rendered by τί or τίς (2:2b, 2:22a, 3:19f, 5:10b, 6:11b, 6:12a, 6:12e, 7:10a, 8:4b, 11:2b, 1:3a, 2:12c, 3:9a, 5:15c, 6:8a, 11:5a). The case of 7:24a appears to read ‫( מׁשהיה‬miššehāyâ) as parent text instead of ‫מה־ׁשהיה‬.68 The variegated treatment of ‫ מה‬by the translator suggests another option: the text of the Greek translator may be ὅ τι, understood as indefinite relative rather than as conjunction. In this way, ὅ τι may be understood as a literal translation of both ‫ כי‬and ‫מה‬. Soon after the translation was made, scribes construed the text as ὅτι and hence efforts to bring the text into quantitative alignment with the Hebrew text resulted in the later textual transmission. This solution follows the principle of literal translation technique of the same text as MT, but also respects the import and weight of the textual witnesses. One possible objection is that the Indefinite Relative Pronoun is neuter and περισσεία is feminine. Lack of concord in constructions like this, however, is native to the translator. Another example just two verses away is 6:10a: καὶ ἐγνώσθη ὅ ἐστιν ἄνθρωπος. Another problem is found in 6:8b. One may note that there I propose διὰ τί as original text based on only 125II 542 543 549 and Syh. The principles enunciated for evaluting the textual witnesses all along seem to be completely ignored! Yet this problem is not as serious as it appears. First, διότι is not only uncharacteristic as a rendering for ‫מה‬, it is a conjunction never used by the translator. Second, διότι could easily result from διὰ τί by palaeographic confusion. Third, once 6:8a reads ὅτι τίς περισσεία τῷ σοφῷ, “for what advantage has the wise,” the change from διὰ τί to διότι is quite natural for scribes who have no access to the parent text and are attempting to make some sense of this verse. The text I have reconstructed has 68  Yun Yeong Yi, “Translation Technique of the Greek Ecclesiastes,” 89-94

226

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

both lines as questions as in the Hebrew text, and the second line makes as much sense in Greek as it does in the Hebrew: “Whatever is an advantage of the wise over the fool? Why does the needy know to walk before life?” The Old Greek, like the Hebrew, questions whether the wise has an advantage over the fool. This text was quickly construed as a statement that the wise did have an advantage over the fool and the following line construed as a support rather than as a parallel question. Thus, one problem in 8a led to another in 8b. The Three were used to bring the text back into line with the Hebrew. 6:11b

Rahlfs’ Edition Influenced by Symmachus

𝔐 6:11

‫ׁש־ד ָב ִרים ַה ְר ֵבה ַמ ְר ִבים ָ ֑ה ֶבל‬ ְ ֵ‫ִכי י‬ ‫ַמה־י ֵֹתר ָל ָא ָ ֽדם׃‬

6:11

ὅτι εἰσὶν λόγοι πολλοὶ πληθύνοντες ματαιότητα. τί περισσὸν τῷ ἀνθρώπῳ;

NETS

For many words are multiplying vanity; what surplus has one?

Ra

Edition 6:11

ὅτι εἰσὶν λόγοι πολλοὶ πληθύνοντες ματαιότητα. τίς περισσεία τῷ ἀνθρώπῳ;

ܵ ܵ Syh ‫ܬܐ܂‬ ܼ ‫ܡܛܠ ܕܐܝܬ ܡܐܠ ܣܓܐ‬ ܵ ‫ܕܡܣܓܝܢ ܣܪܝܩܘܬܐ܂‬ ‫ܡܢܐ ܗܝ ܝܬܝܪܘܬܐ ܠܒܪܢܫܐ܂‬ mg Syh : ‫ ܂ܣ܂ ܡܢܐ ܝܬܝܪܐ ܠܒܪܢܫܐ܂‬Index super ‫ܡܢܐ‬ ܿ ‫ܒܚ‬ ܵ ‫ܡܢܘ ܓܝܪ ݁ܝ ݂ܕܥ ܿܗܝ ݁ܕܡܘܬܪܐ ܠܒܪܢܫܐ‬ ‫ܝܐ܂‬ ܵ ‫ܕܡܢܝܢܐ‬ ܵ ‫ܕܝܘܡܬܐ‬ ‫ܕܚܝܐ ܕܣܪܝܩܘܬܐ ܕܝܠܗ‬ ܿ ܿ ‫ܒܕܝܘܗܝ ܡܛܠ ܕܐܦ ܐܠ ܚܕ ܐܡܪ‬ ‫ܕܢܥ‬ ܼ ܿ ‫ܠܒܪܢܫܐ ܡܢܐ ܗܘܐ ܒܬܪܗ ܬܚܝܬ ܫܡܫܐ ܀‬ App I τίς περισσεία O–V-411 L–(125) Syh: cf 13a] absc Didlem; οτι περισσον S* C´(-σων 609) 336 547 645 OlΑΓ = Ald: cf (8a); τι περισσον A C B-Sc-68´’-998 V cII d k 125II 248´’ 296´ 311 338 339 443 542 543 549 698 706 728 766 795(-σων) Didcom 193,1 193,12 Met VI.2,3 Ol–ΑΓ Hi = Ra ↓ App II ὅτι εἰσὶν λόγοι πολλοί] οἱ τῶν ἑτεροδόξων 161 (s nom) 248mg (s nom) = Scholion | (12a)] ὅτι τίς οἶδεν τί ἀγαθόν] σʹ τίς γὰρ οἶδεν ὃ συμφέρει 161 248 | ἀριθμόν] ἐπὶ σπανίου 252mg (s nom)

Basic Principles for the Constitution of the Text

227

Field σʹ τί περισσὸν τῷ ἀνθρώπῳ; τίς γὰρ οἶδεν ὃ συμφέρει τῷ ἀνθρώπῳ ἐν τῇ ζωῇ ἀριθμοῦ ἡμερῶν ζωῆς ματαιότητος αὐτοῦ, ἵνα ποιήσῃ αὐτὸν (σκέπην); ὅτι οὐδὲ εἷς ἐρεῖ τῷ ἀνθρώπῳ, τί ἔσται ὀπίσω αὐτοῦ ὑπὸ τὸν ἥλιον.

In 6:11a, the main problem in the text is to choose between τίς περισσεία and τί περισσόν as the rendering of the translator for ‫מה־י ֵֹתר‬. ַ The textual variant ὅτι περισσόν in S and the Catena group is secondary, since influence from 6:8a has apparently produced it. Once again, the role of the Three have a major part to play in the text tradition as τί περισσόν is also the reading of Symmachus. Manuscript support for τίς περισσεία is weak—only the O and L groups and the text of Syh. Rahlfs may have also considered the text τίς περισσεία influence from 1:3a, where OG also has τίς περισσεία τῷ ἀνθρώπῳ, but renders ‫יִתרֹון‬ ְ instead of ‫י ֵֹתר‬ as in 6:11a. Rahlfs thus appears to be on firm ground in choosing τί περισσόν as original text. The biggest difficulty in accepting Rahlfs’ text is finding adequate motivation from common factors in transcription to change from τί περισσόν to τίς περισσεία. Motivation for changing τίς περισσεία to τί περισσόν is easily found in the reading from Symmachus who also has τί περισσόν for OG τίς περισσεία in 5:15 and 6:8 and τὶ πλέον is attributed to αʹ σʹ for τίς περισσεία in 1:3. An analysis of translation technique is necessary to clarify matters and the evidence from the concordance by Jarick is sufficient for this.69 All forms of the root ‫ יתר‬in Hebrew in the parent text are rendered by Greek words from the περισσ- stem. A form construed as a verb in 3:19 is translated by the verb περισσεύω. Furthermore, all ten instances of the noun ‫יִתרֹון‬ ְ are rendered by the noun περισσεία (1:3, 2:11, 2:13bis, 3:9, 5:8, 5:15, 7:12, 10:10, 10:11). With the form ‫יֹותר‬, ֵ however, a participle used as a noun, the Old Greek translator shows some variation between τίς περισσεία and τί περισσόν. This is displayed by Jarick as follows: περισσεία 6:8 7:11

Jarick’s Concordance for ‫י ֵֹתר‬

‫ן־ה ְּכ ִסיל‬ ַ ‫ה־ּיֹותר ֶל ָח ָכם ִמ‬ ֵ ‫ּכי ַמ‬ ִ ὅτι τίς περισσεία τῷ σοφῷ ὑπὲρ τὸν ἄφρονα ‫וְ י ֵֹתר ְלר ֵֹאי ַה ָ ּֽׁש ֶמׁש‬  καὶ περισσεία τοῖς θεωροῦσιν τὸν ἥλιον

περισσός 2:15 6:11

‫יֹותר‬ ֵ ‫וְ ָל ָּמה ָח ַכ ְמ ִּתי ֲאנִי ָאז‬  καὶ ἵνα τί ἐσοφισάμην; ἐγὼ τότε περισσόν ‫מה־ּי ֵֹתר ָ ֽל ָא ָדם‬  ַ τί περισσὸν τῷ ἀνθρώπῳ

69  John Jarick, ed. A Comprehensive Bilingual Concordance of the Hebrew and Greek Texts of Ecclesiastes (SBL SCS 36; Atlanta: Scholars Press, 1993).

228

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

7:16 12:9 12:12

‫ל־ּת ִהי ַצ ִּדיק ַה ְר ֵּבה‬ ְ ‫ ַא‬  μὴ γίνου δίκαιος πολὺ καὶ μὴ σοφίζου ‫יֹותר‬ ֵ ‫ל־ּת ְת ַח ַּכם‬ ִ ‫ וְ ַא‬περισσά ‫וְ י ֵֹתר ֶׁש ָהיָה ק ֶֹה ֶלת ָח ָכם‬  καὶ περισσὸν ὅτι ἐγένετο ἐκκλησιαστὴς σοφός ‫וְ י ֵֹתר ֵמ ֵה ָּמה ְּבנִי ִָ̣ה ֵהר‬  [καὶ περισσὸν ἐξ αὐτῶν] υἱέ μου, φύλαξαι

Consideration of the context in each instance reveals that when ‫ י ֵֹתר‬is followed by ‫ ל‬plus person, the translator employed περισσεία, but otherwise the neuter of περισσός is used. The only passage in the edition by Rahlfs that does not match this pattern is 6:11. Had any sources witnessed the neuter in 6:8 and 7:11 Rahlfs would have made all of the cases neuter, but he had to respect the textual tradition in these two instances. This means that translation technique supports reading τίς περισσεία in 6:11. This kind of attention to the context shows that the Old Greek translator is not mechanical in his task and no Aquila. Yun Yeong Yi’s 450-page dissertation providing an exhaustive analysis of translation technique in the Greek Ecclesiastes shows that the pattern described here is in tune with the character of the work as a whole. 7:22

Cases Where Separating the Text of the Old Greek and the Three is Impossible

𝔐 7:22

Ra

Edition

‫ם־ּפ ָע ִמים ַרּבֹות יָ ַדע ִל ֶּבָך‬ ְ ַ‫ִּכי ּג‬ ‫ם־א ָּת ִק ַּל ְל ָּת ֲא ֵח ִרים׃‬ ַ ַ‫( ֲא ֶׁשר ּג‬Qr ‫)אתה‬

ὅτι πλειστάκις πονηρεύσεταί σε καὶ καθόδους πολλὰς κακώσει καρδίαν σου, ὅπως καί γε σὺ κατηράσω ἑτέρους.

ὅτι [πλειστάκις πονηρεύσεταί σε] καὶ ⟨ γε ⟩ καθόδους πολλὰς κακώσει καρδίαν σου, ὅπως καί γε σὺ κατηράσω ἑτέρους. App I om (22a) Hi = 𝔙 | om ὅτι Fa | πλειστάκις] πολλακις PsChr; + καιρου O-411 d–357 | πονηρεύσεται] πονηρευεται 299-390-540*-425*-cII–260 371-752 337 248* 543 547 OlΔΙΚ Syh = Compl; πορευσεται 571* 357 | σε] σου 125΄; > 125II-542 Didlem et com 222,20 223,13 223,14 Syh | (22b)] etenim frequenter scit cor tuum Hi = 𝔐 𝔖; > Fa1 | καί 1°] οτι O Hi; > 337 766B OlΑ; + σου SaI | πολλάς] πολλα S* 645 766; πολλακις 475; πολλους 357 | κακώσει] κακωση 157; praes. Syh; πονευρευσει 698 | καρδίαν] καρδια B-68-998* SaI (sed hab Fa2 Arm Did 223,17 PsChr Met VII.2,5 = Ald); η καρδια 125΄ | ὅπως C΄–298 = Ald Gra. Ra.] absc 998; οτι Sc cII d Fa2 SaI II PsChr Met VII.2,5 = 𝔙; ως Fa1; > Arm; οτι

Basic Principles for the Constitution of the Text

229

ως rel (CPA Did 223,17) | καί / γε] tr 547 609; om γε O 390 k 411 SaI II Fa1 PsChr Olcom = 𝔙; + ως S cII d Met VII.2,5 | ἑτέρους] ετεροις 766B; πολλους 147 798; > 425c PsChr; + πολλους C΄–147 798 298 = Ald App II (22ab)] αʹ ὅτι πρὸς πλεονάκις (-νακης 252) καιροῦ πονηρεύσεται (πορευσεται 161) καρδία (-διαν 252) σου 161 248 252 Field: (σʹ) ὅτι πρὸς πλεονάκις καιροῦ πονηρεύσεται καρδία σου. Auct.: αʹ ὅτι πρὸς πλεονάκις καιροῦ πονηρεύσεται καρδία σου

A major crux in OG Eccl is 7:22. This verse has a double translation of the first line in Hebrew and influence from the Three lurks in the background. We will begin by citing in full the astute analysis of McNeile published in 1904: v. 22 (23). On this verse Field quotes a note of Montef. to the effect that two versions are here combined—the former that of 𝔊: ὅτι πλειστάκις πονηρεύσεταί σε καρδία σου, the latter that of Aq.: ὅτι καθόδους πολλὰς κακώσει καρδίαν σου. But the evidence suggests another explanation: πονηρεύσεται MSS. Σ. σε MSS. | om M. S.H. καρδία Β Σ | καρδίαν MSS. rel. ὡς καί γε MSS. (exc. foll.) S.H. | ὡς καί γε ὡς S*. καί γε Sc.a. καί γε ὡς 254. It is probable that 𝔊 originally ran ὅτι καί γε καθόδ. πολλ. κακ. καρδίαν σου, which is certainly Aquilean; while the first clause seems to be made up from other sources. πονηρεύσεται and καρδία are from Σ (see Field). πλειστάκις may be from Θ (see Field on Ps cxix (cxviii.) 64 [mend pro 164]; and if he also had κακώσει, σε would easily arise from the doubling of the σει. The following words ὅτι ὡς are a doublet. ὅτι καί γε was apparently ousted from the prec. line by the καὶ which was placed before καθόδους to combine the two renderings. Pesh. alone follows M. ‫ידע‬.70 70  A. H. McNeile, An Introduction to Ecclesiastes (London: Cambridge University Press, 1904), 163. McNeile’s observation “Pesh. alone follows M. ‫ ”ידע‬deserves comment. κακώσει in LXX 7:22b is based on reading ‫( יָ ַרע‬Hiph‘il Prefix form ex ‫ )רעע‬instead of ‫יָ ַדע‬. Confusion between daleth and resh in these two words is common: 1 Kgdms 10:24 καὶ ἔγνωσαν (ἄλλος ἐβόησαν, σʹ ἠλάλαξαν) ex ‫ ;וַ ּיֵ ְדעּו‬MT ‫וַ ּיָ ִרעּו‬, 2 Esdr 5:17 καὶ γνούς ex ‫;ּודעּות‬ ְ MT ‫ּורעּות‬, ְ Ezek 19:7 καὶ ἐνέμετο ex ‫ ;וַ ּיֵ ַרע‬αʹ καὶ ἐκάκωσε ex ‫ ;וַ ּיָ ַרע‬σʹ καὶ ἔγνωσεν θʹ καὶ ἔγνω ex MT ‫וַ ּיֵ ַדע‬, Mic 4:9 ἔγνως ex ‫;ּת ַדע‬ ֵ αʹ ἐκάκωσας ex MT ‫יעי‬ ִ ‫ּת ִר‬, ָ Dan 12:4 οʹ ἀδικίας ex ‫ ;? ָה ָר ָעה‬ἡ γνῶσις ex MT ‫ה ָ ּֽד ַעת‬,ַ Prov 13:20 γνωσθήσεται ex ‫ ;יִ ּוָ ַדע‬αʹ θʹ κακωθήσεται σʹ κακουργηθήσεται ex MT ‫רֹוע‬ ַ ֵ‫י‬, Prov 15:14 γνώσεται ex ‫ ;? יִ ְד ֶעה‬σʹ ποιμαίνει ex MT ‫יִ ְר ֶעה‬. Cf. Eccl 8:6b γνῶσις ex ‫ ַ;ּד ַעת‬σʹ κάκωσις ex MT ‫ר ַעת‬.ָ I am indebted to J. Ziegler for the above references. Note that Hi has scit in 7:22a so that the Peshitta is not alone in supporting MT.

230

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

A debate existed, already a hundred years ago, as to whether 22a was original and 22b Aquila or 22a constructed from Symmachus and Theodotion and 22b Aquilanic OG. Let us briefly consider evidence first from the text history and then from translation technique. One may well suspect a double translation, but it is difficult to get behind our present textual witnesses. ὅτι ὡς is not a doublet, but rather a palaeographic error for ὅπως arising as an inner Greek corruption by scribes having no access to the parent text. ὅπως is employed for ‫ אׁשר‬in four instances in Eccl (3:11, 7:21, 7:22, 8:12), but ὅτι ὡς never. The only other variant of interest is καρδία in B 998* and two or three minuscules, but I return to this later. From the perspective of translation technique, the following should be noted. First, the ‫ ּגַ ם‬in 7:22a is the only occurrence out of a total of 58 not rendered by καί γε in OG. McNeile’s suggestion that the Old Greek rendered ‫ ִּכי ּגַ ם‬by ὅτι καί γε and that the καί γε was ousted from the preceding line by the καί which was placed before καθόδους to combine the two renderings makes sense but has no basis in the present textual tradition. The only argument that 7:22b is Aquila/Aquilanic is the use of κάθοδος for ‫ּפ ָע ִמים‬. ְ Analysis above indicated this argument is good, but other possibilities exist. It may also be part of the καί γε tradition as precursor to Aquila. Analysis must extend beyond lexical equivalences to syntax. No one seems to compare and contrast 22a and 22b in terms of syntax. Although the Auctarium of Field is at variance with the main work, one of the reasons given in the main work for attributing 22a in part to Symmachus is that he alone employs πρός in the sense of addition for ‫ּגַ ם‬.71 The instances cited by Field are listed below and the most recent information on hexaplaric sources supplied. ‫ » גם‬πρός (insuper)72 Eccl 1:17c 𝔐

‫רּוח‬ ַ ‫ֶׁשּגַ ם־זֶ ה הּוא ַר ְעיֹון‬

71  See F. Field, Origenis Hexaplorum quae supersunt, II, ad Eccl 1:17, 7:23, 9:1 and pp. 382, n. 38 and 394, n. 48. 72  The characterisation of Symmachus provided by Field includes a brief description of how particles are rendered. Equivalences for ‫ גם‬apart from καίγε are ἀλλὰ καίγε (Prov 14:13), ἔτι (Ps 108:10), ἔτι καί (Ps 82:9), καὶ ἔτι (Prov 25:1), πρός, insuper (Eccl 1:17, 7:23, 9:1) and μέντοιγε (Job 18:5?). See F. Field, Origenis Hexaplorum quae supersunt, I:xxxii.

Basic Principles for the Constitution of the Text

231

Edition/Ra

ὅτι καί γε τοῦτ᾽ ἔστιν προαίρεσις πνεύματος App II (17c)] αʹ ὅτι πρὸς τοῦτό ἐστιν νομὴ (νοημα 161 248) ἀνέμου (ανομου 161 248) 161 (s nom) 248 (s nom) 252 Field: (σʹ) ὅτι πρὸς τοῦτό ἐστι νομὴ ἀνέμου.73 Auct.: αʹ ὅτι πρὸς τοῦτό ἐστι νομὴ ἀνέμου. Eccl 7:23

𝔐

‫ם־ּפ ָע ִמים ַרּבֹות יָ ַדע ִל ֶּבָך‬ ְ ַ‫ִּכי ּג‬ Field (σʹ) ὅτι πρὸς πλεονάκις καιροῦ πονηρεύσεται καρδία σου

Eccl 9:1ef 𝔐 Ra Field

‫יהם‬ ֶ ֵ‫יֹוד ַע ָ ֽה ָא ָדם ַהּכֹל ִל ְפנ‬ ֵ ‫ם־ׂשנְ ָאה ֵאין‬ ִ ַ‫ם־א ֲה ָבה ג‬ ַ ַ‫ּג‬ καί γε ἀγάπην καί γε μῖσος οὐκ ἔστιν εἰδὼς ὁ ἄνθρωπος· τὰ πάντα πρὸ προσώπου αὐτῶν, (1ef)] σʹ πρός τι (ετι 252) δὲ οὐ φιλίαν οὐδὲ ἔχθραν ἐπίσταται ὁ ἄνθρωπος 161 248 252 σʹ πρός τε δὲ οὐ φιλίαν οὐδὲ ἔχθραν ἐπίσταται ὁ ἄνθρωπος, τὰ πάντα ἔμπροσθεν αὐτοῦ ἄδηλα.74

What strikes one right away is that the use of πρός in 1:17 and 9:1 is good Greek, while a search in the Thesaurus Linguae Graece of 577 instances of πλεονάκις showed no other occurrence preceded by πρός or any construction similar. The construction is bad grammar, to be frank and this is characteristic of Aquila and not Symmachus. I judge that the use of πρός in 7:22 is not related to the instances in 1:17 and 9:1 and no reason exists to attribute this use of πρός to Symmachus. The only rendering of ‫ ְּפ ָע ִמים ַרּבֹות‬similar to πλειστάκις in 7:22a is πλεονάκις in Ps 106(105):43. πλειστάκις is also a variant (for Hebrew ‫)ׁש ַבע‬ ֶ in the text history in Ps 119(118):164, but attribution to Theodotion is speculation. The text of 7:22a may well be inspired by the reading attributed to Aquila. It might also be a revision of Aquila by Symmachus, but the use of a verbal modifier in the accusative case with πονηρεύομαι is not good style. One can only speculate about the source of 7:22a.

73  Nobil. affert: Schol. ὅτι πρὸς, [insuper. Cf. ad Cap. vii. 23. ix. 1] τοῦτό ἐστι ν. ἀ. Montef. a Drusio deceptus edidit: ᾽ʹΑλλος· ὅτι πρὸς τοῦτό ἐστι προίρεσις πνεύματος. Field II: 382, n. 38. 74  “Haec prodeunt ex Ed. Rom. et ex MSS. duobus Regiis, qui Olympiodorum referunt.”— Montef. Nobil. ed Codd. 161, 248, 252 afferunt: Σ. πρός τε (πρός τι Cod. 161, προσέτι Cod. 252). Cf. Hex. ad Psal. xlviii. 3). Collation at the Septuaginta-Unternehmen which has been checked thoroughly shows MSS 161 and 248 reading τι while 252 reads ετι. No MS reads τε.

232

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

Three manuscripts attribute the hexaplaric reading in 7:22 to Aquila and this must be taken seriously. We have no reading attributed to Symmachus. We do have a reading attributed to Aquila that resembles 7:22a, but we have no way of getting behind our present witnesses to reconstruct the OG. It is tempting to argue that B and 998* reading καρδία is original. I would then reconstruct the OG as follows:

ὅτι καί γε καθόδους πολλὰς κακώσει καρδία σου, ὅπως καί γε σὺ κατηράσω ἑτέρους.



For, in fact your heart will cause harm many times, so that even you cursed others.

This makes good sense, and fits what we know of the translator. But such a reconstruction is speculative. Reconstructing both OG and the text of the Three is a challenge beyond the available evidence. 9:1b

LXX Has Different Hebrew Parent Text

𝔐 9:1b ‫ת־ּכל־זֶ ה‬ ָ ‫וְ ָלבּור ֶא‬ Vorlage LXX?? 9:1b ‫ולבי ראה את כל זה‬ Edition 9:1b καὶ καρδία μου εἶδεν σὺν πᾶν τοῦτο App I (1b) comma] ut considerarem universa Hi: cf 𝔐 Vulg ↓ ; > 161 PsChr | σὺν πᾶν (aut συμπαν) εἶδε(ν) τοῦτο] τουτο ειδε συμπαν 766; ειδε(ν) συμπαν τουτο 411 k Met VIII.1,15 Geo = 𝔖; ειδεν τουτο συμπαν Arm Fa Sa; συμπαν τουτο ειδεν LatPros Dem 12 Voc 1,24; τουτο συμπαν ειδεν Syh | σὺν πᾶν S A 155 336 698 = Gra Ra] συμπαντα 601: c cf 1214a; συμπαν rel: cf 89a 91a | εἶδε(ν)] …]ν 998; ιδε 336´ DamH T; ιδεν A V OlΖ = Gra; c ειδον B (ιδον) 252 534 (ιδον); οιδε DamL App II καὶ καρδία μου σὺν πᾶν εἶδεν τοῦτο] σʹ ut ventilarem universa Hi 123 132 231 321 312 213

εἶδεν σύμπαν τοῦτο k 411 Geo Met = 𝔖 εἶδεν τοῦτο σύμπαν Arm Fa Sa σύμπαν τοῦτο εἶδεν Prosper = Vetus Latina τοῦτο σύμπαν εἶδεν Syh τοῦτο εἶδεν σύμπαν 766 σύμπαν εἶδεν τοῦτο rel (= 998 O usw.)

Basic Principles for the Constitution of the Text

233

‫ וְ ָלבּור‬in 𝔐 is difficult and unique. Goldman notes that scholars have frequently proposed ‫ ולבי ראה את כל זה‬as the Vorlage of LXX, but rejects this possibility since the Greek translator follows the word order of his parent text so woodenly. Clearly this is one of the many places where σὺν πᾶν for ‫ את כל‬became σύμπαν extremely early as an inner Greek scribal error in the textual transmission. One might argue that the word order εἶδεν σύμπαν τοῦτο is a correction to the Hebrew but this word order was only available to the Greek translator and not in later sources whether in Hebrew or in the Jewish revisors as far as we know. Again, one might argue that εἶδεν σύμπαν τοῦτο is the more natural word order, but it could be equally argued that σύμπαν εἶδεν τοῦτο is a stylistic improvement introduced extremely early in Egypt so that it affected most of the text tradition. The list k 411 Geo Met are a curious assortment of unrelated witnesses and the testimony of Sa Fa Arm is also related and similar. This creates the possibility that they attest what one might expect from the translator while the rest are affected by a stylistic improvement. 8:1a Conjecture

𝔐 8:1a ‫יֹודע ֵּפ ֶׁשר ָּד ָבר‬ ֵ ‫ּומי‬ ִ ‫ִמי ְּכ ֶה ָח ָכם‬ Edition 8:1a Τίς ὧδε σοφός; καὶ τίς οἶδεν λύσιν ῥήματος; ————— τίς ὧδε σοφός Ge. = ‫מי כֹה חכם‬: cf quis ut sapiens Hilem 313,455te = 𝔐; quis ita ut sapiens Hicom 314,14; quis talis ut sapiens Hilem 313,455ap = Vulg ↓ ] τις οιδε(ν) (ειδε(ν) 631 609 125´ 548 645* OlΑ) σοφους (σοφιαν C´–159 161-248c Fa1 = Ald) rel (quis novit sapientes HiLXX 314,15 Geo Arab Arm Fa2 SaI II 2 𝔖mss) ————— τίς ὧδε σοφός] αʹ τίς ὧδε (οιδεν 248´) σοφός (σοφον 248) 248´ 252; σʹ τίς οὕτως (ουτος 252) σοφός 252 788

It is doubtful that the Greek translator had a different parent text from 𝔐 and highly unlikely that he was so free in his translation as the entire textual tradition attests. The Editor proposes that τίς ὧδε σοφός is derived from ‫מי כֹה חכם‬, a simple issue of different division of words from 𝔐 and that ὧδε was changed to οἶδεν early in the tradition so that it affected all known manuscripts. This would be an example of an inner Greek scribal error: scribes who had no access to the original guessed οἶδεν given the context.

234

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

6.  Relation to Early Ancient Versions (e.g. Peshitta)75 Literature John D. Meade and Peter J. Gentry. Evaluating Evaluations: The Commentary of BHQ and the Problem of ‫הֹוללֹות‬ ֵ in Ecclesiastes 1:17. SOPHIA—PAIDEIA: SAPIENZA E EDUCAZIONE (Sir 1,27): Miscellanea di Studi offerti in onore del prof. Don Mario Cimosa edited by Gillian Bonney and Rafael Vicent. Nuova Biblioteca di Scienze Religiose 34; Rome: Libreria Ateneo Salesiano, 2012, 197-217.

1:17b 𝔐 1:17b

Ra

1:17b

‫הֹוללֹות וְ ִׂש ְכלּות יָ ַד ְע ִּתי‬ ֵ

παραβολὰς καὶ ἐπιστήμην ἔγνων,

Edition

1:17b περιφορὰς καὶ ἐπιστήμην ἔγνων, ————— txt A B S 998 O L(–106 125) C´’ d k min verss ————— 17 περιφορὰς καὶ ἐπιστήμην] errores (erroresque PsSalo Vulg) et stultitiam Hi PsSalo = Vulg ↓ | περιφοράς Ge.] περιφοραν Gra.; παραφορας Gord.; παραβολην k SaI; παραβοtxt λας rel (= 𝔖): cf 212b 725d; pr και L(–106 125) SaI | ἐπιστήμην Ol] επιστημας 147-159-503560 OlM | ἔγνων] γνων 336 ————— 17 περιφοράς] αʹ πλάνας 161 248 Syh; θʹ παραφοράς 161 248 Syh Syh ܵ ‫ܘܣܟܘܠܬܢܘܬܐ ݁ܝܕܥܬ܂‬ ‫ܐܘܚܕܬܐ‬ ܼ ܵ ܵ mg Syh : ‫܂ܐ܂ ܛܥܝܘܬܐ܀‬Index super ‫ܐܘܚܕܬܐ‬ ܵ ܵ mg Syh : ‫ܦܗܝܐ܀‬ ܼ ‫܂ܬ܂‬Index super ‫ܐܘܚܕܬܐ‬ ܵ Peshitta ‫ܘܡܬܐܠ ܘܣܟܘܠܬܢܘܬܐ܂ ܘܝܕܥܬ‬

75  What appears here is essentially a reprint from the Cimosa FS with extremely minor alterations. Reprinted by permission.

Basic Principles for the Constitution of the Text

235

On this massive problem I collaborated with John Meade.

Well known to textual critics in Ecclesiastes 1:17 is the problematic ‫הֹוללֹות‬ ֵ in MT. It is appropriate to begin by citing the apparatus of Y.A.P. Goldman’s edition in Biblia Hebraica Quinta in order to compare his interpretation of the problem with the analysis given here. Goldman lists the evidence as follows: ‫הֹוללֹות‬ ֵ (interp) ‫ ׀‬παραβολάς G (S) (interp) ‫ ׀‬πλάνας α΄ V (interp) ‫ ׀‬παραφοράς θ΄ ‫׀‬ ‫ וחולחולתא דמלכותא‬T* ‫ ׀ והולהולתא דמלכותא‬T (err) ‫ ׀׀‬pref ‫הֹוללּות‬ ֵ (origin)76

The apparatus of BHQ thus gives in order the readings of MT, Septuagint and Syriac, Aquila and Vulgate, Theodotion, and the Aramaic Targum (listing an original as well as a secondary reading attested for the last of these witnesses). Goldman prefers a proposed emendation over the text of all the extant witnesses. Of particular interest, Goldman concluded that MT, LXX, Peshitta (𝔖), Aquila, and Jerome’s Vulgate (he does not list Hi, the Bible text of Jerome’s Commentary) all represent interpretations of a putative original Hebrew text—unattested in any witness—and not the original itself. The following will argue pace Goldman, that the text in MT is identical to the respective Vorlagen of the ancient versions, and that this reading is original. Furthermore, the reading in MT does not represent a Masoretic interpretation or modification of an original text that is unattested by any witness, for the lemma in MT is able to explain how all of the other readings arose. Furthermore, Goldman’s emendation lacks the explanatory power it claims to have.

7.  Diachronic Development of Greek (Hellenistic / Byzantine Periods) Editors of the Göttingen Editions do not know who will use the critical texts they create and for what purposes. As a far-reaching example, a revision of the grammar of Byzantine Greek could easily be written using the materials in the apparatus and Grammatica and Orthographica sections of the editions. Again, one example will have to suffice. One question in the diachronic development from Classical Greek through Ptolemaic, Roman and Byzantine Periods is the use of ἄν or ἐάν in indefinite relative sentences. This is discussed by Thackeray77 and documented thoroughly by Mayser.78 The matter can be summarised succinctly: 76  A. Schenker, et al., eds. Biblia Hebraica Quinta. Fascicle 18: General Introduction and Megilloth, (Stuttgart: Deutsche Bibelgesellschaft, 2004), 26. 77  H. St. John Thackeray, A Grammar of the Old Testament in Greek, Vol. 1 (Cambridge: University Press, 1909), p. 65. 78  E. Mayser, Grammatik der griechischen Papyri aus der Ptolemäerzeit (Berlin: De Gruyter, 1926), II.1, pp. 261-267.

236

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

By the first century, ἐάν is beginning to be used in relative sentences, whether with Aorist or Present Subjunctive, but the rule of ἄν is still strong from the classical period. The manuscripts in the textual transmission of the Septuagint show the same tendencies: ἐάν is beginning to be used, but the hold of ἄν from the classical period is strong. Like other phenomena in Hellenistic Greek, both could be used side by side, even by the same author in the same stretch of text. Thus, ἐάν is the lectio difficilior and must be considered seriously in each instance. There are five instances in Ecclesiastes of this phenomenon (3:22, 5:3, 8:3, 8:17e, 8:17g). The evidence is given here according to the Göttingen Edition. This is followed by a chart in which the manuscript support for the lemma is provided by subtracting the evidence for the variant from the Kopfleiste. 3:22 MT

‫יאּנּו ִל ְראֹות ְּב ֶמה ֶׁשּיִ ְהיֶ ה ַא ֲח ָריו‬ ֶ ‫ִּכי ִמי ִיְב‬

Edition

3:22 ὅτι τίς ἄξει αὐτὸν τοῦ ἰδεῖν ἐν ᾧ ἂν γένηται μετ᾽ αὐτόν; ————— A C O B-S-68´’-998 L C´’ d k(46s) al–359 verss ————— 22 ἄξει] εξει 155 ↓ | om τοῦ 252 | τοῦ ἰδεῖν] ειδεναι 252 | om ἐν—γένηται 637 | ἐν ᾧ] εως V; quod Hi Aeth; om ἐν 357 252 | ἄν] εαν C B-68 298 357 = Ald Sixt Ra | γένηται] γενοιτο 125´ | om μετ᾽ αὐτόν 766 | αὐτόν] αυτου V 357 125´ 155; αυτων S* 609-797* 252* 336´ 534´ 539 602; αυτο Ald ————— 22 ὅτι—αὐτόν] τίς γὰρ αὐτὸν ἄξει θεάσασθαι τὰ ἐσόμενα μετὰ ταῦτα 161 248 | τοῦ ἰδεῖν—αὐτόν] σʹ ut videat ea quae futura sunt post haec Hi | ἄξει αὐτόν] θʹ habebit eum Syh ἕξει αὐτόν (‫ܢܩܢܝܘܗܝ‬ ܼ ‫)܂ܬ܂‬ 5:3 MT

Edition

‫ר־ּתּד ֹר ַׁש ֵּלם‬ ִ ‫ֵאת ֲא ֶׁש‬

5:3 σὺ οὖν ὅσα ἐὰν εὔξῃ ἀπόδος. ————— A C (969) O B-S-68´’-998 L–(125) C´’ d k al–359 verss ————— 3 σὺ οὖν ὅσα = ‫]א ָּת ֲא ֶׁשר‬ ַ συ οταν 475; συ ος δ᾽ αν 357; συν οσα = Ra. (ex Klostermann) M (‫)את ֲא ֶׁשר‬: ֵ cf quaecumque Hi (sed hab tu itaque quae An Scrip 1,22 Fulg Ep 1,11 Spec

Basic Principles for the Constitution of the Text

237

556,9); γουν οσα OlΗΝ; om οὖν Syh = Pesch ↓ ; om ὅσα C´–298 797 411; inc C | ἐάν B-S-68´ O–V 475-411 C´–298 539 = Ald Ra] > 475 261-545 260 248´ 336´ 542 766 Hi AGeo OlΗΝ An Scrip 1,22 Fulg Ep 1,11 Spec 556,10 = Vulg; αν rel (PsChr Ol–ΗΝ Met IV.6,3) ————— 3 σὺ οὖν ὅσα ἐὰν εὔξῃ] αʹ ‫ =( ܐܢܬ ܟܠܗܝܢ ܗܠܝܢ ݁ܕܢܕܪ ܐܢܬ‬σὺ ὅσα εὔξῃ); σʹ ‫ =( ܐܢ ܬܢܕܘܪ‬ἐὰν εὔξῃ); θʹ ‫ =( ܟܠܗܝܢ ݁ܕܢܕܪ ܐܢܬ‬ὅσα εὔξῃ) Syh 8:3 MT

‫ל־א ֶׁשר יַ ְחּפֹץ יַ ֲע ֶׂשה‬ ֲ ‫ִּכי ָּכ‬

Edition

8:3 ὅτι πᾶν, ὃ ἐὰν θελήσῃ, ποιήσει, ————— A C O B-S-68´’-998 L(-125) C´’ d k al–155 359 verss ————— 3 ἐάν] αν A 545 cII–260 k 248´ 252 338 549 OlΑΓΔΖΙΚΜ; > 609 357 Hi = Vulg 𝔐 | θελήσῃ] OV θελησει A-C B*-S-534 540* 161-248* 543 795 DamV ; θελησοι 296´; θελη L–125 k 338 P 766 Anton 1000 DamR; θελει 125; λαλησει DamM 8:17d MT

‫ְּב ֶׁשל ֲא ֶׁשר יַ ֲעמֹל ָה ָא ָדם ְל ַב ֵּקׁש‬

Edition

8:17 ὅσα ἂν μοχθήσῃ ὁ ἄνθρωπος τοῦ ζητῆσαι, ————— A C O B-S-68´’-998 L–125 C´’ d k al–359 verss ————— 17 ὅσα 1°] διοπερ O–V 475 ↓ | ἄν 1°] ∩ 2° 253txt; εαν A-C 637 L cII 155 248´ 252 296´ 311 698 706 795 Ol Dam (sed hab PsChr Met VIII.1,17 Anast 525 684 Syn 348 = Compl Ra) | μοχθήσῃ] ‑σει 253mg 540-609 252 728 795 OlΖ; ποιησει k (-σει 337) ————— ὅσα ἄν] διόπερ 248 8:17f MT

Edition

‫אמר ֶ ֽה ָח ָכם ָל ַד ַעת‬ ַ ֹ ‫וְ גַ ם ִאם־י‬

8:17 καί γε ὅσα ἂν εἴπῃ ὁ σοφὸς τοῦ γνῶναι, —————

238

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

A C O B-S-68´’-998 L–125 C´’ –157 d k al–359 verss ————— txt txt 17 ὅσα 2°] ο O(–253 ) (253mg) 542 766 Syh; οσας 698 | ἄν 2°] εαν O(–253 ) (253mg litt ε superscr) L k 443 795 Dam–C Syn 348 (sed hab Anast 525) | εἴπῃ] ποιηση C´’–139* 260 540 571* 609; ποιησει 139*-540-571*-609

The evidence for the five cases may be simplified as follows for the sake of clarity: 3:22 ἄν] ἐάν B-68 C 357 = Ald Sixt 5:3 ἐάν B-S-68´ 253-637-411 C´ 539 = Ald Ra] ἄν A V 106-130 cII d–357 k 155 252 296´ 311 338 339 443 543 547 549 613 645 698 706 795 Met IV.6,3 Ol–Η PsChr > 261-545 248´ 260´ 336´ 542 766 OlΗ An Scrip 1,22 Fulg Ep 1,11 Spec 556,10 8:3 ἐάν] ἄν A 545 cII k 248´ 252 338 549 OlΑΓΔΖΙΚ > 357 609 Hi = Vulg 8:17d ἄν B-68´’ S 998 V-411 C´ d k 336´ 338 339 443 542 543 547 549 645 766 Anast 525 684 Met VIII.1,17 PsChr Syn 348 = Compl Ra ἐάν A C 637 L cII 155 248´ 252 260´ 296´ 311 698 706 795 Dam Ol 8:17f ἄν Anast 525 txt εαν O–(253 ) (253mg litt ε superscr) L k 443 795 Dam–C Syn 348 The LXX reveals a slight Tendenz towards ὅς ἄν and ᾧ ἐάν. Probably one should go with the oldest witnesses or majority, when only a few scattered witnesses support the other, e.g. choose 998 and B or 998 and S etc.

Basic Principles for the Constitution of the Text

239

Excursus 1.  List of Witnesses for Eccl. 1:17 Witnesses

Lemmata

BHQ79

‫הֹוללֹות וְ ִׂש ְכ ֑לּות יָ ַד ְע ִּתי‬ ֵ ‫ָל ַד ַעת ָח ְכ ָמה וְ ַד ַעת‬

DSS

Not Extant

Medieval MSS.

De Rossi: ‫הֹוללּות‬ ֵ in 193, 384, 420, 441, 60680

OG

τοῦ γνῶναι σοφίαν καὶ γνῶσιν, παραφορὰς καὶ ἐπιστήμην txt ἔγνων Apparatus: παραφοράς 788sup lin = Gord.] pr και L(–106 125) SaI; περιφοραν Gra.; παραβολην k Fa1 2 SaI; παραβολας rel (788* Geo = 𝔖): cf 212b 725d

Aquila

α΄: πλάνας = 〈‫〉הֹוללֹות‬ ֵ

Symmachus

σ΄: Not Extant

Theodotion Vulgate Old Latin82 Peshitta83 Targum84

̈ .‫ܬ‬.) = 〈‫〉הֹוללֹות‬ θ΄: παραφοράς (= 161-248 788 Syh81 ‫ܦܗܝܐ‬ ֵ ̣

ut scirem prudentiam atque doctrinam erroresque et stultitiam et agnovi = 〈‫〉הֹוללֹות‬ ֵ ut nossem sapientiam et scientiam, errores et stultitiam; cognovi ̈ ‫( = ܠܡܕܥ ܚܟܡܬܐ ܘܐܝܕܥܬܐ‬παραβολάς) ‫ܘܡܬܐܠ ܘܣܟܘܠܬܢܘܬܐ‬ ‫כּותא‬ ָ ‫הּול ָתא ְד ַמ ְל‬ ְ ‫הּול‬ ְ ְ‫( ו‬Sperber) or ‫כּותא‬ ָ ‫חּול ָּתא ְד ַמ ְל‬ ְ ‫חּול‬ ְ ְ‫( ו‬Merino) = 〈‫〉הֹוללֹות‬ ֵ

79  See the next section “Morphology and Meaning of ‫”הֹוללֹות‬ ֵ for our understanding of the text. 80  J. B. de Rossi, Varia Lectiones Veteris Testamenti Librorum, 4 Vols. (Parma, 1784-88; Reprint Amsterdam: Philo Press, 1969). De Rossi comments, “Nonnuli per vav surech in sing., insaniam, ex meis praesertim 193, 384, 420, 441, 606.” ܳ ‫ ܶܦ‬in Thesaurus Syriacus. The evi81  Noun mp emph. meaning “vagatio, aberratio.” s.v. ‫ܗܝܐ‬ dence of Ra 788 updates the provisional edition prepared by Phillip Marshall, “A Critical Edition of the Hexaplaric Fragments of Ecclesiastes”, who presented the Th fragment as περιφοράς in Eccl. 1:17 on the basis of a retroversion of Syh and evidence of περιφοράν in Hi in Eccl 2:2. 82  The text for the Old Latin given by Sabatier is, in fact, the lemma of Jerome’s Commentary on Ecclesiates. This does not reproduce the Old Latin, but is rather Jerome’s revision of the Old Latin based on the Hebrew and the Three, here via Aquila (πλάνας). Thus, we do not, in fact, have the Old Latin pure and simple for Ecclesiastes 1:17. See Peter J. Gentry, “Propaedeutic to a Lexicon of the Three: The Priority of a New Critical Edition of Hexaplaric Fragments,” Aramaic Studies 2.2 (2004), 145-74. 83  Translation: To know wisdom and knowledge and proverbs/parables and understanding. 84  I set my mind to know wisdom, the intrigues of government, knowledge and understanding, and I investigated so that I know that this too is breaking of spirit for a man who tries to

240

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

This presentation of the evidence supposes that all the versions depend on the consonantal text of MT and even possibly on the same vocalisation of this text (see infra). The key difference to this approach and that of Goldman is that this reading of the evidence does not posit interpretational activity on the part of the scribes or translators. As will be shown below, Goldman suggests an interpretative motive for the deviation of all the witnesses from his putative reconstructed text, while we shall posit that the lemma of MT is original and that common copyist errors are able to explain the textual transmission without recourse to explanations based on a copyist’s or translator’s interpretation. Before turning to the textual problem in MT, two other issues must be addressed. First, the lemma in MT is difficult and the issues pertaining to its morphology and meaning must be analyzed. Second, our research will defend the conjecture of Gordis adopted in the Edition against the text as established by Alfred Rahlfs. If the Edition can be defended, then Goldman’s theory of an interpretive motive on the part of the Greek translator may be dismissed in favour of an easier explanation, and consequently the OG reading has faithfully rendered its Hebrew source. Once the OG reading is explained, a comment on the Peshitta is also in order, since this version contributes to understanding the antiquity of the OG itself.

2.  Morphology and Meaning of ‫הֹוללֹות‬ ֵ in Eccl. 1:17 Before commencing an analysis of the textual transmission, it is crucial to understand the Hebrew text on its own terms. This Hebrew word poses certain morphological and semantic problems since it only occurs in Qoheleth. We will deal first with the morphology and second with its meaning.

2.1 Morphology

The morpheme ‫ֹות‬- marks either an abstract feminine noun or is an allomorph for the ‫ָ◌ה‬/‫ ַ◌ת‬feminine singular endings.85 Phoenician marks feminine singular nouns with ‑ôt, and there is some evidence of this morphology as a northern dialect in Biblical Hebrew. The following examples show some of the evidence from the Hebrew Bible for ‫ֹות‬- as a northern morpheme marking a feminine singular noun. There are examples in Biblical Hebrew where the noun marked by ‫ֹות‬- is the subject of a 3fs verb and the antecedent of a 3fs pronominal suffix: Prov. 1:20, 9:1, 14:1.86 find them out. C. Mangan, J. F. Healey, and P. S. Knobel. The Targum of Job, Proverbs and Qohelet, The Aramaic Bible, vol. 15, (Collegeville, Minnesota: The Liturgical Press, 1991). 85  See Joüon-Muraoka, § 88Mk. Gesenius-Kautzsch, §86k, l. See also Gary Rendsburg, “Morphological Evidence for Regional Dialects in Ancient Hebrew,” In Linguistics and Biblical Hebrew, ed., Walter R. Bodine, 65-88, (Winona Lake, IN: Eisenbrauns, 1992). 86  According to some scholars, in these examples, one would expect the vocalisation to be ‫*ח ָכמֹות‬ ֳ (from *qatalat) if the form was plural, but instead the form seems to be constructed by analogy with the singular ‫ח ְכ ָמה‬,ָ see Bauer-Leander, § 506t, Joüon-Muraoka, § 88Mk.

Basic Principles for the Constitution of the Text

241

There are examples where nouns in ‫ֹות‬- form a hendiadys with other singular nouns87: Ps. 45:16 (‫)ב ְׂש ָמחֹת וָ גִ יל‬, ִ which may be compared to Is. 16:10: ‫ִׂש ְמ ָחה‬ ‫וָ גִ יל‬. Gary Rendsburg also includes Ecclesiastes 1:17, 2:12, 7:25, and 9:3 in this category since ‫הֹוללֹות‬ ֵ is paired with another singular noun in these examples 88 (‫ׂש ְכלּות‬/‫ס‬ ִ or ‫)רע‬. ָ To return to ‫הֹוללֹות‬ ֵ in Qoheleth 1:17, scholars remain divided over this morphological problem. Charles Whitley contends that this form should be understood as a plural noun, even though there is no evidence from the Hebrew Bible.89 Other scholars argue that ‫ֹות‬- is an incorrect vocalisation for ‫ּות‬-.90 The most recent comprehensive discussion of this morphological problem is the work by Antoon Schoors in The Preacher Sought to Find Pleasing Words.91 After considering the previous options, Schoors concludes with Dahood that although it is “impossible” to conclude decisively, ‫הֹוללֹות‬ ֵ is best read as a feminine singular and the fact that ‫ֹות‬- and ‫ּות‬- occur along side each other only demonstrates the coexistence of northern and southern dialectical forms in Qoheleth.92 Therefore, this word is probably best understood as a feminine singular and not a feminine plural. This evidence provides a basis for reading ‫הֹוללֹות‬ ֵ in Eccl. 1:17 either as a feminine singular or plural, and accounts for why some of the versions understood the Hebrew text as a plural, but others understood the same text as a singular. It may also explain why some of the versions did not consistently render ‫הֹוללֹות‬ ֵ as singular or plural in Ecclesiastes. In contrast, Goldman understands the ‫ֹות‬- termination to mark only the feminine plural throughout his entire discussion, and this analysis leads him to a different conclusion than the one adopted here.

2.2 Meaning

The root III ‫ הלל‬means “to be infatuated” in the Qal, “to make foolish, to make a mockery of ” in the Poel, “foolish, senseless” in the Poal Participle and “to feign 87  According to Rendsburg, there is usually number accord in these types of constructions, therefore both nouns should be understood as singular. See Rendsburg, “Morphological Evidence,” 79. 88  Rendsburg, “Morphological Evidence,” 79-80. 89  Charles F. Whitley, Koheleth: His Language and Thought, (New York: Walter de Gruyter, 1979), 16. He supposes that ‫הֹוללֹות‬ ֵ is the plural of ‫הֹול ָלה‬ ֵ even though the latter never occurs in the Hebrew Bible but only in Rabbinical Hebrew. 90  George Aaron Barton, The Book of Ecclesiastes, (Edinburgh: T&T Clark, 1912), 87. Barton argues that the noun should be read as ‫ּות‬- since it is vocalised that way in 10:13 and the LXX translates it as a singular in 2:12 and 7:25. As will be demonstrated below, Goldman’s proposal attempts to defend this type of conclusion. 91  Antoon Schoors, The Preacher Sought to Find Pleasing Words: A Study of the Language of Qoheleth Part 1 Grammar, Leuven: Peeters, 1992. 92  Ibid., 66-7.

242

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

madness, to act like a madman” in the Hithpoel.93 ‫הֹוללֹות‬ ֵ is an abstract noun from this root and means either “madness”94 or “ignorance.”95 Having established that the word means either “madness” or “ignorance” and that the form of the word can be feminine singular or plural, we turn next to the reconstruction of the Old Greek.

3.  The Reconstruction of the Old Greek Antecedent to making claims about the original Hebrew text, correct reconstruction of the Old Greek should be ascertained, and even prior to reconstruction of OG, we must be certain about the reading of Theodotion. At Eccl 1:17 Field gives παραφοράς (sive περιφοράς) as the reading of Theodotion. According to the footnote, Field’s reading follows that of Nobilius (i.e. ̈ normally a rendering of περιφορά. Morinus96) and he notes that Syh has ‫ܦܗܝܐ‬, ̣ He further refers to 2:2 in which Theodotion has περιφοράν as attested by Syh (‫ )ܦܗܝܐ‬and more clearly by Jerome, who gives the reading in Greek. Although Field had collations from 161 and 248 for the Auctarium, the reading of these manuscripts is not given there, nor does he note that the rendering παραφοράς is now attested in Greek and not simply based upon retroversion. Field is the basis for the reading supplied by Goldman. In 2007 Phillip Marshall completed a doctoral dissertation supervised by Peter Gentry entitled “A Critical Edition of the Hexaplaric Fragments of Ecclesiastes.”97 Marshall gives περιφοράς as his critical reconstruction of the text of Theodotion based on the fact that this is the term which is elsewhere rendered by ‫ ܦܗܝܐ‬in Syh. He may also have been influenced by an earlier proposal given by Gentry.98 MS. 788 forces us to reconsider the proposal of Marshall. Three Greek manuscripts now attest παραφοράς for Theodotion at 1:17: 161-248 and 788. If 788 is related to the pair 161-248, it is as a much earlier ancestor in the textual tradi-

93  HALOT s.v. III ‫הלל‬. 94  BDB s.v. ‫הֹול ָלה‬. ֵ HALOT lists “foolishness, blindness” for ‫הֹוללֹות‬. ֵ “Blindness” may be acceptable, if by blindness, one envisions a person walking with no aim or wandering in circles. “Frenzy” may also capture the meaning of this word. 95  Antoon Schoors, The Preacher Sought to Find Pleasing Words: A Study of the Language of Qoheleth Part II Vocabulary, (Leuven: Peeters, 2004), 442-3. Schoors argues that the word means “ignorance” or “stupidity” because it is in opposition with roots such as ‫ ידע‬and ‫חכם‬. 96  Morinus and not Nobilius was the one who collected the hexaplaric materials, see Gérard Norton with Carmen Hardin, Frederick Field’s Prolegomena to Origenis Hexaplorum quae Supersunt sive veterum interpretum Graecorum in totum Vetus Testamentum fragmenta (Paris: Gabalda, 2005), 16-17. 97  Phillip Marshall, “A Critical Edition of the Hexaplaric Fragments of Ecclesiastes,” Ph.D. Dissertation, The Southern Baptist Theological Seminary, 2007. 98  i.e. Peter J. Gentry, “Propaedeutic to a Lexicon of the Three.”

Basic Principles for the Constitution of the Text

243

tion.99 This must be taken seriously. We cannot use Syh as an argument here since ̈ would be used to render either παραφοράς or περιφοράς; the translators of ‫ܦܗܝܐ‬ ̣ Syh could not distinguish between the two synonyms. The only other relevant rendering of Theodotion is Eccl 2:2 where Old Greek has περιφοράν for ‫הֹולל‬ ָ ‫ ְמ‬and for Theodotion, Syh has ‫ ܦܗܝܐ‬and Jerome attests περιφοράν. We must accept Jerome’s witness as the correct retroversion for the word in Syh. How can we reconcile the fact that our witnesses attest παραφοράς for Theodotion at 1:17 and περιφοράν at 2:2? Let us remember that for Ecclesiastes, Theodotion is a reviser. Since OG has περιφοράν at 2:2, he leaves it at that, and, we will argue, since OG had παραφοράς at 1:17, he is also content with that translation there. Both of these renderings suit his approach to translation technique. With the reading of Theodotion established, we can now defend παραφοράς as the original text at 1:17, since this reading is 1) supported by translation technique, 2) is supported by Ra 788 and is no longer a conjecture, 3) can be shown as the source of the inner-Greek error παραβολάς without any intermediate step(s), and 4) evidence from the text history of Eccl 2:12 and 7:25 shows that scribes reverted to the original text of 1:17 when they changed περιφοράν to παραφοράν there. Afterwards, we will interact with Goldman’s objections to Robert Gordis’ conjecture and comment on the reading of the Peshitta in 1:17 and its relationship to the LXX tradition.

3.1  The Evidence of Translation Equivalents100

‫הֹוללֹות‬ ֵ

Greek Equivalents

Reference

παραφοράς (παραβολάς Rahlfs)

1:17

περιφέρεια

9:3

περιφοράν

2:12; 7:25

‫ הלל‬verbal root Greek Equivalents

Reference

περιφοράν

2:2 (for Polal ptc.)

περιφέρει

7:7

99  See stemma for Ecclesiastes in the Appendices. 100  The data for this section comes from John Jarick, ed. A Comprehensive Bilingual Concordance of the Hebrew and Greek texts of Ecclesiastes, Atlanta: Scholars Press, 1999.

244

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

‫הֹוללּות‬ ֵ Greek Equivalents περιφέρεια

Reference 10:13

παραβολή Hebrew Source Terms

Reference

‫הֹוללֹות‬ ֵ

1:17 (?)

‫ָמ ָׁשל‬

12:9

περιφέρεια Hebrew Source Terms

Reference

‫הֹוללֹות‬ ֵ

9:3

‫הֹוללּות‬ ֵ

10:13

περιφορά Hebrew Source Terms

Reference

‫הֹול֑ל‬ ָ ‫ְמ‬

2:2

‫הֹוללֹות‬ ֵ

2:12

‫הֹוללֹות‬ ֵ

7:25

περιφέρω Hebrew Source Terms ‫הֹולל‬ ֵ ְ‫י‬

Reference 7:7

3.2  Analysis of Translation Technique

Although it is possible to posit that the LXX had ‫ ְמ ָׁש ִלים‬for its Vorlage in 1:17, and thus a different Vorlage than MT, this explanation is not able to account for how ‫הֹוללֹות‬ ֵ could be derived from original ‫מ ָׁש ִלים‬, ְ nor is it able to account for how ‫ ְמ ָׁש ִלים‬could be derived from original ‫הֹוללֹות‬. ֵ One could suggest that these readings illustrate a “double reading” which reflects independent traditions of the Hebrew Bible. However, if the OG had the same Vorlage to which MT witnesses and the reading enshrined in the LXX tradition resulted from a very early innerGreek error, a better and simpler solution is at hand. Consideration from translation technique causes one to reject παραβολάς as the OG. The translator does not use this word to render ‫;הֹוללֹות‬ ֵ however, we have six cases where the translator rendered ‫הֹוללֹות‬ ֵ or the verbal root ‫ הלל‬with a sin-

Basic Principles for the Constitution of the Text

245

gular form of περιφορά, the cognate noun περιφέρεια, or a form of the verb περιφέρω. These data support the probability that the translator used the plural form of the synonym παραφοράς in 1:17 and not παραβολάς, which would introduce a very different meaning into the text. Also, one finds the converse in Ecclesiastes 12:9 where παραβολάς renders ‫מ ָׁש ִלים‬. ְ These considerations have given rise to several conjectures / emendations for the OG, and Gordis’ conjecture now finds manuscript support in MS. 788.101 Translation technique also considers the general habits and patterns of the translator by seeking to discover if the translator is given to invent divergent meanings from the source. If the translator had ‫הֹוללֹות‬ ֵ for his Vorlage, would he attempt to interpret the text by inventing or introducing an entirely different meaning (παραβολάς) than what is in his source? Although the translator may have intentionally altered the meaning of the text in 1:18,102 one strains to find any further examples of an inventive approach to the source.103 The translator usually attempts to give a consistent and faithful rendering of his source, and therefore the textual critic may posit a reasonable conjecture based on the habits and tendencies of this translator. Although the translator is not completely consistent with his equivalents in every context, he chooses words from the same Greek root or a synonym for both the verb and the noun in order to render the Hebrew verb and derivatives.104 Before considering the transmission of the LXX text, one must ascertain the meanings of περιφορά and παραφορά. περιφορά is related to περιφέρω which means go/carry round > turn around, make dizzy > turn mad (as in Plut. Caesar 32 and Eccl. 7:7), hence the noun could mean “madness” or “dizziness.”105 In the passive, the verb περιφέρω can mean “to wander about”, or “to be unstable” thus one could argue that περιφορά means “error.” LSJ lists “error” under περιφορά, but only lists references to Ecclesiastes as support. The evidence from Ecclesiastes suggests that the Hebrew word ‫הֹוללֹות‬ ֵ means “madness,” and that meaning is 101  See Robert Gordis, “Ecclesiastes 1:17—It’s Text and Interpretation,” JBL 56 (1937), 323330. Gordis appeals to translation technique to demonstrate an emerging pattern where ‫הֹוללֹות‬ ֵ is correctly translated in six other contexts by a word other than παραβολή. He proposes the original LXX read ΠΑΡΑΦΟΡΑΣ (326). See also Peter J. Gentry, “Propaedeutic to a Lexicon of the Three: The Priority of a New Critical Edition of Hexaplaric Fragments,” Aramaic Studies 2.2 (2004), 145-74. Gentry proposed ΠΕΡΙΦΟΡΑΣ for the OG, but in this earlier paper he simply posited the mechanism of a copyist’s error of hearing and sight without demonstration. 102  γνώσεως for ‫ּכ ַעס‬.ָ Errors of sight and sound are not impossible in this case, but they are improbable. 103  Goldman lists 1:18 and possibly 5:5 as examples of intentional change due to theology. 104  The one occurrence of περιφέρεια for ‫הֹוללֹות‬ ֵ in 9:3 is curious. In 10:13 the translator renders ‫הֹוללּות‬ ֵ with περιφέρεια. There are two possible reasons for this change: 1) an error of hearing between ô and û may have taken place, and accordingly the translator chose different equivalents. 2) The translator did not attempt an overly rigid translation, but was content to vary his equivalents with the use of cognates. 105  LSJ s.v. περιφέρω and περιφορά.

246

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

what the translator intended to convey. The meaning “error” is more explicitly intended by Aquila (πλάνας) and then Jerome (errores), following Aquila. The verb παραφέρω may mean “carry past or beyond.” Other meanings such as “to turn aside or away” > “to mislead, lead away” > “to err, go wrong” derive from this meaning. The noun παραφορά has the basic meaning of “going aside.” From this meaning the word can have the sense of “error,” but the word may also denote a quick side-to-side motion such as the waving of a hand or a sword and thus may mean “frenzy”106 from which “madness” obtains. The words are synonyms and both are used by the translator to render forms from the ‫ הלל‬root. The new evidence of Ra 788 shows that παραφοράς is original and that παραβολάς is the variant, even though the OG translator chose περιφορά for the rest of the instances. Some explanation is required to show how παραβολάς derived directly from παραφοράς according to common scribal errors of sight and sound. In summary, the OG translator employed two words for the ‫ הלל‬root, παραφοράς in 1:17 (as will be shown below) and περιφορά in the other occurrences. This stylistic change is probably attributable to variatio, which is the unsystematic employment of synonyms.107 It is also interesting to note that the translator used the plural of παραφορά at 1:17, the only instance of παραφορά in Ecclesiastes, and the singular of περιφορά in the other occurrences.

3.3  The External and Internal Evidence

Since the manuscript evidence of 788 and translation technique indicate that the OG most probably did not have παραβολάς but παραφοράς, how did παραβολάς obtain? With the manuscript evidence of 788, we may now posit an immediate derivation of παραβολάς from παραφοράς, and 2:12 and 7:25 now furnish corroborating evidence that the reading preserved in 788 is original. A brief examination of 2:12 and 7:25 confirms the original reading of 788. In 2:12, the OG is περιφοράν, but B and 998 among other Greek witnesses read παραφοράν instead of περιφοράν. In 7:25, A and Sa3 have παραφοράν where the original text is περιφοράν. What mechanism can explain the change of περιφοράν to παραφοράν in these witnesses? The original text of 1:17 provides evidence for an explanation of this change. Previously, a solid explanation for the variants in 2:12 and 7:25 was not available, but, if the reconstruction of 1:17 is accepted, 106  LSJ s.v. παραφορά. 107 Regarding variatio see Heinrich Lausberg, Handbook of Literary Rhetoric: A Foundation for Literary Study, edited by David E. Orton and R. Dean Anderson, translated by Matthew T. Bliss, Annemiek Jansen, and David E. Orton, (Leiden: Brill, 1998) §257b. Regarding taedium see idem, § 269. Linguistic ornatus is either in verbis singulis or in verbis coniunctis. The former has several manifestations, but vocalitas is probably the reason why a scribe would change περιφορά to παραφορά in Ecclesiastes, since it “consists, roughly speaking, of synonyms that are more or less equivalent in meaning and suitability but which nonetheless bear fine semantic or other differences” (idem, § 542).

Basic Principles for the Constitution of the Text

247

one may explain the variants in these texts as due to harmonisation with the original text of 1:17. The derivation from παραφοράς to παραβολάς is more plausible than it may appear.108 The Egyptian papyri attest to a confusion of the labials β and φ.109 The papyri also attest to a confusion of the liquids λ and ρ when these letters are in intervocalic position.110 These changes cannot be explained from the history of the Greek language alone, but rather they are caused in part to bilingual interference with Coptic in Egyptian papyri.111 In summary, based on translation technique, the translator would have used περιφοράς or παραφοράς, not παραβολάς. MS. 788 provided the needed manuscript evidence to posit that the OG was παραφοράς and the textual transmission of 2:12 and 7:25 probably confirm this decision. The immediate derivation of παραβολάς from παραφοράς was shown to be very probable since the linguistic environment was conducive for causing confusion of labials and liquids.

3.4  Goldman’s Proposal for the OG

Goldman lists four reasons why Robert Gordis’ thesis112 should be rejected: 1) the inner-Greek scribal error is graphically unlikely; 2) it ignores the fact ‫ הוללות‬plural is an exception in 𝔊 Qoh; 3) it ignores the agreement between MT ‫ שכלות‬and 𝔊 ἐπιστήμη; 4) it ignores the literary inclusio made by this word in 𝔊 Qoh (1:1712:9).113 Of these four objections (1) may be dismissed because Gignac provides sufficient evidence of the errors from the papyri based on phonological factors relevant to Eccl. 1:17. Also, Goldman does not mention the textual evidence from Eccl. 2:12 and 7:25, which indicates that the scribes were probably changing original περιφοράν to παραφοράν, because the original text of 1:17 was παραφοράς. These two factors make an inner Greek scribal error in 1:17 probable. This paper also rejects (4) since Goldman does not provide a sufficient reason for why the LXX would invent a reading for the sake of the literary structure. Instead, one finds evidence of faithfulness to the Vorlage often times at the expense of intelligibility for the target audience in Greek Ecclesiastes. One may also reject (3) since it has no bearing on the present problem, that is, the two problems should be 108  ΠΑΡΑΒΟΛΑΣ // ΠΑΡΑΦΟΡΑΣ. 109  See F.T. Gignac, A Grammar of the Greek Papyri of the Roman and Byzantine Periods, I: Phonology (Cisalpino-Goliardica, 1976): 97-98. 110  Ibid., 102-107 (see 104-5 for examples of the change in intervocalic position). 111  See Francis T. Gignac, “The Papyri and the Greek Language,” Yale Classical Studies 28 (1985): 157 and Carlo Consani, “La koiné et les dialectes grecs dans la documentation linguistique et la réflexion métalinguistique des premiers siecles de notre ère,” in La Koiné grecque antique I: une langue introuvable? ed. Claude Brixhe (Presses Universitaires de Nancy, 1993): 28. 112 That ΠΑΡΑΒΟΛΑΣ would be an inner-Greek corruption of ΠΑΡΑΦΟΡΑΣ (followed by Gentry). 113  Goldman, 68*.

248

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

treated separately and the evaluation of the evidence of one problem should not necessarily influence the evaluation of the evidence of another. In the case of ἐπιστήμη for ‫שכלות‬, the LXX was faithful to the text of MT as it was received, which supports the method and assumption of this paper that OG does not invent readings from the source. Although Goldman’s reconstruction of MT for this textual problem is intriguing, it does not have relevance to the present problem in OG. Of Goldman’s four objections, (2) presents the greatest challenge to Gordis’ conjecture of the OG: why does the translator use the plural παραφοράς in 1:17 instead of the singular περιφοράν to render Heb ‫ ?הוללות‬The translator renders the same Hebrew word with the singular in the other occurrences; therefore, the proposed reading of the OG would be an exception to the translator’s pattern. However, there are two reasons why Goldman’s objection is not sustainable: 1) Goldman misses the more evident conclusion for an obscure one. The conjecture of Gordis and now the critical text of Gentry still have the strength of translation technique on their side. Although the translator uses the plural, he still uses either περιφορά or παραφορά to render the source text. Goldman’s proposal that OG is interpreting ‫ הוללות‬with παραβολάς is simply not plausible in Greek Ecclesiastes where the translator is not given to invent new meanings from his source. 2) Goldman has not adequately shown whether the form in Hebrew (‫ות‬-) is plural or singular. Perhaps the Greek translator is puzzled over this point, and that is why in 1:17 he reads ‫ הוללות‬as a plural and he translates it with the plural of παραφορά, but he reads ‫ הוללות‬as a singular in the rest of the occurrences of the book and thus he uses the singular of περιφορά throughout the rest of the book. The Hebrew form is ambiguous, and the translator has used Greek synonyms to render the same Hebrew root, but he has distinguished also between singular and plural by using παραφοράς in the plural for the Hebrew plural and περιφορά in the singular for the Hebrew singular. The solution to the OG outlined in this paper is to be preferred over Goldman’s, since his solution only attempts to understand the OG in terms of an alleged interpretive translation technique, an approach which the Greek translator employs once or twice.114 A better solution begins with positing a text resembling MT as the Vorlage of the OG translator and attempts to explain how the Greek reading arose from this text in a way consistent with translation technique and the classical canons of textual criticism.

4.  The Significance of the Peshitta’s Reading for the OG

̈ The Peshitta reads ‫ܡܬܐܠ‬, which means “proverbium, parabola.”115 This reading is dependent on the LXX and not on a Hebrew Vorlage. At this point, only general 114  See note 24. ܰ 115 s.v. ‫ܬܐܠ‬ ܴ ‫ ܡ‬in Thesaurus Syriacus.

Basic Principles for the Constitution of the Text

249

statements regarding the character of Peshitta Qoheleth can be made.116 A new study by Sebastian Brock has recently been published to introduce the Syriac Bible. Concerning the character of the Peshitta, Brock says: The translators all worked basically from the Hebrew text, and this Hebrew text was essentially the same as the consonantal Hebrew text of our printed Hebrew Bibles … In some books the translators seem to have consulted or made use of other translations: thus at various places in the Pentateuch (Genesis, Deuteronomy), there are some remarkable links between the Peshitta and the Jewish Aramaic Targums and for some of the Prophets and Wisdom books the translators probably consulted the Septuagint on occasion, in order to seek help over difficult passages in Hebrew.117

Michael Weitzman has articulated a methodology for determining dependence between the versions of 𝔖 and OG.118 First he attempts to locate those instances where 𝔖 and OG agree over and against MT. He then attempts to explain the parallel via (1) polygenesis, (2) common Vorlage, and (3) common exegetical tradition. If the parallel cannot be explained through these means, then 𝔖 is dependent on OG.119 In Ecclesiastes 1:17, there seems to be no explanation for the agreement of OG and 𝔖 except that 𝔖 is dependent on OG, since there is no way ̈ for 𝔖 to translate Hebrew ‫ הוללות‬with ‫ ܡܬܐܠ‬apart from the influence of OG.120 Two other examples of dependency from Qoheleth chapters 1 and 2 may illustrate this point further. Qoh. 1:10 provides us with an interesting example of dependence of 𝔖 on OG. The Hebrew text (‫אמר‬ ַ ֹ ‫ )יֵ ׁש ָּד ָבר ֶׁשּי‬is rendered by the following in OG: ὅς λαλήσει καὶ ἐρεῖ. The LXX has smoothed out the Hebrew syntax by (1) moving the relative pronoun to the front, (2) reading the Hebrew noun (‫)ּד ָבר‬ ָ as a verb, (3) inserting καί, and (4) leaving ‫ יֵ ׁש‬untranslated.121 𝔖 reads ‫“( ܟܠ ܕܢܡܠܠ ܘܢܐܡܪ‬everyone who will speak and say”). 𝔖 agrees with OG but is in116  John Meade has completed a paper which includes a full-scale comparison of MT, OG, and P of Ecclesiastes with the intent of locating those instances where P depends on OG and not on MT. This paper isolated some 47x where P is dependent on OG in the book of Qoheleth and will be submitted for publication in the future. 117  Sebastian Brock, The Bible in the Syriac Tradition, (Georgias Press, 2006), 23. 118  Michael P. Weitzman, “Peshitta, Septuagint and Targum,” in VI Symposium Syriacum 1992: University of Cambridge, Faculty of Divinity, 30 August-2 September, 1992 edited by René Lavenant (Rome: Pontifical Biblical Institute, 1994), 57. 119  Weitzman, “Peshitta, Septuagint and Targum,” 51-55. 120 Antoon Schoors also draws this conclusion in this instance. Antoon Schoors, “The Peshitta of Koheleth and Its Relation to the Septuagint,” in After Chalcedon. Studies in Theology and Church History. edited by C. Laga et al. (Uitgeverij: Peeters, 1985), 354. 121  The last of these changes is surprising since OG translates ‫ יֵ ׁש‬with εἰμί 15 out of 16 times with 1:10 remaining the only exception (2:13, 21; 4:8, 9; 5:12; 6:1, 11; 7:15(2x); 8:6, 14(3x); 9:4; 10:5). Syriac ‫ ܐܝܬ‬occurs in the same 15 places as Hebrew ‫יֵ ׁש‬. Syriac uses the particle of existence in a few other places where ‫ יֵ ׁש‬is not used: 2:15; 3:15; 6:8; 7:19, 26; 8:13; 12:12.

250

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

dependent of OG with respect to the addition of ‫ܟܠ‬. Perhaps the best explanation is that 𝔖 consults OG but makes its own stylistic improvement for the purpose of smoothing out the Hebrew syntax or making explicit what is implicit. Qoh. 2:15(16) provides us with one last example of dependence of 𝔖 on OG. OG reads διότι ἄφρων ἐκ περισσεύματος λαλεῖ. There is no known Hebrew Vorlage for this text, and it is probably because this reading is an inner-Greek scribal error due to the accidental scribal addition of a marginal gloss related to Matt. 12:34 (ἐκ γὰρ τοῦ περισεύματος τῆς καρδίας τὸ στόμα λαλεῖ).122 𝔖 reads along with the inner-Greek scribal error: ‫ܡܛܠ ܕܣܟܐܠ ܡܢ ܝܬܝܪ ܡܡܠܠ‬. There is no extant Hebrew Vorlage for 𝔖 to translate, and polygenesis is not probable since 𝔖 follows the same word order of OG and has the same exact paraphrase. One could imagine that 𝔖 might have access to the text of Matthew, but it is difficult to think that 𝔖 would arrive at the same paraphrase as OG independently. These three examples of dependence of 𝔖 on OG do not have uniform significance. The second example shows that 𝔖 depended on the reading of the OG, while the first and third demonstrate that 𝔖 often had access only to inner-Greek corrupttions. 998 is the earliest MS. witness to the Greek text, and it has παραβολάς at 300 CE rather than the OG παραφοράς. 𝔖 has the same reading as 998. Peter Gentry drew this implication in his paper, “The interesting point is that the most recent position of scholars of the Peshitta is that this version may be as early as the third century, since it was quoted by fourth century Fathers. If so the Peshitta is as early a witness as our earliest Greek manuscript (e.g. 998).”123 Although the implication Gentry draws is accurate, his conclusion may have more significance than originally perceived, since it is more probable to date the Pentateuch of the Peshitta to c. 150 and the last parts of the OT to c. 200 CE.124 If 𝔖 Qoheleth is dated between 150 and 200 CE, 𝔖 actually witnesses to an early inner Greek scribal error one hundred years before the earliest manuscript of Ecclesiastes (e.g. 998). At this juncture 𝔖 provides strong textual warrant for dating Greek Ecclesiastes before the second century CE. The fact that παραβολάς has dominated the 122  Schoors, “The Peshitta of Koheleth and Its Relation to the Septuagint,” 351. 123  Gentry, “The Distinctive Aims of the Gottingen Apparatus.” 124  Michael P. Weitzman, The Syriac Version of the Old Testament: An Introduction, (United Kingdom: Cambridge, 1999), 258. Weitzman generally dates the Pentateuch of P to c. 150 CE and the later books of the OT such as Ezra-Nehemiah and Chronicles to c. 200 CE. Therefore, Qoheleth would probably be dated somewhere between these books. Also, the archaic use of ‫ ܝܬ‬in P as the marker of the direct object, and the subsequent ignorance of this meaning of the particle by later church Fathers such as St. Ephrem (4th century) also aid in dating P generally. See Weitzman’s acute observation of Ephrem’s ignorance of the use of ‫ ܝܬ‬as marker of the direct object in his interpretation of Genesis 1:1 in Weitzman, The Syriac Version…, 253. P Qoheleth also contains instances of this archaic use of the direct object marker (cf. 3:17 and 4:1) which is positive evidence that this book was translated before the third century.

Basic Principles for the Constitution of the Text

251

entire textual tradition, except the sup lin reading of 788, indicates that this error occurred very early in the transmission of the text. The earliest witnesses contain the error (𝔖 and 998), confirming that the error entered before 300 CE and possibly c. 200 CE. Therefore Greek Ecclesiastes would need to be translated between the first century BCE and the first century CE.125

Conclusion

The inner-Greek scribal error of Eccl 1:17 entered very early into the transmission of 𝔊 so that no Greek witnesses preserve the OG, and even 𝔖 between 175200 CE is dependent on the corruption of the Greek text providing further evidence that the corruption from παραφοράς to παραβολάς happened soon after the original translation.

5.  The Hebrew Text Having analyzed closely the morphology of MT and having reconstructed the OG, we now proceed to establish the Hebrew text. The external evidence will be evaluated and then the internal evidence. We will then continue to interact with Goldman’s conjecture that the original Hebrew was ‫הֹוללּות‬ ֵ and not ‫הֹוללֹות‬ ֵ as preserved in MT.

5.1  External Evidence

‫הֹוללֹות‬: ֵ MT OG (𝔖) α΄ (OL V) θ΄ T // ‫הֹוללּות‬: ֵ De Rossi mss. Goldman’s conjecture.126 Whether one reads MT as a feminine plural or feminine singular, the versions testify to the consonantal text and the vocalisation of that text. Those versions within parentheses depend upon the version to the left, and thus they are not in125  Pace James K. Aitken, “Rhetoric and Poetry in Greek Ecclesiastes,” BIOSCS 38 (2005): 55-77. Aitken has provided scholars with a very intriguing and well written article on the literary features of Greek Ecclesiastes. He argues that Ecclesiastes was probably translated in the second century CE based on rhetorical and poetic features that the book possesses. However, this type of evidence is circumstantial, since the same features could be found earlier or later than the second century CE. What we have proposed here provides a more stable foundation for dating Greek Ecclesiastes, since we have taken into account the text history of the book of Ecclesiastes itself combined with the external evidence of the Syriac Version; thus, dateable sources form the basis of this proposal. Further, see James K. Aitken, “Phonological Phenomena in Greek Papyri and Inscriptions and Their Significance for the Septuagint” in Studies in the Greek Bible: Essays in Honor of Francis T. Gignac, S.J., eds. J. Corley and V. Skemp (CBQ Monograph Series 44. Washington: The Catholic Biblical Association of America, 2008). 126  Goldman makes a conjecture that this reading is original on p. 68* on the basis of 10:13 and because of the singular nouns in the immediate context. Of course, if MΤ has a singular noun, Goldman has no reason to emend the text.

252

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

dependent witnesses to the Hebrew text. In the case of 𝔖, it consulted an erroneous text, which obtained from an early inner-Greek error. In the case of the Latin versions, Jerome made use of Aquila’s revision in an attempt to remain faithful to the Hebrew text. ‫הֹוללֹות‬ ֵ is the original reading according to the external evidence.

5.2  Internal Evidence

MT has the lectio difficilior, and the original ‫ֹות‬- termination probably caused the problems present in the transmission of the text as evidenced by the Versions. OG translates the word with a singular in all the instances except 1:17. That the Greek translator used both the singular and plural may indicate that the form in his Vorlage was difficult. There is some difficulty in deciding whether Aquila read ‫ֹות‬- as a singular or a plural in 2:12, though the best interpretation of the evidence is to conclude that Aquila read the form as a plural in all the occurrences in Ecclesiastes.127 Symmachus translates ‫הֹוללֹות‬ ֵ with a variety of lexemes in the singular.128 Theodotion translates the Hebrew with a plural form of the noun.129 T probably read the Hebrew as a feminine singular. Jerome and the Old Latin testify to a Hebrew plural form though these sources probably depend on Aquila. 𝔖 uses the singular form to render the Hebrew.130 In summary, Aq OL V Th attest to reading ‫הֹוללֹות‬ ֵ as a plural form, while Symmachus and T attest to reading the same form as a singular. OG and 𝔖 are mixed witnesses which prefer the singular, but in 1:17 both have the plural. This evidence indicates that ‫הֹוללֹות‬ ֵ was read in two different ways, but ‫הֹוללּות‬ ֵ could only be read as a singular noun, and therefore does not account for the readings in the Versions. The later Mediaeval MSS. probably assimilated the reading of MT to harmonise with ‫ ִׂש ְכלּות‬or they simply misread holem waw for shureq.

127  Phillip Marshall, “A Critical Edition of the Hexaplaric Fragments of Ecclesiastes,” (Ph.D. Dissertation The Southern Baptist Theological Seminary, 2007), 83-85. Marshall concludes that Aquila read the Hebrew form as a plural, and the Greek mss., which contain the variant singular form of πλάνη (7:25 and 9:3) may be explained as assimilating their text to the singular number of the lemma of the OG. In 2:12, there is a double attribution (a reading attributed to Symmachus and Aquila). In this verse, Symmachus probably has the singular and Aquila has the same lexeme but not necessarily the singular form. 128  Pace Goldman who thinks that Symmachus reads the singular ‫הֹוללּות‬. ֵ Symmachus has a singular form in 2:12 (πλάνην), 7:25 (ἔννοιαν θορυβώδη; “turbulent/confusing thought”), and 9:3 (αὐθάδιας; “willfulness, stubbornness”). One has to wonder how Symmachus has a Vorlage with the singular (‫)‑ּות‬, if the evidence indicates that ‫ֹות‬- is read as a plural at this time. The latter termination may be read as either plural or singular, and Symmachus chose to read it as a singular. In addition, Symmachus is only revising ΟG, which also has the singular form in these same instances. Thus, ΟG may influence the choice of Symmachus. Therefore, the evidence of Symmachus does not necessarily demand that he read ‫ּות‬- as Goldman surmises (Goldman, 103*). 129  He uses περιφοράς in 1:17 and πλανάς in 2:12. ܵ 130  In 2:12 and 7:25 P reads ‫ܒܪܢܘܬܐ‬ ܼ ‫ܬܥ‬ ܸ In 9:3, P reads the fem. sing. ‫ܦܗܝܘܬܐ‬. ܲ� ‫ܡ‬.

Basic Principles for the Constitution of the Text

253

If ‫הֹוללּות‬ ֵ was the original reading, it would not be possible to account for the plural reading which arises in the transmission of the text unless one adopts Goldman’s suggestion (infra). If ‫הֹוללֹות‬ ֵ is the original reading, one is able to explain how both the singular and plural forms arise in the transmission of the text.

5.3  Goldman’s Conjecture

We must now return to the conjecture of Goldman that the original text is ‫הֹוללּות‬ ֵ ‫וְ ִס ְכלּות‬, and that MT simply preserves an early interpretation of this reading (‫הֹוללֹות‬ ֵ ‫)וְ ִס ְכלּות‬. Goldman argues on the grounds for an intentional interpretational change on the part of MT: 1) He posits that Eccl. 10:13 probably has the original vocalisation (fem. sing ‫ּות‬-) for ‫הֹוללֹות‬. ֵ The plural ending in 2:12 and 7:25 follows the same interpretational approach as 1:17 in MT according to Goldman, since in these three places there was an original hendiadys of singular nouns ending in ‫ּות‬-. However, Goldman does not offer the same explanation of MT’s vocalisation in 9:3 (‫ֹות‬- is paired with a singular adjective ‫)רע‬, ָ though he still claims MT to be secondary in this context. There is no hendiadys in this text for MT to conform the terminations in 9:3 for the sake of assonance. There is also no reason for MT to soften the language in 9:3 since the context is entirely negative. Thus, one would expect MT to leave ‫הֹוללּות‬ ֵ “madness.” This instance is problematic for Goldman’s thesis since there appears to be no reason for MT to change the text from singular to plural. 2) Goldman argues for an interpretive mechanism on the part of MT, which accounts for the “plural” reading. He says, “In effect it was difficult to accept that Qoheleth had ‘applied his heart’ to madness and stupidity as well as to wisdom and knowledge. In MT the pl. was enough to lead the reader to a meaning other than ‘madness.’ Breaking with the preceding sg. of ‫חכמה ודעת‬, the pl. ‫ הוללות‬may refer to attitudes or actions explored by Qoheleth.”131 Goldman does not venture to say what those “attitudes” or “actions” are, but instead he asserts that there is a substantive change in meaning between the two forms and that MT chose one over the other. We will revisit this conclusion below. Goldman also suggests that MT has intentionally changed the second word of the hendiadys from ‫ ִס ְכלּות‬to ‫ׂש ְכלּות‬. ִ The former term means “folly” in its occurrences in Ecclesiastes (2:3, 12, 13; 7:25; 10:1, 13), and OG attests that the meaning of ‫ ִׂש ְכלּות‬is “understanding” (ἐπιστήμη). Thus, Goldman’s suggestion creates a variant reading. It is difficult to decide which reading is original, but that is not the point at present. Rather, what is under examination is Goldman’s putative mechanism of intentional change on the part of MT. He suggests that MT found it difficult for Qoheleth to apply his heart to both wisdom and knowledge, and madness and folly; therefore, MT simplified the text. Rather than adopt Goldman’s suggestion that MT made the text easier intentionally (not a characteristic of MT generally nor specifically in Ecclesiastes), a better way forward would be 131  Goldman, 68*.

254

Methodology to Establish the Earliest Attainable Text of LXX Ecclesiastes

to understand the variant resulting from confusion of sibilants on the part of the scribe (‫ש‬/‫)ס‬. However, there seems to be no reason to conclude that the reading in MT is secondary, even though this word is a hapax legomenon in the Hebrew Bible, but this question remains open. Although Goldman has mounted an interesting case, his conjecture should be rejected for two reasons: 1) He has not adequately considered all of the options for reading ‫הֹוללֹות‬ ֵ in Eccl. 1:17. He has not considered the possibility that this word should probably be read as a feminine singular, and thus Qoheleth would have contained an original hendiadys which agreed in number without the use of the ‫ּות‬- termination. 2) He has not succeeded in providing a sufficient mechanism which explains why MT would make a difficult reading easier. He fails in this respect for three reasons: a) he does not supply the semantic difference between ‫ֹות‬- and ‫ּות‬-, which causes his mechanism to lack explanatory power. These two words have the same meaning and Qoheleth simply contains two allomorphs (‫ּות‬- in 10:13; ‫ֹות‬- elsewhere) which arose due to dialectical interference. b) Goldman has not considered the possibility that if MT has intentionally changed the text from a singular to a plural due to theological reasons, then MT has made the reading more difficult linguistically, since now the hendiadys contains a plural noun joined to a singular noun according to his analysis. c) Regarding ‫ׂש ְכלּות‬, ִ if this word represents a textual variant, Goldman bypasses an easier explanation of an unintentional scribal error due to homophony of sibilants for a much more complicated solution which attempts to ascribe an interpretive motive to MT scribes.

Conclusion

Our research has argued (1) that ‫הֹוללֹות‬ ֵ in Qoheleth 1:17 as preserved in MT is the original reading and that the textual critic cannot plausibly ascertain an earlier reading in this case and (2) that the Versions support this reading in the final analysis. We also evaluated Y.A.P. Goldman’s textual decision and commentary in BHQ, and although Goldman’s proposal remains possible, we concluded that (1) it did not properly commence with the right understanding of the morphology of MT, (2) it did not accurately evaluate the evidence of the Versions generally and the OG and P in particular, and (3) the conjectured reading ‫הֹוללּות‬ ֵ might actually cause more problems linguistically than it would solve theologically.

D.  On the Arrangement of the Edition, Its Sigla and Abbreviations I.  The Introduction The “Einleitung” or Introduction presents and explains those details that are immediately necessary for the user of the Edition. Detailed description on how to use the apparatus and information on the textual history is found above in sections B. and C. Discussion of the Grammatica and Orthographica is given in an appendix.1 They are limited, however, to the treatment of a few orthographical and grammatical phenomena that contribute to the understanding of individual passages. In general, the statements in the appendices of the Editions of Genesis and Ruth also apply to Ecclesiastes. When preparing the Edition, the editor had to decide between variants that are genuine, and variants that merely represent Orthographica (i.e. mistakes in spelling due to confusion of letters or sounds). Only genuine variants are taken up in the First Apparatus.

II.  The Critical Text The principles followed by the Editor to arrive at the critical text are explained above in the section “The Methodology to Establish the Earliest Attainable Text of the Septuagint Ecclesiastes.”

III.  Kopfleiste (List of Witnesses) A C O B-S-68´’-998 L C´’ d k alii verss Indirect Between the reconstructed text of Ecclesiastes and the First Apparatus is a list of all manuscript witnesses and ancient translations that are extant for the text on that page in exactly the same order as they are mentioned in the apparatuses. This saves the reader from having to refer to the Introduction to see whether or not every witness is actually extant for the portion of the text in question. Cf. the following paragraphs on this. A witness whose text is fragmentary will appear in brackets or parentheses, e.g. (818). If the text of a witness begins on the page in question, the siglum of the witness is given in parentheses followed by a hyphen, e.g. (155-). If the text of a witness ends on the page in question, the hyphen precedes the siglum in parentheses, e.g. (‑299). 1  See Peter J. Gentry, ed. Ecclesiastes (Septuaginta Vetus Testamentum Graecum, Vol. 11.2; Göttingen: Vandenhoeck & Ruprecht, 2019), 98-115.

256

On the Arrangement of the Edition, Its Sigla and Abbreviations

Also noted in the Kopfleiste or List of Witnesses are cases where a manuscript is lacking or missing a stretch of text for other reasons. For example, when MS. 125 is missing a folio or page, the notation L –125 appears in the List of Witnesses. This indicates that the entire L-Group (to which MS. 125 belongs) has the text in question except for MS. 125. The abbreviation ‘al’ (for alia manuscripta) designates codices mixti, i.e. Greek manuscripts, that cannot be assigned to any Text Group, and also manuscripts too fragmentary to permit any classification (cf. above B. VIII/IX). The expression ‘verss’ (for versions) designates all translations whose text for the book of Ecclesiastes is extant in full. Translations whose text is not extant in full, such as CPA, will be listed at the end of the Kopfleiste or List of Witnesses.

IV.  Apparatus I The supporting witnesses for a variant are always given in a precise order: a) the majuscule witnesses in alphabetical order, b) the papyri in numerical order, c) the witnesses of the O-MSS., d) the witnesses of the Egyptian text-type of codices B-S and their congeners/descendants (B-S-68´’-998), e) the witnesses of the L-MSS., the witnesses of the C´’-MSS., f) the witnesses of the remaining Text Groups (cf. sections VI and VII above) in alphabetical order, h) the codices mixti, i) the remaining ancient translations in the order described in the Introduction, j) important Greek commentaries in chronological order, k) the other remaining Greek Fathers in alphabetical order. After the = sign are placed the early Printed Editions and sources that as a general rule witness to the parent text of the translator, e.g. MT 𝔔 Pesh Vulg.

V.  Apparatus II The principles underlying the Second Apparatus were already elucidated above in section XI. Unless otherwise specified, variants listed in the Second Apparatus always indicate marginal readings. Readings of the Three Revisors that derive from the text of the commentary of Catena MSS. will be designated with a superscript com. If the commentary can be assigned to a specific church father, the name is also given (e.g. Olcom). Readings within the catena that are not in a commentary are designated with a superscript cat (e.g. 260cat).

VI.  The Colometry or Stichometry of the Edition Colometry or stichometry, especially in poetry, refers to the division of the text into cola or lines. Joseph Ziegler, who began preparation of the Edition, essentially followed the colometry in Rahlfs’ Text. Nonetheless, the colometry of this Edition has been corrected to follow that of Codex Alexandrinus. Felix Albrecht,

The Colometry or Stichometry of the Edition

257

Mitarbeiter at the Septuaginta-Unternehmen who is charged with revision of Duodecim Prophetae edited earlier by Joseph Ziegler, noted a number of places in the Twelve Prophets where Ziegler had made a conjecture, i.e. a reading against all witnesses. When the colometry of the text, however, was corrected to match that of Codex Alexandrinus, Albrecht discovered that the majority reading made good sense and the conjectures were completely unnecessary. There is evidence in Ecclesiastes as well that the stichometry in Codex Alexandrinus is an old tradition. For example, the colometry in A at 817 follows the accents in MT while the stichometry provided by Rahlfs (and notes left by J. Ziegler) does not. In order to correct the stichometry to that of A and maintain a colometric system essentially equal to that of Rahlfs, lines that belong to the previous line in the colometric system of A have been indented in this Edition. This maintains the division into lines in Rahlfs’ Text and allows the reader to note the lineation in A. In the following places this was not possible and the division of the lines in the text was changed: Rahlfs’ Edition 213ab 214cd 224bc 226a-g 31b-2a 311c-d 316a-d 318c-d-19a 322a-d 58ab 815d-f 817b-f

> > > > > > > > > > > >

Gentry’s Edition 213a-c 214cd 224bc 226a-f 31b-2a 311c-d 316a-c 318c-e-19a 322a-e 58ab 815ef 817b-e

By following the colometry in Codex Alexandrinus, readers interested in the reception history of the text can easily observe how it was construed at an earlier time.

E.  Sigla / Symbols and Abbreviations - the MSS. belonging to a group become through “-” related to one another * designates the original reading of a MS. c designates later correctors c1 designates a first corrector c2 designates a second corrector s designates a suppletor of a MS. If there is no note otherwise, there is no * corresponding to an s. mg designates the reading of the margin of a MS., is opposed to txt txt designates the reading of the text of a MS. is opposed to mg ap designates readings in the apparatus of an edition te designates readings in the text of an edition sup lin designates readings between the lines of a MS. ǀ designates line change in a MS. ǁ designates page change in a MS. α̣ a dot under a letter indicates it cannot be read with certainty in the MS. +, add means additions >, om means omissions ◠ indicates an omission due to parablepsis (e.g. through homoioteleuton) (ν) means that the letter is lacking in different MSS., e.g. εστι(ν) [ ] designates attributions, letters, or words to be deleted against the textual tradition or portions of a MS. that cannot be read marked as lacunae: […] or supplied by conjecture: [ωσ]ει. The number of dots corresponds to the number of unreadable letters. ⟨ ⟩ designates attributions, letters, or words inserted against the textual tradtion ↑ refers the reader to App I ↓ refers the reader to App II gr words which stand in the margin of Syh in Greek letters. 𝔊, LXX means the Greek Text (i.e. Septuagint) 𝔎 means the Cairo Genizah Text = Translation of LXX into Byzantine Greek absc al app cf cod(d)

= = = = =

abscissus alia manuscripta apparatus confer codex (-dices)

n p paenult pers plur

= = = = =

nota pagina paenultimus persona pluralis, pluraliter

260

Sigla / Symbols and Abbreviations

corr

= = corr pr m = del = dittog = dubl = ed = eras ex = fin = fort = gen = gr = hab =

haplogr homoiot homoiar ib inc ind init lat leg lib lin lit mend

= = = = = = = = = = = = =

correxit (-xerunt) correctura correxit prima manus delevit dittography doublet edition erasum except finis fortasse genetivus graecus habe(n)t (for positive mention; always means: the witness has the reading of the lemma haplography homoioteleuton homoiarcton ibidem incertus index initium latinus legendum libere (free translation) linea(e) littera(e) mendose

pr praec praef ras (3) rel s ss

scr sec sim

sing spat sq sup ras superscr tr trah ult v var lect verss vid

= praemittit (-mittunt) = praecedens (for text references within the verse) = praefatio = rasura (3 litterarum) = reliqui = sequens, sequentes (c page numbers: s = the following page, ss = the following pages = scripsit = secundum = similiter (similar translation, which suggests a variant of the LXX as a template = singularis, singulariter = spatium = sequens (for text references within the verse) = supra rasuram = superscripsit = transponit (-nunt) = trahit (-hunt) = ultimus = versus = vario lectio = versions = ut videtur

Recensions and Manuscripts Textual witnesses are arranged in the following order: A C 818 870 969 992 3010 O B-S-68´’-998 L C´’ d k al verss patres The A Text: A­C The Hexaplaric recension: O = V­253­475­637

Sigla / Symbols and Abbreviations

261

The Egyptian Text-type: B-S-68´’-998 68´’ = 68´ = 68’ = 534´ =

68-534-613 68-534 68-613 534-613

The Lucianic Recension: L = 106-125´-130-545 125´ = 125-261 The Catena Group: C

= 139-147-159-299(fin 729)-390-415-503-504-522-540-560­563­5 71-574­732­798 cI = 157´-425-601-609 157´ = 157-797 cII = 260­295(ab 312)­371­561­752(ab 312) C´’ = C + cI + cII C´ = C + cI C ’ = C + cII The text families d and k: d = 254´-342(fin 729)­357 254´ = 254-754 k = 46´-631 46´ = 46-631 Codices mixti (al = alia manuscripta): 155 248´’ 296´ 311 336´ 338 339 359 411 443 539 542 543 547 549 645 698 705(fin 312) 706 766 770(fin 118) 788 795 3011 248´’ = 248´ = 296´ = 336´ = 766 =

161-248-252 161-248 296-548 336-728 766A ­766B The Secondary Translations (verss = versiones)

La (La94 95 160 Hi) Syh

= Old Latin Translation = Syro-Hexapla Translation

262

Sigla / Symbols and Abbreviations

CPA Co Sa (SaI II 1 2 3 4 5 6) Fa (Fa1 2 3) Aeth (Aethte A2 ap ap(A2))  Geo (GeoO J S) Arm Arab

= = = = = = = =

Christian Palestinian Aramaic Translation Coptic Translation (Sa Fa) Sahidic Translation Fayyumic Translation Ethiopic Translation Old Georgian Translation Armenian Translation Arabic Translation

The Indirect Tradition (patres) See above pp. ??-?? (Greek Fathers) and pp. ??-?? (Latin Fathers) The Hebrew Text 𝔐 𝔔 K Q

Masoretic Text (also MT) Hebrew Text from Qumran means Ketib means Qere Translations Based Upon the Hebrew Text

𝔖 = Peshitta 𝔗 = Targum 𝔙 = Vulgate The Later Greek Translators / Revisors α´ = Ἀκύλας (Aquila) σ´ = Σύμμαχος (Symmachus) θ´ = Θεοδοτίων (Theodotion) γρ´ = γράφεται οἱ λ ´ = οἱ λοιποί (the Rest: i.e. α´, σ´ and / or θ´) (οἱ) ο´ = (οἱ) ἑβδομήκοντα (the Seventy) π´ = οἱ πάντες ἄλλος aut ἄλλως aut ἄλλοι aut ἕτερος = unknown source(s) Early Printed Editions Ald = Compl = Sixt =

Aldina (Venice 1518) Complutensis (Alcalá 1520, printed 1514-1517) Sixtina (Rome 1586/87)

Sigla / Symbols and Abbreviations

Gra H.­P. Ra

= = =

263

Grabe (Oxford 1709) Holmes-­Parsons (Oxford 1827) Rahlfs (Stuttgart 1935) Abbreviations for Periodicals and Series1

AGWG.PH = Abhandlungen der Königlichen Gesellschaft der Wissenschaften zu Göttingen, Phil.­hist. Klasse AraSt = Aramaic Studies ASEs = Annali di Storia dell’Esegesi Aug. = Augustinianum BHQ = Biblia Hebraica Quinta Bib. = Biblica BJRL = Bulletin of the John Rylands Library Br.­M. = The Old Testament in Greek according to the text of Codex Vaticanus ed. by A. E. Brooke, N. McLean and H. St. J. Thackeray, Cambridge 1906-1940 Byz. = Byzantion CBQ = Catholic Biblical Quarterly CC = Corpus Christianorum CCCM = Corpus Christianorum—Continuatio Mediaevalis CCSG = Corpus Christianorum—Series Graeca CCSL = Corpus Christianorum—Series Latina CHB = Cambridge History of the Bible ConBOT = Coniectanea biblica: Old Testament Series CSCO = Corpus Scriptorum Christianorum Orientalium. Scriptores Syri CSEL = Corpus Scriptorum Ecclesiasticorum Latinorum DJD = Discoveries in the Judaean Desert GCS = Die griechischen christlichen Schriftsteller der ersten (drei) Jahrhunderte (Berliner Ausgabe) GOF.S = Göttinger Orientforschungen, Reihe 1, Syriaca Gord. = R. Gordis, “Ecclesiastes 1:17—Its Text and Interpretation,” JBL 56 (1937): 323-330 H.-P. = Vetus Testamentum Graecum cum variis lectioni bus, ed. R. Holmes et J. Parsons, tom.V, Oxonii 1827 JAWG = Jahrbuch der Akademie der Wissenschaften zu Göttingen JBL = Journal of Biblical Literature 1  Cf. S. M. Schwertner, IATG3—Internationales Abkürzungsverzeich­nis für Theologie und Grenzgebiete. Zeitschriften, Serien, Lexika, Quellenwerke mit bibliographischen Angaben, 3., überarbeitete und erweiterte Auflage, Berlin / Boston 2014.

264 Kl. LIE Mayser

Sigla / Symbols and Abbreviations

= Kleronomia = Lessico Intellettuale Europeo = E. Mayser, Grammatik der griechischen Papyri aus der Ptolemäerzeit. I 1 (2. Aufl. bearb. von H. Schmoll) Berlin 1970. I 2 u. 3 (2 Aufl.) 1938, 1936. II 1 1926. II 2 1933/34. II 3 1934. MO = Monde Oriental MPSW = Mitteilungen aus der Papyrussammlung der Nationalbibliothek in Wien MSP = Monumenta Sacra et Profana MSU = Mitteilungen des Septuaginta­Unternehmens NGWG.PH =  Nachrichten der Königlichen Gesellschaft der Wissenschaften zu Göttingen, Phil.­hist. Klasse OBO = Orbis Biblicus et Orientalis OrChr = Oriens Christianus PG = Patrologiae cursus completus. Accurante J.­P. Migne Series Graeca PL = Patrologiae cursus completus. Accurante J.­P. Migne Series Latina PLS = Patrologiae cursus completus. Accurante J.­P. Migne Series Latina. Supplementum PSI = Pubblicazioni della Società Italiana PTA = Papyrologische Texte und Abhandlungen PTS = Patristische Texte und Studien RdQ = Revue de Qumran REJ = Revue des études juives RivBib = Rivista Biblica SC = Sources Chrétiennes SeL = Storia e Letteratura StPatr = Studia Patristica StPP = Studien zur Palaeographie und Papyruskunde StT = Studi e Testi Thack. = H. St. J. Thackeray, A Grammar of the Old Testament in Greek according to the Septuagint, vol. I, Cambridge 1909 THGE = Text History of the Greek Ecclesiastes ThSLG = Theologische Studien der Leo­Gesellschaft TSAJ = Texte und Studien zum antiken Judentum TU = Texte und Untersuchungen zur Geschichte der altchristlichen Literatur UMS.H = University of Michigan Studies, Humanistic Series VT = Vetus Testamentum

Sigla / Symbols and Abbreviations

Walters ZAW

265

= P. Walters, The Text of the Septuagint. Its Corruptions and Their Emendation. By the late Peter Walters (formerly Katz), Edited by D. W. Gooding, Cambridge 1973. = Zeitschrift für die alttestamentliche Wissenschaft Select Secondary Literature2

Aitken, James K. “Phonological Phenomena in Greek Papyri and Inscriptions and Their Significance for the Septuagint” in Studies in the Greek Bible: Essays in Honor of Francis T. Gignac, S.J. J. Corley and V. Skemp, eds. CBQ Monograph Series 44. Washington: The Catholic Biblical Association of America, 2008. Albrecht, F. Codex Ephraemi Syri rescriptus. Neue Lesarten zum Septuagintatext des Koheletbuches, ZAW 122 (2010), 272-79. Barthélemy, D. Les Devanciers d’Aquila, Supplements to Vetus Testamentum 10, Leiden 1963. Cannon, W. W. Jerome and Symmachus: Some Points in the Vulgate Translation of Koheleth, ZAW 4 (1927), 191-199. de Waard, J. The Translator and Textual Criticism (with Particular Reference to Eccl 2,25), Bib. 60 (1979), 509-29. Euringer, Sebastian. Der Masorahtext des Koheleth kritisch untersucht. Leipzig: J. C. Hinrichs’schen Buchhandlung, 1890. Géhin, Paul. Évagre le Pontique: Scholies a l’Ecclésiaste. Sources Chrétiennes 397. Paris, 1993. Gentry, P. J. “Origen’s Hexapla,” in A. G. Salvesen and T. M. Law (eds), The Oxford Handbook of the Septuagint. Oxford: OUP, 2021, 553-571. Gentry, P. J. The Role of the “Three” in the Text History of the Septuagint. II. Aspects of Interdependence of the Old Greek and the Three in Ecclesiastes,” Aramaic Studies, 4.2 (2006), 153-192. Gentry, P. J. and F. Albrecht, The Amazing History of MS Rahlfs 159—Insights from Editing LXX Ecclesiastes, in JSCS 44 (2011), 31-50. Ginsburg, C. D. The Writings. London: British and Foreign Bible Society, 1926. Hagedorn, Ursula and Dieter. Die Älteren griechischen Katenen zum Buch Hiob, Bd. I. Berlin: de Gruyter, 1994. Klostermann, Erich. De libri Coheleth versione Alexandrina. Diss. Kiel, 1892. Klostermann, Erich. Analecta zur Septuaginta, Hexapla und Patristik. Leipzig: A. Deichert’sche Verlagsbuchhandlung, 1895. Leanza, Sandro. Le tre versioni geronimiane dell’Ecclesiaste, Annali di storia dell’esegesi 4 (1987), 87-108. McNeile, A. H. An Introduction to Ecclesiastes. London: Cambridge University Press, 1904. Meade, J. D. and P. J. Gentry, “Evaluating Evaluations:The Commentary of BHQ and the Problem of ‫ ֹהולֹלות‬in Ecclesiastes 1:17, in: G. Bonney/R. Vicent (eds.), Sophia—Paideia: Sapienza

2  Cf. also the bibliography listed at the beginning of each section under “Literature.”

266

Sigla / Symbols and Abbreviations

e Educazione (Sir 1,27): Miscellanea di Studi offerti in onore del prof. Don Mario Cimosa, Nuova Biblioteca di Scienze Religiose 34, Rome 2012, 197-217. Mercati, G. Nuove Note di Letteratura Biblica e Cristiana Antica. Studi e Testi 95. Città del Vaticano: Biblioteca Apostolica Vaticana, 1941. Reider, J. An Index to Aquila. Completed and Revised by N. Turner. Supplements to Vetus Testamentum 12. Leiden: Brill, 1966. Salters, R. B. Notes on the History of the Interpretation of Koh 5 5, ZAW 90 (1978), 95-101. Salters, R. B. The Book of Ecclesiastes: Studies in the Versions and the History of Exegesis. Diss. St. Andrews, 1973. Strothmann, Werner. Konkordanz des syrischen Koheletbuches nach der Peshitta und der Syrohexapla. GOF I.4. Wiesbaden: Harrassowitz, 1973. Strothmann, Werner. Das syrische Fragment des Ecclesiastes-Kommentar von Theodor von Mopsuestia: Syrischer Text mit vollständigem Wörterverzeichnis. GOF I.28. Wiesbaden: Harrassowitz, 1988. Strothmann, Werner. Syrische Katenen aus dem Ecclesiastes-Kommentar des Theodor von Mopsuestia: Syrischer Text mit vollständigem Wörterverzeichnis. GOF I.29. Wiesbaden: Harrassowitz, 1988. Strothmann, Werner. Kohelet-Kommentar des Johannes von Apamea: Syrischer Text mit vollständigem Wörterverzeichnis. GOF I.30. Wiesbaden: Harrassowitz, 1988. Strothmann, Werner. Kohelet-Kommentar des Dionysius bar Ṣali-bi-: Auslegung des Septuaginta-Textes. GOF I.31. Wiesbaden: Harrassowitz, 1988. Wevers, John William. A Note on Scribal Error, Canadian Journal of Linguistics 17 (1972), 185-190. Ziegler, Joseph. “Der Gebrauch des Artikels in der Septuaginta des Ecclesiastes” in Studien zur Septuaginta–Robert Hanhart zu Ehren. Ed. D. Fraenkel, U. Quast u. John Wm Wevers. MSU XX. Göttingen: Vandenhoeck & Ruprecht, 1990. Ziegler, Joseph. ‘Hat Lukian den griechischen Sirach rezensiert?’ Biblica 40 (1959), 210-229.

F.  Differences Compared to Rahlfs 1.11c

αὐτοῖς] αὐτῶν

1.13c

γινομένων] γενομένων

1.15a

τοῦ] om

1.15b

τοῦ] om

1.17b

παραβολάς] παραφοράς

2.3a

κατεσκεψάμην] pr καί

2.3a

ἐν καρδίᾳ μου] εἰ ἡ καρδία μου

2.3b

τοῦ ἑλκύσαι εἰς] ἑλκύσει ως

2.10b

οὐχ ὑφεῖλον] οὐκ ἀφεῖλον

2.12d

τὰ ὅσα] pr σύν

2.12d

ἐποίησεν] ἐποίησαν

2.15f

διότι ἄφρων ἐκ περισσεύματος λαλεῖ] om

2.16b

αἰῶνα] pr τόν

2.16c

αἱ ἡμέραι αἱ ἐρχόμεναι] ταῖς ἡμέραις ταῖς ἐρχομέναις

2.18c

γινομένῳ] γενησομένῳ

2.19b

ἐξουσιάζεται] ἐξουσιάσεται

2.20a

τῇ καρδίᾳ] τὴν καρδίαν

2.20b

τῷ] om

2.21a

οὗ] ὅ τι

2.21c

ὅς] ᾧ

268

Differences Compared to Rahlfs

2.24b

ὃ πίεται] πίεται

2.24b

ὃ δείξει] δείξει

2.25

φείσεται] πίεται

3.16c

ἀσεβής] εὐσεβής

3.22d

ἐὰν] ἄν

5.3a

καθώς ἂν] om ἄν

5.3d

σὺν ὅσα] σὺ οὖν ὅσα

5.6a

ματαιότητες καὶ λόγοι πολλοί] ματαιοτήτων καὶ λόγων πολλῶν

5.8a

ἐν παντί ἐστι] ἐπὶ παντί ἐστι

5.10c

ἀλλ’ ἢ] ἀρχὴ

5.16a

καὶ πένθει] καὶ ἐν πένθει

6.3c

οὐκ ἐμπλησθήσεται] οὐ πλησθήσεται

6.8a

ὅτι τίς] ὅ τι

6.8b

διότι] διὰ τί

6.11b

τί περισσὸν] τίς περισσεία

7.1b

γενέσεως] γεννήσεως

7:2d

δώσει] δώσει ἀγαθὸν

7.6a

ὅτι ὡς] ὡς

7.7b

ἀπόλλυσι] ἀπολλύει

7.14b

καὶ ἐν ἡμέρᾳ κακίας ἰδέ] καὶ ἰδὲ ἐν ἡμέρᾳ κακίας

7.21a

τοὺς λόγους] λόγους

Differences Compared to Rahlfs

7.21a

269

οὓς λαλήσουσιν] οὓς λαλήσουσιν ἀσεβεῖς

7.22 Ra ὅτι πλειστάκις πονηρεύσεταί σε καὶ καθόδους πολλὰς κακώσει καρδίαν σου, ὅπως καί γε σὺ κατηράσω ἑτέρους. Ge ὅτι [πλειστάκις πονηρεύσεταί σε] καί ‹ γε › καθόδους πολλὰς κακώσει καρδία[ν] σου, ὅπως καί γε σὺ κατηράσω ἑτέρους. 7.25d

σκληρίαν] ὀχληρίαν

7.26a

καὶ εὑρίσκω ἐγὼ πικρότερον ὑπὲρ θάνατον] καὶ εὑρίσκω ἐγὼ αὐτήν, καὶ ἐρῶ πικρότερον ὑπὲρ θάνατον

7.26d

δεσμοὶ χεῖρες] δεσμὸς εἰς χεῖρας

7.28b

ἄνθρωπον] καὶ ἄνθρωπον

8.1a

Τίς οἶδεν σοφούς] Τίς ὧδε σοφός

8.8c

τοῦ θανάτου] θανάτου

8.8d

ἐν τῷ πολέμῳ] ἐν ἡμέρᾳ πολέμου

8.10b

ἐπορεύθησαν] καὶ ἐπορεύθησαν

8.12b

αὐτῷ] αὐτῶν

8.15f

ὅσας] ἅς

8.16a

σοφίαν] τὴν σοφίαν

9.2f

ὧς] ὥς

9.2g

ὧς] ὥς

9.9a

ἰδέ] καὶ ἰδέ

9.9b

πάσας ἡμέρας] πάσας τὰς ἡμέρας

9.9d

πάσας ἡμέρας] πάσας τὰς ἡμέρας

270 9.12d

Differences Compared to Rahlfs

οἱ υἱοί] υἱοί

10.12b καταποντιοῦσιν] καταποντίσουσιν 10.14c ὀπίσω] ὅ τι ὀπίσω 10.15a αὐτούς] αὐτόν 10.16c πρωίᾳ] πρωΐᾳ 10.20b ταμιείοις] ταμείοις 10:20d ὁ ἔχων τὰς πτέρυγας ἀπαγγελεῖ λόγον] ὁ τὰς πτέρυγας ἔχων ἀπαγγελεῖ λόγον σου 11.6a

πρωίᾳ] πρωΐᾳ

12.6b

ἀνθέμιον] τὸ ἀνθέμιον

12.9b

λαόν] ἄνθρωπον

12.11c συναγμάτων] συνθεμάτων 12.12a ποιῆσαι] τοῦ ποιῆσαι

G.  Corrections to the Published Edition Correction to Einleitung, p. 4: 260 Kopenhagen, Kgl. Bibl., Gamle Kgl. Saml. 6; X. Cent. Eccl. (Cat.) (zu Hs. 260 siehe ausführlich F. Albrecht, Psalmi Salomonis, 13-15).

Corrections to Edition Loc.

Text—July 15, 2019

Correction

1:9b

πεποιημένον,

πεποιημένον;

3:19e

Sc

Sc

5:5

angeli Hi 292,45 294,96 = 𝔙 ↓

angeli Hi 292,45 294,96 = 𝔐 𝔙 ↓

6:5b

αναπαυσεις A

αναπαυσεις A O–475 637

7:22b

καρδια B-68-998* Fa1 SaI (sed hab Arm

καρδια B-68-998* SaI (sed hab Fa2 Arm

9:9b

ματαιότητός 1°] νεοτητος C–298:

ματαιότητός 1°] νεοτητος C–(298):

9:9d

> A 475-637 Bc-68 797txt cII–260 d k 338

> A 475-637 Bc-68 797txt cII–260 d 338

11:6b

στοιχήσει αʹ εὐθετεῖ

στοιχήσει] αʹ εὐθετεῖ

12:13c

Jb.Cat Β 33

Jb.Cat A 10

O–637

Reconsidering Eccl 7:26 Published Version 26

καὶ εὑρίσκω ἐγὼ αὐτήν, καὶ ἐρῶ πικρότερον ὑπὲρ θάνατον, σὺν τὴν γυναῖκα, ἥτις ἐστὶν θηρεύματα καὶ σαγῆναι καρδία αὐτῆς, δεσμὸς εἰς χεῖρας αὐτῆς· ἀγαθὸς πρὸ προσώπου τοῦ θεοῦ ἐξαιρεθήσεται ἀπ᾽ αὐτῆς, καὶ ἁμαρτάνων συλλημφθήσεται ἐν αὐτῇ.

26 ἐγὼ / αὐτήν] tr 298 | ἐγώ] καγω 296; > 147-159-299-503-560-797* 357 AGeo Fa1 2 | αὐτήν] sub ÷ et «Origenes» sup lin Syh; > Hi = Ra 𝔐; + και ευφροσυνη (-νην Sc 754; αφροσυνη 475) πλανα (-ας Sc 475) O Sc d–357 ↓ | καὶ ἐρῶ B-S*-68-998 357 (+ εγω) 698 PsChr Spec 532,8 Arm] και (+ εγω SaI) ειπον O Sc (ειπα) d–357 SaI II; > rel (absc Fa1) = 𝔐 | πικρότερον] πικροτέραν C L C–298 797 k 248 296 311 339 411 443 542 547 698

272

Corrections to the Published Edition

706 766 795 Ol Metlem et com VII.4,5,51 Damap (sed hab Did 227,1 PsChr Anast 628 684 Antioch 1485 Dam)

App 2: πικρότερον] αʹ σʹ θʹ ὁμοίως Syh (= πικρότεραν) Proposed Correction 26

καὶ εὑρίσκω ἐγὼ αὐτήν πικροτέραν ὑπὲρ θάνατον, σὺν τὴν γυναῖκα, ἥτις ἐστὶν θηρεύματα καὶ σαγῆναι καρδία αὐτῆς, δεσμὸς εἰς χεῖρας αὐτῆς· ἀγαθὸς πρὸ προσώπου τοῦ θεοῦ ἐξαιρεθήσεται ἀπ᾽ αὐτῆς, καὶ ἁμαρτάνων συλλημφθήσεται ἐν αὐτῇ.

26 ἐγὼ / αὐτήν] tr 298 | ἐγώ] καγω 296; > 147-159-299-503-560-797* 357 AGeo Fa1 2 | αὐτήν] sub ÷ et «Origenes» sup lin Syh; > Hi = Ra 𝔐; + καὶ ἐρῶ B-S*-68-998 357 (+ εγω) 698 PsChr Spec 532,8 Arm (contra 𝔐); + και (+ εγω SaI) ειπον O Sc (ειπα) d–357 SaI II (absc Fa1) (contra 𝔐); + και ευφροσυνη (-νην Sc 754; αφροσυνη 475) πλανα (-ας Sc 475) O Sc d–357 ↓ | πικροτέραν C L C–797 k 248 296 311 339 411 443 542 547 698 706 766 788 795 Ol Metlem et com VII.4,5,51 Damap Ald] πικρότερον A O B-S*-68-998 d 252 336 338 543 549 645 Did 227,1 PsChr Anast 628 684 Antioch 1485 Dam

App 2: πικρότερον] αʹ σʹ θʹ ὁμοίως Syh (= πικρότεραν) Analysis Rahlfs omitted αὐτήν because it is not in MT and Hi. He believed, erroneously, that Hi was the Old Latin pure and simple, whereas we know that the Bible Text of the Ecclesiastes Commentary was corrected by Jerome towards the Hebrew. All of our witnesses attest this αὐτήν. So we do not have evidence from the manuscripts that it is secondary. It may have been inserted by the translator to anticipate the direct object which follows πικροτέραν ὑπὲρ θάνατον later on in the text. Note that the direct object is also preceded by σύν so the insertion clarifies the syntax. The additions καὶ ἐρῶ, Early Egyptian Group, and καὶ εἶπον, O and Late Egyptian Group, look now suspiciously like a correction designed along with πικρότερον to make sure that the reader does not connect the comparative phrase necessarily with “the woman.” It is difficult to think that additions such as these were derived from the translator. Witnesses that do not have either καὶ ἐρῶ or καὶ εἶπον are as follows: A C L C–797 k 248 296 311 336 338 339 359 411 443 539 542 543 547 549 645 698 706 766 788 795. When I began to establish the critical

Corrections to the Published Edition

273

text, I was overly influenced by B-S and 998 and had not given consideration to the fact of the Hesychian recension. In any case, there is a mistake in the Edition since 788, an extremely important manuscript, supports πικρότεραν and 298 does as well. Rahlfs thought πικρότερον was original, but remember that he omitted αὐτήν.

H. Appendices

1.  Computer Generated Stemma for Ecclesiastes Genealogical Stemma of Witnesses of the Greek Ecclesiastes This appendix contains the stemma of the genealogical history of the Greek translation of the Book of Ecclesiastes. The stemma was generated by a computer program developed by Dr. James D. Price. The computer program is based on principles of a genealogical theory of textual criticism. For detailed description of how the computer program produces its results see https://www.jamesdprice. com/. Here only brief instructions are provided on how to read the stemma. First, genealogical relations between manuscripts and witnesses used for the Edition were determined without the aid of computer, i.e. by patiently assembling the lists provided here in THGE. It was remarkable to discover, after determining the text groups the old way, that the computer program generated the precisely same text groups. Perhaps the computer program requires minor refinements, but the results were remarkably similar to the earlier results. The stemma is displayed vertically rather than horizontally. That is, the autograph in the upper left corner with succeeding generations indented from the left progressively downward. Sibling daughter descendants are linked by vertical lines. For example, the first-generation descendants of the autograph are Ex312#,1 Ex-316#,2 and Ex-315#. Only the primary exemplars are displayed, so no mixture connections are shown. The diagram spills over onto succeeding pages, but the lower-case letters at the page breaks show where the lines from one page connect to those of the next. The head exemplar of each of the text-groups is identified at its right. The format of the information on each line is as follows: (1) the name of the witness; (2) the genealogical affinity of the witness with its primary parent exemplar, enclosed in square brackets []; (3) generation from the autograph, enclosed in angular brackets ; (4) date, enclosed in curly brackets {}; (5) the number of variants the witness differs from its primary parent, enclosed in slant marks //; (6) the number of readings in the sibling gene, also enclosed in slant marks //; and (7) the number of parents the witness has. 1  The names of exemplars created by the software have the prefix “Ex-” followed by a number; extant witnesses have the numbers or sigla employed in the Edition. 2  The character # at the end of the name of a witness indicates that its place in the tree was determined numeric affinity without the benefit of genetic affinity. The character ‘%’ indicates the witness had less than 60% of the text, and an additional ‘!’ indicates that it had less than 20%.

276

Appendices

Appendices

Autograph[0.00]{AD 0}/0/0/0 |-Ex-312#[0.99]{AD 250}/22/200/1 | |-818%[0.99]{AD 275}/1/0/1 | |-Ex-207[0.95]{AD 300}/108/6/1 | | |-S^1[1.00]{AD 550}/1/1/1 | | |-S*[0.97]{AD 350}/58/1/1 | | |-S^c[0.98]{AD 1150}/47/1/1 | | |-S^2[1.00]{AD 650}/1/1/1 | |-Ex-285[0.99]{AD 850}/19/5/1 | | |-Ex-284[0.99]{AD 900}/31/1/1 | | | |-311[0.98]{AD 1150}/50/5/1 | | | |-706[0.98]{AD 914}/44/4/1 | | | |-Ex-211[0.95]{AD 1186}/112/10/1 | | | | |-795*[1.00]{AD 1200}/0/6/1 | | | | |-795^c[1.00]{AD 1300}/1/6/1 | | | |-Ex-254[0.94]{AD 1000}/133/11/1 | | | |-548^c[1.00]{AD 1500}/0/23/1 | | | |-548*[0.99]{AD 1400}/12/23/1 | | | |-Ex-240[0.96]{AD 1023}/76/16/1 | | | |-296^tx[1.00]{AD 1050}/3/4/1 | | | |-296^mg[1.00]{AD 1050}/2/3/1 | | |-Ex-277[0.98]{AD 1150}/35/3/1 | | |-248*[0.99]{AD 1250}/27/1/1 | | |-Ex-265[1.00]{AD 1186}/3/1/1 | | |-248^c[1.00]{AD 1350}/8/1/1 | | |-248^mg[0.99]{AD 1250}/13/3/1 | | |-Ex-237[0.93]{AD 1200}/145/2/1 | | | |-547^mg[1.00]{AD 1250}/2/1/1 | | | |-Ex-236[1.00]{AD 1202}/3/1/1 | | | |-547*[1.00]{AD 1250}/5/15/1 | | | |-547^c[1.00]{AD 1350}/0/15/1 | | |-Ex-257[0.99]{AD 1286}/16/1/1 | | |-161^c[1.00]{AD 1450}/6/1/1 | | |-161*[0.99]{AD 1350}/11/1/1 | | |-161^mg[0.99]{AD 1350}/14/1/1 | | |-Ex-217[0.94]{AD 1300}/126/4/1 | | |-338*[1.00]{AD 1350}/1/2/1 | | |-338^c[1.00]{AD 1450}/1/2/1 | |-Ex-283[0.98]{AD 1050}/40/8/1 | |-Ex-224[0.92]{AD 1202}/173/5/1 | | |-645^c[1.00]{AD 1350}/2/43/1 | | |-645*[0.99]{AD 1250}/20/45/1

277

278

Appendices

| | |-645^mg[1.00]{AD 1250}/1/43/1 | |-Ex-274[0.90]{AD 1100}/214/15/1 | | |-766^Ic[1.00]{AD 1250}/2/47/1 | | |-766^I[0.99]{AD 1150}/19/47/1 | | |-766^II[0.95]{AD 1150}/97/54/1 | |-Ex-260[0.94]{AD 1200}/131/2/1 | | |-602[0.99]{AD 1450}/17/7/1 | | |-Ex-238[0.99]{AD 1202}/19/6/1 | | |-613*[1.00]{AD 1250}/4/21/1 | | |-613^c[1.00]{AD 1350}/1/21/1 a b c

Ba

279

Appendices

a b | |-Ex-272[0.89]{AD 1075}/229/12/1 | | |-728[0.97]{AD 1100}/58/13/1 | | |-Ex-241[0.97]{AD 1300}/61/11/1 | | |-336*[1.00]{AD 1350}/4/74/1 | | |-336^c[1.00]{AD 1350}/2/74/1 | |-Ex-270[0.90]{AD 1250}/221/12/1 | | |-Ex-230[0.99]{AD 1300}/16/17/1 | | | |-631*[1.00]{AD 1350}/3/3/1 | | | |-631^c[1.00]{AD 1450}/0/3/1 | | |-Ex-262[1.00]{AD 1286}/8/16/1 | | |-46^s[1.00]{AD 1400}/6/13/1 | | |-46*[1.00]{AD 1300}/6/8/1 | | |-Ex-248[0.98]{AD 1300}/41/5/1 | | |-337*[1.00]{AD 1350}/5/2/1 | | |-337^c[1.00]{AD 1450}/2/1/1 | |-Ex-279[0.93]{AD 1100}/149/11/1 | | |-Ex-247[0.96]{AD 1286}/96/26/1 | | | |-261*[1.00]{AD 1300}/7/7/1 | | | |-261^c[1.00]{AD 1400}/2/7/1 | | | |-125%[0.94]{AD 1350}/79/0/1 | | |-Ex-273[0.99]{AD 1202}/17/21/1 | | | |-545*[0.99]{AD 1250}/11/11/1 | | | |-545^c[0.99]{AD 1350}/20/10/1 | | | |-545^mg[1.00]{AD 1250}/4/7/1 | | |-Ex-271[0.99]{AD 1150}/20/18/1 | | |-Ex-210[0.99]{AD 1186}/27/4/1 | | | |-130*[1.00]{AD 1200}/1/2/1 | | | |-130^c[1.00]{AD 1300}/0/2/1 | | |-Ex-216[0.99]{AD 1300}/28/4/1 | | |-106^c[1.00]{AD 1450}/0/7/1 | | |-106*[1.00]{AD 1350}/5/7/1 | | |-106^mg[1.00]{AD 1350}/2/7/1 | |-Ex-282[0.97]{AD 1186}/64/1/1 | |-Ex-239[0.94]{AD 1450}/131/6/1 | | |-411^c[1.00]{AD 1650}/1/3/1 | | |-Ex-215[1.00]{AD 1500}/4/3/1 | | |-411*[1.00]{AD 1550}/0/21/1 | | |-411^mg[1.00]{AD 1650}/2/21/1 | |-Ex-259[0.96]{AD 1200}/84/1/1 | |-Ex-227[0.99]{AD 1500}/18/1/1 | | |-371*[1.00]{AD 1550}/3/3/1

k

L

Oa

cIIa

280

Appendices

| | |-371^c[1.00]{AD 1650}/0/3/1 | |-Ex-258[1.00]{AD 1202}/6/1/1 | |-295[0.99]{AD 1450}/10/1/1 | |-561[0.99]{AD 1250}/11/2/1 | |-Ex-219[0.96]{AD 1500}/79/2/1 | |-752*[1.00]{AD 1550}/1/5/1 | |-752^c[1.00]{AD 1650}/1/5/1 |-Ex-315#[0.98]{AD 250}/36/200/1 | |-Ex-263[0.95]{AD 275}/101/28/1 | | |-Ex-225[0.98]{AD 350}/40/6/1 | | | |-68*[1.00]{AD 1450}/3/8/1 | | | |-68^c[1.00]{AD 1550}/0/8/1 | | | |-357^s%[0.86]{AD 1400}/4/0/1 a b c d

Bb

281

Appendices

a b c d | | | |-870%[0.90]{AD 400}/5/0/1 | | |-Ex-221[0.99]{AD 300}/24/9/1 | | |-B^1[1.00]{AD 350}/1/1/1 | | |-B*[0.99]{AD 350}/13/1/1 | | |-B^c[1.00]{AD 600}/2/1/1 | | |-B^mg%[0.99]{AD 350}/5/0/1 | |-Ex-269[0.96]{AD 750}/88/28/1 | | |-125^II[0.99]{AD 1350}/22/4/1 | | |-Ex-228[0.99]{AD 800}/17/3/1 | | |-542^tx[1.00]{AD 850}/3/8/1 | | |-542^mg[1.00]{AD 850}/0/8/1 | |-Ex-305[0.96]{AD 275}/80/25/1 | | |-998*[0.94]{AD 300}/123/22/1 | | |-998^c[0.95]{AD 400}/113/20/1 | | |-Sa^I[0.88]{AD 950}/232/20/1 | | |-Ex-245[0.91]{AD 1150}/186/10/1 | | |-543*[1.00]{AD 1186}/6/3/1 | | |-543^c[1.00]{AD 1286}/2/2/1 | |-Ex-311[0.98]{AD 275}/45/12/1 | |-Dam[0.92]{AD 700}/169/13/1 | |-Did^lem[0.94]{AD 350}/97/8/1 | |-359%[0.95]{AD 1300}/19/0/1 | |-Ex-297[0.94]{AD 914}/137/12/1 | | |-GregAg^lem[0.99]{AD 950}/18/12/1 | | |-GregAg^com[0.99]{AD 950}/22/16/1 | |-Ex-310[0.97]{AD 300}/54/8/1 | |-698[0.92]{AD 1000}/162/9/1 | |-Did^com[0.90]{AD 350}/151/15/1 | |-Ex-306[0.91]{AD 850}/199/24/1 | | |-OL^E[0.95]{AD 900}/102/31/1 | | |-OL^Ecom[0.94]{AD 900}/121/31/1 | | |-Ex-294[0.94]{AD 914}/119/12/1 | | |-260*[0.99]{AD 1000}/15/4/1 | | |-Ex-286[1.00]{AD 950}/6/6/1 | | |-260^c[1.00]{AD 1100}/0/7/1 | | |-149[0.99]{AD 1050}/17/17/1 | | |-260^mg[0.99]{AD 1000}/11/8/1 | |-Ex-302[0.93]{AD 650}/157/18/1 | |-Ex-229[0.93]{AD 700}/140/40/1 | | |-V*[1.00]{AD 750}/3/5/1 | | |-V^c[1.00]{AD 850}/1/5/1

La cIIb

Oa

282

Appendices

| |-Ex-293[0.96]{AD 950}/91/28/1 | |-475[0.95]{AD 1300}/108/52/1 | |-Ex-292[0.99]{AD 1000}/21/44/1 | |-Ex-232[0.96]{AD 1023}/91/41/1 | | |-253*[1.00]{AD 1050}/4/16/1 | | |-253^c[1.00]{AD 1150}/0/16/1 | |-Ex-204[0.95]{AD 1023}/110/46/1 | |-637^c[1.00]{AD 1150}/0/25/1 | |-637*[1.00]{AD 1050}/2/25/1 | |-637^mg[1.00]{AD 1050}/0/25/1 |-Ex-316#[0.94]{AD 50}/132/200/1 |-Ex-288[0.99]{AD 1050}/13/131/1 | |-Ex-250[0.94]{AD 1500}/124/6/1 a b c

cIa

283

Appendices

a b c | | |-609^c[1.00]{AD 1650}/0/16/1 | | |-609*[1.00]{AD 1550}/8/17/1 | |-Ex-261[0.96]{AD 1286}/80/8/1 | | |-797^c[1.00]{AD 1400}/3/7/1 | | |-797*[1.00]{AD 1300}/9/8/1 | |-Ex-214[0.99]{AD 1100}/22/2/1 | | |-157^c[1.00]{AD 1250}/1/4/1 | | |-157*[1.00]{AD 1150}/6/3/1 | | |-157^mg[1.00]{AD 1150}/0/3/1 | |-Ex-256[0.98]{AD 1075}/42/3/1 | |-571*[0.99]{AD 1550}/28/4/1 | |-Ex-251[0.99]{AD 1500}/19/3/1 | | |-571^c[0.98]{AD 1650}/36/5/1 | | |-571^mg[1.00]{AD 1550}/5/2/1 | |-Ex-246[0.99]{AD 1100}/24/2/1 | |-563*[1.00]{AD 1150}/2/6/1 | |-299[0.92]{AD 1202}/114/5/1 | |-563^c[1.00]{AD 1250}/6/6/1 |-Ex-304[0.98]{AD 650}/35/131/1 | |-390[0.98]{AD 1075}/47/4/1 | |-Ex-206[0.99]{AD 700}/18/1/1 | | |-522*[1.00]{AD 750}/1/1/1 | | |-522^c[1.00]{AD 850}/0/1/1 | |-Ex-253[0.97]{AD 1400}/60/5/1 | | |-601^c[1.00]{AD 1550}/0/3/1 | | |-601*[0.99]{AD 1450}/13/3/1 | |-Ex-205[0.99]{AD 1300}/27/2/1 | | |-425^1[1.00]{AD 1450}/1/5/1 | | |-425*[0.99]{AD 1350}/17/1/1 | | |-425^c[1.00]{AD 1450}/7/1/1 | | |-425^2[1.00]{AD 1550}/1/4/1 | | |-425^mg[0.99]{AD 1350}/12/4/1 | |-Ex-264[0.99]{AD 914}/16/10/1 | |-540^c[1.00]{AD 1350}/0/3/1 | |-540*[0.99]{AD 1250}/11/4/1 | |-Ex-255[0.99]{AD 950}/20/1/1 | |-415[0.99]{AD 1500}/12/3/1 | |-732[1.00]{AD 1000}/9/1/1 | |-Ex-220[1.00]{AD 1500}/3/2/1 | |-504*[1.00]{AD 1550}/3/2/1 | |-504^c[1.00]{AD 1650}/0/2/1

cIb

284

Appendices

| |-574^tx%[0.97]{AD 1550}/18/0/1 | |-574^mg%[0.97]{AD 1550}/16/0/1 |-Ex-301[0.98]{AD 850}/43/130/1 | |-Ex-244[0.99]{AD 1202}/27/4/1 | | |-560*[1.00]{AD 1250}/5/4/1 | | |-560^c[1.00]{AD 1350}/2/3/1 | |-Ex-300[1.00]{AD 900}/2/5/1 | |-Ex-243[0.99]{AD 1100}/22/8/1 | | |-147*[1.00]{AD 1150}/4/2/1 | | |-147^c[1.00]{AD 1250}/4/2/1 | |-Ex-299[1.00]{AD 914}/4/5/1 | |-Ex-208[0.99]{AD 950}/22/4/1 | | |-159^tx[1.00]{AD 1000}/0/2/1 a b c

Appendices

a b c | | |-159^mg[1.00]{AD 1000}/1/2/1 | |-Ex-226[0.99]{AD 1186}/28/2/1 | | |-503*[1.00]{AD 1200}/3/1/1 | | |-503^c[1.00]{AD 1300}/0/1/1 | |-Ex-242[0.99]{AD 950}/19/1/1 | |-139^tx[1.00]{AD 1000}/3/1/1 | |-139^mg[1.00]{AD 1000}/3/1/1 | |-Ex-213[0.98]{AD 1300}/51/3/1 | |-798*[1.00]{AD 1350}/0/2/1 | |-798^c[1.00]{AD 1450}/1/2/1 |-Ex-314[0.93]{AD 100}/159/120/1 |-Ex-313[0.98]{AD 150}/36/12/1 | |-Ex-218[0.97]{AD 300}/59/6/1 | | |-GregNy^lem[1.00]{AD 350}/0/4/1 | | |-GregNy^com[1.00]{AD 350}/2/4/1 | |-Ex-233[1.00]{AD 200}/8/3/1 | | |-Dion^lem[0.98]{AD 250}/29/4/1 | | |-298[0.98]{AD 1150}/35/1/1 | | |-Hi[0.82]{AD 400}/398/19/1 | | |-Chr[0.98]{AD 400}/53/4/1 | | |-Dion^com[0.98]{AD 250}/31/4/1 | | |-969%[0.98]{AD 600}/2/0/1 | |-Ex-235[0.97]{AD 1000}/70/10/1 | |-539*[1.00]{AD 1050}/3/3/1 | |-Ex-234[1.00]{AD 1023}/1/1/1 | |-539^c[1.00]{AD 1150}/3/12/1 | |-539^com[1.00]{AD 1050}/1/12/1 |-Ex-249[0.97]{AD 300}/72/9/1 | |-252^c[1.00]{AD 1050}/3/5/1 | |-252*[0.99]{AD 950}/11/6/1 | |-252^mg[1.00]{AD 950}/6/5/1 | |-Ex-223[0.98]{AD 350}/51/1/1 | |-549*[0.99]{AD 1050}/26/3/1 | |-339[0.97]{AD 1050}/69/4/1 | |-549^c[0.99]{AD 1150}/27/3/1 | |-788[0.97]{AD 950}/72/1/1 | |-155*%[0.92]{AD 1050}/75/0/1 | |-155^c%[0.92]{AD 1150}/62/0/1 | |-Ex-212[0.96]{AD 400}/80/3/1 | |-C*[1.00]{AD 450}/1/3/1 | |-C^c[1.00]{AD 550}/0/3/1

285

286

Appendices

| |-705%[0.83]{AD 1300}/22/0/1 |-Ex-308[0.97]{AD 300}/54/14/1 | |-Aeth[0.95]{AD 800}/108/9/1 | |-Syh^tx[0.93]{AD 850}/146/18/1 | |-Syh^mg[0.93]{AD 850}/151/19/1 | |-Ex-289[0.93]{AD 914}/145/14/1 | | |-443^c[1.00]{AD 1050}/9/15/1 | | |-443*[0.99]{AD 950}/15/16/1 | |-Ex-307[0.95]{AD 350}/116/21/1 | |-Sa^II[0.89]{AD 600}/231/5/1 | |-Ex-209[1.00]{AD 400}/0/22/1 | |-A*[1.00]{AD 450}/1/4/1 | |-A^c[1.00]{AD 550}/0/4/1 a b

Appendices

a |-Ex-309[0.97]{AD 550}/68/16/1 |-Arm[0.90]{AD 600}/219/19/1 |-Ex-231[0.94]{AD 1023}/127/12/1 Bc | |-534*[1.00]{AD 1050}/1/4/1 | |-534^c[1.00]{AD 1150}/3/4/1 |-Ex-291[0.95]{AD 900}/115/15/1 d | |-357[0.89]{AD 1150}/239/39/1 | |-Ex-278[0.99]{AD 1023}/18/18/1 | | |-342[0.99]{AD 1250}/16/11/1 | | |-754[0.99]{AD 1050}/19/19/1 | |-Ex-222[0.99]{AD 914}/17/17/1 | |-254*[1.00]{AD 950}/3/4/1 | |-254^c[1.00]{AD 1050}/0/3/1 |-Ex-303[0.93]{AD 900}/149/12/1 OL |-Ex-290[0.97]{AD 1000}/56/5/1 | |-OL^M[1.00]{AD 1023}/7/12/1 | |-OL^Mcom[0.99]{AD 1023}/24/14/1 | |-Ex-281[0.97]{AD 1186}/55/11/1 | |-Ex-252[0.97]{AD 1550}/59/2/1 | | |-OL^Q[1.00]{AD 1600}/4/2/1 | | |-OL^Qcom[0.99]{AD 1600}/24/2/1 | |-Ex-275[1.00]{AD 1200}/4/4/1 | | |-OL^D[1.00]{AD 1250}/10/1/1 | | |-Ex-266[0.99]{AD 1202}/16/1/1 | | |-OL^I[0.99]{AD 1350}/19/4/1 | | |-OL^K[1.00]{AD 1250}/8/2/1 | |-Ex-276[0.99]{AD 1200}/26/3/1 | |-OL^Dcom[0.99]{AD 1250}/12/3/1 | |-Ex-267[0.99]{AD 1202}/15/2/1 | |-OL^Icom[0.99]{AD 1350}/19/4/1 | |-OL^Kcom[1.00]{AD 1250}/9/2/1 |-Ex-298[0.99]{AD 914}/30/2/1 |-OL^N%[0.95]{AD 1450}/18/0/1 |-OL^Ncom%[0.94]{AD 1450}/20/0/1 |-Ex-296[0.97]{AD 950}/57/7/1 | |-OL^B[1.00]{AD 1000}/7/23/1 | |-OL^Bcom[0.99]{AD 1000}/29/26/1 | |-Ex-268[0.97]{AD 1300}/60/24/1 | |-OL^H[1.00]{AD 1350}/5/2/1 | |-OL^Hcom[0.99]{AD 1350}/25/1/1

287

288

Appendices

|-Ex-295[0.97]{AD 950}/58/12/1 |-OL^Z[1.00]{AD 1000}/8/9/1 |-OL^Zcom[0.99]{AD 1000}/24/19/1 |-Ex-287[0.97]{AD 1075}/74/10/1 |-OL^G[1.00]{AD 1100}/6/9/1 |-OL^Gcom[0.99]{AD 1100}/23/8/1 |-Ex-280[0.94]{AD 1186}/122/15/1 |-OL^A[1.00]{AD 1200}/9/9/1 |-OL^Acom[0.99]{AD 1200}/22/9/1

289

Appendices

2.  Complete List of Patristic Quotations for Ecclesiastes Citations from Greek Fathers (List Prepared by Andrew D. Case) Note: if the quotation in this column is in brackets, it is only for reference. It represents the full line of the lemma and need not be quoted for this table. 1,1a

υἱοῦ Δαβἰδ υἱοῦ Δαβἰδ comma comma υἱοῦ Δαυἰδ comma

1,1b

1,2a

3  PG 22.

ματαιότης ματαιοτήτων ματαιότης ματαιοτήτων comma ματαιότης ματαιοτήτων ματαιότης ματαιοτήτων ματαιότης ματαιοτήτων comma ματαιότης ματαιοτήτων ματαιότης ματαιοτήτων ματαιότης ματαιοτήτων ματαιότης ματαιοτήτων 

Dionlem Dioncom GregNy V Didlem Didcom

210 210 280,21 5,1 6,4 8,13

Dionlem Did

210 5,1 6,4 8,13

CyrHier Eus Gen3 GregNy V

4,8 1128 281,3 283,6 429 8,18 10,10 10,31 12,25 13,10 154,13

GregNy Bapt Didcom Didlem Didcom

290

Appendices

1,2b

[+1,14c]

Clem II Or II Eus VIII Eus VIII Eus Ps4 τὰ πάντα ματαιότης τὰ πάντα ματαιότης ματαιότης ματαιοτήτων ματαιότης ματαιοτήτων comma comma comma comma

1,3a

1,3b ᾧ μοχθεῖ ὑπὸ τὸν ἥλιον ᾧ μοχθεῖ ὑπὸ τὸν ἥλιον comma ἐν παντὶ μόχθῳ μοχθεῖ ὑπὸ τὸν ἥλιον

4  PG 23.

Bas III GregNa I Did Hiob Dionlem Eus VIII GregNy V

τίς περισσεία

1,4a

Eus Ps GregNy V

Didlem Dionlem Eus VIII GregNy V Didlem

Dionlem Or XII GregNy V Didlem Did Ps Did Ps.CatA

385 201,2 2,22 2,199 353 596 449 281,3 295,1 282,11 283,6 291,2 541 780 216,17 210 2,22 291,3 285,13 11,12 12,16 210 2,22 291,3 285,13 11,12 12,16 12,27 210 198 286,25 12,3 56,9 200

291

Appendices

1,4b

Dionlem GregNy V ἡ γῆ εἰς τὸν αἰῶνα comma comma comma

GregNy IV,1 Bas III Didlem Didcom

210 288,15 289,13 925 1445 12,3 12,10 12,20

1,5a

Didlem

12,24 12,30 327,15

1,5b

Didlem

12,24 13,22

1,6a

Didlem

13,18

1,6b

Didlem

12,31 13,18

1,6c

Or II GregNy V

104,17 287,21 288,2 12,31

εἰς τὸν τόπον

πορεύεται τὸ πνεῦμα, comma 1,6d

Didcom Or II GregNy V Didcom

104,17 288,2 12,31 13,5

Didlem Didcom

14,2 14,5 14,17 14,25 14,28 14,15 14,22

1,7b

Didlem

14,2

1,7c

Didlem

14,2

καὶ ἐπὶ κύκλους αὐτοῦ ἐπιστρέφει ἐπὶ κύκλους αὐτοῦ  1,7a

εἰς τὴν θάλασσαν

πάντες εἰς τὴν θάλασσαν οἱ χείμαρροι, πορεύονται εἰς τὴν θάλασσαν

5  Reference is from PG 44.

292

Appendices

1,7d

Didlem ἐπιστρέφουσιν τοῦ πορευθῆναι

Didcom

14,2 15,23 15,7 15,8

1,8a

GregNy V Didlem

291,15 15,26

1,8b

GregNy V

291,15 292,7 293,14 294,4 15,26 16,9

Didlem 1,8c

Didlem Didcom

16,19 16,26 16,31

1,8d

Bas III

464

1,9a

Or I Eus VIII

282,6 2,24 2,22 2,25 294,19 57,6 18,11

[οὐ πλησθήσε]ται [ὀφθαλμὸς] [οὐ π]λησθήσεται

ὃς λ[αλήσει καὶ ἐρεῖ ᾽Ιδὲ τοῦτο καινόν ἐστιν] 1,9b

GregNy V Did Hiob Didcom Or I Eus VIII GregNy V

[] 1,9c

Did Hiob Didcom Eus VIII GregNy V

[τοῖς γενομένοις ἀπὸ] ἔ[μπροσθεν ἡμῶν]

Proc CatP Did Hiob Didcom

282,6 2,24 2,22 2,25 294,20 296,3 57,6 18,11 2,24 2,22 296,20 297,1 34,112 57,6 18,11

293

Appendices

1,10a

Eus VIII GregNy V

2,24 297,1

1,10b

Eus VIII GregNy V

2,24 297,7

1,10c

Eus VIII GregNy V

2,24 297,7

GregNy V Didcom

297,17 21,2 27,1 27,10

1,11b

GregNy V

297,17

1,11c

GregNy V

297,18

Didcom

27,1 27,14

1,12a

GregNy V Didcom

299,10 22,22

1,12b

GregNy V

299,10

1,13a

GregNy V

299,22 300,12 12,166

1,11a

1,11d

1,13b

ὃς λαλήσει 

οὐκ ἔστιν μν[ήμη τοῖς πρώτοις] [οὐκ ἔστ]ιν μνήμη τοῖς πρώτοις οὐκ ἔστιν μ[νήμη]

μετὰ τῶν γενη[σομένων] μετὰ τῶν γενησομένων

ἐκζητῆσαι ἔδωκα τὴν καρδίαν μου

CatH GregNy V

299,23 300,12

1,13c

GregNy V

299,24 300,13

1,13d

GregNy V CatH Didcom

301,6 13,171 82,21 156,9 165,14

1,13e

GregNy V CatH Didcom

301,6 13,171 82,21 156,9 165,14

GregNy V CatH

301,7 13,171

τοῦ κατασκέψασθαι

ἔδωκεν ὁ θεὸς 1,13f

294 1,14a

Appendices

σὺν πάντα comma πάντα τὰ ποιήματα

1,14b

GregNy V Did

303,3 25,9 25,10

GregNy V Did

303,3 25,9 25,10

1,14c [+1,2b] τὰ πάντα ματαιότης καὶ προGregNa I αίρεσις πνεύματος προαίρεσις πνεύματος CatH καὶ ἰδοὺ τὰ πάντα ματαιότης καὶ Did πρ[ο]αίρεσι[ς πνεύματο]ς

780

1,15a

PetA GregNy V Bas I Didlem

473 303,12 245 25,28

Didcom

26,5

PetA GregNy V Didlem

473 304,6 305,8 25,28

Didcom

26,30

GregNy V

307,18 308,1 27,20

Διεστρα[μμένον οὐ δυνήσεται ἐπικο]σμηθῆναι διεστραμμένον 1,15b καὶ ὑστέρημα οὐ [δυνήσετ]αι ἀριθμηθῆναι. ὑστέρημα ... δυνήσεται  1,16a ἐλάλησα ἐγὼ [ἐν καρδίᾳ μου τῷ λέγειν] 1,16b

GregNy V [᾽Εγ]ὼ ἰδοὺ ἐμεγαλύνθην ἐμεγαλύνθην

1,16c

1,16d

Didlem

καὶ προσέθηκα σοφίαν καὶ προσέ[θη]κα σοφίαν ἐπὶ πᾶσιν [οἳ ἐγ]ένοντο ἔμπροσθέν μου ἐν Ιερουσαλημ,

Didlem Didcom

13,177 25,9

307,19 308,2 27,20 29,4

Clem II GregNy V Didlem

37 308,3

Clem II Didlem

37 29,2

295

Appendices

1,16e

Clem II Dioncom

37 212

GregNy V Didlem

308,14 29,10

Dionlem GregNy V Didlem Didcom

212 308,17 309,1 29,28 29,29

Clem II Dionlem Dioncom GregNy V Didlem

37 212 213 309,9 30,19

Clem II Dionlem Dioncom Didlem

37 212 213 30,23

1,18b

Dionlem GregNy V

212 309,16

2,1a

Dionlem GregNy V

213 309,19

2,1b

Dionlem GregNy V

213 309,19

Dionlem GregNy V Dioncom Dioncom

213 309,20 214 214

2,2a

Dionlem GregNy V Bas III

214 310,16 961

2,2b

Dionlem Dioncom GregNy V

214 214 311,2

1,17a

καὶ ἔδωκα [καρδίαν μου τοῦ γνῶναι σοφίαν καὶ γνῶσιν]

1,17b περιφορὰ[ς καὶ ἐπιστήμην ἔγ] νων περιφορὰς καὶ ἐπιστήμην [ἔγ] ν[ων] 1,17c προαίρεσις πνεύματος [ὅτι καί γ]ε τοῦτ᾽ ἔστιν [προαίρε]σις πνεύματος 1,18a

2,1c

comma comma πλῆθος δλ γνώσεως ὅτι ἐν [πλ]ήθει σοφί[ας πλῆθ]ος γνώσεως

καὶ γε τοῦτο ματαιότης εὐφροσύνῃ ἀγαθῷ 

296 2,3a

Appendices

GregNy V Didlem

311,15 33,1

Dioncom GregNy V Didlem

214 311,16 33,1

Dionlem GregNy V Didlem

215 312,7 33,24

Dionlem Didlem Didcom

215 34,26 34,27

ἀνθρώπου

Dionlem Didcom

215 36,11

ὑπὸ τὸν ἥλιον

Dionlem GregNy V Didcom

215 314,3 35,18

2,3g

Dionlem GregNy V Didlem

215 314,4 35,6

2,4a

Dionlem GregNy V

215 319,12 327,17 327,25 328,1 350,12 35,21 36,10

2,3b

καὶ κατεσκεψάμην ἐν καρδ[ί] ᾳ μου ὡς οἶνον comma ἑλκύσαι ὡς οἶνον τὴν σάρ[κα] μ[ου]

2,3c καὶ καρδία μου ὡδήγησεν ἐν σοφίᾳ 2,3d

2,3e

καὶ τοῦ κρατῆ[σαι] ἐπ᾽ ἀφ[ρ] οσύνῃ comma

2,3f

ἐμεγάλυνα comma comma comma 2,4b

Didlem Didcom Dionlem GregNy V Didlem

215 319,13 327,26 35,21

297

Appendices

2,4c

Dionlem GregNy V

ἀμπελῶνας 2,5a

Bas II Didlem Didcom Dionlem Bas II

215 327,26 330,17 369 36,12

Didlem Didcom

215 369 228 36,12 37,7

2,5b

GregNy V Bas II Didlem

333,2 369 36,19

2,6a

Dionlem GregNy V Didlem

216 333,13 37,15

Dionlem GregNy V Didcom Didlem

216 333,14 37,26 37,15

Dionlem GregNy V

Didlem

216 334,15 335,5 336,6 338,16 38,2

2,7b

Dionlem GregNy V Didlem

216 334,16 38,2

2,7c

Dionlem

216 38,8

Dionlem Didcom Didlem

216 28,2 39,1

παραδείσους

ἐποίησά μοι κ[ολυμ]β[ήθρα]ς ὑδάτων 2,6b δρυμὸν βλ[αστῶ]ντα ξύλα τοῦ ποτίσαι ἀπ᾽ αὐτῶν δρυμὸν βλαστῶντα ξύλ[α] 2,7a

2,7d

καί γε κτ[ῆσι]ς βουκολίου καὶ ποιμνίου πολλὴ ἐγένετό μοι ὑπὲρ πάντας τοὺς comma

298

Appendices

2,7e

Dionlem Didlem

216 39,1

2,8a

Dionlem GregNy V Didlem

216 339,3 346,6 39,16

Dionlem Didlem Didcom

216 39,16 39,18

Dionlem GregNy V Didcom Didlem Didcom

216 347,4 34,19 40,25 44,12

Dionlem Didlem

216 41,15

Dionlem Didlem

216 40,25 41,18

Dionlem Didlem

216 42,23

2,9b

Dionlem Didlem

216 42,23

2,9c

Dionlem ἔμπροσθέν μου ἐν Ιερουσαλη[μ] Didlem

216 42,23

2,9d

comma comma σοφία μου ἐστάθη μοι

Dionlem Dioncom Dioncom

comma

Didlem

216 217 218 43,13 43,9

2,10a

Dionlem Dioncom Didlem

216 216 43,22

2,10b

Dionlem Dioncom Didlem

216 216 43,22

2,8b καὶ περιουσιασμοὺς βασιλέων 2,8c ᾄδοντας ... ᾀδούσας ἐποίησά μοι [ᾄδον]τας καὶ ᾀδούσας ᾄδοντας καὶ ᾀδού[σ]α[ς] 2,8d

κα[ὶ ἐ]ντρυφήματα υἱῶν τοῦ ἀν[θρώπ]ου

2,8e

2,9a

ἐμεγαλύν[θη]ν κ[αὶ] προσέθηκα

299

Appendices

2,10c

Dionlem GregNy V

216 351,8

2,10d

Dionlem GregNy V

216 351,9

GregNy V Didlem

351,9 44,22

Didcom

45,7

GregNy V Didlem

351,10 44,27

2,10g

GregNy V Didlem

351,11 44,27

2,11a

Didlem Didcom

45,2 46,6

2,11b

Didlem

45,2

2,11c

Didlem

45,2

2,10e

2,10f

[ὅ]τι [καρδία μου] εὐφράνθη ἐν παντὶ μόχθῳ μου εὐφράνθη ἐν παντὶ μόχθῳ μου [καὶ τοῦτο ἐγένετο] μερίς μου

2,11d

ματαιότης καὶ προαίρεσις πνεύματος πάντα ματαιότης καὶ ἰδο[ὺ τ]ὰ πάντα ματαιότης καὶ προαίρεσι[ς πνεύματος]

Dioncom GregNy V Didlem

217 352,2 45,16

2,11e

οὐκ ἔστιν περισσεία comma comma comma ὑπὸ τὸν ἥλιον οὐκ ἔστιν περισσεία ὑπὸ τὸ[ν ἥλιο]ν

Dioncom Dioncom GregNy V Didlem Didcom Didcom

217 218 353,5 45,25 46,1 46,6

Dionlem GregNy V Didlem

218 355,1 46,15

Dionlem Dioncom Didlem

218 218 46,15

2,12a Καὶ ἐπέβλεψ[α ἐγὼ] τοῦ ἰδεῖν σοφίαν 2,12b

300 2,12c

Appendices

Dionlem Dioncom [ὅτι τίς ὁ ἄνθρωπος, ὃς ἐπ]ελεύ- Didlem σεται ὀπίσω τῆς βουλῆς

218 218 46,25

Dionlem Didlem

218 47,25

Dionlem Didlem

219 47,25

2,13b

Dionlem Didlem

219 48,10

2,14a

Dionlem GregNy V

Didlem

219 356,23 358,11 360,17 537 384 2,15 669a 1222 29,12 64,14 105,6 262,17 301,23 48,19

2,14b

Dionlem GregNy V

219 358,11

2,14c

Dionlem GregNy V

220 361,14 364,10

2,14d

Dionlem GregNy V

220 361,14 364,11

2,15a

Dionlem GregNy V

220 361,15 364,12

2,12d 2,13a

[σ]ὺν τὰ ὅσα ἐποίησαν αὐτήν κ[αὶ εἶδ]ον ἐγὼ ὅτι ἔστιν περισσεί[α] τ[ῇ σοφίᾳ ὑ]πὲρ τὴν ἀφροσύνην

Bas II Bas I Bas Creat Did Ps.CatA Did Ps

[τοῦ σο]φ[οῦ οἱ ὀ]φθαλμοὶ αὐτοῦ ἐν κεφαλῇ αὐτοῦ

καὶ εἶπα ἐγὼ 

301

Appendices

2,15b

Dionlem GregNy V

220 361,16 364,12

2,15c

Dionlem GregNy V

220 361,16 364,13

Dionlem Dioncom GregNy V

220 221 361,17 364,13

Dionlem Dioncom Dioncom GregNy V

220 221 221 361,17 364,18

2,15f

GregNy V

361,18 364,20

2,16a

Dionlem GregNy V

222 361,19 364,20 365,18

2,16b

Dionlem GregNy V

222 361,19 364,21 365,19

2,16c

GregNy V

361,20 366,2

2,16d

GregNy V

361,21 366,3

2,16e

Dioncom GregNy V

222 361,21

2,17a

GregNy V

366,19

GregNy V Didcom

366,20 83,9

2,17c

GregNy V Didcom

366,20 83,9

2,17d

GregNy V

367,4

2,18a

GregNy V

368,4

καί γε ἐμοὶ συναντήσεταί  2,15d

2,15e

2,17b

ἵνα τί ἐσοφισάμην comma ἐγὼ περισσὸν ἐν καρδίᾳ 

τὸ ποίημα

302

Appendices

2,18b

GregNy V

368,5

2,18c

GregNy V Bas III

368,5 300

2,19a

GregNy V Bas III Did Ps

368,6 300 276,22

2,19b

GregNy V

368,7

2,19c

GregNy V

368,8

2,20a

GregNy V

363,1 369,7

2,20b

GregNy V

363,1 369,8

2,21a

GregNy V

363,2 369,9

2,21b

GregNy V

363,3 362,16 369,10

2,21c

GregNy V

363,4

2,21d

GregNy V

363,4

2,21e

GregNy V

363,5

Dionlem

223

GregNy V

363,5 369,19

2,22b

GregNy V

363,6 369,20

2,22c

GregNy V

363,7 369,21

2,23a

GregNy V

363,7 369,21

2,23b

GregNy V

363,8 369,22

2,23c

GregNy V

363,9 369,24

GregNy V

363,9 370,11

οἶδεν εἰ σοφὸς ἔσται

2,22a

2,23d

ὅτι τί γίνεται τῷ ἀνθρώπῳ ἐν παντὶ μόχθῳ comma

καί γε τοῦτο ματαιότης 

303

Appendices

2,24a

Dionlem GregNy V

223 363,21 370,19

2,24b

καὶ ὃ δείξει τῇ ψυχῇ αὐτοῦ comma

Dioncom GregNy V

224 363,22 370,20

2,24c

ἐν μόχθῳ 

Dioncom GregNy V

224 363,24 370,20

2,24d

GregNy V

363,24 370,21

2,25a

Dionlem GregNy V

223 364,1 370,22

2,26a

GregNy V

364,2 371,2

Cat 2,26b

GregNy V

364,2 371,9

2,26c

GregNy V

364,3 371,10

2,26d

GregNy V

364,4 372,12

2,26e

GregNy V

364,4 372,12

2,26f

GregNy V

364,5 372,13

2,26g

GregNy V

364,6 372,14

3,1a

Or X GregNy V

11,16 372,21 373,15 373,18 424

Bas III

304

Appendices

3,1b

Or Jb.Cat PG 17 Or X καὶ καιρὸς τῷ παντὶ πράγματι καιρὸς τῷ παντὶ πράγματι καιρὸς τῷ παντὶ πράγματι καὶ καιρὸς τῷ παντὶ πράγματι καιρὸς τῷ παντὶ πράγματι καὶ καιρὸς τῷ παντὶ πράγματι 

3,2a

Bas III GregNa Briefe GregNa Disc CyrHier GregNy V Bas III Didlem

καιρὸς τ[οῦ] τε[κεῖ]ν [καὶ καιρὸ]ς τ[οῦ] ἀποθανεῖν καιρὸς τοῦ τεκεῖν καιρὸς τοῦ τεκεῖν  3,2b

καιρὸς τοῦ φυτεῦσαι [κα]ιρὸς τοῦ φυτεῦσαι καὶ καιρὸς τοῦ ἐκτῖλαι πεφυτευμένον καιρὸς τοῦ ἐκτῖλαι πεφυτευμένον

3,3a Κα[ιρ]ὸς τοῦ ἀποκτεῖναι καὶ καιρὸς τοῦ ἰά[σασ]θαι καιρὸς τοῦ ἰάσασθαι 3,3b

καιρὸς τοῦ καθελεῖν καὶ κ[α] ιρὸς τοῦ οἰκοδομῆσαι

Κ 192 93 11,16 372,21 373,16 374,18 1012 1157 424 118,1 27,4 2,4 378,7 378,24 379,14 424 65,25 66,10 67,15

Didcom

66,26 67,1

GregNy V Didlem

381,19 383,4 68,7

Didcom

69,2

Dionlem GregNy V Didlem

224 383,14 69,10

Didcom

69,24

Didlem

70,12

305

Appendices

3,4a καιρὸς τοῦ γελάσαι 3,4b

Dionlem GregNy V GregNy IX Didcom Didlem

224 385,20 476,23 66,7 71,3

Dionlem GregNy V Didlem

225 388,15 72,18

3,5a

καιρὸς τοῦ συναγαγεῖν λίθους comma καιρὸς τοῦ βαλεῖν λίθους καιρὸς τοῦ βαλεῖν λίθ[ο]υς καὶ καιρὸς τοῦ συναγαγεῖν λίθους

Or X Or Sel in Gen6 GregNy V Didlem

11,17 128 391,9 73,21

3,5b

comma καιρὸς τοῦ μακρυνθῆναι ἀπὸ περιλήμψεως comma

GregNy V

397,18 399,8

Bas III Didlem

416 75,3

GregNy V

400,10 403,16 77,17

3,6a

Didlem 3,6b

Dionlem GregNy V Didlem

225 403,22 405,22 78,14

Didcom

78,24

καιρὸς τοῦ ῥῆξαι καὶ καιρὸς τοῦ Didlem ῥά[ψαι]

78,21

καιρὸς [τοῦ] φυ[λάξαι] καὶ καιρὸς τοῦ ἐκβαλεῖν κ[αι]ρὸς τοῦ ἐκβαλεῖν 3,7a

6 Origenes, Selecta in Genesim (PG 12).

306

Appendices

Bas Spir Bas Ep III Didlem

Κ 192 226 93 10,51 414,10 415,18 415,7 29,75 223,1 79,3

Did Hiob

338,18

GregNy V

Didlem

416,12 417,4 420,21 422,11 426,11 427,10 424,16 80,19

3,8b

GregNy V Didlem

429,1 81,20

3,9a

GregNy V Didlem

436,23 436,23 437,20 82,1

3,10a

Dionlem GregNy V Didlem

226 437,21 82,20

3,10b

Dionlem GregNy V Didlem

226 438,1 82,20

3,11a

Dionlem GregNy V

226 438,2 440,2 1605 82,27

3,7b

καιρὸς τοῦ σιγᾶν καιρὸς τοῦ σιγᾶν comma καιρὸς τοῦ σιγᾶν comma [καιρὸς] τοῦ σιγᾶν κ[αὶ] καιρὸς τοῦ λαλεῖν καιρὸς τοῦ σιγᾶν 3,8a

καιρὸς τοῦ φιλῆσαι comma

Or Jb.Cat Dionlem PG 17 Or Sel in Gen7 GregNy V

Bas III Didlem

7 Origenes, Selecta in Genesim, P. Glaue, Ein Bruchstück des Origenes über Genesis 1,28 (P. Bibl. univ. Giss. 17).

307

Appendices

3,11b

Dionlem GregNy V Didlem

226 438,2 440,2 82,27

Dionlem GregNy V Didlem

226 438,3 83,8

Dionlem Dioncom GregNy V Didlem

226 249 438,4 83,8 83,21

3,12a

GregNy V Didlem Didcom

441,1 83,21 85,28

3,12b

GregNy V Didlem

441,2 83,21

3,13a

GregNy V Didlem

441,17 84,9

3,13b

GregNy V Didlem

441,18 84,9

3,13c

GregNy V Didlem

441,19 84,9

Didlem Didcom

86,26 87,17

Didlem Didcom

86,26 87,5 87,17

3,14c

Didlem

87,26

3,14d

Didlem

87,26

3,14e

Didlem Didcom

88,25 89,15

3,15a

Didlem Didcom

89,21 89,22

3,15b

Didlem Didcom

89,21 89,27 89,23

Didlem

90,1

3,11c

3,11d

3,14a 3,14b

3,15c

comma ἀρχῆς μέχρι τέλους comma comma ἀπ᾽ ἀρχῆς καὶ μέχρι τέλους

πάντα, ὅσα ἐποίησεν ὁ θεός εἰς τὸν αἰῶνα εἰς τὸν αἰῶνα

ἀπὸ προσώπου αὐτοῦ

308 3,16a

Appendices

Καὶ ἔτι εἶδον ὑπὸ τὸν ἥλιον τόπον [τ]ῆ[ς] κρ[ίσ]εως εἶδον ὑπὸ τὸν ἥλιον 

Didlem

90,15

Didcom

91,16

Didlem

90,15

Didlem

90,15

3,16d

Didlem

90,15

3,17a

Didlem

Didcom

92,27 92,27

3,17b

Didlem

92,27

3,17c

Didlem

94,1

3,17d

Didlem

94,1 94,20

3,16b 3,16c

καὶ τόπον τοῦ [δ]ικαίου

3,18a

ἐκεῖ ε[ἶ]π[α] ἐγὼ ἐν καρδίᾳ

Didlem

3,18b

περὶ λαλιᾶς ῶν τοῦ ἀνθρώπου

Didlem

94,20

3,18c

ὅτι διαεῖ αὐτοὺς ὁ θεός

Didlem

94,20

3,18d

καὶ τοῦ δεῖξαι ὅτι αὐτοὶ κ[τ]ήνη εἰσὶν καί γε αὐτοῖς

Didlem

94,20

3,19a

98,10 Didlem

98,10 98,10

3,19b

καὶ συνάντημα τοῦ κτήνου[ς]

Didlem

3,19c

[συνάντημα ἓν] αὐτοῖς

Didlem

98,10

Bas I Didlem

728 98,11 100,17

3,19e

Didlem

100,24

3,19f

Didlem

100,24 102,3

3,19d ὡς ὁ θάνατος το[ύτου], οὕτως ὁ θάνατος τούτου

[καὶ τί ἐπερί]σσευσεν ὁ ἄνθρωπος παρὰ τὸ κτῆνος;

3,19g

οὐδέν, ὅτι τὰ πάντα ματ[αιότης] ὅτι τὰ πάντα ματαιότης

Didlem Didcom

102,3 102,10

3,20a

[τὰ] πάντα πορεύεται εἰ[ς τόπον ἕνα]

Didlem

102,17

3,20b

[τὰ π]άντ[α] ἐγένετο ἀπὸ τοῦ χοός

Didlem

102,17

309

Appendices

3,20c

Didlem Didcom

102,17 103,1

Didlem

103,22

Didcom

104,7

Didlem

103,22

Didlem

103,23

Didlem

103,23

Didlem

106,12

Didlem

106,12

ἐ[ν ποιήμασιν αὐτοῦ], ὅτι αὐτὸ μερὶς αὐτοῦ ὅτι αὐ[τ]ὸ μερὶς αὐτοῦ

Didlem

106,12

Didcom

106,21

3,22d

[ὅτι τίς ἄξει αὐτὸν τοῦ ἰδεῖν] ἐν ᾧ ἂν γένηται μετ᾽ αὐτόν;

Didlem

107,2

4,1a

[Καὶ ἐπέστρεψα ἐγὼ καὶ εἶδον]

Didlem

107,13

4,1b

[σὺν] πάσας τὰς συκοφαντίας

Didlem

107,13

4,1c

τὰς [γι]νομένας ὑπὸ τὸν ἥλιον

Didlem

107,13

4,1d

[καὶ ἰδοὺ δάκρυον τῶν συκοφαντουμ]ένων

Didlem

107,22

4,1e

καὶ οὐκ ἔστι[ν] [αὐτοῖς παρακαλῶν]

Didlem

107,22

4,1f

[καὶ ἀπὸ χειρὸς συκοφαντούντ] ων αὐτοὺς ἰσχύς

Didlem

107,22

4,2a

Or I

19

4,2b

Or I

19

4,2c

Or I

19

Didlem

Did Hiob

113,1 62,9

Didlem Did Hiob

113,1 62,9

3,21a

καὶ τὰ πάντα [ἐπιστρέφει εἰς τὸν] χοῦν καὶ [τίς] οἶδεν τὸ π[νεῦμα υἱ]ῶν τοῦ ἀνθρώπου κα[ὶ] τίς ο[ἶ]δεν τὸ πνεῦμα υἱῶν [το]ῦ ἀνθρώπου

3,21b 3,21c

[.].ὶ τὸ πνεῦμα τοῦ κ[τήνους]

3,21d 3,22a

καὶ εἶδον ὅτι οὐκ ἔστιν ἀ[γα] θ[ὸν]

3,22b 3,22c

4,3a

4,3b

[καὶ ἀγαθὸς ὑπὲρ τοὺς δύο τούτους] ἀγαθὸς ὑ[π]ὲρ τοὺς δύο τουτ[ους] [ὅστις] οὔπω ἐγέ[νε]το

310 4,3c 4,3d 4,4a

Appendices

[ὃς οὐκ εἶδεν σὺν τὸ ποίημα τὸ πον]ηρὸν τὸ πεποιη[μένον ὑπὸ τὸν ἥλιον] [Καὶ εἶδον ἐγὼ σὺν πάντα τὸν μό]χθον

4,4b 4,4c

ὅτι αὐτὸ [ζῆλος ἀνδρὸς ἀπὸ τοῦ] ἑταίρου αὐτοῦ

4,4d

Didlem

114,7

Didlem Didcom

114,7 115,14

Didlem

116,22

Didlem

116,22

Bas III Didlem

380 116,22

Didlem

116,22 117,24

4,5a

[ὁ] ἄφρων περιέλαβεν τὰς χεῖρας αὐτοῦ

Didlem

4,5b

καὶ [ἔφ]αγεν τὰ[ς σάρκας αὐτοῦ]

Didlem

117,24

4,6b

[ὑπὲρ πλήρω]μα δύο δρακῶν μόχθου

Didcom

119,8

Didcom

119,8

[Καὶ] ἐπέστρεψα ἐγὼ [καὶ εἶδον μα]ταιότητα ὑπὸ τὸν ἥλιον [Κα]ὶ ἐπέστρεψα ἐ[γὼ καὶ εἶδον ματ]αιότη[τα ὑπὸ τὸν ἥλιον]

Didcom

122,5

Didlem

121,15

4,8a

[ἔστιν εἷς, καὶ οὐκ] ἔστιν δεύτερος

Didlem

121,15

4,8b

καί γε υἱὸς καὶ ἀδελφὸς οὐκ ἔστιν [αὐ]τῷ

Didlem

121,26

4,8c

Κα[ὶ οὐκ ἔστιν περασμὸς τῷ παντὶ] μόχθῳ αὐτοῦ

Didlem

121,26

4,8d

[καί γε ὀφθαλμὸς αὐτοῦ οὐ]κ ἐμπίπλαται πλού[του] καί γε ὀφθαλμὸς αὐ[τοῦ οὐκ ἐμπίπλαται πλούτου]

Didlem

122,4

Didcom

123,4

4,8e

[καὶ τίνι] ἐγὼ μοχθῶ

Didlem

123,18

4,8f

καὶ στερίσκω τὴν ψυχήν μου ἀ[πὸ ἀγα]θωσ[ύνης;]

Didlem

123,18

4,8g

[καί γε τοῦτο ματα]ιότης

Didlem

123,18

Didlem

123,19

4,6c 4,7a

4,8h

311

Appendices

4,9a

[ἀγαθοὶ οἱ] δύο ὑπὲρ τὸν ἕνα ἀγαθοὶ οἱ δύο [ὑπ]ὲ[ρ τὸν ἕν]α 4,9b

οἷς ἔστιν α[ὐτ]οῖς μισ[θὸς ἀγαθὸς ἐν μόχθῳ αὐτῶν]

4,10a [ὅτι ἐὰν πέσωσιν, ὁ εἷς ἐγε]ρεῖ [τὸ]ν μ[έ]τοχον αὐτο[ῦ] [ὅτι ἐὰν πέσωσιν, ὁ εἷ]ς ἐγερεῖ τὸν μέτοχ[ο]ν [αὐτοῦ] 4,10b

4,11a

4,11b

227 290,12 27,2 1228 23,3 123,23 124,10

Dionlem Didlem

227 123,23

Dionlem Did Ps.CatA Didlem

227 1103 124,11 126,9

Bas III

καὶ οὐαὶ αὐτῷ τῷ ἑν[ί, ὅταν πέσῃ] 4,10c

Dionlem Or IX Epiph I Bas III GregNy Virg8 Didlem Didcom

καὶ μὴ ᾖ δεύ[τερος τοῦ ἐγεῖραι αὐτόν] [καὶ μὴ ᾖ δεύτερος τοῦ ἐγεῖραι αὐτόν] [καί γε ἐὰν κοιμηθ]ῶσιν [δ]ύ[ο], καὶ θέρμη αὐτοῖς κ[αὶ ὁ εἷς πῶς θερμανθῇ;]

4,12a

GregNy VIII,I Didlem

Didlem

929 1228 337,2 124,11 126,10 124,11 126,10

Didlem

124,18

Didlem

124,18 127,2

Bas I

761

Didcom

128,2

Eus Gen

1125 1128 1128 (2x)

4,12b 4,12c

οὐ ταχέως ἀπορραγήσεται

4,13a [+4,14a]

ἀγαθὸς καὶ σοφὸς παῖς 

8  Reference is from SC 119.

312

Appendices

4,13b

Eus Gen

1125 1128 (2x)

4,13c

Eus Gen

1125

4,14a [+4,13a]

ἐξ οἴκου δεσμίων

Or Her Eus Gen

102 1128

4,14b

ἐξ οἴκου τῶν δεσμίων

Eus Gen

1128

5,1a

μὴ σπεύσῃς

Or Phil

1,28

5,1b

μὴ ταχυνάτω

645cat

17

5,1c

645cat

Or Eu Or Phil GregNy I

17 23,4 1,28 254,18 257,24

5,1d

Or Eu Or Phil Bas I GregNy III, I GregNy VII,29 Did Ps.CatA

23,4 1,28 308 214,12 1145 622a

Or Phil

1,28

ὁ θεὸς ἐν τῷ οὐρανῷ ὁ θεὸς ἐν τῷ οὐρανῷ comma 5,1e 5,3a

εὔξῃ εὐχὴν εὔξῃ εὐχὴν

Eus Ps

628 885

5,3b

μὴ χρονίσῃς μὴ χρονίσῃς τοῦ ἀποδοῦναι μὴ χρονίσῃς τοῦ ἀποδοῦναι 

645cat Eus Ps Eus Ps

16 628 885

Dioncom Eus Ps

251 628 885

Dioncom Eus Ps

251 628 885

Didlem

145,1

5,4a

ἀγαθὸν τὸ μὴ εὔξασθαί ἀγαθὸν τὸ μὴ εὔξασθαί 

5,4b

5,8a

[καὶ περισσεία γῆς], ἐπὶ παντί, ἐστι

9  Reference is from PG 44.

313

Appendices

5,8b 5,9a

Βασιλ[ε]ὺς τοῦ ἀγροῦ εἰργασμένου ᾽Αγαπῶν ἀρ[γ]ύριον οὐ πλησθή[σ]εται ἀργυρίου

Didlem

145,1

Bas II Didlem

213 147,4

5,9b

[κ]αὶ τίς ἠγάπησεν ἐν π[λή]θ[ει] αὐτῶν γένημα;

Didlem

148,12

5,9c

καί γε τοῦτο ματα[ιότ]ης

Didlem

148,12

5,10a

ἐν πλήθει ἀγαθωσύ[νης] ἐπληθύνθησαν [ἔσ]θοντες αὐτήν

Didlem

150,2

5,10b

καὶ ί ἀνδρεί

Didcom

152,9

5,10c

ὅτι ἀρχὴ τοῦ ὁρᾶν ὀφθαλμο[ῖς α]ὐτοῦ; ἀρχὴ τοῦ ὁρᾶν ὀφθαλμοῖς

Didlem

151,2

Didcom

152,13

Didcom

153,6

Didlem

155,9

5,11a 5,12a

ἔστιν ἀρρωστία, [ᾖ]ν εἶδον ὑπὸ τὸν ἥλιον

5,12b

Bas III πλοῦτον φυλασσ[ό]με[ν]ον [τ]ῷ Didcom παρ᾽ αὐτοῦ πλοῦτον φυλασσόμενον τῷ παρ᾽ Didlem αὐτ[ο]ῦ εἰς κακίαν [αὐτοῦ]

5,13a

5,13b

5,14a

ἀπολεῖται ὁ πλοῦτος ἐκεῖνος ἐν περισπασμῷ πονηρ[ῷ] καὶ ἐγέννησεν υἱόν, καὶ οὐκ ἔστιν ἐν χειρὶ αὐτοῦ κα[θ]ὼς ἐξῆλθεν ἀπὸ γαστρὸς μητρὸς αὐτοῦ γυμνός κα[θ]ὼς ἐξῆλθεν ἀπὸ γαστρὸς μητρὸς

5,14b

300 155,16 155,9

Didlem Didcom

155,9 156,3

Didlem Didcom

156,13 158,5

Didlem

156,13

Didcom

158,5

Didlem Didcom

156,13 159,10 158,5

Didlem

160,3

5,14d 5,15c

314

Appendices

5,16a

Didlem

158,3 160,7

5,16b

καὶ θυμῷ πολλῷ καὶ ἀρρωστίᾳ κ[αὶ χ]όλῳ

Didlem

158,3

5,17a

᾽Ιδοὺ ὃ εἶδον ἐγὼ [ἀ]γαθόν, ὅ ἐστιν καλ[όν]

Didcom

161,2

5,17f

ἀρ[ι]θμὸν ἡμερῶν ζωῆς αὐτοῦ ἀριθμ[ὸ]ν ἡμερῶν ζωῆς αὐτοῦ

Didlem Didcom

161,27 162,5

5,17g

ὧν ἔδ[ω]κεν αὐτῷ ὁ θεός

Didlem

161,27

5,17h

Didlem

161,27

5,18a

Didcom Didlem

164,1 163,6

Didlem

163,6

Didlem

163,6

Didlem

163,6

Didlem

163,6

Didlem

165,6

5,19b

Didlem

165,6

6,1a

GregNa Orat B Didlem

2,50 166,14

6,1b

Didlem

166,14

6,2a

Didlem

166,14

6,2b

Didlem

Didcom Didcom

166,14 168,20 171,20

Didlem

168,16 168,16

5,18b

π[λοῦτο]ν καὶ ὑπάρχοντα καὶ ἐξουσίασεν αὐτὸν τοῦ φ[α]γεῖν ἀπ᾽ αὐτοῦ καὶ τοῦ λαβε[ῖν τ]ὸ μέρος αὐτοῦ

5,18c 5,18d

τοῦτο δόμα θεοῦ [ἐ]στιν

5,18e 5,19a

ὅτι οὐ πολλὰ μνησθήσεται τὰς ἡμέρας τῆς [ζ]ωῆς αὐτοῦ

πλοῦτον καὶ [ὑπάρχο]ντα καὶ δόξαν πλοῦτον κα[ὶ ὑπάρχοντ]α καὶ δόξα[ν]

6,2c 6,2d

ἀπὸ π[ά]ντων, ὧν ἐπιθυμήσει

Didlem

6,2e

καὶ [οὐ]κ [ἐ]ξουσιάσει αὐτῷ ὁ θεὸς τοῦ φαγεῖν ἀπ᾽ αὐτοῦ

Didlem

168,16

Didlem

169,9 169,16 175,27

6,3a

καὶ ἔτη πολλὰ ζήσεται ἐὰν γεννή[σῃ ἀ]νὴρ ἑκατὸν

Didcom

315

Appendices

6,3b

καὶ πλῆθος ὅ τι ἔ[σ]ονται ἡμέραι ἐτῶν αὐτοῦ

Didlem

6,3c

καὶ ψυχὴ αὐτοῦ οὐ πλ[η]σθήσεται ἀπὸ τῆς ἀγαθωσύνη[ς]

Didlem

171,19

6,3d

[καί γε] τ[α]φὴ οὐκ ἐγένε[το] αὐτ[ῷ] κ[αί γ]ε [ταφὴ] οὐκ ἐγένετ[ο α] ὐτῷ καί γε [τα]φὴ οὐκ ἐγέν[ετο] αὐτῷ 

Didlem Didcom

171,19 171,26 172,17

6,3e

εἶπα [᾽Αγαθὸν ὑπ]ὲρ αὐτὸν τὸ ἔκτρ[ωμα]

Didlem

172,18

6,4a

[ὅτι] ἐν μαται[ό]τητι ἦ[λθ]εν καὶ Didlem ἐν [σ]κότει πορεύετ[αι] ὅτι ἐν ἦλθεν καὶ ἐν σκότει πορεύεται

172,18

6,4b

καὶ [ἐν σ]κότει ὄν[ομ]α αὐτοῦ καλυφ[θήσετ]αι

Didlem

174,26

6,5a

καί γε ἥλιον οὐκ εἶδεν κ[αὶ οὐκ ἔγ]νω καί γε ἥλιον οὐκ εἶδεν

Didlem

174,26

6,5b

[ἀνάπ]αυσις τούτῳ ὑπὲρ τοῦτον τούτῳ ὑπὲρ τοῦτον

Didlem

174,26 175,5

6,6a

ἔζησεν χιλίων ἐτῶν καθόδους ἔζησεν χιλίων ἐτῶν καθόδους comma

Didcom Didlem

169,27 171,26 175,26

6,6b

καὶ ἀγαθωσ[ύνην οὐκ] εἶδεν

Didlem

175,26

6,9b

καί γε τοῦτο ματαιότης

Didcom

169,7

6,11b

τίς πε[ρ]ι[σσε]ία τῷ ἀνθρώπῳ

Didcom

193,1

6,12c

Didlem

Didcom

193,20 195,9

6,12d

Didlem Didcom

193,20 194,14

τίς ἀπαγγελε[ῖ] τῷ ἀνθρώπῳ

169,9 170,1

173,4

175,2 Didcom

6,12e

τί ἔσται ὀπ[ί]σω α[ὐ]τοῦ ὑπὸ τὸν ἥλιον;

Didlem

193,20

7,1a

᾽Αγαθὸν ὄνομα ὑπὲρ ἔλαιον

Didlem

196,11

316 7,1b

Appendices

καὶ ἡμέρα τοῦ θανάτου ὑ[πὲρ ἡ] Didlem μέραν γεννήσεως [αὐτοῦ.

7,2a

7,2b 7,2c

7,2d

εἰς οἶκον πότου comma τοῦτ[ο τέλο]ς παντὸς τοῦ ἀνθρώπου τέλος παντὸς τοῦ ἀνθρώπου

196,11

Dioncom Epiph I Bas III Didlem

224 26,30 257 197,18

Bas III Didlem

257 197,18

Didlem Didcom

198,21 199,17 199,17

Didlem

198,27 198,21

καὶ ὁ ζῶν δ[ώ]σει ἀγαθ[ὸν] εἰς καρδίαν αὐτοῦ δώσει ἀγαθὸν εἰς καρδίαν αὐτοῦ Didcom

199,2

7,3a

Didlem

Didcom

199,21 200,14

7,3b

Didlem Didcom

199,21 199,27

7,4a

Didlem

200,25

7,4b

Didlem

200,25

7,5a

Didlem

Didcom

202,1 203,12

Didlem

202,1

Bas III Didlem

961 203,1

7,6b

Bas III Didlem

961

7,6c

Didlem

203,1

7,7a

Didlem

203,26

7,5b 7,6a

ὅτι ἐν κακίᾳ προσώπου ἀγαθυνθήσεται

ἀγαθὸν τὸ ἀκοῦσαι ἐπιτίμη[σιν σο]φοῦ ὑπὲ[ρ ἄνδρα ἀ]κούοντα ᾆσμα ἀφρόνων ὡς φωνὴ τῶν ἀκαν[θῶν ὑπ]ὸ τὸν λέβητα

317

Appendices

7,8a

ἀγαθὴ ἐσχάτη λόγων ὑπὲρ ἀρχὴν ἀγαθὴ ἐσχάτη λόγων ὑπὲρ ἀρχὴν ἀγαθὴ ἐσχ[άτη] λόγων ὑπὲρ ἀρχὴν αὐτοῦ comma

Epiph I Epiph III Didlem

27,3 205,20 204,21

Mel Hom

303,8

7,8b

Didlem

204,21

7,9a

Didlem

205,27 205,27

7,9b

ὅτι θυμὸς ἐν κόλπῳ ἀφρ[όνω]ν ἀναπαύσεται comma

Didlem

7,10a

μὴ εἴπῃς comma

Bas II Didlem

145 207,7

Bas II Didlem

145 207,7

7,10c

Bas II

145

7,11a

Didlem

208,2

7,11b

Didlem

208,15 208,2

7,10b

ὅτι αἱ ἡμέραι αἱ πρότεραι ἦσαν ἀγαθαὶ ὑπὲρ αύτας;

καὶ περι[σσεία τ]οῖς θεωροῦσιν τὸν ἥλιον

206,21

7,12a

ὅ[τι ἐν σκιᾷ αὐτῆς ἡ] σοφία ὡς σκιὰ τοῦ ἀργυρ[ίο]υ

Didlem

208,22

7,12b

και—γνώσεως

Clem II

37

7,13a

Didlem

209,25 210,22

7,13b

Didlem

209,25 211,9

7,13c

Didlem

209,25

7,14a

Didlem

212,21 212,25

Didlem

212,21 212,27

7,14b

ὅτι τίς δυν[ήσ]ετ[α]ι τ[ο]ῦ κοσ[μῆσαι]

ἐν ἡμέρᾳ ἀγαθωσύνης ἐν ἡμέρᾳ κακίας

Didcom

318

Appendices

7,14c

Didlem

212,21

7,14d

Didlem

213,7

Didlem

213,7

Didlem Didcom

213,7 213,19 214,4

7,15b

Bas II Didcom

408 214,1

7,15c

Bas II

408

7,16a

Dioncom

64 32,6 58,3

σοφίζου περισσά μὴ σοφίζου περισσά μὴ σοφίζου περισσά μὴ σοφίζου περισσά μὴ σοφίζου περισσά comma καὶ μὴ σοφίζου περ[ισ]σά, μ[ήποτ]ε ἐκπλαγῇς

GregNy I GregNa II Didcom Did Trin Didlem

76,28 181 216,5 217,21 217,14 620 58,3

μὴ γίνου σκληρός μὴ ἀσεβήσῃς πολὺ καὶ μὴ [γίνου σκ]ληρός καὶ μὴ γίνου σκληρός μὴ γίνου σκληρός μὴ γίνου σκληρός

Didcom Didlem

217,14 216,6

Did Hiob

217,10 153,17 405,27

ἵ[να μ]ὴ [ἀποθάν]ῃ[ς] ἐν οὐ καιρῷ σου comma comma

Didcom Did Hiob

217,14 153,17 405,27

7,18a

Didlem

217,19

7,18b

Didlem

217,19

Didlem

218,5

7,14e

ἵνα μὴ εὕρῃ ὁ ἄνθρωπο[ς ὀπίσω αὐτο]ῦ μηδέν

7,15a

7,16b

[+7,17a]

7,17a [+7,16b]

7,17b

7,18c

GregNa Disc Didlem

ὅτι φοβούμενος τὸν θεὸ[ν ἐξελ] εύσεται τὰ πάντα

319

Appendices

7,19a

Or Jer10 Or Rom11 Didlem

8,1 230,8 218,22

7,19b

Or Jer Or Rom Didlem Didcom

8,1 230,8 218,22 219,21

EPH.CAT (v. 3) Didlem Or John12 Or John Didcom

404 219,24 20,36,328 20,36,335 221,8

EPH.CAT (v. 3) Didlem Or John Or John Didcom

404 221,2 20,36,328 20,36,335 221,8

Didlem

222,19

Didcom

225,5 222,19

Didlem

222,19

7,22a

Didlem

222,20 223,13

7,22c

Didlem

223,20

Didlem

224,25

τοὺς ὄντας ἐν τῇ πόλει 7,20a δίκαιος ἐν τῇ γῇ, οὐκ ἔστιν δίκαιος ἐν τῇ γῇ, ὅτι ἄνθρωπος οὐκ ἔστιν 7,20b

7,21a

καί γε [εἰς πάντας] λόγους, οὓς λαλήσουσιν

7,21b 7,21c

7,23a

ὅπως μ[ὴ ἀκούσῃς τοῦ] δούλου σου

[Πάν]τα ταῦτα ἐπείρασα ἐν τῇ σοφίᾳ

10  Reference is from SC 232, 354. 11  Reference is from Scherer, Le Commentaire d’Origène sur Rom. III. 5-V.7 (Cairo, 1957). 12  Or IV = SC 120bis, 157, 222, 290, 385.

320

Appendices

7,23b

Or Jer Mel Hom GregNa Disc13 Bas III Didlem Did Hiob Did Ps.CatA

8,7 307,19 481 684 224,25 298,6 405 910 1212

7,24a

Or Jer Mel Hom GregNa Disc Bas III Didlem

8,7 307,19 481 684 224,25

Did Hiob Did Ps.CatA

298,7 405 910 1212

Or Jer Mel Hom Did Hiob

Didlem

8,7 307,19 298,8 106,21 405 910 1212 225,4

7,25a

Didlem

225,13

7,25b

Didlem

225,13

Didlem

226,9

καὶ αὐτὴ ἐμ[ακρύνθη ἀπ᾽] ἐμοῦ [μακρ]ὰν ὑπὲρ ὃ ἦν comma comma

7,24b

Did Ps.CatA

7,25c

[καὶ] το[ῦ] γνῶν[α]ι ἀσεβοῦς ἀφροσύνην

226,18

7,25d

Didlem

226,9

7,26a

Didlem

227,1

7,26b

Didlem

227,1

Didlem

228,3

7,26c

καὶ σαγῆνα[ι κ]αρδία αὐτῆς

13  Bas I = GregNa Disc [CPG 3010]. Reference is from SC 247.

321

Appendices

7,26d

Didlem

228,3 229,8

7,26e

Didlem

228,3

7,26f

Didlem

228,3

Didlem

230,4

Didlem Didcom

230,4 231,15 230,10 230,11

Didlem Didcom

230,4 230,19 230,25

7,28b

Didlem

231,1

7,28c

Didlem

231,1

ἐποίησεν ὁ θεὸς ὁ θεὸς ὃ ἐποίησεν ὁ θεὸς ὃ ἐποίησεν ὁ θεὸς

Didlem Epiph III Epiph II Bas III Did Ps.CatA

231,11 439,10 247,3 1045 963

σὺν τὸν ἄνθρωπον εὐθῆ, τὸν ἄνθρωπον εὐθῆ 

Didlem Did Ps.CatA Epiph III Epiph II Bas III

231,11 963 439,10 247,3 1045

Epiph III Epiph II Didlem

439,10 247,4 231,11

8,1a

Didlem

232,27

8,1b

Didlem

232,27 233,14 837

7,27a

ἰδὲ τοῦτο εὗρ[ο]ν, [εἶ]πεν ὁ ᾽Εκκλησιαστής

7,27b μία τῇ μιᾷ 7,28a οὐχ εὗρον

7,29a

7,29b

7,29c κα[ὶ α]ὐτοὶ ἐζήτησα[ν] λογισμοὺς πολλούς

Did Ps.CatA 8,1c

Didlem

234,2 234,1

8,2a

Didlem

234,1 234,10

322

Appendices

8,2b

Didlem μὴ σπουδάσῃς

Didcom

234,1 234,14 234,17

8,3a

Didlem Didcom

234,22 234,24 235,2

8,3b

Didlem

234,22 235,3

8,3c

Didlem

234,22

8,4a

Didlem

234,23 235,10

8,4b

Didlem

235,24

8,5a

Or Matth Or XII Didlem Did Ps.CatA

63,29 199 235,24 332

8,5b

Didlem

237,11

8,6a

Didlem

237,11 237,23

8,6b

Didlem

237,12

8,7a

Didlem

239,4

8,7b

Didlem

239,4 239,14

Didlem

240,19

Didlem

240,19

8,8a

οὐκ [ἔστι]ν ἄνθρωπος ἐξουσιάζων ἐν πνεύματι

8,8b 9,4b

ὁ κύων ὁ ζῶν, αὐτὸς ἀγαθὸς

Epiph I

27,1

9,4c

Epiph I

27,2

9,7a

CyrHier

4,8

9,7b

CyrHier

4,8

9,7c

CyrHier

4,8

9,8a

Or X CyrHier Bas II

650,26 4,8 453

323

Appendices

9,8b

ἔλαιον μὴ ὑστερησάτω

CyrHier Didcom

4,8 273,19

μετὰ γυναικός, ἧς ἠγάπησας

Didlem Didcom

274,5 275,6

9,9b

Didlem

277,1

9,9c

Didlem

277,1

Didlem

277,1

Cat frag Prov (PG12) Didlem

1224

9,9f

Didcom Didlem

278,9 278,4

9,10a

Didlem Didcom

278,4 278,15

9,10b

Didlem Didcom

278,4 278,15

9,10c

Didlem Didcom

279,2 280,25

Didlem

279,2

εἶδον ὑπὸ τὸν ἥλιον

Didlem Didcom

281,25 282,16

ὅτι οὐ τοῖς κούφοις ὁ δρ[όμος]

GregNa II Didlem

300 281,25

GregNa II Didlem

300 283,4 281,25

9,11d

Didlem

281,26

9,11e

Didlem

283,26 284,3

9,11f

Didlem

283,26

9,9a

9,9d

πάσας τὰ[ς ἡμέ]ρας ματαιότητός σου

9,9e [ὅτι αὐτὸ] μερίς σου ἐν τῇ ζωῇ σου

9,10d 9,11a 9,11b

ὅτι οὐκ ἔστιν ποίημα καὶ [λογισμὸς καὶ γνῶσι]ς καὶ σοφία ἐν ᾅδῃ, ὅπου σὺ πορεύῃ [ἐκεῖ]

9,11c καὶ οὐ [τοῖς δυνατοῖς] ὁ πόλεμος

278,4

324 9,11g

Appendices

καιρὸς καὶ ἀπάντημα

9,12a

Didlem Didcom

283,26 285,15

Didlem

285,18 285,18

9,12b

ὡς [οἱ ἰχθύες οἱ θηρευόμενοι] ἐν ἀμφιβλήστρῳ κακῷ

Didlem

9,12c

ὡς αὐτὰ παγιδε[ύονται] οἱ υἱοὶ τοῦ ἀνθρώπου

Didlem

287,4

Didlem

287,4

Didlem

287,4

Didlem

287,5 287,20

9,12d 9,12e 9,12f

[εἰς καιρὸν] πονηρόν [ὅταν ἐπι]πέσῃ ἐπ᾽ αὐτοὺς ἄφνω

Eus Gen Καί γε τοῦτο εἶδον σοφίαν ὑπὸ Didlem τ[ὸν ἥλιον] [Καί γε τοῦτο εἶδον σοφίαν] ὑπὸ τὸν ἥλιον

1128 287,5

9,13b

Eus Gen Didcom Didlem

1128 288,3 288,1

9,14a

Eus Gen Didlem

1128 288,1

9,14b

Eus Gen Didlem

1128 288,1

9,14c

Eus Gen Didlem

1128 288,2

9,15a

Eus Gen Didlem

1128 288,3

Eus Gen Didlem

1128 289,10

Eus Gen Didlem

1128 289,12 289,10

9,13a

9,15b

9,15c

[καὶ διασώσει α]ὐτὸς τὴν πόλιν ἐν τῇ σοφίᾳ αὐτοῦ καὶ ἄνθρωπο[ς οὐκ ἐμνήσ]θη καὶ ἄνθρωπος οὐκ ἐμνήσθη σὺν τοῦ ἀ[νδρὸς τοῦ πέν]ητος ἐκείνου

287,20

325

Appendices

9,16a

᾽Αγαθὴ σοφία ὑπὲρ δύναμιν comma [καὶ εἶπ]α ἐγώ ᾽Αγαθὴ σοφία ὑπὲρ δύναμιν

9,16b

[καὶ σοφία τοῦ πένη]τος ἐξουδενωμένη 9,16c καὶ λόγοι αὐτοῦ οὔκ εἰσιν ἀκουόμεν[οι]

Eus Gen Didlem

1128 1129 289,11 289,14 289,18

Eus Gen Epiph I Epiph III Mel Hom Didlem

1128 53,8 170,5 303,11 289,18

Eus Gen Mel Hom Didlem

1128 303,12 289,18

9,17a

[λόγοι σοφῶν ἐν ἀνα]παύσει ἀκούονται

Didlem

289,18

9,17b

ὑπὲρ κραυγὴν ἐξουσιαζόντ[ων ἐν ἀφροσύναις]

Didlem

289,19

9,18a

[ἀγαθὴ σο]φία ὑπὲρ σκεύη πολέμου

Didlem

290,10

9,18b

καὶ ἁμαρτάνων εἷς ἀπολέσει ἀγ[αθωσύνην πολ]λήν

Didlem

290,10

Epiph II Didlem

83,11 290,11

10,1a

Μυῖαι θανατοῦσαι σαπριοῦσιν σκευασίαν ἐλαίου ἡ[δ]ύσματ[ος] Μυῖαι θανατοῦ[σαι σαπριοῦσιν σκευασίαν ἐλαίου ἡδύσματος]

290,21

10,1b

Didlem

291,7

10,2a

Didlem

291,7 904

10,2b

Did Ps.CatA Didlem

904 291,7 293,1

Didlem

293,3

Did Ps.CatA

καρδία ἄφρονος εἰς ἀριστερὸν 10,3a

326

Appendices

10,3b

Didlem Didcom

293,3 293,6 293,20

10,3c

Didlem

293,21

10,4a

Cyr I Cyr II Cyr III

792 1088 41 93 509 880 336 401 790 808 1748 385 412 324 768 1504 925 293,21

Cyr IV Chr XV Did Ps.CatA Antioch CyrHier Bas I Bas II Ath IV Max I Didlem

ἐὰν πνεῦμα τοῦ ἐξουσιάζοντος ἀ[ναβῇ ἐπὶ σέ] ἐὰ[ν πνεῦμα τοῦ ἐξου]σιάζοντος ἀναβῇ ἐπὶ σέ ἐὰν πνεῦμα τοῦ ἐξου[σιάζοντος Didcom ἀναβῇ ἐπὶ σέ] ἐὰν πνεῦμα τοῦ ἐξουσιάζοντος 

293,24 27,23 295,16

327

Appendices

10,4b

Cyr I Cyr II Cyr III

Cyr IV Chr XV Did Ps.CatA Antioch CyrHier

[τόπον] σου μὴ ἀφῇς comma 10,4c

Bas I Bas II Ath IV Max I Didlem Didcom Cyr I Cyr III

Cyr IV ὅτι ἴαμα καταπαύσει ἁμαρτίας καταπαύσει ἁμαρτίας μεγάλας comma [ὅτι ἴαμα καταπα]ύσει ἁμαρτίας μεγάλας

Chr XV Bas I Bas III Didcom Didlem

792 1088 41 93 509 880 336 401 790 808 1748 385 412 324 768 1504 925 293,21 27,23 294,8 792 41 93 509 880 336 401 790 209 461 295,16 296,1 295,6

328 10,5a

10,5b

10,6a

Appendices

Didlem

295,6 295,22

[ὡς ἀκούσιον, ὃ ἐ]ξῆλθεν ἀπὸ προσώπου τ[ο]ῦ ἐξουσιάζοντος ὡς ἀκούσιον, ὃ ἐξῆλθεν ἀπὸ [π] ροσώπου τοῦ ἐξουσιάζον[το]ς ἀπὸ προσώπου τοῦ ἐξουσιάζον[τος] ἀκούσιον...ἐξῆλθεν

Didlem

295,7

Didcom

297,1 297,2

ἐδόθη ὁ ἄφρων ἐν ὕψεσι [μεγάλοις] comma

Didlem

295,6 295,23

Didlem Didcom

300,23 306,20

ἔστιν [πο]νηρ[ία, ᾖν] εἶδον ὑπὸ τὸν ἥλιον

295,23

10,6b 10,8a 10,8b

καὶ καθαι[ρ]οῦν[τα φρα]γμόν, δήξεται [α]ὐτὸν ὄφις

Didlem

301,10

10,9a

ἐξαίρων λίθους διαπονηθή[σ] ε[τ]αι ἐν αὐτοῖς

Didlem

301,23

10,9b

Didlem

301,23 302,16

10,10a

Didlem Didcom

301,24 302,16 303,6

10,10b

Didlem Didcom

301,24 303,11

10,10c

Didlem

303,26

Didlem

304,3 303,26

10,11a

Didlem Didcom

303,26 304,12

10,11b

Didlem

303,27

10,12a

Didlem

305,2 305,4

ἐὰν [ἐκ]πέσῃ τὸ σιδήριον

10,10d

καὶ περισσεία τ[οῦ ἀ]νδρείου [σοφ]ία

Didcom

329

Appendices

10,12b

Didlem

306,2 305,2

Did Ps.CatA

554

10,13a

Didlem Didcom

305,2 306,15

10,13b

Didlem

305,3

10,14a

Didlem

305,3 307,1

10,14b

Didlem

307,7

καὶ χείλη ἄφρονος καταποντιο[ῦσι]ν αὐτόν comma

10,14c

καὶ τί τ[ὸ] ἐσόμεν[ον] ὀπίσω αὐτοῦ, τίς ἀναγγελεῖ αὐτῷ

Didlem

307,7

10,15a

μόχθος τῶν ἀφρόνω[ν] κοπώσει αὐ[τού]ς comma comma

Didlem

307,8

Didcom

307,21 307,25

Didlem Didcom

307,8 308,1

Bas II GregNa I Bas Ep I Didlem

293 481 61 308,7

10,16b

Bas II Bas Ep I Didlem

293 61 308,7

10,17a

Didlem Didcom

308,7 309,12

Didcom Didlem

310,1 312,10 309,24

καὶ οὐκ αἰσχυνθήσον[ται]

Didlem

309,24

ταπεινωθήσεται ἡ δόκωσις

GregNy VI Didlem Didcom

110,4 309,25 311,23

10,15b

ἔγνω τοῦ πορευθῆναι εἰς πόλιν

10,16a οὐαί σοι, πόλις, ἧς ὁ βα[σιλ]εύς σου νεώ[τερ]ος

10,17b

10,17c

πρὸς καιρὸν φάγονται ἐν δυνάμει comma

10,18a

330 10,18b

Appendices

GregNy VI Didlem Didcom

110,5 309,25 310,21 311,23

10,19a

Didlem

311,25

10,19b

Didlem

311,25

10,19c

Didlem

311,25 312,19

Hipp14 Didlem

4,29 313,10

10,20a

10,20b

καὶ ἐν ἀργίᾳ χειρῶν σ[τάξει ἡ οἰκία] comma καὶ ἐν ἀργίᾳ χειρῶν

καί γ[ε] ἐ[ν συν]ειδήσει σου βασι[λ]έα μὴ καταράσῃ ἐν συνειδήσει σ[ου βασιλέα μὴ καταράσ]ῃ καὶ ἐν ταμιείο[ις] κ[οι]τώνων σου μὴ καταρ[άσ]ῃ π[λού]σιον

314,21 Didlem

313,10

Didlem

313,11 315,9

Didcom

315,20

καὶ ὁ ἔχων τὰς πτέρ[υγ]ας ἀπαγγελεῖ λόγον ἔχων τὰς πτέρυγας Didcom

313,11

Didlem Didcom

316,1 316,3 316,10

11,1b

Didlem Didcom

316,1 316,19 317,1

11,2a

Or IV GregNa I Bas II GregNa II Didlem

547,23 885 336 612 316,1 317,14

11,2b

Didlem

319,21

10,20c

10,20d

11,1a

[ὅτι] πετεινὸν τοῦ οὐ[ρ]ανοῦ ἀποίσει σὺν τὴν [φ]ωνήν πετεινὸν τοῦ οὐρανοῦ

Didlem

ἄρτον ... ἐπὶ πρόσωπον τοῦ ὕδ[ατ]ος ἐπὶ πρόσωπον τοῦ ὕδατος

14  Reference here is from SC 14.

315,20

331

Appendices

11,3a

Didlem

319,21 320,4

11,3b

Didlem

319,22 320,4 320,23

ἐπὶ τὴν γῆν 11,3d

Didcom Didcom

321,22 322,5 322,6 322,8

11,4a

Didlem

322,23

11,4b

Didlem

322,23

11,5a

ἐν οἷς οὐκ ἔστιν γινώσκων τ[ίς ἡ Didlem ὁ]δὸς τοῦ πνεύματος comma

322,23

ἐκεῖ ἔσται ἐκεῖ ἔσται τόπῳ, οὗ πεσεῖται ... ἐκεῖ ἔσται

11,5b

324,15

GregNy III, II Didlem Didcom Did Ps.CatA

102,13 324,20 326,3 1216

11,5c

Didlem Did Ps.CatA

324,20 1216

11,6a

Didlem Didcom

326,6 326,15 327,20 331,18

Didlem Didcom

326,6 327,20 328,9 331,18

11,6c

Didlem

326,6 327,23

11,6d

Didlem

328,20

11,7a

Didlem Didcom

328,20 330,6 330,7

Didlem

328,20 331,14 331,14

ἐν γαστρὶ τῆς κυοφορούσης comma

11,6b

σπεῖρον τὸ σπέρμα σου πρωίᾳ σπεῖρον

μὴ ἀφέτω ἡ χείρ σου ἑσπέραν μὴ ἀφέτω ἡ χείρ σου

καὶ ἀγαθὸν τοῖς ὀφθαλμοῖς 11,7b 11,8a

ὅτι καὶ ἐὰν ἔτ[η πολ]λ[ὰ ζήσεται ὁ ἄνθρωπο]ς

Didlem

11,8b

[ἐν π]ᾶσιν...εὐφρανθήσεται

Didlem

332

Appendices

11,8c

Didlem

331,14 332,11

11,8d

Didlem

331,15 332,11

11,8e

GregNy VI Didlem Didcom

22,12 333,3 333,18

Didlem

333,3 334,16 335,5

τὸ ἐρχόμενον 11,9a Εὐφραίνου, νεανίσκε, 11,9b

Didcom Didlem

ἀγαθυνάτω σε ἡ καρδία σου 11,9c

Didcom

333,3 335,20 336,7

Didlem

333,4 335,26 338,6

11,9d

Didlem

333,3

11,9e

Didlem

336,24 337,1

περιπάτει ἐν ὁδοῖς καρδίας σου

11,10a

ὅτι ἐπὶ πᾶσι τούτοις ἄξει σε ὁ θεὸς ἐν κρίσει ἄξει σε ὁ θεὸς ἐν κρίσει

Didcom

337,6 Didlem Didcom

336,24 337,7 338,2

11,10b

Didlem

336,25 337,24

11,10c

Didlem

338,4

12,1a

Didlem

338,4 338,17

12,1b

Didlem

338,4 338,17

12,1c

Didlem

338,5 338,21 338,22 339,5

Didcom 12,1d

Didlem

340,7

12,1e

Didlem

340,7

333

Appendices

12,2a

Didlem σκοτισθῇ ὁ ἥλιος

Didcom

340,7 340,17 347,14

12,2b

Didlem

340,8 340,17

12,2c

Didlem

340,8

12,3a

Didlem

Didcom

342,3 345,12

σαλευθῶσιν φύλακες τῆς οἰκίας

12,3b

καὶ διασ[τρα]φῶσιν ἄνδρες τῆς δυν[ά]με[ως]

Didlem

345,16

12,3c

[καὶ ἤρ]γησαν αἱ ἀλήθουσαι, ὅτ[ι ὠλιγ]ώθησα[ν] ἀλήθουσαι, ὅτι ὠλιγώθησαν

Didlem

345,16

Didcom

347,5

καὶ σκοτάσουσιν αἱ βλέπουσαι ἐ[ν ταῖς ὀπ]αῖς

Didlem

345,16

12,4a

Didlem

349,4

12,4b

Didlem Didcom

349,4 351,1 351,17

12,4c

Didcom

352,26 352,15

12,4d

Didlem

353,25

12,5a

Didlem

355,9

12,5b

Didlem Didcom

355,9 357,27 358,6 4,15

12,3d

εἰς φωνὴν τοῦ στρουθίου

ἀνθήσῃ τὸ ἀμύγδ[αλον] ἀνθήσῃ τὸ ἀμύγδαλον 12,5c

Hipp15 Didlem

καί παχυνθῇ ἡ ἀκρίς

Hipp

355,9 357,18 4,15

12,5d

καὶ [δι]ασκεδασθῇ ἡ κάππαρις ἡ κάππαρις

Didlem Hipp

358,9 4,15

12,5e

ὅτι ἐπορεύθ[η ὁ ἄν]θρωπ[ος] ε[ἰ]ς οἶκον αἰῶνος αὐτοῦ

Didlem

358,9

15  Reference here is from SC 14.

334

Appendices

12,5f

Didcom

358,25 359,19 360,28 360,19

Didlem Didcom

360,2 361,1

Didlem

360,2 360,11

Didlem Hipp

361,8 4,15

καὶ συ[ντρ]οχάσῃ ὁ τροχὸς ἐπὶ τὸν λάκκο[ν] κα[ὶ συντροχά]σῃ ὁ τροχὸς ἐπὶ τὸν λάκκον 

Didlem

361,8

Hipp

4,15

12,7a

καὶ ἐπιστρέψῃ ὁ χοῦς ἐπὶ τὴν γῆν

Dion16

310,3

12,7b

[τὸ πνεῦμα ἐπιστρέφει πρὸς τὸν θεὸν ὃς ἔδοκεν αυτό] comma

Or Matth Cl Nr.

144

ματαιότης ματαιοτήτων ματαιότητα ματαιοτήτων ματαιότητα ματαιοτήτων ματαιότης ματαιοτήτων ματαιότης ματαιοτήτων ματαιότης ματαιοτήτων

CyrHier Eus Gen Eus Ps Bas III GregNy Bapt Didcom

4,8 1128 596 541 429 154,13

12,8b

Eus Ps Bas III

596 541

12,11a

Or X Bas I

281,7 401

12,11b

Or X

281,7

Or X

281,7

ἐν ἀγορᾷ οἱ κοπτόμενοι 12,6a 12,6b

σχοινίον τοῦ ἀργυρίου καὶ συν[θλιβῇ ἀνθ]έμιον τοῦ χρυσίου

12,6c 12,6d

12,8a

12,11c

ἐκ ποιμένος ἑνὸς

362,12

Dion

310,6

16  W. A. Bienert, Neue Fragmente des Dionysius und des Petrus von Alexandrien aus Cod. Vatop. 236, Klèronomia 5 (1973) 310.

335

Appendices

12,12a

Or Phil

5,1 5,2 5,4

12,12b

Or Phil

5,1

12,13a [+12,14]

Eus Ps GregNa I GregNa I17

41 780 1246

12,13b

Chr Jb.Cat Did Ps.CatA GregNa I GregNa I

Β 33 315 780 1246

12,13c

Chr Jb.Cat

Β 33

12,14a

[+12,13a]

Cat frag Prov18 1664 Eus Ps 80 428 Eus Ps II19 532 41

12,14b [+12,13a]

Eus Ps Eus Ps II

80 41

12,14c [+12,13a]

Eus Ps Eus Ps II

80 41

Τὸν θεὸν φοβοῦ Τὸν θεὸν φοβοῦ 

Citations from Latin Fathers Note: if the quotation in this column is in brackets, it is only for reference. It represents the full line of the lemma and need not be quoted for this table. 1,1a

verba Ecclesiastae filii David Solomon filius David

Ruf Cant

83,24 83,23

1,1b

regis Istrahel in Hierusalem qui regnavit in Istrahel

Ruf Cant

83,25 83,23

17  GregNa I = PG 35 = GregNa Disc SC 284. 18  PG 12. 19  PG 24.

336

Appendices

vanitatibus vanitatum vanitatem vanitatum comma

Ruf Num Ruf Cant Ruf Or princ

11,8 78,3 1,7,5

comma omnia uanitas omnia uanitas comma comma comma

Ruf Num Ruf Or princ Amb Fuga Amb Luc Amb Ps Ambrst Ques

92,3 1,7,5 1,4 Pr,2 39,4 A,2,2

1,3a

Amb Ps 118

12,21

1,3b

Amb Ps 118

12,21

1,4a

Or Matth Ruf Rom Amb Ps 118

122 6,5 12,21 (2x)

1,4b

Or Matth Ruf Rom Amb Ps 118

122 6,5 12,21 (2x)

Amb Ps 118

12,22 (2x) 12,23

1,5b

Amb Ps 118

12,22 (2x)

1,6a

Amb Ps 118

12,22 12,23

1,6b

Amb Ps 118

12,22 12,23

1,6c

Amb Ps 118

12,22

1,6d

Amb Ps 118

12,22

1,7a

Amb Exam Amb Tob

3,2,10 13,44

1,7b

Amb Exam Amb Tob

3,2,10 13,44

1,8c

PsCyp Ruf Ex Amb Mort Amb Tob

205 11,6 7,28 13,44

1,8d

Ruf Ex Amb Tob

11,6 13,44

1,2a [+ 1,14] 1,2b [+ 1,14]

1,5a

comma Ipse oriens

337

Appendices

1,9a

Ruf Or princ Ruf Or princ Amb Tob

1,4,5 3,5,3 13,44

1,9b

Ruf Or princ Ruf Or princ

1,4,5 3,5,3

1,9c

Ruf Or princ Ruf Or princ Amb Mort

1,4,5 3,5,3 7,28

1,10a

Ruf Or princ Ruf Or princ

1,4,5 3,5,3

1,10b

Ruf Or princ

1,4,5

1,10c

Ruf Or princ

1,4,5

1,14a [+ 1,2]

Vidi universa opera Respexi enim, et uidi universa

Cyp Ruf Or princ

122 1,7,5

1,14b [+ 1,2]

quae facta sunt sub sole quae sub sole sunt,

Cyp Ruf Or princ

122 1,7,5

1,14c [+ 1,2]

et ecce omnia uanitas et ecce omnia uanitas

Cyp Ruf Or princ

122 1,7,5

Hil Ps

126,13

2,1a

dixi ego in corde meo:

Ruf Cant

94,2

2,1b

veni et tentabo te in laetitia, et vide in bono

Ruf Cant

94,2

2,2a

Ruf Reg S Bas

8,28

2,4a

Ruf Cant

94,22

2,4b

Ruf Cant

94,22

2,4c

Ruf Cant

94,22

2,5a

Ruf Cant

94,23

2,6a

Amb Isa

4,25

2,6b

Amb Isa

4,25

2,8c

Ruf Cant

94,25

2,8d

Ruf Cant

94,25

2,8e

Ruf Cant

94,25

1,18b

338 2,14a

Appendices

Ruf Cant Or Matth Amb Exam Amb Helia Amb Jb Amb Myst Amb Noe Amb Ps 118 Amb Ps Amb Sacr Amb Ep Amb Exh virg

65,19 268 6,9,55 10,36 4,7,28 6,30 7,17 15,12 20,1 35,26 37,29 43,61 3,1,1 13,11 10,63

3,1a

Amb Ps 118 Amb Ep

19,17 11,15 13,6

3,1b

Amb Ps 118

19,17

3,2a

Or Matth Amb Tob

188 13,43

3,2b

Amb Tob

13,43

3,3a

Amb Tob

13,43

3,6a

Amb Tob

13,43

3,6b

Amb Tob

13,43

3,7b

Or Matth Amb Ps 118 Amb Inst virg

1 16,36 1,4

3,8a

PsCyp Amb Ps 118

188 15,17 15,20

3,8b

Amb Ps 118

15,17

339

Appendices

4,2a

laudavi omnes mortuos comma comma comma comma comma

Ruf Num Ruf Rom Ruf Ps Amb Sat Amb Ps 118 Hil Ps

7,3 3,2 38,1,11 2,30 18,3 119,18

4,2b

magis quam vivos eos, comma comma comma comma comma

Ruf Num Ruf Rom Ruf Ps Amb Sat Amb Ps 118 Hil Ps

7,3 3,2 38,1,11 2,30 18,3 119,18

4,2c

qui vivunt usque nunc comma comma comma comma comma

Ruf Num Ruf Rom Ruf Ps Amb Sat Amb Ps 118 Hil Ps

7,3 3,2 38,1,11 2,30 18,3 119,18

4,3a

et optimus super comma comma comma comma

Ruf Num Ruf Rom Ruf Ps Amb Mort Amb Sat

7,3 3,2 38,1,11 2,4 2,30

4,3b

hos qui nondum natus est comma comma comma

Ruf Num Ruf Ps Amb Mort Amb Sat

7,3 38,1,11 2,4 2,30

4,3c

Amb Mort Amb Sat Amb Ps 118

2,4 2,30 18,3

4,3d

Amb Sat

2,30

4,4a

Amb Sat

2,30

4,4b

Amb Sat

2,30

4,4c

Amb Sat

2,30

4,4d

Amb Sat

2,30

4,5a

Amb Ep

7,33 27,8

340

Appendices

4,5b

Amb Ep

7,33 27,8

4,8a

Amb Inst virg

10,64

4,8c

Amb Inst virg

11,69

4,8d

Amb Inst virg

11,70

4,9a

Ruf Reg S Bas Amb Ep

174,7 17,3 17,8

4,9b

Amb Ep

17,3

4,10a

Amb Ep

17,3 17,6 11,74

Amb Inst virg 4,10b

Ruf Reg S Bas Amb Ep Amb Inst virg

3,7 174,8 17,6 11,74

4,10c

Amb Ep Amb Inst virg

17,6 11,74

4,11a

Amb Ep Amb Inst virg

17,6 11,74

4,11b

Amb Inst virg

11,74

4,12c

Ruf Ex Amb Luc Amb Ep

9,3 7,208 17,8

4,13a

Amb Inst virg

12,78

4,13b

Amb Ep

17,9

4,13c

Amb Ep

17,9

4,15a

Amb Inst virg

12,76

4,15b

Amb Inst virg

12,76

4,15c

Amb Inst virg

12,76

4,15d

Amb Inst virg

12,76

4,16a

Amb Inst virg

12,78

5,1d

Ruf Ps

36,5,4

5,3a

Cyp Amb Cain

143 1,7,25

5,3b

Amb Cain

1,7,25

341

Appendices

5,3c

Amb Cain

1,7,25

5,3d

Amb Cain

1,7,25

5,4a

Amb Luc

9,36

5,4b

Cyp Amb Luc

143 9,36

5,9a

Cyp Amb Cain Amb Nab Amb Ps

164 1,5,21 6,28 1,28

5,9b

Amb Cain

1,5,21

5,9c

Amb Cain

1,5,21

5,11a

Amb Nab

6,29

5,11b

Amb Nab

6,29

5,11c

Amb Nab Amb Ps

6,29 1,28

5,11d

Amb Nab Amb Ps

6,29 1,28

5,12a

Amb Cain Amb Ps

1,5,21 1,28

5,12b

Amb Cain Amb Nab Amb Ps

1,5,21 5,25 1,28

5,15b

Amb Nab

6,28

5,16a

Amb Nab Amb Ps

6,28 1,28

5,16b

Amb Nab Amb Ps

6,28 1,28

6,3d

Amb Luc

10,140

6,3e

Ruf Num Or Matth Amb Jb

7,3 71 2,4,15

6,4a

Ruf Num Or Matth

7,4 71

6,4b

Or Matth

71

342

Appendices

7,1b

Or Matth

71

7,2a

Amb Ep

14,106

7,2b

Amb Ep

14,106

7,4a

Amb Ps Amb Paenit Amb Tob

1,25 2,6,51 1,5

7,4b

Amb Ps Amb Paenit Amb Tob

1,25 2,6,51 1,5

7,5a

Amb Ps

1,25

7,5b

Amb Ps

1,25

7,6a

Ruf Reg S Bas

8,29

7,6b

Ruf Reg S Bas

8,29

7,11a

Amb Ep

34,12

7,15b

Ambrst Rom Ambrst Eph

12,19te 4,26

7,16a

Ruf Rom Hi Luc Ambrst Rom Ambrst Eph Amb Paenit Ambrst Ques Ambrst Ques

9,2 151,16 12,19te 4,26 1,1,2 4,12 41,22

7,16b

Cyp Hi Luc Amb Ep

140 151,16 1,6

7,20a

Ruf Cant Ruf Rom Ruf Lev

222,14 6,3 459

7,20b

Ruf Cant Ruf Rom Ruf Lev

222,14 6,3 459

Ruf Pamph Ruf Or princ

570 4,3,14

Noli esse multum sapiens

7,23b

Dixi, sapiens efficiar, comma

343

Appendices

et ipsa sapienta longe facia est a me, longe plus quam erat: comma

Ruf Pamph

570

Ruf Or princ

4,3,14

et altitudinem profundam quis inveniet. comma

Ruf Pamph Ruf Or princ

570 4,3,14

7,25a

Amb Mort

7,28

7,25b

Amb Mort

7,28

7,25c

Amb Mort

7,28

7,25d

Amb Mort

7,28

7,26a

Amb Mort

7,28

7,26c

Clem.R-Ps20

2,10,4

7,26d

Clem.R-Ps

2,10,4

Ruf Ex

2,1

Or Matth Ruf Rom IVD.HOM21 Amb Jb

127 7,6 470 4,7,27

Ambrst Ques

6,19 66,11

9,4c

Ambrst Ques

66,11

9,7a

Amb Exh virg

17,110

9,7b

Ruf Ps

36,4,7

9,7c

Amb Exh virg

17,110

9,8a

Amb Exh virg

10,62 (2x)

Amb Exh virg

10,62 10,63

7,24a

7,24b

7,28b

unum ex mille

8,5a

9,4b

9,8b

spes est comma

in capite tuo

20  For Clem.R-Ps, see CLEM.R-PS VIRG in Biblia Patristica 2. 21  GCS 30.

344

Appendices

10,4a

Ruf Num Ruf Cant Ruf Or princ Ruf Or princ Amb Jb Amb Ps

27,12 211,18 3,2,1 3,2,3 4,7,28 37,35

10,4b

Ruf Num Ruf Cant Ruf Or princ Ruf Or princ Amb Jb Amb Ps Amb Ep

27,12 211,18 3,2,1 3,2,3 4,7,28 37,35 17,13

10,4c

Ruf Cant Ruf Or princ Ruf Or princ Amb Ps

211,18 3,2,1 3,2,3 37,35

10,8b

Amb Ep Amb Inst virg

39,1 9,60

10,9b

Scinders ligna periclitabitur in cis

Cyp

174

10,10a

si exiderit ferrum

Cyp

174

10,19b

vino laetitiae

Ruf Cant

51,25

10,20a

Amb Inst virg

1,7

10,20b

Amb Inst virg

1,7

11,1a

Amb Jb

4,2,7

11,2a

Amb Ep Amb Iacob Amb Ep

68,8 2,11,53 31,6

Amb Ep

36,4

11,9a

Amb Inst virg

10,69

11,9b

Amb Inst virg

10,69

11,9c

Amb Inst virg

10,69

11,9d

Amb Inst virg

10,69

11,9e

Amb Inst virg

10,69

11,3a

12,7b

effundunt nubes

convertitur ad deum qui dedit eum Or Matth [spiritus quidem eius revertitur ad Or Matth deum qui dedit eum]

132 144

345

Appendices

Ruf Num Ruf Num Or VIII Amb Ps

11,8 25,3 78 39,4

12,8b

Ruf Num Amb Ps

25,3 39,4

12,11a

Or VIII22 Amb Ep

328 36,5

12,11b

Ruf Ez

328

12,11c

Ruf Ez

328

12,11d

Ruf Ez

328

12,14a

Amb Fide

3,3,20 5,11,142

12,8a

vanitatibus vanitatum vanitas vanitatium vanitatem vanitatum uanitas uanitatum

22  GCS 33.