From Scribal Error to Rewriting: How Ancient Texts Could and Could Not Be Changed [1 ed.] 9783666522093, 9783525522097

Citation preview

De Septuaginta Investigationes (DSI) Edited by Anneli Aejmelaeus, Kristin De Troyer, Wolfgang Kraus, Emanuel Tov In Co-operation with Kai Brodersen (Erfurt, Germany), Cécile Dogniez (Paris, France), Peter J. 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 12

Anneli Aejmelaeus / Drew Longacre / Natia Mirotadze (eds.)

From Scribal Error to Rewriting How Ancient Texts Could and Could Not Be Changed

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. © 2020, Vandenhoeck & Ruprecht GmbH & Co. KG, Theaterstraße 13, D-37073 Göttingen 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: NEUNPLUS1, Berlin Vandenhoeck & Ruprecht Verlage | www.vandenhoeck-ruprecht-verlage.com Online ISSN 2197-0912 ISBN 978-3-666-52209-3

Table of Contents Introduction�������������������������������������������������������������������������������������������������������������������7 Part I Ancient Scribal and Editorial Practices Anna Kharanauli Origen and Lucian in the Light of Ancient Editorial Techniques������������������������15 Amneris Roselli Galen’s Practice of Textual Criticism������������������������������������������������������������������������53 Julio Trebolle Pre-Lucianic Readings of 3–4 Reigns in Marginal Notes of the Syrohexapla and in the Syriac Text of Jacob of Edessa�����������������������������������������������������������������73 Kristin De Troyer The Scribe of the Marginal Notes of Manuscript 344 (Ra 344; BM v)�������������������������������������������������������������������������������������������������������������99 Part II Textual History of the Hebrew Bible Peter J. Gentry and John D. Meade MasPsa and the Early History of the Hebrew Psalter�������������������������������������������113 Emanuel Tov The Possible Revision of Hebrew Texts According to MT����������������������������������147 Anneli Aejmelaeus Rewriting David and Goliath?���������������������������������������������������������������������������������165 Drew Longacre Multilinear Genealogical Networks: Expanding the Scope of Textual History����������������������������������������������������������������������������������������������������������181 Part III Writing and Rewriting in Translation Katja Kujanpää Adjusted to the Argument: Tracing Paul’s Motives for Modifying the Wording of Scriptural Quotations������������������������������������������������201

6

Table of Contents

Andrés Piquer Otero Creative Philology and Glosses: Secondary Versions of Kingdoms and Lexical Accumulation or Mutation����������������������������������������������221 S. Peter Cowe Scribe, Translator, Redactor: Writing and Rewriting Scripture in the Armenian Versions of Esther, Judith, and Tobit����������������������������������������237 Jean-Marie Auwers The Intermediate Version of the Book of Tobit in its Greek Dress����������������������������������������������������������������������������������������������������������������271 Natia Dundua What Can the Georgian Translation of the Book of Tobit Tell about GIII?�����������������������������������������������������������������������������������������������������������289 Natia Mirotadze The Old Georgian Version of the Book of Esther—All in One��������������������������321 Magda Mtchedlidze A Translation, Paraphrase, or Metaphrasis? Regarding Euthymius the Hagiorite’s Versions of the Orations by Gregory the Theologian�������������������������������������������������������������������������������������������361 Anna Kharanauli Septuagint Text Types in the Georgian Translations��������������������������������������������391

Introduction The present volume focuses on ancient literary cultures and the work of copyists, editors, and translators. The contributions included in it represent the work of a diverse group of senior and junior scholars from North America, Europe, Israel, and Georgia, who were gathered to an interdisciplinary symposium in Tbilisi, Georgia from 30 April–3 May 2015. On behalf of all the participants, the editors would like to express their sincere gratitude to the initiator and organizer of the symposium Professor Anna Kharanauli and her many colleagues and students for their generous hospitality and excellent organization of the symposium and its accompanying program. We would also like to thank the Ivane Javakhishvili Tbilisi State University and the Shota Rustaveli National Science Foundation for making the event financially possible. The symposium was a wonderful opportunity to meet and exchange ideas between scholars and students of different backgrounds and specialties who might not regularly have other occasions to see each other, and it was a great chance to get acquainted with the country and culture of Georgia. The papers published in this volume have been organized into three parts. The first part Ancient Scribal and Editorial Practices focuses on scribal and editorial techniques in Greek, Latin, and Syriac sources. In her paper “Origen and Lucian in the Light of Ancient Editorial Techniques”, Anna Kharanauli argues that Origen and Lucian were heirs of a conservative Alexandrian philology and explores the ramifications of that for their editorial work. She argues that the Alexandrian grammarians were not concerned with producing critical texts for popular dissemination, but rather created annotated editions and commentaries for scholarly reference. In light of this, Origen’s Hexapla should not be understood as creating a new composite Greek text in his fifth column, but merely presenting a synopsis of readings without the need for the critical signs attributed to him. The Hexapla was used by Origen and his students as a resource for creating new editions, where differences were occasionally indicated as needed with the use of the asterisk and obelus. This scenario helps explain the inconsistencies in hexaplaric texts and marginal annotations in the tradition. Kharanauli explains the work of Lucian similarly as not creating a new composite text or revision, but rather an edition with marginal annotations that only inconsistently come to be worked into the text in the tradition. In “Galen’s Practice of Textual Criticism”, Amneris Roselli surveys the principles of textual criticism as utilized by Galen in light of his commentaries on Hippocrates and his recently discovered work De indolentia. Galen pays close attention to the quality of the text when working with stylistically difficult texts. He makes a point to seek out ancient copies and commentaries and considers the plausibility of their readings. In this process, Galen shows awareness of

8

Introduction

many ways in which the text could be corrupted, and he was confident in his ability to reconstruct it. He tries to explain the text by the author’s own words elsewhere in order to produce a medically useful text, including emendations where necessary. Julio Trebolle argues in his paper “Pre-Lucianic Readings of 3–4 Reigns in Marginal Notes of the Syrohexapla and in the Syriac Text of Jacob of Edessa” that many marginal notes in the Syrohexapla and readings from Jacob of Edessa stem from a pre-Lucianic or Old Greek (OG) version of 3–4 Reigns. These readings occur in parts where the majority B text reflects the kaige recension, but they faithfully preserve the characteristics of the OG translation based on a different Hebrew text. Many of these readings are also paralleled in pre-Lucianic sources like Josephus, the Old Latin, and the Ethiopic. Thus, these readings provide important evidence for the reconstruction of the OG text. Kristin De Troyer’s article “The Scribe of the Marginal Notes of Manuscript 344 (Ra 344; BM v)” examines the marginal notes in Ra 344 to investigate how the scribes read and annotated the text. The main scribe of the manuscript (MS) faithfully copied a model manuscript and occasionally wrote down hexaplaric variants. At least one further scribe (MN) subsequently inserted more marginal notes based on different sources. Within these notes, De Troyer argues that the siglum ο' refers to the fifth column of Origen’s Hexapla, which consisted of an Old Greek substratum revised and annotated by Origen, though the MN is not entirely consistent in how he references sources for his readings. The second part Textual History of the Hebrew Bible focuses on scribal and editorial aspects of the textual history of the Hebrew Bible. In their article “MasPsa and the Early History of the Hebrew Psalter”, Peter J. Gentry and John D. Meade compare the Masada Psalms scroll MasPsa to the Aleppo Codex and early Greek codices with regard to their stichometric layout. The text of MasPsa agrees almost completely with the Aleppo Codex, and the divisions marked by blank space and line breaks in MasPsa agree very closely with the Masoretic terminal markers (pausal forms and accents). The Aleppo Codex also employs a system of division by blank space, but this does not correspond well with meaningful semantic breaks and the pattern of MasPsa. This suggests that the Masoretic tradition of the Psalter retained the visual concept of the stichometric layout of earlier scribal praxis, but without necessarily preserving the ancient content divisions. The divisions of MasPsa agree more closely with the stichometric divisions evident in the Greek codices Sinaiticus and Vaticanus, implying a common tradition going back earlier than MasPsa. In “The Possible Revision of Hebrew Texts according to MT”, Emanuel Tov argues that the correction towards the Masoretic Text so evident in the Greek tradition cannot be substantiated within the Hebrew Dead Sea Scrolls. Though many examples can be cited of corrections in agreement with the MT, most of these were likely corrected according to their exemplars, rather than a different

Introduction

9

proto-MT text. This holds true for scrolls that were closely related to the MT as well as those that were quite different. Anneli Aejmelaeus suggests in her article “Rewriting David and Goliath?” that the longer MT version of the story of David and Goliath was the result of processes of rewriting, where the story was expanded with interpretive insertions and bridges to the Torah. She sees the shepherd motif as the key to explaining the rewritten narrative, building on other references to David’s early career. This motif was especially appropriate, since it was often used metaphorically in reference to kings. By drawing from other passages in 1 Samuel and building bridges to similar motifs in the Pentateuch, the reviser rewrites the earlier story to fill narrative gaps. In “Multilinear Genealogical Networks: Expanding the Scope of Textual History”, Drew Longacre concludes the section and calls for expanding the horizons of textual history beyond the limits of literary works by exploring multilinear networks of genealogically related texts. The genealogical structure of textual history can be conceptualized with series of source-recipient relationships that cross the literary boundaries of works or compositions. In this way, the scope of textual history can be expanded to include all interrelated texts and portions thereof, while at the same time allowing for literary distinctions within that structure. Longacre illustrates this perspective with examples from the book of Exodus. The third part Writing and Rewriting in Translation deals with a variety of writings from the Old Testament, New Testament, Apocrypha, and Patristic texts in various languages, focusing on issues of textual and literary criticism and including a number of papers with an emphasis on Georgian translations. In her article “Adjusted to the Argument: Tracing Paul’s Motives for Modifying the Wording of Scriptural Quotations”, Katja Kujanpää demonstrates how the Apostle Paul purposefully reworked scriptural citations in the course of making his arguments. He freely adjusted the wording of his quotations to best fit within their new literary contexts as part of his argument. In some cases Paul quoted from a text already revised towards the Hebrew, and in others Paul’s quotations subsequently influenced the textual transmission of the LXX texts he quoted. Kujanpää concludes that it was more important for Paul to highlight the significance of the cited verses than to replicate their precise wording. In “Creative Philology and Glosses: Secondary Versions of Kingdoms and Lexical Accumulation or Mutation”, Andrés Piquer Otero cites numerous examples of how he sees translators and scribes in the Coptic and Arabic traditions creatively engaging with and building complex textual traditions. By examining the phenomena of doublets, onomastics, toponomy, glosses, adaptations, expansions, and eclecticism, he argues that these versions reflect creative interaction between various Vorlagen and the literary and philological interests of the translators. Understanding the work of these translators is important for understanding the full textual tradition.

10

Introduction

S. Peter Cowe in his article “Scribe, Translator, Redactor: Writing and Rewriting Scripture in the Armenian Versions of Esther, Judith, and Tobit” constructs a typology of the types of creative changes made by Armenian translators in their work, surveying the evidence from the books of Esther, Judith, and Tobit. He concludes that translators occupy the middle ground between scribes and redactors, attempting to communicate the text in meaningful ways to their readers. They do this not in isolation, but in the context of interpretive traditions which influence the ways they read and reformulate the texts they translate. Cowe further argues that the translators were influenced by theological and rhetorical concerns that shaped their presentation. In “The Intermediate Version of the Book of Tobit in its Greek Dress”, JeanMarie Auwers describes the characteristics of the Greek III version of the book of Tobit and stresses that it should be studied in its own right. He concludes that Greek III is a revision based on Greek II to make it more readable, coherent, and succinct. Thus, Greek III is intermediate in size between the short Greek I and long Greek II text-forms, but not actually a mix of the two. This creative rewriting yields a story with its own distinctive forms and emphases. Asking “What Can the Georgian Translation of the Book of Tobit Tell about GIII?” Natia Dundua analyzes the sources for the Old Georgian translation of Tobit, concluding that it is a good witness to the complete Greek III version of the book. The Old Georgian does not consistently agree with any Greek manuscripts, but rather attests more fully to Greek III than any surviving Greek manuscripts, which preserve it only in part. Thus, the Old Georgian (often supported by the Old Latin) is an important indirect witness to the partially lost Greek III version. In her article “The Old Georgian Version of the Book of Esther—All in One”, Natia Mirotadze discusses structural and compositional peculiarities of the additional (apocryphal) sections of the Book of Esther and argues that the composite nature of the Old Georgian version (GeII) of Esther reflects an aim to collect and document a wide variety of texts. The editor probably tried to maintain the smoothness of the resulting story, but in some cases chose to sacrifice the coherent character of the text in favor of completeness. Using the Septuagint as its base text, the distinctive Greek Vorlage of GeII regularly added texts from L, GrLa, and a postulated additional source GrX wherever they differ. The result is a full inventory of all the various Greek versions of Esther, with frequent redundancies and inconsistencies. Magda Mtchedlidze’s paper “A Translation, Paraphrase, or Metaphrasis? Regarding Euthymius the Hagiorite’s Versions of the Orations by Gregory the Theologian” illustrates the liberties taken by the prominent 10th–11th century Georgian scholar Euthymius the Hagiorite in his version of the Orations by Gregory the Theologian to adapt the text to communicate with his readers. He freely translated the language and metaphors into comprehensible Georgian. He also

Introduction

11

gave interpretive translations of difficult passages in order to bring across his perceived meaning. At times, Euthymius even inserts his own creative contributions. In “Septuagint Text Types in the Georgian Translations”, Anna Kharanauli surveys the relevance of Georgian translations for the textual criticism of a number of Old Testament books. She argues that the Georgian tradition of Ezekiel gives evidence that pre-Origenian variants were added in the margins of manuscripts and subsequently (inconsistently) incorporated into the main text. The Georgian text of Jeremiah is a rare witness to the O text, but also demonstrates that the marginal variant readings were inconsistently incorporated into this tradition, rather than being systematically included in a single hexaplaric recension. So also the Georgian texts of Isaiah and 1 Esdras, which support the manuscript family L, but not a Lucianic recension. Thus, the Georgian manuscript tradition provides both important evidence for early forms of the Greek text, as well as supporting evidence for the process of creating ancient ekdoseis and the gradual infiltration of marginal readings argued for in the first paper in this volume. We owe a special debt of gratitude to the editors of De Septuaginta Investigationes and the staff at Vandenhoeck & Ruprecht for their help in bringing this volume to publication, as well as to Susanna Asikainen (Th.D.) for performing the technical editing with the support of the Centre of Excellence “Changes in Sacred Texts and Traditions” (University of Helsinki). It is our hope that these contributions will prove valuable both in relation to their specific concerns and to the broader question of how authoritative texts do and/or do not change. Anneli Aejmelaeus Drew Longacre Natia Mirotadze

Part I Ancient Scribal and Editorial Practices

Anna Kharanauli Origen and Lucian in the Light of Ancient Editorial Techniques Ἀμαθέστατε καὶ κακέ, ἄφες τὸν παλαιόν, μὴ μεταποίει Codex Vaticanus, 1512

The history of the Septuagint starts in Alexandria. This was the place where Ptolemaios Lagos brought from Athens Aristotle’s pupils, the Peripatetics (Demetrius Phalereus among them) and founded professional philology. Thus, the beginning of the history of philology and the beginning of the text history of the Septuagint coincide. This is, at least, according to tradition. Even in later times, Alexandrian philology and the philology of the Septuagint never actually separated from each other; the subsequent textual history of the Septuagint has been related to the Alexandrian grammarians and Antiochean critics. Therefore, my thesis is that Hellenistic philology represents the context in which both the formation of the Septuagint and the subsequent history of its text must be considered. The present article is an attempt to generally and schematically draw upon this context.1

*  This paper was written with the support of the Helsinki Collegium for Advanced Studies and Kone foundation (2014) and with encouragement of Prof. Anneli Aejmelaeus, whom I thank for many useful suggestions. I would like to express my gratitude to Drew Longacre as well for his interesting critical questions and notes, and also to Natia Dundua, who helped me to write this article in English. I want to thank Peter Gentry, who read the paper and supported ideas expressed therein. 1  Recently, in the study of the Hebrew or the Greek Bible a special emphasis has been placed on the phenomenon of the Alexandrian library. However, special attention is paid to historical and exegetical parallels and is not (or is less) concerned with the relationship between Alexandrian Philology and the Septuagint—its creation as well as its subsequent text history. See M.R. Niehoff, Jewish Exegesis and Homeric Scholarship in Alexandria (Cambridge: Cambridge University Press, 2011); M.R. Niehoff (ed.), Homer and the Bible in the Eyes of Ancient Interpreters (Jerusalem Studies in Religion and Culture 16; Leiden: Brill, 2012); N.L. Collins, The Library in Alexandria and the Bible in Greek (VTSup 82; Leiden: Brill, 2000); S. Honigman, The Septuagint and Homeric Scholarship in Alexandria: A Study in the Narrative of the Letter of Aristeas (London: Routledge, 2003); J. Wyrick, The Ascension of Authorship: Attribution and Canon Formation in Jewish, Hellenistic, and Christian Traditions (Harvard Studies in Comparative Literature 49; Cambridge, MA: Harvard University Press, 2004).

16

Anna Kharanauli

Alexandrian Philology and the Work of a Grammarian2 Five generations of philologists worked in the Library of Alexandria to edit and interpret the texts of the poets and prosaists. They called themselves “grammarians”. According to the definition of grammar by Dionysius Thrax:3 Γραμματικὴ ἐστιν ἐμπειρία τῶν παρὰ ποιηταῖς τε καὶ συγγραφεῦσιν ὡς ἐπὶ τὸ πολὺ λεγομένων. Μέρη δὲ αὐτῆς ἐστιν ἕξ· πρῶτον ἀνάγνωσις ἐντριβὴς κατὰ προσῳδίαν, δεύτερον ἐξήγησις κατὰ τοῦς ἐνυπάρχοντας ποιητικούς τρόπους, τρίτον γλωσσῶν καὶ ἱστοριῶν πρόχειρος ἀπόδοσις, τέταρτον ἐτυμολογίας εὕρεσις, πέμπτον ἀναλογίας ἐκλογισμός, ἕκτον κρίσις ποιημάτων, ὅ δὲ καλλιστόν ἐστι πάντων τῶν ἐν τῇ τέχνῃ.4

We are aware of the criteria applied by grammarians during the criticism of poetical productions (κρίσις ποιημάτων): ἀνάγνωσις—reading aloud—is the first requirement. Here Dionysius means mainly reading according to the rules of prosody which apart from the intonational-tonic modulation of the voice also meant other things such as the division of words, considering diacritical signs and reading various poetic meters. Thus, ἀνάγνωσις is related to the understanding of the text, which is also embodied in the stem (γνῶσις) of this term.

2  Selected Bibliography: R. Devreesse, Introduction à l’étude des manuscrits grecs (Paris: Klincksieck, 1954); E.G. Turner, Greek Papyri: An Introduction (Princeton, N.J.: Princeton University Press, 1968); R. Pfeiffer, History of Classical Scholarship from the Beginnings to the End of the Hellenistic Age (Oxford: Clarendon Press, 1968); H. Erbse, “Über Aristarchs Iliasausgaben”, Hermes 87 (1959) 275–303; H. Hunger/O. Stegmueller/H. Erbse, Die Textüberlieferung der antiken Literatur und der Bibel (München: Deutscher Taschenbuch, 1975); E. Pöhlmann, Einführung in die Überlieferungsgeschichte und in die Textkritik der antiken Literatur (1 vol.; Darmstadt: Wissenschaftliche Buchgesellschaft, 1994); K. Nickau, Untersuchungen zur textkritischen Methode des Zenodotos von Ephesos (Untersuchungen zur antiken Literatur und Geschichte 16; Berlin: de Gruyter, 1977); B. Neuschäfer, Origenes als Philologe (Schweizerische Beiträge zur Altertumswissenschaft, H. 18; 1–2 vol.; Basel: Reinhardt, 1987); S. Matthaios/F. Montanari (ed.), Ancient Scholarship and Grammar: Archetypes, Concepts and Contexts (Berlin: de Gruyter, 2011). Of course, the history of the philology—and particularly the textual criticism—from the Hellenistic period to the Late Antiquity is very nuanced and complex, so the picture that I aim to propose here is simplified to a certain extent. 3  Dionysios Thrax (ca. 170 – ca. 90 b.c.e.), one of latest grammarians who worked in the Alexandrian Library, studied in Alexandria in the school of Aristarchus of Samothrace. The definition of grammar in his Τέχνη Γραμματική as well as its subsequent comments determined the development of philology throughout the Middle Ages and was applied for the editing and hermeneutics of both pagan and Christian texts. 4  G. Uhlig (ed.), Dionysii Thracis, Ars Grammatica: qualem exemplaria vetustissima exhibent subscriptis discrepantiis et testimoniis (Lipsiae: Teubner, 1883 [reprint Hildesheim: Olms, 1979]), 5–6; Pfeiffer, History of Classical Scholarship, 268–72.

Origen and Lucian in the Light of Ancient Editorial Techniques

17

According to the commentators of Dionysius, the terms ἐξήγησις, εὐρέσις, ἀπόδοσις all mean the study of the language (γλῶσσων καὶ ἱστοριῶν, ἐτυμολογίας and ἀναλογίας) and style (ποητικοὺς τρόπους) of the text, the ultimate goal of which is literary criticism—interpretation of literary works (κρίσις ποιημάτων). So the grammarian, philologist or critic (as the Pergamonian philologists referred to themselves) was both textual and literary critic. Literary criticism, i.e. finding out the divinely inspired writer’s5 οἰκονομία (plan, arrangement) or προνοῖα (foreknowledge, foresight), is practically, first of all, related to the establishment of the author’s text.6 And the establishment of the author’s text, in turn, is a task, which could be “accomplished not only through internal and linguistic analysis, but also through an extensive collection and collation of authoritative manuscripts”.7 Several terms such as ἔκδοσις, σημεῖα, ὑπομνήματα, διόρθωσις were connected with the process of working with text. The term ἔκδοσις (Latin editio, German Ausgabe) meant the edition of the text.8 These editions were either anonymous or named after their authors (e.g. πλείονας ἐκδόσεις τῆς Ἀρισταρχείου διορθώσεως),9 being more or less accepted and wide-spread (e.g. κοιναὶ ἐκδόσεις). However, an ἔκδοσις of the Alexandrians did not include just the text. According to recent studies, an ἔκδοσις is a copy of a text, with critical signs (σημεῖα), and marginal and interlinear annotations (ὑπομνήματα).10 There were also other concepts related to ἔκδοσις: προέκδοσις (preliminary or “previous edition”)11

5  “Apotheosis of Homer” (British Museum in London), a marble stela made in Alexandria, is a symbol of the approach towards the poet in Alexandria. For other examples, see F. Pontani, “‘Only God Knows the Correct Reading!’ The Role of Homer, the Quran and the Bible in the Rise of Philology and Grammar”, in M.R. Niehoff (ed.), Homer and the Bible in the Eyes of Ancient Interpreters 43–86, on pp. 54–5 and pp. 65–7. 6  About the growing interest towards the author’s text, see Turner, Greek Papyri, 106–10. The concept of οἰκονομία in Scholia on Homer see Neuschäfer, Origenes als Philologe, II, 396, notes 234–6. 7  Pontani, “‘Only God Knows the Correct Reading!’”, 45. 8  It is known that ἔκδοσις had existed even before Alexandria, e.g. ἐκδόσεις κατὰ πόλεις, ἐκδόσεις κατ᾿ ἄνδρα, ἡ κατ’ Ἀντίμαχον. 9  The same grammarian could be the author of several ekdoseis of the same text. 10  F. Montanari, “Alexandrian Homeric Philology: The Form of the Ekdosis and the Variae Lectiones”, in M. Reichel/A. Rengakos (ed.), Epea Pteroenta: Beitrage zur Homerforschung. Festschrift für Wolfgang Kullmann zum 75 (Stuttgart: Franz Steiner, 2002) 119–40; F. Montanari, “Correcting a Copy, Editing a Text: Alexandrian Ekdoseis and Papyri”, in F. Montanari/L. Pagani (ed.), From Scholars to Scholia: Chapters in the History of Ancient Greek Scholarship (TCSV 9; Berlin: De Gruyter, 2011) 1–15; Nickau, Untersuchungen zur textkritischen Methode, 18. Due to such intricate structure it is no wonder that in different contexts this term may imply only its particular element(s)—e.g. text or/and correction or/and interpretation, cf. Erbse, “Über Aristarchs Iliasausgaben”, 291; Pfeiffer, History of Classical Scholarship, 216–17. 11 Pfeiffer, History of Classical Scholarship, 141–2.

18

Anna Kharanauli

and ἐπέκδοσις (the “re-issued recension”, “a revised text drawn up ... by a pupil ... from material left by the master”).12 A system of critical signs called σημεῖα13 apparently had existed even before Alexandria. Zenodotus applied this system in the most simple and easily understandable way: the only sign he used was obelus, which marked a suspicous text, and it did not require any further explanation. With later philologists critical signs not only increased in number but also became ambiguous, providing no information to an ordinary reader without the help of the commentaries.14 In Aristarchus’s case this circumstance—the ambiguity of signs—allows Pfeiffer to assume that “marginal sigla in Aristarchus’s ἐκδόσεις were the link to his ὑπομνήματα.”15 The commentaries—ὑπομνήματα16—apparently represented brief notes, similar to the modern apparatus criticus. They consisted of: 1. σημεῖον, 2. the reading under question, i.e. lemma17 and 3. its alternative reading(s) (the preferred reading—διόρθωσις—of a grammarian among them) with brief18 or large explanatory notes dealing with the problems of the author’s language, style and especially vocabulary, as well as the value of the manuscripts and their paleographic features. It is assumed19 that in the case of scrolls ὑπομνήματα were included in separate volumes, while in the case of codices ὑπομνήματα were placed in the margins of the texts. Therefore, ὑπομνήματα and the text with σημεῖα are meant to exist simultaneously; one cannot be understood without the other. Based on ancient witnesses Montanari describes the whole process of the preparation of an ekdosis as follows:

12 Pfeiffer, History of Classical Scholarship, 217; see also Montanari, “Alexandrian Homeric Philology”, 126–7. 13  See M. Stein, “Kritische Zeichen”, RAC 22 (2008) 133–63; F. Schironi, “The Ambiguity of Signs: Critical Semeia from Zenodotus to Origen”, in M.R. Niehoff (ed.), Homer and the Bible in the Eyes of Ancient Interpreters 87–112. 14  For instance, an asterisk as denoting a repeated line (versus iterati) already in Aristarchus’s case required a comment to indicate the place with the same line and to specify where it was more appropriate (Pfeiffer, History of Classical Scholarship, 178). Apart from this, in the works of later philologists an asterisk had various meanings, namely, the beginning or the end of a verse in Lyrics, or a change in meter; an asterisk was also applied at the end of a paragraph, etc, see Stein, “Kritische Zeichen”; Devreesse, Introduction à l’étude des manuscrits grecs, 74, 87. 15 Pfeiffer, History of Classical Scholarship, 218. Visualisation of this idea see Schironi, “The Ambiguity of Signs”, 94. 16  See Erbse, “Über Aristarchs Iliasausgaben”, 278–86; Pfeiffer, History of Classical Scholarship, 29, 161, 175, 212–33; Turner, Greek Papyri, 112–24; Schironi, “The Ambiguity of Signs”, 92–96. 17  For methods of distinguishing the lemma from the commentaries and the accuracy of the use of σημεῖα see Turner, Greek Papyri, 114–16. 18  E.g.: ‘μή σε’, ἀλλὰ ‘μή τι’ αἱ Ἀρισταρχου καὶ αἱ ἄλλαι σχεδὸν πᾶσαι διορθώσεις (Schol. A 522), see Pfeiffer, History of Classical Scholarship, 216. 19  See esp. Schironi, “The Ambiguity of Signs”, 92–4.

Origen and Lucian in the Light of Ancient Editorial Techniques

19

A philologist chose, according to his own preferences, an exemplar that he considered suitable as a basis for his work. When he rejected the text, he noted in the place in question the preferred reading in the free spaces or between the lines. His own text resulted from the original text chosen together with the changes suggested and contained in the paratext created... This copy, bearing the traces of the work of diorthosis, resulted materially in the philologist’s own ekdosis of Homer. This was his own personal copy, it bore his name for purposes of identification and contained the fruit of his work and insights; ekdosis in that it was ekdotheisa, i.e. available for consultation by scholars, poets and intellectuals.20

The entire process could be described even in more detail. Generally, the need for an ekdosis is determined by differences between different sources—diaphony— both in oral and written tradition. The inventory of diaphony, first of all, implied collecting the available and/or important sources—simple copies of the text and the manuscripts (ἀντίγραφα / ἀπόγραφα / μετάγραφον) of both the previous ἐκδώσεις and ὑπομνήματα. The copies were classified according to their city of origin, the editor, or their age and quality.21 The next phase of the preparation of ekdosis was an evaluation of the available manuscripts and choosing of the basic text among them. Afterwards, the collected and selected sources were collated: texts of manuscripts or textual data scattered in ὑπομνήματα were compared with each other.22 As a result of collation the textual variants were identified. Oral or written tradition of ὑπομνήματα allowed the grammarians of every subsequent generation not only to identify the textual variant, but also to find out (or at least suppose) the reason why a particular variant had been formed. The analysis of variants used to commence by defining a type of variant, i.e. defining whether there were unintentional errors or intentional interventions in the text. The collation of sources leads to διόρθωσις. Διόρθωσις can be the result of proof-reading (checking and correcting a scribe’s work)23 as well as philologists’ “own conjectures” or corrections “from better readings in other manuscripts”.24 The procedure of dealing with readings considered doubtful, unauthentic, or secondary was called ἀθετεῖν. It did not mean erasing a questionable reading automatically.25 Such a doubtful reading remained in the text but was marked by a critical sign (obelus) and could be supplied with a commentary.

20  Montanari, “Correcting a Copy”, 2–3. 21 Turner, Greek Papyri, 110. 22 Turner, Greek Papyri, 93–6; Erbse, “Über Aristarchs Iliasausgaben”, 287–8. 23  Montanari, “Alexandrian Homeric Philology”, 122. 24  See for example Pfeiffer about Zenodotus, History of Classical Scholarship, 110. 25 Nickau, Untersuchungen zur textkritischen Methode, 6–19 and 25–30; Montanari, “Alexandrian Homeric Philology”, 122.

20

Anna Kharanauli

A main difference between the grammarians was whether one should adopt εὐλαβεία (‘caution, reverence’)26 or τόλμη τῆς διορθώσεως (‘audacity of conjectural criticism’). In the history of textology, a tendency developed in favor of the former.27 Since as early as Aristophanes of Byzantium’s time (ca 265–190 or 258–180 b.c.e.), μήτε προσθεῖναι μήτ’ ἀφελεῖν (‘do not add or remove anything’)28 was at least the theoretical working principle. Semantic or stylistic correction of the text was not aimed at, either in Hellenistic (at least since Aristarchus) or in Roman and Late Byzantine philology. To sum up, according to the recent studies an ekdosis was a unique manuscript of a grammarian, containing the text and scholia written on free spaces (margins and between lines). Those variant readings that were commented on in scholia were taken from manuscripts against which an original manuscript was collated. Ekdosis was a tool for literary criticism and was intended for scholars (for the grammarian himself as well as for his pupils), unlike the so-called vulgata which was meant for public use.29 An exact copy of this ekdosis might have never been made, but it may have been used as working material for the production of a manuscript, which would have been adapted in accordance with various needs of scholars or ordinary readers.30 The reworking of the original form of ekdosis, its extension or reduction, as it seems, was accomplished by the pupils of the author of an ekdosis. Afterwards, the original itself—the ekdosis—would become useless and, therefore, disappear.31

Are the ‘Recensions’ of the Septuagint ‘Ekdoseis’? Such philological activities are the background against which the Septuagint translation was carried out and against which its textual history developed. It 26  For example, Didymus blames Aristarchus for his excessive caution (περισσῆς εὐλαβείας) against an incorrect reading attested in numerous manuscripts (Schol A I 222), see Neuschäfer, Origenes als Philologe, II, 394, n. 225. 27  “That the result of his [i.e., an Alexandrian grammarian’s] work was a completely new exemplar, containing his own text, i.e. the entire work re-written according to the way he believed was appropriate and right, is not a widely held view today” (Montanari, “Alexandrian Homeric Philology”, 120). 28  See C. Schäublin, “Μήτε προσθεῖναι μήτ’ ἀφελεῖν”, MH 31 (1974) 144–9. 29  See Turner, Greek Papyri, 94–6, 112–13, 118; Pfeiffer, History of Classical Scholarship, 215; Erbse, “Über Aristarchs Iliasausgaben”, 303. 30  In his article Montanari clearly shows how ekdosis is being transformed and, first of all, how comments disappear from Aristarchus’s selected variants in scholia (i.e. the comment is lost and only the selected variant is left). See F. Montanari, “The Fragments of Hellenistic Scholarship”, in G.W. Most (ed.), Collecting Fragments / Fragmente Sammeln (Göttingen: Vandenhoeck & Ruprecht, 1997) 273–88. 31  E.g. the ekdosis by Zenodotus was not available for Aristarchus himself.

Origen and Lucian in the Light of Ancient Editorial Techniques

21

is interesting that the sources reflecting on the creation of the Septuagint—Letter of Aristeas, Josephus, Philo—assess the whole translation-process from the perspective of Alexandrian philology,32 but now I will skip these parallels, as well as the parallels between Alexandrian philology and the ongoing activity in the Judean milieu, that were aimed not only at the Hebrew text, but also at the correction of the Septuagint and creation of new Greek translations. Instead, I will try to figure out how the textual history of the Septuagint developed in the Christian Church and present this history in the light of ancient editorial technique. As far as we know, the first text-critical approach to the Scriptures is attested in the second half of the 2nd century c.e. According t o Eusebius,33 several persons worked on the text of the Scriptures. They were Christians fascinated by the philosophy of the Greeks—Theodotus the Tanner and his followers. Eusebius mentions some characteristics of their activities: the members of the circle had 32  See Sylvie Honigman: “The depiction of the origins of the LXX in Book of Aristeas is deeply influenced by both the practice and the ideology of Homeric scholarship in the royal library” (Honigman, The Septuagint and Homeric Scholarship in Alexandria, 119). This influence is evidenced by concepts and terms used in the sources describing the process and evaluation of the translation, such as ἀνάγνωσις (see above p. 16 and A. Van der Kooij, “Zur Frage der Exegese im LXX-Psalter: Ein Beitrag zur Verhältnisbestimmung zwischen Original und Übersetzung”, in A. Aejmelaeus/U. Quast [ed.], Der Septuaginta-Psalter und seine Töchterübersetzungen: Symposium in Göttingen 1997 [MSU 24; Göttingen: Vandenhoeck & Ruprecht, 2000] 366–79, on p. 374); διασάφησις (explanation, interpretation): ἐτρέποντο πρὸς τὴν ἀνάγνωσιν καὶ τὴν ἑκάστου διασάφησιν (Pelletier, A. [ed.], Lettre d’Aristêe à Philocrate [Sources chrétiennes 89; Paris: Les éditions du CERF, 1962], 305); ἀντιβολή (collation): σύμφωνα ποιοῦντες πρὸς ἑαυτοὺς ταῖς ἀντιβολαῖς (Lettre d’Aristêe, 302). The principle of the translators is expressed by the known formula: οὔτε προσθεῖναι τις οὐδὲν, οὕτε ἀφελεῖν αὐτῶν, οὕτε μεταθεῖναι τετόλμηκεν (Niese, B. [ed.], Flavii Iosephi Opera 5, De Judaeorum vetustate, sive contra Apionem libri 2 [Berlin: Weidmann, 1889], 1, 42); μήτ᾿ ἀφελεῖν τι μήτε προσθεῖναι ἤ ­μεταθεῖναι δυναμένους (Colson F.H. [ed.], Philo, De Vita Mosis, I–II [LCL 289, vol. 6; Cambridge, MA: Harvard University Press, London: William Heinemann, 1984], II, 6, 34). This formula was used not only in regard to translators but also regarding those ones who would make any changes in the text in future: ἐκέλευσαν διαράσασθαι ... εἴ τις διασκευάσει προστιθεὶς ἤ μεταφέρων τι τὸ σύνολον τῶν γεγραμμένων ἤ ποιούμενος ἀφαίρεσιν (Lettre d’Aristêe, 311). For more about this formula in Aristeas, Philo, Josephus, see W.C. Van Unnik, “De la regle Μήτε προσθεῖναι μήτ’ ἀφελεῖν dans l’histoire du canon”, Vigilae Christianae 3 (1949), 16–18. Τhe concept of the divine inspiration underlined in relation to the seventy translators is also familiar to the Alexandrian philologist. The focus on the authenticity of the Hebrew manuscript that was employed as the Vorlage for translation is an especially interesting parallel with the Alexandrian philology as well. 33  K. Lake (ed.), Eusebius: Ecclesiastical History, Books I–V (LCL 153; London: William Heinemann, New York: G.P. Putnam, 1926), (HE) 5.28.16–18, on p. 522; E. Schwartz (ed.), Eusebius, Kirchengeschichte (GCS 9/1; Leipzig: Hinrichs, 1903). See A. Grafton/M. Williams, Christianity and the Transformation of the Book: Origen, Eusebius, and the Library of Caesarea (Cambridge, MA: Belknap Press of Harvard University Press, 2008), 190–1; B. Metzger, The Text of the New Testament: Its Transmission, Corruption, and Restoration (Oxford: Oxford University Press, 1968), 150–1.

22

Anna Kharanauli

produced their own antigrapha, which were copied and multiplied by their pupils; the antigrapha of the same editor at various times differed from each other; the antigrapha were a result of diorthosis—a corrected text based on the editors’ own considerations (τὰ ἑαυτοῦ δοκοῦντα). Traces of such ekdoseis do not appear in the extant Septuagint manuscripts. Despite this, this information is still very interesting in order to conceptualise the text-history of the Septuagint. Firstly, it reflects the procedure of correcting the text within one circle of scholars and the possibility of a variety of such corrections. Secondly, it clearly indicates the reason why such corrections were unacceptable for the church and particularly for Eusebius, the pupil of the Alexandrian philologist Origen. The reason is the type of correction—conjecture, i.e. emendatio ope ingenii. A watershed in the text history of the Septuagint as the Christian Old Testament text is Origen’s Hexapla.34 This phase of Christian textual criticism lasted approximately one century starting with Origen (d. 254 c.e.) and being completed probably with the Constantinian Pandect Bibles (ca 330 c.e.). Jerome’s information still remains the only (though vague) landmark for characterisation of the early Christian phase of the Septuagint’s text history: Alexandria et Aegyptus in Septuaginta suis Hesychium laudat auctorem, Constantinopolis usque Antiochiam Luciani martyris exemplaria probat, mediae inter has prouinciae Palaestinos codices legunt, quos ab Origene elaboratos Eusebius et Pamphilus uulgauerunt; totusque orbis hac inter se trifaria uarietate conpugnat.35

This is the well-known passage based on which text critics of the Septuagint look for the text forms related to Origen, Lucian and Hesychios, forms which are typologically similar to each other. Today, all these forms are called ‘recensions’, the term that means an intentional change of the text according to some principle. But is it feasible to call such textual forms of the Scripture a ‘recension’ if we consider biblical philology in the 3rd–4th centuries to be the continuation of Alexandrian philology? Another question is: Were those textual forms that had been discussed by Jerome really typologically homogeneous? Let us examine the much-discussed problems from a new perspective and try to deal with the trifaria varietas within the context of Alexandrian philology.

34  However, papyrus Ra 967, which seems to have been copied in Christian circles, supposedly in Alexandria, in the late 2nd or early 3rd century, already has traces of philological work, manifested in the influence of the Masoretic text type; see I. E. Lilly, Two Books of Ezekiel: Papyrus 967 and the Masoretic Text as Variant Literary Editions (VTSup 150; Leiden: Brill, 2012), 257–61. 35  Praefatio in Paralipomena, see D. De Bruyne (ed.), Prefaces to the Latin Bible (Introductions by Pierre-Maurice Bogaert and Thomas O’Loughlin; Studia Traditionis Theologiae 19; Turnhout: Brepols, 2015), 30.

Origen and Lucian in the Light of Ancient Editorial Techniques

23

As generally recognised, in relation to Origen, Lucian and Hesychius trifaria varietas can be understood as: 1. the exemplars which contain the texts accomplished by Origen, Lucian and Hesychios themselves; or 2. the exemplars based on the ekdoseis of the texts of Origen, Lucian and Hesychius which were created by their pupils; or 3. the exemplars containing the texts, which represent even a later stage of the text history, a mixture of the text-forms over a long time of textual transmission. According to Jerome’s letter 106 to Sunnia and Fretela, at his time the textform, which by Origen, Eusebius and all other commentators was called koine, was widespread in Constantinople and Antioch and was attributed to Lucian.36 As for the text related to Origen, in one case he says that it is the Septuagint, which he has found in the Hexaplaric codices (ἑξαπλοῖς codicibus),37 and in the other case (Praefatio in Paralipomena) that the codices, which were widespread in Palestine, were produced by Origen and edited by Eusebius and Pamphilus.38 In letter 106 Jerome characterises in more detail the following two textual types of the Septuagint: Κοινή and Septuaginta represent one and the same text, however the first is an older edition that has been corrupted in the transmission process, while the other is included in the Hexapla and represents the translation by the seventy translators in its pure and immaculate form that could be found among the manuscripts of erudites.39 Likewise, Jerome discusses the corrupt vs. pure Septuagint in letter 112 to Augustine. Here he contrasts, on the one hand, the emended or, better, corrupted editions of Origen with asterisks and obeli and, on the other hand, the ‘Septuagint’. He writes that it is already at his time impossible to find the authentic edition of the seventy translators among the manuscripts used in the Church and

36  “.... aliam esse editionem, quam Origenes et Caesariensis Eusebius omnesque Graeciae tractatores κοινήν (/κοινά)—id est communem—appellant atque uulgatam et a plerisque nunc Λουκιάνειος dicitur”. I. Hilberg (ed.), Sancti Eusebii Hieronymi Epistulae. Pars II; Epistulae LXXI–CXX (CSEL 55; Vienna/Leipzig: Tempsky/Freytag, 1912), 248. 37  “Aliam Septuaginta interpretum, quae et in ἑξαπλοῖς codicibus repperitur” (Ep. 106,2; CSEL 55, 248). About the Septuagint in the Hexaplaric codices, “ἑξαπλοῖς codices”, see O. Munnich, “Les Hexaples d’Origène à la lumière de la tradition manuscrite de la Bible Grecque”, in G. Dorival/A. Le Boulluec (ed.), Origeniana Sexta: Origen et la Bible. Actes du Colloquium Origenianum Sextum, Chantilly, 1993 (Leuven: Peeters, 1995) 167–85, on pp. 174–5. 38  See above, Praef. in Par., in D. De Bruyne (ed.), Prefaces to the Latin Bible, 30. 39 “Κοινὴ autem ista, hoc est Communis, editio ipsa est quæ et Septuaginta. Sed hoc interest inter utramque, quod κοινὴ pro locis et temporibus et pro uoluntate scriptorum, uetus corrupta editio est. Ea autem, quæ habetur in Ἑξαπλοῖς et quam nos uertimus, ipsa est, quæ in eruditorum libris incorrupta et inmaculata Septuaginta Interpretum translatio reseruatur” (Ep.106,2, CSEL 55, 249).

24

Anna Kharanauli

that these manuscripts are full of the additions borrowed from another translation made by a Jewish man, Theodotion.40 In the same sources Jerome clarifies the essence of two editions related to Origen. The first is the Hexapla. In Praefatio in Paralipomena he provides a description of this edition: In his work Origen composed (conposuit) manuscripts (exemplaria) of four editions (i.e. Septuagint, Aquila, Symmachus and Theodotion) writing a single word on a separate row. As Jerome notes, in such editions, a different variant of any edition is clearly visible in comparison with the homogeneity of the rest. The other type of edition related to Origen represents rather a mixed text. The edition by Theodotion has been mixed into the edition of the Septuagint and pluses and minuses are marked by obelus and asterisk.41 Such treatment of a text is regarded by Jerome as audacity and thus, seemingly, he

40  From this letter one can assume that Jerome knew the meaning of the signs and that for him the authentic Septuagint is the one which lacks the asterisked passages: “Quod autem in aliis quaeris epistulis, cur prior mea in libris canonicis interpretatio asteriscos habeat et uirgulas praenotatas et postea aliam translationem absque his signis ediderim. ... Illa enim interpretatio septuaginta interpretum est et, ubicumque uirgulae, id est obeli, sunt, significatur, quod Septuaginta plus dixerint, quam habetur in Hebraeo, ubi autem asterisci, id est stellae praelucentes, ex Theodotionis editione ab Origene additum est. ... Et miror, quomodo septuaginta interpretum libros legas non puros, ut ab eis editi sunt, sed ab Origene emendatos siue corruptos per obelos et asteriscos et Christiani hominis interpretatiunculam non sequaris, praesertim cum ea, quae addita sunt, ex hominis Iudaei atque blasphemi post passionem Christi editione transtulerit. Uis amator esse uerus septuaginta interpretum? Non legas ea, quae sub asteriscis sunt, immo rade de uoluminibus, ut ueterum te fautorem probes. Quod si feceris, omnes ecclesiarum bibliothecas condemnare cogeris. Uix enim unus aut alter inuenietur liber, qui ista non habeat” (Ep. 112,19, CSEL 55, 389). Jerome’s assessment of the New Testament text which has been related to Lucian and Hesychios is also notable. Here too, his attitude towards amendments made by Lucian and Hesychius is sharply negative: “Praetermitto eos codices, quos a Luciano et Hesychio nuncupatos paucorum hominum adserit peruersa contentio, quibus utique nec in Ueteri instrumento post LXX interpretes emendare quid licuit, nec in Nouo profuit emendasse, cum multarum gentium linguis scriptura ante translata doceat falsa esse quae addita sunt” (Praefatio in Evangelio, in D. De Bruyne (ed.), Prefaces to the Latin Bible, 154). 41  He mentions such an edition in his other writings too, e.g. in Ep. 112 (see above, n. 40), in Praefatio in Pentateuchum: “Origenis me studium prouocavit, qui editioni antiquae translationem Theodotionis miscuit, asterisco et obelo, id est stella et ueru (/uirgula), opus omne distinguens, dum aut inlucescere facit quae minus ante fuerant aut superflua quaeque iugulat et confodit” (Bruyne [ed.], Prefaces to the Latin Bible, 7); in Praefatio in Job: “Quasi non ... omnia ueteris instrumenti uolumina Origenes obelis asteriscisque distinxerit, quos uel additos uel de Theodotione sumptos, translationi antiquae inseruit, probans defuisse quod additum est” (Bruyne [ed.], Prefaces to the Latin Bible, 38); and in In Isaiam, I,2,22, where he considers Aquila and not Theodotion as an author of added texts marked by asterisk: “Hoc praetermisere LXX, et in graecis exemplaribus ab Origene sub asteriscis de editione Aquilae additum est” (CSEL 73, 39,2-4).

Origen and Lucian in the Light of Ancient Editorial Techniques

25

highlights an activity that looked extraordinary in comparison with the traditional philological work.42 Thus, Jerome juxtaposes the editions: I. according to their localization (the texts widespread in different geographical areas); II. according to their attribution (1. the edition of the seventy translators—the koine or vulgar—which in his time was related already to Lucian, and 2. two editions of Origen—the Hexapla and the mixed texts); and III. according to their authenticity (pure Septuagint and Septuagint mixed with translations of Theodotion and Aquila). Hence, it seems to me that Jerome means different forms of the text edition: 1. the traditional one (vulgar text); 2. the Hexapla; 3. the text that is the part of the Hexapla— presumably the text of its fifth column (in relation to Origen);43 and 4. the mixed text equipped with asterisks and obeli. In short, what we get from Jerome’s witnesses is that the manuscripts of his time contained texts of various traditions, in almost none of which was the Septuagint preserved authentically. We also learn that the main differences of the text of the Septuagint are related to those influences, which we call Hexaplaric. Therefore, the witnesses of Jerome confirm what we already know from the surviving early manuscripts and fragments of the Septuagint—the diversity of the textual forms containing Hexaplaric material.

Origen’s Ekdosis What was the reason for such change of the Septuagint, the inspired text? Barthélemy notes, “Jamais non plus personne n’exerça sur l’histoire de son texte [i.e. Greek Bible] une influence aussi décisive, ni aussi catastrophique.” Usually citations of Barthélemy end here, and the main responsibility for the textual variety of the Septuagint turns out to be Origen’s. But actually Barthélemy continues, “Et pourtant Origène était un érudit d’une honnête scrupuleuse, animé d’un profond sens de la tradition.”44 What does Origen’s “sens de la tradition” mean? And why are the changes to the text and his “sens de la tradition” incompatible with each other? 42  “Et certe Origenes non solum exemplaria (exempla//exapla) conposuit quattuor editionum, e regione singula uerba describens, ut unus dissentiens, statim ceteris inter se consentientibus arguatur, sed, quod maioris audaciae est, in editione Septuaginta Theodotionis editionem miscuit, asteriscis designans quae minus fuerint, et uirgulis, quae ex superfluo uidebantur adposita” (In Par, in Bruyne (ed.), Prefaces to the Latin Bible, 30). 43  “Septuaginta” must be designated the 5th, Septuagint, o’ column. See Munnich, “Les Hexaples d’Origène”, 175. 44  D. Barthélemy, “Origène et le texte de l’Ancien Testament”, in Études d’histoire du texte de 1’Ancien Testament (OBO 21; Freibourg/Göttingen: Universitätsverlag/Vandenhoeck & Ruprecht, 1978) 203–17, on p. 203.

26

Anna Kharanauli

Origen is a representative and successor of the Alexandrian grammar school. This is true also literally, since for some period he was a teacher of grammar in Alexandria. Besides the biographical note, “the Alexandrian roots” of Origen are shown in various details of his life and work: in creating a working infrastructure (i.e. founding a library in Caesarea, creating a team of assistants and supporters), in collecting the working material (manuscripts), in preparing excerpts from different texts and in the mode of commenting them, as well as in the terminology used in his writings. Based on these well-approved work organisation and philological methods Origen establishes the Hexapla, the innovative form of the collation—synopsis of various sources—which can be considered as a new form of ekdosis.45 Instead of the traditional form (i.e. the main text and marginal notes with variant readings) there are parallel columns containing the continuous texts. Besides, instead of the old content of the traditional ekdosis (some alternatives of the same texts) the new form of ekdosis reflects the new content—different Greek translations against its Hebrew original. But the Hexapla was not only a simple form of comparative research, it was a multifunctional work: in contrast to the traditional ekdosis intended only for the particular, inner circle of grammarians and reflecting their specific interests, the Hexapla was material for exegesis that could be used for different purposes and in different ways by its author and by others.46 What do Origen’s “propres plans” mean in regard to the Hexapla as the “collection d’informations”? In the case of both Alexandrian philology and Origen “collection d’informations” is the first stage of text criticism: “Origène avait fait des hexaples un merveilleux instrument permettant de porter un regard critique sur le Bible grecque”.47 The final goal of the Hexapla was the exegesis of Scripture on the basis of all its sources and not only of the Greek text. The specificity of the information chosen by Origen is conditioned and, moreover, required by that historical-theological context in which he worked and which is wider than the context of his Christian predecessors or the context of the church fathers 45  This innovation does not oppose the Alexandrian tradition, on the contrary, it is the further developement of it. See “Schon das Problem, das Origenes in der Hexapla sich stellte, ihn als würdigen Epigonen der grossen hellenistischen Philologen des 3. und 2. Jahrhundert legitimiert” (E. Schwarz, “Zur Geschichte der Hexapla”, in Zum Neuen Testament und zum frühen Christentum [Berlin: de Gruyter, 1963] 183–91, on p. 191). 46  “Origène avait conçu les hexaples comme une ample collection d’informations” (Barthélemy, “Origène et le Text de AT”, 203); “Les Hexaples constituent donc un remarquable instrument de travail, une précieuse synopse qu’ Origène s’était fait constituer sur ses propres plans” (Barthélemy, “Origène et le Text de AT”, 212). See also D. Barthélemy, “La place de la Septante dans l’Eglise” in Études d’histoire du texte de l’Ancien Testament (OBO 21; Fribourg/Gottingen: Universitätsverlag/Vandenhoeck & Ruprecht, 1978) 111–26, on p. 114. 47  Barthélemy, “Origène et le Text de AT”, 203.

Origen and Lucian in the Light of Ancient Editorial Techniques

27

working far from the centres of Judaism. For Judeo-Christian dialogue or polemics he had to accomplish a specific aim—his “propres plans”—to present texts in commensurable form.48 It is known from Origen himself how “professional” his attitude is while collecting and evaluating the manuscripts (not only of the Septuagint), for example, that he finds these manuscripts in different places49 and indicates this information in his comments. Origen organises and evaluates the sources in the traditional manner as well: he distinguishes old manuscripts (e.g. ἀρχαῖα ἀντίγραφα [Sel. in Ps. 3:8, PG 12, 1129 B]) and groups them from different points of view (e.g. ἐν τοῖς πλείστοις τῶν ἀντιγράφων [in Jer. hom. 14:3, GCS 6, 107]). Origen evaluates the translations of the Three as well: ἐν τοῖς ἀκριβεστάτοις καὶ συμφωνοῦσι τοῖς Ἑβραικοῖς (in Jer. hom. 14:3, GCS 6, 107); Ἀκύλας, δουλεύων τῇ ἑβραϊκῇ λέξει (ad Afr. 2 SCh 302, 526), ἐμφατικώτερον—about Aquila (Sel. in Ps. 2:10, PG 12, 1112 B), σαφέστερον—about Symmachus (Sel. in Ps. 4:1, PG 12, 1133 B).50 The collection and evaluation of sources is followed by their collation. Origen discusses the results of the collation with the traditional termini technici. For textual corresponences and divergence he uses terms συμφωνία and διαφωνία or διαφορά. Διαφωνία, which Origen mentions in his various works, are, on the one hand, the divergences between the Greek manuscripts themselves and, on the other hand, between the Greek and the Hebrew texts. Origen often notes how hard he labours to mark in detail divergences between all these sources.51 Following the principle of analogy, well-approved by grammarians, Origen groups the examples of diaphony and in his commentaries and homilies discusses their origin in the form of assumption (εἰκὸς οὖν...[Sel. in Ps. 3:8 PG 12.1129B]). One part of the differences he explains due to errors in the Hebrew manuscripts themselves (ἔστι πλάνῃ Ἑβραίου γραφέως [Sel. in Gn. 2:4; PG 12. 97C])52 or errors of the Septuagint copists (τὰ ἀντίγραφα ἁμάρτηται [Sel. in Ps. 2:12, PG 12, 1116D]), among which some are unintentional changes (μὴ νοήσαντες [in Jer. hom 16:5, GSC 6, 137]), the results of, for example, misspelling (γραφικὸν ἁμάρτημα [in Jer. hom 15:5, GCS 6, 129]), copyist’s carelessness (ῥαθυμία [In Mt. 48  See Letter to Africanus: “We try not to be ignorant of what they have, in order that while having dialogue with the Jews we do not bring forth [to them] what is not present in their manuscripts and, at the same time, in order that we make use of what is present in their [manuscripts], even if it’s not present in our books. Being so prepared for the inquiry against them, they neither will disregard us, nor, as is their habit, will they laugh at us, the gentile believers, as at those who are ignorant of the true writing preserved by them [Jews].” Greek text: Ep. ad Afr. 9, SCh 302, 534]. 49  ε' ekdosis in Nicopolis, near Actium and ekdosis ζ' in Jar near Jericho under Antoninus, son of Severus. See P. Nautin, Origène: Sa vie et son œuvre (Paris: Beauchesne, 1977), 310, 313. 50  G. Sgherri, “Sulla valutazione origeniana del LXX”, Biblica 58 (1977) 1–28, on p. 21. 51  See also Ep. ad Afr. 4, SCh 302, 524, 530; In Joh 6:41, GSC 10, 150; See Sgherri, “Sulla valutazione”, 6–15. 52  See Sgherri, “Sulla valutazione”, 13–14.

28

Anna Kharanauli

15:14, GCS 40, 387]), etc. Another part of the differences are due to the divine inspiration of the seventy-two interpreters (κατ᾿οἰκονομίαν προσέθηκαν [Sel in Ps 2:12, PG 12 1116D]).53 In some cases Origen is hesitant between two possible explanations of the reading: τὸ ‘πρῶτον’ εἲτε μὴ νοήσαντος ἐξεῖλάν τινες τῶν γεγραμμένων εἲτε καὶ οἰκονομήσαντες ἐξελεῖν οἱ Ἑβδομήκοντα, θεὸς ἂν εἰδείη [in Jer hom. 16:5, GSC 6, 137].54 Another reason for changes is impiety (ἀσέβεια) for the Scriptures, audacity of alterations or corrections (τόλμη ἀλλοιοῦν, τόλμη τῆς διορθώσεως, πρόφασις διορθώσεως) and wicked daring of some scribes (τόλμη τινῶν μοχθηρὰ τῆς διορθώσεως γραφομένων).55 One part of the intentional changes come from grammarians and can be explained not by a wicked intention but by a wrong attitude to their work, which gives the corrector the audacity, while emending, to make additions and omissions (ἐν τῇ διορθώσει προστιθέντων ἢ ἀφαιρούντων) in his own consideration (τὰ ἑαυτοῦ δοκοῦντα) [in Mt. 15:14, GCS 40, 388].56 Now let us recall what Origen says about his working method as a textual critic (if he speaks of his own work indeed). I mean two well-known, still controversial witnesses of Origen. In the letter to Africanus (we should not forget that the ­pathos of this letter is the assertion of the primacy of the Septuagint to the addressee), Origen speaks of quantitative differences, which he marked between “our copies” and the “Hebrew” in various books (Job, Jeremiah, Genesis, Exodus) and then, of the differences in Genesis, which he had indicated by marks (σημεῖα), called obelisk and asterisk by the Greeks. Origen also clarifies the meaning of the latter: by asterisk are marked those things which are included in the

53  See also in Cant.: “we are certain that the Holy Spirit willed that the figures of the mysteries should be roofed over in the Divine Scriptures, and should not be displayed publicly, and in the open air” (translation of R.P. Lawson, Origen: The Song of Songs: Commentary and Homilies (ACW 26; New York: Newman, 1956), 74; latin translation: in Cant , Lib. I, GCS 33, 101. See also Neuschäfer, Origenes als Philologe, 136–7. 54  For Origen’s opinion about possible origin of different variants see Sgherri, “Sulla valutazione”, 13–14; Nautin, Origène, 349–53; C.P. Bammel, “Die Hexapla des Origenes: Die Hebraica Veritas im Streit der Meinungen”, Aug 28 (1988) 125–49, on pp. 130–1; Neuschäfer, Origenes als Philologe, 112–13. 55  “Dass Origenes selbst peinlich darauf bedacht ist, auch nur den geringsten Anschein “textkritischer” τόλμη zu vermeiden, hängt mit der von ihm selbst übernommenen kirchlichen communis opinio zusammen, textkritische Kühnheit resultiere aus häretischen Neigungen” (Neuschäfer, Origenes als Philologe, 395, n. 228). Concerning Origen’s attitude towards such an approach to the Scriptures and parallels with Alexandrian philologists, see Neuschäfer, Origenes als Philologe, 85–138. 56  νυνὶ δὲ δῆλον ὅτι πολλὴ γέγονεν ἡ τῶν ἀντιγράφων διαφορά, εἴτε ἀπὸ ῥᾳθυμίας ­τινῶν γραφέων, εἴτε ἀπὸ τόλμης τινῶν μοχθηρᾶς τῆς διορθώσεως τῶν γρα­φομένων, εἴτε καὶ ἀπὸ τῶν τὰ ἑαυτοῖς δοκοῦντα ἐν τῇ διορθώσει προστιθέντων ἢ ἀφαιρούν­των [in Mt. 15:14, GCS 40, 387–8]; Nautin, Origène, 348.

Origen and Lucian in the Light of Ancient Editorial Techniques

29

Hebrew, but not in “our [text]”.57 If we rely only on this passage, it is not clear whether Origen marks the distinctions between the Greek and the Hebrew texts only in the case of Genesis or also in the case of other books which he discusses in the mentioned section of the letter. Origen tells even more in Comm. in Mt. Here he contrasts the carelessness of copyists with his own careful approach toward the text, which consists in the following: While evaluating disagreements of the texts and ‘treating’ or ‘fixing’ them, he relies not on himself, but uses other editions—‘ekdoseis’—as criteria (κριτηρίῳ χρησάμενοι ταῖς λοιπαῖς ἐκδόσεσιν): 1. he keeps what was in agreement; 2. He does not dare to erase the Greek text, which is not evidenced in the Hebrew, but marks it with obelus (ὠβελίσαμεν)58 (here again, the Alexandrian understanding of ἀθετεῖν is evident—just marking the doubtful readings and not erasing them); 3. He uses asterisk (μετ’ ἀστερίσκων προσεθήκαμεν) in order to mark clearly the text corresponding to the Hebrew, which is omitted in the Septuagint. Besides, Origen underlines that he has not restored himself the text under asterisk, but has copied from other sources—from the ekdoseis of the Three, which agree with the Hebrew.59 It should be also noted that Origen’s judgement is not categorical even here—he does not consider his position as uniquely acceptable and leaves the reader free to make a decision.60 Caution towards the extant witnesses—not correcting the text and just marking the differences—as well as appellation of the sources reveals the method of documentation used by Origen,61 the method applied by Alexandrians and their successors. 57  Καὶ ἄλλα δὲ ἔστιν εὑρεῖν ἐν τῇ Γενέσει, οἷς ἡμεῖς [σημεῖα] παρεθήκαμεν τοὺς καλουμένους παρ’ Ἕλλησιν ὀβελούς, ἵν’ ἡμῖν γνώριμον ᾖ τὸ τοιοῦτον· ὡς πάλιν ἀστερίσκους τοῖς κειμένοις μὲν ἐν τῷ Ἑβραϊκῷ, παρ’ ἡμῖν δὲ μὴ εὑρισκομένοις (Ep. ad Afr. 7, SCh 302, 530–2). 58  See Neuschäfer, Origenes als Philologe, 387–8, n. 172. 59  τὴν μὲν οὖν ἐν τοῖς ἀντιγράφοις τῆς παλαιᾶς διαθήκης διαφωνίαν θεοῦ διδόντος εὕρομεν ἰάσασθαι, κριτηρίῳ χρησάμενοι ταῖς λοιπαῖς ἐκδόσεσιν·τῶν γὰρ ἀμφιβαλλομένων παρὰ τοῖς Ἑβδομήκοντα διὰ τὴν τῶν ἀντιγράφων διαφωνίαν τὴν κρίσιν ποιησάμενοι ἀπὸ τῶν λοιπῶν ἐκδόσεων τὸ συνᾷδον ἐκείναις ἐφυλάξαμεν, καὶ τινὰ μὲν ὠβελίσαμεν ἐν τῷ Ἑβραϊκῷ μὴ κείμενα (οὐ τολμήσαντες αὐτὰ πάντη περιελεῖν), τινὰ δὲ μετ’ ἀστερίσκων προσεθήκαμεν, ἵνα δῆλον ᾖ ὅτι μὴ κείμενα παρὰ τοῖς Ἑβδομήκοντα ἐκ τῶν λοιπῶν ἐκδόσεων συμφώνως τῷ Ἑβραϊκῷ προσεθήκαμεν (in Mt.15:14, GCS 40, 388). See also Jerome, above, n. 40. 60  καὶ ὁ μὲν βουλόμενος προῆται αὐτά, ὧ δὲ προσκόπτει τὸ τοιοῦτον ὃ βούλεται (περὶ τῆς παραδοχῆς αὐτῶν ἢ μὴ) ποιήσῃ (in Mt. 15:14, GCS 40, 388). See Neuschäfer, Origenes als Philologe, 373 n. 30; Nautin, Origène, 348. 61  In his commentary of Hosea 12:4(5) Origen condemns who has an alleged motive of correction (προφάσει διορθώσεως) and corrections of the solecisms (Philocalia 8, Sc 302, 336, see Neuschäfer, Origenes als Philologe, 108 and n. 108); J. Ziegler (ed.), Duodecim prophetae. Vetus Testamentum Graecum, XIII (3 Auflage; Göttingen: Vandenhoeck & Ruprecht, 1984), 129; J. Schaper, “The Origin and Purpose of the Fifth Column of the Hexapla”, in A. Salvesen (ed.), Origen’s Hexapla and Fragments. Papers Presented at the Rich Seminar on the Hexapla, Oxford Centre for Hebrew and Jewish Studies, 25th July–3rd August 1994 (TSAJ 58; Tübingen: Mohr Siebeck, 1998) 3–15, on p. 6. Origen’s argument is Prov 22:28: “Οὐ μεταθήσεις ὅρια αἰώνια, ἃ

30

Anna Kharanauli

The use of signs—obelus and asterisk—is one more parallel between Origen and the Alexandrian philologists. Origen himself witnesses about their origin “παρ’ Ἕλλησιν” (Ep. Ad Afr. 7). Nevertheless, the use of these signs by Origen is just a formal parallel to Alexandrian philology. Firstly, unlike the Hellenic philologists, Origen uses them only for text criticism and not for literary criticism. Secondly, he changes their meaning. He marks by obelus not the doubtful readings, but the readings that are not evidenced in the Hebrew. Similarly, by asterisk he marks not the repeated line or stylistically or semantically remarkable text, but the text parallel to the Hebrew added by him to the Septuagint.62 Therefore, once more Origen can be considered as an innovator in the Alexandrian tradition.63 In sum, Origen is a continuer of the traditions of Alexandrian philology, who developes the very tradition taking into consideration the new reality. He creates a special form of ekdosis, the Hexapla, which is intended especially for one type of users—readers working in the Judean context. Thus, the Hexapla, on the one hand, is the traditional Alexandrian form of ekdosis, where the text is presented together with its variants, and on the other hand, it is an innovative form, where the alternative text is placed not in the margins or in a separate volume in the form of comments or scholia, but is presented in the form of parallel texts. As for the text-critical and literary-critical comments of the editor, instead of hypomnemata Origen creates commentaries in monograph form, in which he comments not only on his preferable readings, but also other variants. Now for the text in which, according to Origen, asterisks and obeli were used and which is usually considered to be a “recension”—a text corrected according to the Hebrew. Once more I stress the fact that conservatism as a text-critic and uncategorical judgement in the evaluation of variants are the most remarkable characteristics of Origen as an Alexandrian philologist. This conservatism ἔστησαν οἱ πρότεροί σου”, based on which he refuses the change of the text of the Septuagint— the text used in the Church (ἀθετεῖν τὰ ἐν ταῖς Ἐκκλησίαις φερόμενα ἀντίγραφα [Ep. ad Afr. 8, SCh 302, 532]). See Barthélemy, “La place de la Septante dans l’Eglise”, 192; H.B. Swete, An Introduction to the Old Testament in Greek (Cambridge: Cambridge University Press, 1914), 61. Following the analysis of Origen’s testimonies Neuschäfer concludes: “Origenes hat offenbar ... einschneidende eigene Emendationen oder Konjekturen in den LXX-Text der Hexapla [nicht] aufgenommen” (Neuschäfer, Origenes als Philologe, 99). For such caution of Origen and the authority of the Septuagint in the Church, see Barthélemy, “La Place de la Septante dans l’Eglise”, 111–26; Barthélemy, “Origène et le texte de AT”, 204–5, 214–17. 62  About the Origenian meanings of the signs see, e.g., Barthélemy, “Origène et le texte de AT”, 209; P. Gentry, The Asterisked Materials in the Greek Job (SBLSCS 38; Atlanta: Scholars Press, 1995), 2–5; P. Gentry, “The Aristarchian Signs in LXX Ecclesiastes”, in K. De Troyer/T.M. Law/M. Liljeström (ed.), In the Footsteps of Sherlock Holmes. Studies in the Biblical Text in Honour of Anneli Aejmelaeus (CBET 72; Leuven: Peeters, 2014) 463–78; Neuschäfer, Origenes als Philologe, 388 n. 175. 63  Schironi, “The Ambiguity of Signs”, 108–9; F. Schironi, “P. Genf.1.5, Origen, and the Scriptorium of Caesarea”, BASP 52 (2015) 181–223, on pp. 210–12.

Origen and Lucian in the Light of Ancient Editorial Techniques

31

is based, on the one hand, on the old principle “not to omit or add anything”, and on the other, on Origen’s respect for the divine pronoia as a philologist and a Christian.64 Therefore, from my point of view, ascribing a ‘recensional’ text to Origen seems to be an anachronistic approach to the issue. It is not acceptable that an Alexandrian philologist would have emended text with a wording of different origin (in Origen’s case the text of the Seventy with the text of the Three). It is also hard to assume that Origen, an author of the letter to Africanus and assured in the divinity of the Septuagint, the one who composed the Hexapla as a tool for polemics against the Jews, could have changed this text because of its differences from the Hebrew, changing ‘our’ (παρ’ ἡμῖν) text by the text ‘of the Jews’ (παρὰ Ἰουδαίοις). It is inconceivable that in one and the same letter65 Origen, on the one hand, speaks about the absurdity of the compromise with the Jews—which is expressed in the change of the Septuagint according to the Hebrew—and, on the other hand, explains how to use the text already corrected by him according to the Hebrew. But, in this case, how did that manuscript mentioned by Origen in Comm. in Mt. and Ep. ad Afr. look physically? From my point of view, Origen does not mean the Hexapla manuscript with the asterisked and obelised passages in one of its columns. This is due to the aim of the Hexapla itself: as it is well known, the first purpose of the Hexapla, as a synopsis of the sources was to show the relations (not only differences) between the texts. But, in such case, why was it needed to change the text of the Septuagint of the fifth column and then to mark these changes, when the form of the Hexapla—the synopsis of different texts—already fulfills its goal and shows the relations (accordances and differences) between the various versions at one glance? Why was it needed to restore the text corresponding to the Hebrew in the fifth column, when an empty space along the Hebrew already clearly indicated the lacuna in the Septuagint? The supplement of the Septuagint by the text of the Three cannot be explained with the fact that Origen wanted to make it easy for the reader to understand the Hebrew text. They could see the Greek corresponding to the Hebrew omitted in the Septuagint in three different variants of the parallel columns. Why should Origen have repeated the 64  See Schaper, “The Origin and Purpose”, 14. 65  “And, forsooth, when we notice such things, we are forthwith to reject as spurious the copies in use in our Churches, and enjoin the brotherhood to put away the sacred books current among them, and to coax the Jews, and persuade them to give us copies which shall be untampered with, and free from forgery! Are we to suppose that that Providence which in the sacred Scriptures has ministered to the edification of all the Churches of Christ, had no thought for those bought with a price, for whom Christ died; whom, although His Son, God who is love spared not, but gave Him up for us all, that with Him He might freely give us all things?” Ep. ad Afr. 6(4), English translation of F. Crombie in A.C. Coxe (ed.), The Ante-Nicene Fathers (vol. 4, Fathers of the Third Century; New York: Christian Literature Publishing, 1885) 386–92, on p. 387.

32

Anna Kharanauli

same thing already written next to it? Moreover, why was it needed to mark by the obelus the text not presented in the Hebrew, when the empty space in the Hebrew columns along the text of the Septuagint already indicated the plus in the Septuagint? Even more questions will arise while attributing to the fifth column characteristics of so-called Hexaplaric recension, e.g. the change of the proper names: why should the proper names have been corrected in the Septuagint column, when, to say nothing of the Three, the transcribed form of the Hebrew was presented also in the second column? And if Origen changed the variant of the Septuagint (let’s say, because it seemed as a damaged form of the Greek manuscript to him), why did he not mark these changes with special signs, the way he was doing it in the cases of additions and omissions? Isn’t it true that without marking the user had no longer access to the text of the Septuagint? This problem is even more acute if we suppose that Origen changed the word order. Could Origen make such change and even not mark it in the synopsis? I still repeat, if the Hexapla is a synopsis of various ekdoseis in order to make evident their accordances and differences, why would Origen make the picture obscure? Why would he make differences hardly recognisable for the reader? Why would he change the text of one of his sources, the Septuagint, especially in the “Septuagint-centred” edition of the Hexapla?66 Such correction of the Septuagint towards the Hebrew would lead to the loss of the peculiarity of the Septuagint, the peculiarity which was the demonstration of the divine pronoia for Origen. In such a case the Hexapla would be neither Septuagint-centred (as far as it no longer shows the Septuagint in its relations to the Jewish alternatives) nor easy-to-use for the Christians.67 In my opinion, in the Comm. in Mt. and Ep. ad Afr. Origen must mean an exemplar designated for scholarly use, an exemplar where quantitative relations of the Septuagint to the Jewish (Hebrew and Greek) editions would be represented by using unambiguous critical signs. Such kind of text fits the Alexandrian philological context. It shows a particular interest in the quantitative composition of the text, as the interest in numerus versuum of the text, which was important for

66  Cf. Schaper, “The Origin and Purpose”, 15. 67  For the further argumentation about the theory that manuscripts with critical marks did not represent the Hexapla see: Schironi, “P.Genf. 1.5”, 196, 214–15 and n. 73, 219, and O. Munnich, “Les révisions juives de la Septante. Modalités et fonctions de leur transmission: Enjeux éditoriaux contemporains”, in R. Gounelle/J. Joosten (ed.), La Bible juive dans l’Antiquité (HTB 9; Prahins: Éditions du Zèbre, 2014) 141–90; R.G. Jenkins, “Colophons of the Syrohexapla and the Textgeschichte of the Recensions of Origen”, in C.E. Cox (ed.), VII Congress of the International Organization for Septuagint and Cognate Studies, Leuven 1989 (Atlanta: Scholars, 1991) 261–77, on p. 269.

Origen and Lucian in the Light of Ancient Editorial Techniques

33

the average readership of Homer68 as well as for readers of the Old and New Testaments, and is reflected in stichometry supplied to the text in the manuscripts. But another thing is where were these asterisked additions placed—in the text or outside the text? Also, where did Origen put obelus—in the text, in the margins, or between the lines—upon the “plus” of the Septuagint?69 Origen says nothing about this. But we take it for granted that the additions from the Three as well as the signs (not only asterisks and obeli but also metobeli/lemnisks) were inserted in the text by Origen himself, and moreover, that this exemplar contained the text form which we call today the “Hexaplaric” or the “recension of Origen”, and which includes also those characteristics, which are nowhere noted by Origen himself—the changes in the word order70 and the proper names transcribed according to the Hebrew. In any case, to sum up, if we look at the problem under discussion from the angle of Alexandrian philology, the Hexapla—the text with its parallel readings as a form of presentation of the information, a synopsis of textual data, and material for exegesis,71 i.e. the Alexandrian ekdosis—is a completed work, but at the same time, it could also be the basis for other ekdoseis as well as for exemplars intended for scholars or for wider use.72 Origen probably speaks exactly about this: one of

68  See Schironi, “The Ambiguity of Signs”, 100. 69  Another problem posed already by Field (see Frederick Field, Prolegomena to Origenis Hexaplorum Quae Supersunt, Sive Veterum Interpretum Graecorum in Totum Vetus Testamentum Fragmenta [Trans. and annot. G.J. Norton, O.P. with the collaboration of Carmen Hardin; CahRB 62; Paris: Gabalda, 2005], 115–16, 118) but still overlooked, is how precisely Origen marked the minuses and the pluses of the Septuagint. Indeed, we count the cases of Origenic corrections against Hebrew and miss the differences not marked by Origen. 70  Cf. Field, Prolegomena, 114–15. My question is, if the Septuagint of the fifth column of the Hexapla already contained the word order corresponding to the Hebrew, and if the text of this column became the substratum for the so-called Hexaplaric manuscripts, why is there not a stable word order in the “Hexaplaric manuscripts” themselves (not to mention those groups of manuscripts, which are considered as dependent on the Hexaplaric)? For example, in the text of Exodus, of 184 cases of transposition corresponding to the Hebrew, only 16 cases occur in most of the Hexaplaric manuscripts (but never in all of them). The O group (also often incomplete), which contains most of the transpositions, reflects only ca half of the cases of transpositions. See J.W. Wevers, Text History of the Greek Exodus (MSU 21; Göttingen: Vandenhoeck & Ruprecht, 1992), List 4, 31–6. 71  Barthélemy, “Origène et le texte de AT”, 203, 209, 214–17. Usage of the Hexapla in exegesis starts already from Eusebius, see also the variants of the Three in the scholia of the Catena mss and their citations by Theodoret. Hexapla, as opus exegeticum see Neuschäfer, Origenes als Philologe, 379 n. 77. 72  See Dörrie: “eine ἒκδοσις ist in der Regel ein Exemplar, das in einer Bibliothek zur Einsicht, zum Abschreiben und zum Kopieren anderer Hss. ausliegt. Dabei liegt es in der Natur der Sache, dass selten das Ganze abgeschrieben, sehr oft (nur mit verschiedener Gründlichkeit) danach korrigiert wird” (H. Dörrie, “Zur Geschichte der Septuaginta im Jahrhundert Konstantins”, ZNW 39 [1940] 57–110, on p. 90).

34

Anna Kharanauli

the ekdoseis (i.e. text supplied by marginal notes) of Genesis produced by himself on the basis of the Hexapla.73

Ekdosis of Origen’s Pupils I am not the first who doubts that the text containing the asterisked additions and free from the obelised variants stems from Origen and who ascribes the features of the O-text to other philologists and copyists from Origen’s circle—not only to those colleagues of Origen, whose names we know, but also to the copyists, glossators, and persons with different interests. Already Barthélemy claims: “Origène avait conçu les hexaples comme une ample collection d’informations. Mais des glossateures et correcteurs s’en servirent sans discernement pour retoucher le texte vulgaire de la Septante, alors qu’ Origène avait donné de celle-ci une édition critique où il la traitait d’une manière autrement respectueuse.”74

And, “Origène eut pour disciples plutôt des pillards que des continuateurs.”75

Who does Barthélemy mean by Origen’s correctors and glossators? Origen’s text passed through the generations—beginning from his immediate colleagues and ending with the philologists and copyists of later periods, who belonged to other philological-theological schools. As it seems, the philological activities of Origen’s pupils and of the scholars of later generations needs to be differentiated.76 First of all, the role of Origen’s immediate circle should be specified. The most eminent personalities from Origen’s circle are Eusebius and Pamphilus. According to Jerome, they spread the text 73  See R.G. Jenkins: “Hexaplaric Marginalia and the Hexapla-Tetrapla Question”, in A. Salvesen (ed.) Origen’s Hexapla and Fragments: Papers Presented at the Rich Seminar on the Hexapla, Oxford Centre for Hebrew and Jewish Studies (TSAJ 58; Tübingen: Mohr Siebeck, 1998) 73–87; and “Colophons of the Syrohexapla”, 261–77. Here the author expresses the same doubt and suggests the theory that Origen’s edition, the Tetrapla, represented “four traditions (i.e., LXX, Aquila, Symmachus, and Theodotion), one complete and three in extracts, woven together into one column” (269). For the Tetrapla as a form of Origen’s edition, see P.J. Gentry, “Did Origen Use the Artistarchian Signs in the Hexapla?” in W. Kraus/M.N. van der Meer/M. Meiser (ed.), XV Congress of the International Organisation for Septuagint and Cognate studies, Munich 2013 (Munich: SBL, 2016) 133–48. 74  Barthélemy, “Origène et le texte de AT”, 203. 75  Barthélemy, “Origène et le texte de AT”, 217. 76  The fact that the Origen’s working methodology and his critical signs were used later on through centuries is also proved by an example of Jerome; see above, n. 40 and 41.

Origen and Lucian in the Light of Ancient Editorial Techniques

35

elaborated by Origen in Palestine: “Origenes elaboratos Eusebius et Pamphilus vulgaverunt.”77 Does this phrase indicate 1. the collaboration of Origen and his colleagues or 2. independent enterprise of Eusebius and Pamphilus?78 And in what exactly is this enterprise represented? Here again we should start with the “Alexandrian” approach to the text.79 It is clear that the activities of pupils in the process of editing, copying and dissemination of texts were connected with various problems—technical difficulties, which were related to deciphering the teacher’s texts, the lack of philologists, little or no access to the manuscripts, etc.80 The complexity of the teacher’s texts and the purposes of the edition required the pupils to intervene in the text—in its grammar, wording, punctuation—which, clearly, could not protect the texts from interpretations. Responsibility towards the teacher’s text was based on the belief in the divine inspiration of the teacher, and this forced the pupils, first of all, to become proficient in the teacher’s ideas.81 According to both Eusebius of Caesarea and Jerome depending on him, Origen’s pupils used to work on their teacher’s writings also in the traditional way. Pamphilus collected Origen’s writings, set them into order, drew up some sort of catalogue and copied them in his own hand.82 We get some additional information about the contribution of the later philologists on Origen’s text from notes of Psalm-catenae: οὗ (Ὠριγένους) τοῖς συγγράμασιν ἐπιμελέσται προσέχεν Εὐσεβίος τε καὶ οἱ πρὸ Εὐσεβίου καὶ μετ’ Εὐσεβίον φιλολόγοι τε καὶ περὶ τὰς ἁγίας γραφὰς φιλόπονοι [ms. Ra 113=Ambros. cod. B 106]. We learn more specifically from colophons in Sinaiticus, Marchalianus and the Syro-Hexapla about the contribution of the first generation.83 These colophons witness various stages 77  Praefatio in Paralipomena, in D. De Bruyne (ed.), Prefaces to the Latin Bible, 30. 78  See, for example: “stellt die sg. hexaplarische Gruppe allem Anschein nach die Rezension von Eusebios und Pamphilios dar, die eine Ausgabe der LXX-Spalte der Hexapla war” (I. Soisalon-Soininen, Der Charakter der asterisierten Zusätze in der Septuaginta [AASF 114; Helsinki: Suomalainen Tiedeakatemia, 1959], 9; Neuschäfer, Origenes als Philologe, 98. 79  Parallels are already made in this case as well, namely, between the circles of Origen and Plotinus, see Grafton/Williams, Christianity and the Transformation of the Book, 32, 226. 80  See, for example, the letter of Longinus to Porphyry, Grafton/Williams, Christianity and the Transformation of the Book, 37. 81 Grafton/Williams, Christianity and the Transformation of the Book, 34, 39–40. 82 Grafton/Williams, Christianity and the Transformation of the Book, 182–3. 83  For the texts and interpretations of the colophons, see Field, Prolegomena, 183–6; Swete, An Introduction, 74–78; Mercati, Nuove note di letteratura biblica e cristiana antica (StT 95; Citta del Vaticano: Biblioteca Apostolica Vaticana, 1941); Devreesse, Introduction à l’étude des manuscrits grecs, 123–4; Nautin, Origène, 322–3; Dörrie, “Zur Geschichte der Septuaginta”, 90–1; J. Ziegler (ed.), Isaias. Vetus Testamentum Graecum, XIV (3 Auflage; Göttingen: Vandenhoeck & Ruprecht, 1983), 50–2; J. Ziegler (ed.), Ezechiel, Septuaginta. Vetus Testamentum Graecum, XVI,1 (3 Auflage, mit einem Nachtrag von Detlef Fraenkel; Göttingen: Vandenhoeck & Ruprecht, 2006), 32–3; R. Hanhart (ed.), Esther, Vetus Testamentum Graecum, VIII,3 (Göttingen: Vandenhoeck & Ruprecht, 1966), 60–2; R. Hanhart, (ed.), Esdrae liber II, Septuaginta.

36

Anna Kharanauli

of philological activities on the texts of the Old Testament. E.g., the corrector of Sinaiticus (ca 6th c.) writes almost the same in the colophons of 2 Esdras and Esther: his ancient parent manuscript (παλαιώτατον λίαν ἀντίγραφον) is collated (ἀντεβλήθη) and corrected (δεδιορθωμένον) by the hand of Pamphilus. The autographic subscription (ἰδιόχειρος ὑποσημείωσις), written in first person, witnesses that this “ancient manuscript” was collated (μετελήμφθη) and corrected (διωρθώθη)84 from the Hexapla of Origen in the following manner: Antoninus collated, and Pamphilus corrected. The colophon to the text of Ezekiel transmitted in codex Marchalianus,85 written by Eusebius in the first person as well, speaks about the same stages of the philological enterprise: transcription, collation and correction. His manuscript was copied from the editions of the Hexapla (μετελήμφθη ἀπὸ τῶν κατά τὰς ἐκδόσεις Ἑξαπλῶν) and corrected (διωρθώθη) from Origen’s own Tetrapla, which was corrected (διώρθωτο) and annotated (ἐσχολιογράφητο) in his own hand. Exactly from this source Eusebius copied the scholia (σχόλια παρέθηκα), and Pamphilus and Eusebius corrected the manuscript. Thus, we get the double correction of the text, first according to the original, and then according to the controlling source.86 A similar picture is shown in the Syro-Hexaplaric colophon to the Proverbs: “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 in which scholia were written in the margins by the hand of Pamphilus and Eusebius and in which were noted also these things: These things that we found were taken from the Hexapla Vetus Testamentum Graecum VIII,2 (Göttingen: Vandenhoeck & Ruprecht, 1993), 8, 14, 249; L.B. Paton, A Critical and Exegetical Commentary on the Book of Esther (New York: Charles Scribners & Son, 1908), 35; Grafton/Williams, Christianity and the Transformation of the Book, 189–92, 340–2 n. 19–23; Jenkins, “Colophons of the Syrohexapla”, 261–8; Gentry, “Did Origen Use”, 135–45. 84  Again, corrected (διωρθώθη) in the scholia, not in the text. 85  See Ziegler, Ezechiel, 32. 86  See also the colophon on the text of Isaiah in the same codex: μετελήμφθη ὁ Ἠσαΐας ἐκ τῶν κατά τὰς ἐκδόσεις Ἑξαπλῶν. ἀντεβλήθη δὲ καὶ πρὸς ἓτερον Ἑξαπλοῦν. In the colophon of Isaiah the process of accomplishing the manuscript is described in a very complex way. Ziegler interprets this colophon in the following way: “Der Kodex also, der die uns in Q aufbewahrte Unterschrift hatte, stammt aus einem Exemplar des Abtes Apollinarius. Das Exemplar des Apollinarius hatte wiederum ein Kolophon, das die Abstammung weiterführen lässt: der Is.Text des Apollinarius stammt aus der Hexapla; zugleich wurde er mit einem ἓτερον ἓξαπλοῦν verglichen. Dieses ἓτερον ἓξαπλοῦν war besonders wertvoll, weil es seiner Unterschrift nach der Hexapla und nach der Tetrapla genau revidiert worden ist. Weiterhin waren zum Vergleich auch der Is.-Kommentar des Origenes bis Kap. 23 (ἓως τοῦ ὁράματος Τύρου) herangezogen. Endlich wurde die Septuaginta-Kolumne mit dem Is.-Kommentar des Eusebius vergliechen” (Ziegler, Isaias, 51). For other interpretations see Dörrie, “Zur Geschichte der Septuaginta”, 90–1 and n. 82; Jenkins, “Colophons of Syrohexapla”, n. 28, 275; Munnich, “Les Hexaples d’Origène”, 183.

Origen and Lucian in the Light of Ancient Editorial Techniques

37

Version of Origen. And again: in their own handwriting ‘Pamphilius and Eusebius corrected’”.87 The unique picture of the philological enterprise on the text-edition described in the mentioned sources can be summarised as follows. It is a collective work of several philologists, which involves different stages: 1. find an old and accurate manuscript as a Vorlage and a manuscript which serves as a controller; 2. transcribe the Vorlage; 3. collate the transcribed text against both the parent and controlling manuscripts; and 4. make corrections. It must be noted, that the parent exemplar as well as its copy contain not only the text, but also the marginal notes. This enterprise—editing the text—ends traditionally with the writing of a colophon, where the whole process is described by the philologists themselves.88 Besides, the colophons witness that Origen’s Hexapla was still in use for the “scholarly” editions and appealing to it, so to speak, was the guarantee of quality even centuries after its creation and far from Palestine. In the so-called “Hexaplaric” text-history most interesting is how “μετελήμφθη καὶ διωρθώθη πρὸς τὰ Ἑξαπλᾶ Ὠριγένους” or “μετελήμφθη ἀπὸ τῶν κατά τὰς ἐκδόσεις Ἑξαπλῶν” should be interpreted. From my point of view, first of all, it has to mean the copying of the fifth column of the Septuagint, which, as I already mentioned, should not have been corrected through the Hebrew. At the same time, it should also mean the copying of the text of the Three, however, copying not in the form of the text, but copying excerpts in margins (σχόλια παρέθηκα).89 Thus, I think the result of the contribution of Eusebius, Pamphilus and Anthonine described in the abovementioned colophons, is still the traditional form of ekdosis. The text was thoroughly copied from a carefully chosen Vorlage, then collated and corrected, and supplied with scholia. The codex archetypus of the Greek Vorlage of Paul of Tela’s Syriac translation as well as the model-exemplar of the codex Marchalianus and an exemplar which served as a base manuscript for the catenarist of the Greek OT should have been exactly such kind of manuscript—the ekdosis of Eusebius and Pamphilus, i.e. text and marginal scholia linked to the text with critical signs.90 87  Translation according to Peter Gentry, “Did Origen Use”, 136. Other colophons in the Syro-Hexapla (colophons of Isaiah, Cantica, Minor Prophets and 3 and 4 Reg.) presents the philological work in the same way, esp. interesting is the colophon to 3 Kingdoms: “this book... was copied from the Hexapla, that is, from the Six Columns that was among the texts of the Six Columns that was among the books of Caesarea in Palestine and was compared to a manuscript in which was noted at the end as follows: I, Eusebius, corrected as accurately as possible” (Gentry, “Did Origen Use”, 144). 88  Cf. Turner, Greek Papyri, 93; Montanari, see above, 19. 89  See Field, Prolegomena, 186; cf. Jenkins, “Colophons of the Syrohexapla”, 263, 267–8. 90  It is noteworthy, that the Hexaplaric character of these sources is evidenced not so much through its text, but more through its margins. See Nautin, Origène, 358; About the Syro-Hexa-

38

Anna Kharanauli

Another source recording the contribution to the biblical text by Origen’s followers is Eusebius himself. Constantine ordered him to provide “fifty volumes with ornamental leather bindings, easily legible and convenient for portable use, to be copied by skilled calligraphists well trained in the art, copies that is of the Divine Scriptures, the provision and use of which you well know to be necessary for reading in church”.91 It is logical to think that such commission should have been followed by the creation of vulgata, of a more-or-less unified, standardised text—textus receptus. It would be logical as well to assume that for this reason Origen’s pupils should have used their teacher’s ‘recension’—asterisked and obelised text, if such has ever existed.92 However, this is not confirmed by the ancient codices. The addition of the Hexaplaric variants in the texts does not standardise these texts, but makes more variegated the textual transmission, as Jerome noted too. The unstable character is caused by the irregular appearance of the Hexaplaric material in the text. A good example of this is already the ancient Greek fragment from the Fayum, almost of Origen’s time, 3rd–4th cc., Papyrus 922 (Bodl. Ms. Gr. Bibl. d.4), which includes Ezekiel 5:12–6:3.93 The text of the papyrus contains the Hexaplaric additions and is the oldest extant Biblical manuscript attesting asterisks. But interesting is that the transmission of the Hexaplaric materials in the papyrus differs from other O group manuscripts of Ezekiel and, among them from the 6th–7th century Codex Marchalianus, which contains the so called Hexaplaric text in Ezekiel.94 With this I want to underline what I already noted and what is well-known: none of the extant manuscripts, including ancient so-called Hexaplaric sources, even within one book contain a ‘Hexaplaric text’ (neither a text of the fifth column, nor a text of the separate edition of Origen’s recension). This means, at least for me, that such text was not diffused in Origen’s time even among his circle, neither has it later become a standard text through the ekdoseis of Pamphilus and Eusebius, nor has it become a koine of the Church, in spite of the fact that exactly Eusebius was obliged to supply the Church with Pandects. Neither do I think that the quantitative and qualitative variety of the Hexaplaric readings and the critical signs in the manuscripts, plaric text of Ecclesiastes, see Gentry, “The Aristarchian Signs in LXX Ecclesiastes”, 466. The examples of the similar type of ekdosis are e.g. S (codex Sinaiticus), Q (codex Marchalianus= Vatican, Vat. gr. 2125), M (Paris, BnF. Coisl 1), Ra 86 (Vatican, Vat. Bab. Gr. 549), Ra 344 (Athos, Pantocratoros, 24: see K. De Troyer, “The Scribe of the Marginal Notes of Manuscript Ra 344 (= Ca v)”, in the present volume), Ra 710 (=Sin. cod. Gr. 5), Ra 64 (=Paris, BnF Gr. 2), Ra 549 (Vatican, Vat. Baber. Gr. 549). 91  Eusebius, VC 4.36.2. English translation see A. Cameron/S.G. Hall (ed.), Eusebius: Life of Constantine (Oxford: Clarendon Press, 1999), 167. 92  Cf. Dörrie, “Zur Geschichte der Septuaginta”, 88, 91–2. 93  See Schironi, who suggests that this papyrus “is likely to be a very old copy of the edition of the Greek Bible by Origen... for which the Hexapla was a preparatory (though extremely important) step” (Schironi, “The Ambiguity of Signs”, 107). 94  For the comparison between these two manuscripts, see Schironi, “P.Genf. 1.5”, 200–5.

Origen and Lucian in the Light of Ancient Editorial Techniques

39

which we have ­today and which existed already in the time of Jerome, is the result of “deterioration” of the standard text, which was a principled revision towards the Hebrew. On the contrary, the incorporation of the Hexaplaric material in the text took place only after Origen, Pamphilus and Eusebius; the process of text transmission was connected more to the gradual filling of the text with the asterisked materials than to the sporadic deleting of these materials from the text or the loss of the critical signs. This very direction of the textual development indicates for me the fact that the texts of the ancient ‘Hexaplaric’ sources, i.e. Greek and Syro-Hexaplaric manuscripts, are less ‘Hexaplaric’. The same direction indicates well the facts, which are more often noted in relation to the ‘Lucianic recension’, but are true also for O manuscripts, that the same ‘Hexaplaric’ additions in the text sometimes (the statistics for this have not yet been collected) are inserted in different places in the manuscripts, or they do not look like organic parts of the text.95 The same thing is indicated by the doublets composed by the variants of the Septuagint and the Hexapla, which are certainly created unintentionally and are the results of the collation of different sources.96 The fact that readings relevant to MT are represented with different variants in the manuscripts, could also serve as an evidence for the various traditions of philological work.97 From my point of view, the filling of the text with the Hexaplaric material should have been made from the margins of the ekdoseis, produced by Eusebius and Pamphilus (maybe under direction of Origen) and by later philologists. These ekdoseis were accomplished not in order to standardise the text, but for the exegetical purposes of the editors. Therefore, it is theoretically possible that on the margins of different ekdoseis, various Hexaplaric materials could have appeared, according to the interests of the editors. Gradually and sporadically the variants of the margins used to be incorporated into the text by the collators and the copyists. The observation of the extant Hexaplaric sources witnesses the very direction of the text development. E.g., the Hexaplaric variants on the margins of the Syro-Hexapla in Cod. Curz (8th c.) are already incorporated into the text in Cod. Ambr. (8th c.).98 This process is more intensively demonstrated by the observation of the connections of Syro-Hexapla and 88, where the Hexaplaric 95  See Soisalon-Soininen, Asterisierte Zusätze, 193–7; J.W. Wevers, “The L Text of Ezekiel”, in J. W. Wevers/D. Fraenkel (ed.), Studies in the Text Histories of Deuteronomy and Ezekiel (MSU 26; Göttingen: Vandenhoeck & Ruprecht, 2003), 115. 96  Dörrie notes interestingly that the doublets are no more created later, in the medieval manuscripts, see Dörrie, “Zur Geschichte der Septuaginta”, 89. By the way, while dealing with the origin of the textual variants, already Galen notes that some emendations are placed between the lines (μέτωπον) and become part of the text while copying, which leads to double readings. 97  Such variant readings have been properly classified by Munnich, see, Munnich, “Les Hexaples d’Origène”, 183. 98 Ziegler, Isaias, 42; see also Wevers, “The L Text of Ezekiel”, 68.

40

Anna Kharanauli

variants are contained in the margins of the Syro-Hexapla, but 88 has them inside the text.99 Finally, what does Jerome mean when he mentions Origen’s text which was widespread in Palestine? I think that he does not mean the text containing variants of the Three, but the text the substratum of which was the “Septuagint” used in Origen’s ekdosis. Now, what does Jerome call Lucianic?

Lucian’s Ekdosis Although Lucianic studies have become very intensive recently, the common sense of the ‘Lucianic’ text still can be best expressed in Rahlfs’ well-known phrase, “Hauptcharakterzug dieser Rezension ist das Fehlen eines klaren Prinzips.”100 Indeed, the principles and tendencies of the text emendations which are considered as common characteristics of the L manuscripts-group 1. differ in various books, 2. are contradictory (if not incompatible) and irregular among one book, and 3. quantitatively differ in various manuscripts. Given all these, quite a legitimate question arises: How feasible is it to ascribe all the tendencies occurring in one family of medieval manuscripts to one person?

99  See “Der grösste Unterschied zwischen Syh und 88 besteht darin, dass 88 im Gegensatz zu Syh keine Randlesarten hat und dass sie weiterhin eine Reihe von Zusätzen bietet, die in Syhtxt fehlen, aber manchmal in Syhmg erhalten sind” (Ziegler, Isaias, 41–2). 100  For the characteristics of the ‘Lucianic’ / ‘Antiochene’ recension see the selected bibliography: A. Rahlfs, Lucians Rezension der Königsbücher. Septuaginta Studien 1–3 (2nd edn; Göttingen: Vandenhoeck & Ruprecht, 1965), 293 [633]; B.M. Metzger, Chapters in the History of New Testament Textual Criticism (NTTS 4; Leiden/Grand Rapids: Brill/Eerdmans, 1963), 1–41; Dörrie, “Zur Geschichte der Septuaginta, 57–110; D. Barthélemy, “Les problemes textuels de 2 Sam 11, 2 – 1 Rois 2, 11 reconsideres à la lumiere de certaines critiques des ‘Devanciers d’Aquila’”, in Études d’histoire du texte de l’Ancien Testament (OBO 21; Fribourg/Göttingen: Universitätsverlag/Vandenhoeck & Ruprecht, 1978), 218–54 [78–88]; S. Jellicoe, The Septuagint and Modern Study (Oxford: Clarendon Press, 1968), 157–71; N. Fernández Marcos, The Septuagint in Context: Introduction to the Greek Versions of the Bible (Leiden: Brill, 2001), 223–46; S. Kreuzer/M. Sigismund (ed.), Der Antiochenische Text der Septuaginta in seiner Bezeugung und seiner Bedeutung (DSI 4; Göttingen: Vandenhoeck & Ruprecht, 2013); E. Tov, “Lucian and Proto-Lucian: Toward a New Solution of the Problem”, in The Greek and Hebrew Bible, Collected Essays on the Septuagint (VTSup 72; Leiden: Brill, 1999) 477–88; K. De Troyer, “Der lukianische Text: Mit einer Diskussion des A-Textes des Estherbuches”, in S. Keuzer/J.P. Lesch (ed.), Im Brennpunkt: die Septuaginta: Studien zur Entstehung und Bedeutung der griechischen Bibel. Band 2 (BWANT 161; Stuttgart: Kohlhammer, 2004), 229–46, on pp. 229–37; T. Kauhanen, The Proto-Lucianic Problem in 1 Samuel (DSI 3; Göttingen: Vandenhoeck & Ruprecht, 2012).

Origen and Lucian in the Light of Ancient Editorial Techniques

41

Witnesses about the enterprise of revising the text of the Septuagint assigned to Lucian the Martyr are still to be collected and analysed.101 Until then, let us summarise what we know from already studied sources. According to Jerome, the text which in his time bore the name of Lucian was a text known already by Lucian’s predecessors (Origen, Eusebius and others) as koine. That text was widespread in Constantinople and Antioch102 and was corrupted already in Jerome’s day.103 Besides Jerome, Lucian’s ekdosis is mentioned in the so-called Synopsis Sacrae Scripurae of pseudo Athanasius, the Synaxarion, Passio S. Luciani of Symeon Metaphrates and Suidas and the Arabic translation of the Syro-Hexapla. These sources call Lucian’s text an ‘ekdosis’ and in some cases, like other ekdoseis, mark it with a special siglum—λ̥. Moreover, they give us notes about the reasons and methods of Lucian’s working-style on the biblical texts. According to the Synaxarion, Lucianic emendations (ἐπανορθώσει) are caused by 1. unintended changes in the Septuagint manuscripts due to copying the text for a long time and 2. intended changes which were made by wicked men.104 According to the Synopsis, he used previous ekdoseis (προγεγραμμέναις ἐκδόσεσι) and the Hebrew (Ἑβραϊκοῖς) and made emendations (διωρθωσάμενος) mainly concerning the quantitative changes (additions and omissions—τὰ λείποντα ἢ περιττὰ) of the text.105 Another witness about Lucian’s ekdosis is the Armenian translation of the Pseudo-John Chrysostom’s Commentary on Isaiah 9:6, where it is said that Lucian did not change anything in his own will; he did not add or take away anything, but he collated the text with the Hebrew and the Three, adjusted and edited it.106 In short, the tradition relates Lucian the Martyr to the working on the biblical text, and it characterises him as a philologist of his time. If we consider Lucian’s 101  For a review of the sources see Field, Prolegomena, 157–64; Swete, An Introduction, 80– 85; Devreesse, Introduction à l’étude des manuscrits grecs, 117–21; Dörrie, “Zur Geschichte der Septuaginta”, 62–87; Fernández Marcos, The Septuagint in Context, 223–6; Metzger, Chapters in the History of New Testament Textual Criticism, 3–7. 102 See Praefatio in Paralipomena, Jerome’s witnesses see above, pp. 22–24. 103  See above, n. 40. 104  Οὗτος, τὰς ἱερὰς βίβλους θεασάμενος πολὺ τὸ νόθον εἰσδεξαμένας τοῦ τε χρόνου λυμηναμένου πολλὰ τῶν ἐν αὐταῖς καὶ τῆς συνεχοῦς ἀφ᾿ ἑτέρων εἰς ἕτερα μεταθέσεως, καὶ μέντοι καί τινων ἀνθρώπων πονηροτάτων οἱ τοῦ Ἑλληνισμοῦ προεστήκεσαν, παρατρέψαι τὸν ἐν αὐταῖς θελησάντων νοῦν καὶ πολὺ τὸ κίβδηλον ἐνσκευασαμένων, αὐτὸς ἁπάσας ἀναλαβὼν ἐκ τῆς Ἑβραΐδος αὐτὰς ἐπανενεώσατο γλώσσης, ἣν καὶ αὐτὴν ἠκριβωκὼς εἰς τὰ μάλιστα ἦν, πόνον τῇ ἐπανορθώσει ὅτι πλεῖστον εἰσενεγκάμενος (Passio Luciani see Dörrie, “Zur Geschichte der Septuaginta”, 71; the same text with small divergences, see PG 114, 401). 105 [Λουκιανός] ταῖς προγεγραμμέναις ἐκδόσεσι καὶ τοῖς Ἑβραϊκοῖς ἐντυχὼν καὶ ἐποπτεύσας μετὰ ἀκριβείας τὰ λείποντα ἢ καὶ περιττὰ τῆς ἀληθείας ῥήματα, καὶ διωρθωσάμενος ἐν τοῖς οἰκείος τῶν γραφῶν τόποις ἐξέδοτο τοῖς Χριστιανοῖς ἀδελφοῖς (Dörrie, “Zur Geschichte der Septuaginta”, 79, see also PG 28, 436). 106 Ziegler, Isaias, 73.

42

Anna Kharanauli

traditional education, which he received in Edessa and Caesarea of Asia Minor, and consider as well, on the one hand, the philological traditions of Antioch107 and, on the other hand, those ambitions which Antioch had in the era of the formation of liturgical texts and biblical exegesis, the text related to Lucian should be nothing else than an ekdosis of the Holy Scriptures, based on the methodology applied by his predecessor philologists—both pagan and Christian. Lucian’s traditional ekdosis should include the basic text (i.e. the text of that concrete manuscript on which Lucian worked) and the ‘critical apparatus’ (i.e. alternative readings in the free spaces). The medieval manuscripts that come down to us present a hybrid of these two layers. Is it possible today to separate these layers? E. Tov notes correctly about the difficulties “in defining criteria for unravelling the three layers of boc2e2, viz. the OG substratum, Lucian’s borrowings from the ‘Three’ and the fifth column of the Hexapla, and Lucian’s own corrections”.108 The substratum for L (i.e. the text of Lucian’s basic manuscript) would probably have been the koine, the only characteristic of which, as Jerome witnesses, was its wide diffusion. On the other hand, as far as it is possible to judge on the basis of extant L manuscripts, this text had different characters in various books, and, consequently, its reconstruction in each book is connected to the study of its common strata with various pre-Hexaplaric sources. However, one thing is obvious based on literary sources and textual evidence: for some books this substratum should have been a more ancient textual form than the extant majuscule codices.109 Following the traditional technique of the formation of an ekdosis, Lucian’s basic manuscript would have been collated with external sources. The number of these sources depended on how big his library was. I am not sure on what Wevers based his statement that “it is highly unlikely that he had a library of texts at his disposal.”110 Anyway, the variety of readings of L manuscripts differing in character as well as in origin clearly indicates the multiplicity of the sources used by their creator. If we focus our attention in this regard, relations between L manuscripts and various old textual traditions of the Septuagint can be found, if only sporadically.111

107  In the time of Antiochus the Great (224–188 b.c.e.) a library—with Euphorion of Chalkis as the head—already existed in Antioch. The rich philological tradition of Antioch is evidenced by the existence of the grammatical school of Libanios (born in 314 c.e.). See R. Cribiore, The School of Libanius in Late Antique Antioch (Princeton: Princeton University Press, 2007). 108  Tov, “Lucian and Proto-Lucian: Toward a New Solutionof the Problem”, 482. 109  The fact that the material, used by Lucian, was “fairly old” is mentioned already by Kahle and Burkitt. See Jellicoe, The Septuagint and Modern Study, 169. 110  Wevers, “The L Text of Ezekiel”, 68. 111  About Lucian’s acquaintance with Kaige-Theodotion see e.g. A. Aejmelaeus, “Where Do Doublets Come from? A Problem of the Septuagint of 1 Samuel”, in J.K. Aitken/T.V. Evans

Origen and Lucian in the Light of Ancient Editorial Techniques

43

In the L textual tradition the readings of the Three are more obvious. In L manuscripts there are variants of the Three: 1. which occur also in other manuscript groups (mainly in O and C), 2. which more-or-less differ from the same readings witnessed elsewhere, 3. which replace the variants of other Jewish translators presented in other sources (e.g. variants of Symmachus instead that of Theodotion), and 4. which do not occur in extant manuscripts containing other text forms (i.e. O and C) at all. It is a common view that the variants of the Three in L are taken from Origen’s so-called ‘recensional’ text. However, in my opinion, the variety of the readings from the Three in L manuscript groups and the irregularity of their attestation are against this view. For instance, if the ‘Lucianic recension’ was based on ‘Origen’s recension’, the provenance of variants of the Three in L manuscripts, not attested in O sources, remains unclear. The absence of a large number of asterisked additions in L manuscripts cannot be explained by their loss in any stage of L history—during the process of the formation of L or during its textual transmission—because the omission does not reflect the character of the L text.112 If we, nevertheless, ascribe the omissions of the asterisked passages to Lucian, in each case it should be explained what was the reason for neglecting some of the pluses, while copying others of similar type. Still, the presumption that the parent of L is Origen’s ‘recensional’ text, cannot explain cases where Theodotion’s variants in O are replaced by Symmachus’ variants in L. That is why, in order to explain the latter cases, Wevers is forced to assume that Lucian’s Vorlage was a manuscript containing Symmachus’ variants in the margins.113 For me too this assumption seems to be absolutely appropriate. The physical form of the manuscripts from which Lucian had taken the texts of the Jewish translators could have been some sort of ekdosis (not necessarily that of Origen), which apart from the text contained marginal notes with excerpts from the Three. At the same time, there is no reason to exclude that Lucian had access to the complete texts of the Three in the form of the antigrapha of each of these translators or in the form of the Hexapla and its corresponding columns. As for the form in which the texts of the Three appeared in Lucian’s ekdosis, again, I think they would have had the form of scholia. Lucian’s interest in the Three was only exegetical,114 and he did not aim to show the textual relations of the Hebrew and Greek texts or to correct the Septuagint. Besides, as in Origen’s (ed.), Biblical Greek in Context: Essays in Honour of John A.L. Lee (BTS 22; Leuven: Peeters, 2015) 9–19. 112  See e.g. Wevers, “The L Text of Ezekiel”, 112, 115. 113  See Wevers, “The L Text of Ezekiel”, 115–16; N. Fernández Marcos, “On Symmachus and Lucian in Ezekiel”, in F.G. Martínez/M. Vervenne (ed.), Interpreting Translation: Studies on the LXX and Ezekiel in Honour of Johan Lust (BETL 192; Leuven: Peeters, 2005) 151–61, on pp. 157, 160. 114  The fact that there was an exegetical interest in these texts in Antioch is clear also from Theodoret’s commentaries.

44

Anna Kharanauli

case, considering Lucian’s attitude to the Septuagint as a Christian and grammarian, I find it impossible that Lucian would have mixed texts of various authors with each other in order to correct the translation of the seventy-two elders according to the texts of the three later Jewish translators. The practical study of the text leads to the same conclusion, namely that the variants of the Three were not incorporated in the text of the Lucianic ekdosis. Firstly, as I already mentioned, the use of the Three is not regular in L manuscripts. 2. When the textual tradition of L splits, the variants of the Three sporadically occur sometimes in one of the manuscripts or manuscript groups and other times in another one. 3. The variants of the Three creating doublets in L manuscripts are also a sign of their secondary character. 4. The divergences of the readings of the Three attested in L from those occurring in Hexaplaric witnesses, as well as the cases of the substitution of variants of Symmachus and Theodotion, also speak against their provenance from one and the same author.115 All these imply that the variants of the Three were not part of the text from the beginning of the L textual tradition, but rather they were gradually inserted into the text only later. Besides the material of L’s substratum and the text of the Jewish translators, in the L manuscripts there are readings which frequently have no parallels in Septuagint sources and, therefore, are regarded as specifically Lucian’s variants, i.e. the emendations made due to his own considerations (i.e. τὰ ἑαυτοῦ δοκοῦντα, according to the term of Origen). This assumption seems to be inappropriate for Lucian as a grammarian of his time.116 However, let us more closely consider this issue. Some specifically ‘Lucianic’ features are Atticisms or stylistic changes towards ‘better Greek’ and a text-form “designed for public reading”.117 But are these features really so markedly traceable in the L manuscripts? Moreover, I can not even exclude that the examples of the ‘better Greek’ scattered in the medieval manuscripts, which could not be the OG readings, are the result of the work of Byzantine archaists, well-educated scholars and copyists. In addition, it is hard to assume that a Christian philologist of the 4th c. made ‘grammatical-stylistic’ changes, if we consider again the theological-philological context of Lucian’s 115  To the same conclusion lead variety of the Three’s readings in L (such as completion of the text according to the Hebrew, stylistic improvement, clarification of the sense, theological interpretation), see Fernández Marcos, “On Symmachus and Lucian in Ezekiel”, 160; See also O. Munnich, “Le texte lucianique d’Isaïe Septante”, in F. García Martínez/M. Vervenne (ed.), Interpreting Translation: Studies on the LXX and Ezekiel in Honour of Johan Lust (BETL 192; Leuven: Peeters, 2005) 269–99, on pp. 279–83, 298. 116  See Galen’s working method in A. Roselli, “Galen’s Practice of Textual Criticism”, in the present volume. 117  S.P. Brock, The Recensions of the Septuagint Version of I Samuel (Quaderni di Henoch 9; Turin: Silvio Zamonari, 1996), 18, 252; Munnich, “Le texte lucianique d’Isaïe Septante”, 279–83, 298.

Origen and Lucian in the Light of Ancient Editorial Techniques

45

time.118 Besides the fact that the need for the preservation of the author’s style was acknowledged, the linguistic and stylistic peculiarities of the Holy Scripture (especially of the Gospels and Paul’s Epistles) compared to Attic prose119 was also well-known, and this stylistic controversy did not typically lead to favoring ‘Atticism’. Another issue is to what extent ‘public reading’ influenced the text of the Scriptures in general. Firstly, the assumption about Lucian’s connection to the formation of liturgical texts is based on the presumed closeness of L texts to the texts of the Byzantine lectionaries. But, I think that this closeness (if any) should be rather explained either according to the fact that the L text and the lectionaries were based on one widespread form of the Greek Scriptures or that in later forms of the Greek lectionaries the L text was used. Further, apart from this, there is no known analogue of adjusting already existing text of the Scriptures for liturgical use (another issue is how the text is developing in the liturgical collections themselves) in the old lectionaries.120 On the contrary, the liturgical reading (melodic or rhythmic) facilitates the perception of the text, making grammatical discrepancies and stylistic imperfections indiscernible. So, the function of liturgical reading does not necessarily lead to adapting the text. Also questionable are the descriptions of the L variants as ‘theological’ ones. I think that in the 3rd–4th cc., in the epoch of sophisticated theological discussions, such qualification of textual variants should be grounded on much deeper study of the theological issues than has been done until now. Thus, we have still no objective criteria which make it feasible to ascribe all the abovementioned types of variants necessarily to the one particular person, Lucian, and to exclude that they belonged either to the proto-Lucianic layer (i.e. Lucian’s base manuscript, as well as the manuscript(s) against which the main text was checked) or to the post-Lucianic text-developement. Consequently, it is not possible to consider the text-form of the extant manuscripts as a “recension” produced by one author, a text which was corrected by this author according to some 118  Cf. Passio S. Luciani: τινων ἀνθρώπων πονηροτάτων οἱ τοῦ Ἑλληνισμοῦ προεστήκεσαν, see above, n. 104. The interpretation of this passage (in particular, the assumption that Ἑλληνισμός here is the synonym of ἀττικισμός) makes Dörrie to assume that Lucian “müsste also gerade Attizismen getilgt, nicht eingeführt haben” (Dörrie, “Zur Geschichte der Septuaginta”, 72, n. 42). 119  This opposition is manifested in termini: ἄτεχνον—τέχνον; ἁλιευτικός (ἁλιευτικὴ ἁπλότης)—ἀττικός; ἁλιευτικῶς οὐκ ἀριστοτελικῶς; ἐκκλησιαστικός, ἀλλ’ οὐκ ἀγοραίως. See E. Norden, Die Antike Kunstprosa vom VI. Jahrhundert V. Chr. bis in die Zeit der Renaissance (2 vols.; Leipzig: Teubner, 1898), 451–572; H.G. Beck, “Antike Beredsamkeit und byzantinische Kallilogia”, Antike und Abendland 15 (1969) 91–101. 120  Anyway, the biblical texts preserved in the 9-10th c. mss of the Georgian lectionaries, which present the early stage of the lectionary development (see M. Tarchnischvili, Le grand lectionnaire de l’église de Jérusalem (Ve–VIIIe siècle) (4 vol.; Louvain: Secrétariat du Corpus SCO, 1959–1960), do not witness the stylistic peculiarities of the oldest textual layers.

46

Anna Kharanauli

principle. Based on the same arguments, I can not agree that it is an “edition”, in the sense of a “standard text”, which was formed after Lucian in the Antiochene school as a competitor textual type to the “Caesarean recension”.121 Indeed, it is a text, as Dörrie claims, created gradually or “geschichtlich”,122 but the question is how this history should be figured out. In brief, if we consider the issue of the “Lucianic text” in the light of philological tradition and take into account the authority of the Septuagint, if we check the established assumptions on the basis of re-selection and re-analysis of the textual data, the enterprise of Lucian regarding the biblical texts, in my opinion, may be determined as follows: the Lucianic ‘recension’, in the sense that we apply that term does not exist.123 Instead, there are text-forms arisen on the basis of the Lucianic ekdosis. That ekdosis would have contained 1. the text—the widespread form of the Septuagint—and 2. its alternative readings (maybe supplemented with commentaries) from different sources in the free spaces. The text of this ekdosis (or ekdoseis of Antiochene school depending on the Lucianic edition) started soon to be filled with the material from commentaries and scholia124 and became the ‘Antiochene’ or the ‘Byzantine’ text widespread in Byzantium.

Conclusion To sum up, Origen created two types of ekdosis of the Septuagint: 1. Innovative— Hexapla, and 2. Traditional—Tetrapla, that is text with marginalia including alternative readings from the Septuagint taken from the texts of younger translators. The marginal variants were related to the text with asterisk and obelos, to both of which Origen gave values different from their traditional meaning (in this sense, it seems improper to refer to those signs as Aristarchian). In the next stage, Origen’s pupils, together with him and without him, compared ekdoseis of their teacher and other, faithful manuscripts of the Septuagint and marked differences, again in the traditional ekdosis, applying the method of their teacher by marking with asterisk and obelos. This form of the Septuagint ekdosis did not stop with Origen’s Ekdosis, but the base text was being changed, alternative sources are multiplying, and the number of signs is increasing (α',

121  Dörrie, “Zur Geschichte der Septuaginta”, 81–7; Barthélemy, “Les problemes textuels”, 248. 122  Dörrie, “Zur Geschichte der Septuaginta”, 105. 123  The same is said by Barthélemy, see Barthélemy, “Les problemes textuels”, 246–7. 124  As typologically similar cases see A. Schenker, “Der Platz der altlateinischen Randlesarten des Kodex von León und der Valvanera-Bibel in der Biblischen Textgeschichte (1–4Kgt)”, in S. Kreuzer/M. Sigismund (ed.), Der Antiochenische Text der Septuaginta in seiner Bezeugung und seiner Bedeutung (DSI 4; Göttingen: Vandenhoeck & Ruprecht, 2013).

Origen and Lucian in the Light of Ancient Editorial Techniques

47

σ', θ', οἱ γ', οἱ ἄλλοι, ἑβρ, συρ, ο', οἱ λοίποι, λ̥). Along with Origen’s signs, other marks were introduced for usage (metobelos [⸓], antisigma [Ͻ], lemniskos [÷], hypolemniscos [⨪], mallet [⸔], colon [:], semicolon [;]). The types of variants also underwent change (along with the quantitative variant readings, also lexical, grammatical changes and transpositions were being noted). Interest in such editions did not fade through the centuries; one and the same manuscript was compared to other sources several times and in chronologically distant periods. Probably comparison implies also other ekdoseis with different marginal notes or several sources at the same time. As examples of such cases could be considered those manuscripts, which on the one hand, represent original ekdoseis or copies of ekdoseis with marginal notes, and, on the other, so to say, double ekdoseis, which are being supplemented by new marginal notes as a result of comparison with ekdoseis including marginalia with other sources.125 Exactly these types of ekdoseis enabled the exegetes (Origen, Eusebius of Cesarea, Theodoret, John of Chrysostomos, Procopius of Gaza)126 to achieve the ultimate, the noblest goal of Ars Grammatica, that is κρίσις ποιημάτων, which implied treating a text not only from its esthetic and moral perspectives, but to go even further than the Alexandrian philologists did—to interpret text also from a theological point of view. However, along and in parallel with these ekdoseis that were intended for an erudite milieu, there existed vulgar editions, which emerged on the basis of the scholarly editions and were intended for ordinary readers. Such (vulgate) textual forms which are attested in the medieval manuscripts and are marked with O and L must have been created in Constantine’s epoch. Exactly these forms were meant by Jerome when speaking about the texts connected to Origen and Lucian and underlining their differences from their archetype—the Septuagint. The formation of these textual forms would have been the result of the process of diffusion of biblical manuscripts, which was intensified as the result of the increasing need for biblical manuscripts after Constantine’s Edict of Milan. This process would have been related to the simplification of one type of the edition—of a grammarian’s (or his school’s) ekdos(e)is—and to its transformation into another type intended for ordinary readership. This transformation meant a gradual and sporadic process of merging scholia with the text of the working exemplar, which, on the one hand, was impossible without the loss of important data for the scholarly edition and, on the other hand, without overloading the text with details of various textual forms. This transformation of the ekdosis would not have been a singular case. Philological activity, which primarily aided the text’s 125  Catenae collections could be regarded as one of the developed forms of the Greek Old Testament editions. Herein, philological scholia marked with signs are supplemented by exegetical commentaries marked with numbers. 126  See Munnich, “Les révisions juives de la Septante”, 160–6.

48

Anna Kharanauli

exegesis, consisted of multiple processes of copying the standard editions—viz., copying model-exemplars, checking them according to other exemplars, making diorthoseis in the margins, and transforming this sort of edition again into exemplars for ordinary readers. The results of such textual development are the well-known Caesarean and Antiochene (and/or Byzantine) textual types of the medieval manuscripts, the archetypes of which would have been the ekdoseis of two theological schools—of Origen and Lucian.

Bibliography Aejmelaeus, A., “Where Do Doublets Come from? A Problem of the Septuagint of 1 Samuel”, in J.K. Aitken/T.V. Evans (ed.), Biblical Greek in Context: Essays in Honour of John A.L. Lee (BTS 22; Leuven: Peeters, 2015) 9–19. Baehrens, W.A., Origenes Werke VIII, Homilien zu Samuel I, zum Hohelied und zu den Propheten, Kommentar zum Hohelied in Rufins und Hieronymus Übersetzungen (GCS 33; Leipzig, 1925). Bammel, C.P., “Die Hexapla des Origenes: Die Hebraica Veritas im Streit der Meinungen”, Aug 28 (1988) 125–49. Barthélemy, D., “Eusèbe, le Septante et ‘les autres’”, in Études d’histoire du texte de l’Ancien Testament (OBO 21; Fribourg/Gottingen: Universitätsverlag/Vandenhoeck & Ruprecht, 1978) 179–93. Barthélemy, D., “La place de la Septante dans l’Eglise”, in Études d’histoire du texte de l’Ancien Testament (OBO 21; Fribourg/Gottingen: Universitätsverlag/Vandenhoeck & Ruprecht, 1978) 111–26. Barthélemy, D., “Origene et le texte de l’Ancien Testament”, in Études d’histoire du texte de 1’Ancien Testament (OBO 21; Freibourg/Göttingen: Universitätsverlag/Vandenhoeck & Ruprecht, 1978) 203–17. Barthélemy, D., “Les problemes textuels de 2 Sam 11, 2 – 1 Rois 2, 11 reconsideres à la lumiere de certaines critiques des ‘Devanciers d’Aquila’”, in Études d’histoire du texte de l’Ancien Testament (OBO 21; Fribourg/Göttingen: Universitätsverlag/Vandenhoeck & Ruprecht, 1978) 218–54. Beck, H.G., “Antike Beredsamkeit und byzantinische Kallilogia”, Antike und Abendland 15 (1969) 91–101. Brock, S.P., The Recensions of the Septuagint Version of I Samuel (Quaderni di Henoch 9; Turin: Silvio Zamonari, 1996). Cameron, A. & Hall, S.G. (ed.), Eusebius: Life of Constantine (Oxford: Clarendon, 1999). Collins, N.L., The Library in Alexandria and the Bible in Greek (VTSup 82; Leiden: Brill, 2000). Colson, F.H. (ed.), Philo, De Vita Mosis, I–II (LCL 289 vol. 6; Cambridge, MA: Harvard University Press, 1984). Coxe, C. (ed.), The Ante-Nicene Fathers (vol. 4; Fathers of the Third Century; New York: Christian Literature Publishing, 1885).

Origen and Lucian in the Light of Ancient Editorial Techniques

49

Cribiore, R., The School of Libanius in Late Antique Antioch (Princeton: Princeton University Press, 2007). De Bruyne, D. (ed.), Prefaces to the Latin Bible (Introductions by Pierre-Maurice Bogaert and Thomas O’Loughlin; Studia Traditionis Theologiae 19; Turnhout: Brepols, 2015). De Lange, N. (ed.), La Lettre à Africanus sur l’histoire de Suzanne (SC 302; Paris, 1983). De Troyer, K., “Der lukianische Text: Mit einer Diskussion des A-Textes des Estherbuches”, in S. Keuzer/J.P. Lesch (ed.), Im Brennpunkt: die Septuaginta: Studien zur Entstehung und Bedeutung der griechischen Bibel. Band 2 (BWANT 161; Stuttgart: Kohlhammer, 2004) 229–46. Devreesse, R., Introduction à l’étude des manuscrits grecs (Paris: Klincksieck, 1954). Dörrie, H., “Zur Geschichte der Septuaginta im Jahrhundert Konstantins”, ZNW 39 (1940) 57–110. Erbse, H., “Über Aristarchs Iliasausgaben”, Hermes 87 (1959) 275–303. Fernández Marcos, N., “On Symmachus and Lucian in Ezekiel”, in F. García Martínez/M. Vervenne (ed.) Interpreting Translation: Studies on the LXX and Ezekiel in Honour of Johan Lust (BETL 192; Leuven: Peeters, 2005) 151–61. Fernández Marcos, N., The Septuagint in Context: Introduction to the Greek Version of the Bible (Leiden: Brill, 2001). Field, F., Prolegomena to Origenis Hexaplorum Quae Supersunt, Sive Veterum Interpretum Graecorum in Totum Vetus Testamentum Fragmenta (Trans. and annot. G.J. Norton, O.P. with the collaboration of Carmen Hardin; CahRB 62; Paris: Gabalda, 2005). Gentry, P.J., “Did Origen Use the Artistarchian Signs in the Hexapla?”, in W. Kraus/M.N. van der Meer/M. Meiser (ed.), XV Congress of the International Organisation for Septuagint and Cognate studies, Munich 2013 (Munich: SBL, 2016) 133–48. Gentry, P.J., “The Aristarchian Signs in LXX Ecclesiastes”, in K. De Troyer/T.M. Law/M. Liljeström (ed.), In the Footsteps of Sherlock Holmes: Studies in the Biblical Text in Honour of Anneli Aejmelaeus (CBET 72; Leuven: Peeters, 2014) 463–78. Gentry, P.J., The Asterisked Materials in the Greek Job (SBLSCS 38; Atlanta: Scholars, 1995). Grafton, A. & Williams, M., Christianity and the Transformation of the Book: Origen, Eusebius, and the Library of Caesarea (Cambridge, MA: Belknap Press of Harvard University Press, 2008). Hanhart, R. (ed.), Esther, Vetus Testamentum Graecum, VIII,3 (Göttingen: Vandenhoeck & Ruprecht, 1966). Hanhart, R. (ed.), Esdrae liber II, Septuaginta. Vetus Testamentum Graecum VIII,2 (Göttingen: Vandenhoeck & Ruprecht, 1991). Hilberg, I. (ed.), Sancti Eusebii Hieronymi Epistulae. Pars II; Epistulae LXXI–CXX (CSEL 55; Vienna/Leipzig: Tempsky/Freytag, 1912). Honigman, S., The Septuagint and Homeric Scholarship in Alexandria: A Study in the Narrative of the Letter of Aristeas (London: Routledge, 2003). Hunger, H. & Stegmueller O. & Erbse H., Die Textüberlieferung der antiken Literatur und der Bibel (München: Deutscher Taschenbuch, 1975). Jellicoe, S., The Septuagint and Modern Study (Oxford: Clarendon, 1968).

50

Anna Kharanauli

Jenkins, R.G., “Colophons of the Syrohexapla and the Textgeschichte of the Recensions of Origen”, in C.E. Cox (ed.), VII Congress of the International Organization for Septuagint and Cognate Studies, Leuven 1989 (Atlanta: Scholar, 1991) 261–77. Jenkins, R.G., “Hexaplaric Marginalia and the Hexapla-Tetrapla Question”, in A. Salvesen (ed.), Origen’s Hexapla and Fragments: Papers Presented at the Rich Seminar on the Hexapla, Oxford Centre for Hebrew and Jewish Studies (TSAJ 58; Tübingen: Mohr Siebeck, 1998) 73–87. Kauhanen, T., The Proto-Lucianic Problem in 1 Samuel (DSI 3; Göttingen: Vandenhoeck & Ruprecht, 2012). Kemp, J.A., “The Tekhne Grammatike of Dionysius Thrax: Translated into English”, Historiographia Linguistica 13(2/3) (1986) 343–63. Klostermann, E. (ed.), Origenes Werke III. Jeremiahomilien, Klagliederkommentar, Erklärung der Samuel- und Königsbücher (GCS 6; Leipzig, 1901). Klostermann, E. & Benz, E., Origenes Werke 10. Origenes Matthäuserklärung, 1 (GCS 40; Leipzig, 1935). Kreuzer, S. & Sigismund M. (ed.), Der Antiochenische Text der Septuaginta in seiner Bezeugung und seiner Bedeutung (DSI 4; Göttingen: Vandenhoeck & Ruprecht, 2013). Lake K. (ed.), Eusebius: Ecclesiastical History, Books I–V (LCL 153; London: William Heinemann; New York: G.P. Putnam, 1926). Lawson, R.P., Origen: The Song of Songs: Commentary and Homilies (ACW 26; New York: Newman Press, 1956). Lilly, I. E., Two Books of Ezekiel: Papyrus 967 and the Masoretic Text as Variant Literary Editions (VTSup 150; Leiden: Brill, 2012). Matthaios, S. & Montanari, F. (ed.), Ancient Scholarship and Grammar: Archetypes, Concepts and Contexts (Berlin: de Gruyter, 2011). Mercati, G., Nuove note di letteratura biblica e cristiana antica (StT 95; Citta del Vaticano: Biblioteca Apostolica Vaticana, 1941). Metzger, B., The Text of the New Testament: Its Transmission, Corruption, and Restoration (Oxford: Oxford University Press, 1968). Metzger, B.M., Chapters in the History of New Testament Textual Criticism (NTTS 4; Leiden/ Grand Rapids: Brill/Eerdmans, 1963). Migne, J.-P., Selecta in Genesis, PG 12, 92–145. Migne, J.-P., Selecta in Psalmos, PG 12, 1051–685. Migne, J.-P., Passio Luciani PG 114, 397–416. Montanari, F., “Correcting a Copy, Editing a Text: Alexandrian Ekdoseis and Papyri”, in F. Montanari/L. Pagani (ed.), From Scholars to Scholia: Chapters in the History of Ancient Greek Scholarship (TCSV 9; Berlin: De Gruyter, 2011) 1–15. Montanari, F., “Alexandrian Homeric Philology: The Form of the Ekdosis and the Variae Lectiones”, in M. Reichel/A. Rengakos (ed.), Epea Pteroenta: Beitrage zur Homerforschung Festschrift für Wolfgang Kullmann zum 75 (Stuttgart: Franz Steiner, 2002) 119–40. Montanari, F., “The Fragments of Hellenistic Scholarship”, in G.W. Most (ed.), Collecting Fragments / Fragmente Sammeln (Göttingen: Vandenhoeck & Ruprecht, 1997) 273–88.

Origen and Lucian in the Light of Ancient Editorial Techniques

51

Munnich, O., “Le texte lucianique d’Isaïe Septante”, in F. García Martínez/M. Vervenne (ed.), Interpreting Translation: Studies on the LXX and Ezekiel in Honour of Johan Lust (BETL 192; Leuven: Peeters, 2005) 269–99. Munnich, O., “Les Hexaples d’Origène à la lumière de la tradition manuscrite de la Bible Grecque”, in G. Dorival/A. Le Boulluec (ed.), Origeniana Sexta: Origen et la Bible. Actes du Colloquium Origenianum Sextum, Chantilly, 1993 (Leuven: Peeters, 1995) 167–85. Munnich, O., “Les révisions juives de la Septante. Modalités et fonctions de leur transmission: Enjeux éditoriaux contemporains”, in R. Gounelle/J. Joosten (ed.), La Bible juive dans l’Antiquité (HTB 9; Prahins: Éditions du Zèbre, 2014) 141–90. Nautin, P., Origène: Sa vie et son œuvre (Paris: Beauchesne, 1977). Neuschäfer, B., Origenes als Philologe (Schweizerische Beiträge zur Altertumswissenschaft, H. 18; 1–2 vol.; Basel: Reinhardt, 1987). Nickau, K., Untersuchungen zur textkritischen Methode des Zenodotos von Ephesos (Untersuchungen zur antiken Literatur und Geschichte 16; Berlin: de Gruyter, 1977). Niehoff, M.R. (ed.), Homer and the Bible in the Eyes of Ancient Interpreters (Jerusalem Studies in Religion and Culture 16; Leiden: Brill, 2012). Niehoff, M.R., Jewish Exegesis and Homeric Scholarship in Alexandria (Cambridge: Cambridge University Press, 2011). Niese, B. (ed.), Flavii Iosephi Opera 5, De Judaeorum vetustate, sive contra Apionem libri 2 (Berlin: Weidmann, 1889). Norden, E., Die Antike Kunstprosa vom VI. Jahrhundert V. Chr. bis in die Zeit der Renaissance (2 vols.; Leipzig: Teubner, 1898). Paton, L.B., A Critical and Exegetical Commentary on the Book of Esther (New York: Charles Scribners & Son, 1908). Pelletier, A. (ed.), Lettre d’Aristêe à Philocrate (Sources chrétiennes 89; Paris: Les éditions du CERF, 1962). Pfeiffer, R., History of Classical Scholarship from the Beginnings to the End of the Hellenistic Age (Oxford: Clarendon Press, 1968). Preuschen, E. (ed.), Origenes Werke IV, Der Johanneskommentar (GCS 10; Leipzig, 1903). Pöhlmann, E., Einführung in die Überlieferungsgeschichte und in die Textkritik der antiken Literatur (1 vol., Darmstadt: Wissenschaftliche Buchgesellschaft, 1994). Pontani, F., “‘Only God Knows the Correct Reading!’ The Role of Homer, the Quran and the Bible in the Rise of Philology and Grammar”, in M.R. Niehoff (ed.), Homer and the Bible in the Eyes of Ancient Interpreters (Leiden: Brill, 2012) 43–86. Rahlfs, A., Lucians Rezension der Königsbücher. Septuaginta Studien 1–3 (2nd edn; Göttingen: Vandenhoeck & Ruprecht, 1965). Schaper, J., “The Origin and Purpose of the Fifth Column of the Hexapla”, in A. Salvesen (ed.), Origen’s Hexapla and Fragments: Papers Presented at the Rich Seminar on the Hexapla, Oxford Centre for Hebrew and Jewish Studies, 25th July–3rd August 1994 (TSAJ 58; Tübingen: Mohr Siebeck, 1998) 3–15. Schäublin, C., “Μήτε προσθεῖναι μήτ’ ἀφελεῖν”, MH 31 (1974) 144–9.

52

Anna Kharanauli

Schenker, A., “Der Platz der altlateinischen Randlesarten des Kodex von León und der Valvanera-Bibelin der Biblischen Textgeschichte (1–4Kgt)”, in S. Kreuzer/M. Sigismund (ed.), Der Antiochenische Text der Septuaginta in seiner Bezeugung und seiner Bedeutung (DSI 4; Göttingen: Vandenhoeck & Ruprecht, 2013). Schironi, F., “P. Genf.1.5, Origen, and the Scriptorium of Caesarea”, BASP 52 (2015) 181–223. Schironi, F., “The Ambiguity of Signs: Critical Semeia from Zenodotus to Origen”, in M.R. Niehoff (ed.), Homer and the Bible in the Eyes of Ancient Interpreters (Leiden: Brill, 2012) 87–112. Schwartz, E. (ed.), Eusebius, Kirchengeschichte (GCS 9/1; Leipzig: Hinrichs, 1903). Schwarz, E., “Zur Geschichte der Hexapla”, in Zum Neuen Testament und zum frühen Christentum (Berlin: de Gruyter, 1963) 183–91. Soisalon-Soininen I., Der Charakter der asterisierten Zusätze in der Septuaginta (AASF 114; Helsinki: Suomalainen Tiedeakatemia, 1959). Stein M., “Kritische Zeichen”, RAC 22 (2008) 133–63. Swete, H.B., An Introduction to the Old Testament in Greek (Cambridge: Cambridge University Press, 1914). Tarchnischvili, M., Le grand lectionnaire de l’église de Jérusalem (Ve-VIIIe siècle) (4 vol.; Louvain: Secrétariat du Corpus SCO, 1959–1960). Tov, E., “Lucian and Proto-Lucian: Toward a New Solution of the Problem”, in The Greek and Hebrew Bible, Collected Essays on the Septuagint (VTSup 72; Leiden: Brill, 1999) 477–88. Turner, E.G., Greek Papyri: An Introduction (Princeton, N.J.: Princeton University Press, 1968). Uhlig, G. (ed.), Dionysii Thracis, Ars Grammatica: qualem exemplaria vetustissima exhibent subscriptis discrepantiis et testimoniis (Lipsiae: Teubner, 1883 [reprint Hildesheim: Olms, 1979]). Van der Kooij, A., “Zur Frage der Exegese im LXX-Psalter: Ein Beitrag zur Verhältnisbestimmung zwischen Original und Übersetzung”, in A. Aejmelaeus/U. Quast (ed.), Der Septuaginta-Psalter und seine Töchterübersetzungen: Symposium in Göttingen 1997 (MSU 24; Göttingen: Vandenhoeck & Ruprecht, 2000) 366–79. Van Unnik, W.C., “De la règle Μήτε προσθεῖναι μήτ’ ἀφελεῖν dans l’histoire du canon”, Vigilae Christianae 3 (1949) 1–36. Wevers, J.W., “The L Text of Ezekiel”, in J. W. Wevers/D. Fraenkel (ed.), Studies in the Text Histories of Deuteronomy and Ezekiel (MSU 26; Göttingen: Vandenhoeck & Ruprecht, 2003). Wevers, J.W., Text History of the Greek Exodus (MSU 21; Göttingen: Vandenhoeck & Ruprecht, 1992). Wyrick, J., The Ascension of Authorship: Attribution and Canon Formation in Jewish, Hellenistic, and Christian Traditions (Harvard Studies in Comparative Literature 49; Cambridge, MA: Harvard University Press, 2004). Ziegler, J. (ed.), Duodecim prophetae. Vetus Testamentum Graecum, XIII (3 Auflage; Göttingen: Vandenhoeck & Ruprecht, 1984). Ziegler, J. (ed.), Isaias. Vetus Testamentum Graecum, XIV (3 Auflage; Göttingen: Vandenhoeck & Ruprecht, 1983). Ziegler, J. (ed.), Ezechiel, Septuaginta. Vetus Testamentum Graecum, XVI,1 (3 Auflage, mit einem Nachtrag von Detlef Fraenkel; Göttingen: Vandenhoeck & Ruprecht, 2006).

Amneris Roselli Galen’s Practice of Textual Criticism Introduction Neuschäfer’s1 book on Origen as philologist, as well as a recent monograph by P.W. Martens2 on the role of philological practice in Origen’s personal experience of salvation, have drawn attention to the philological instrumentation he employed in his impressive exegetical work on Biblical texts—recently enlarged by the discovery of twenty-nine Homelies on Psalms in the manuscript Monacensis Graecus 3143. The discrepancies (διαφωνία, διαφορά) among the copies (ἀντίγραφα), and also among the ἐκδόσεις (i.e. the different translations of the Biblical texts), mostly required that the exegete provided an explanation for the origin of the variants, and invited him to select among them the authentic variant, with the guidance of an appropriate method. Not only did Neuschäfer retrace in Origen’s criticism the methods of Alexandrian Homeric criticism, in the form they had acquired over time, and as they are documented in the Homeric scholia; he also highlighted several affinities between the philological method of Origen (ca 185–254) and that of Galen (129–ca 216), the author of numerous commentaries on Hippocrates, which form the richest exegetical corpus in all antiquity. This paper aims at concisely showing how Galen tackled textual issues during the several years he devoted to commenting on the writings of Hippocrates— whom he considered the most venerable authority in the field of medicine.4 1 B. Neuschäfer, Origenes als Philologe (Schweizerische Beiträge zur Altertumswissenschaft 18/1–2; Basel: Friedrich Reinhard, 1987) especially ch. 3. “Die textkritische Methode (διορθωτικόν)”, ch. 4. “Die exegetische Methode (ἐξηγητικόν)”, and ch. 5. § 16 (276–85). 2  P.W. Martens, Origen and Scripture: The Contours of the Exegetical Life (Oxford: University Press, 2011). 3 Origenes, Die neuen Psalmenhomilien. Eine kritische Edition des Codex Monacensis Graecus 314 (ed. L. Perrone/M. Molin Pradel/E. Prinzivalli/A. Cacciari; Origenes Werke 13; Berlin: Walter de Gruyter, 2015); on philology, see 12–17. 4  The bibliography on Galen’s commentaries on Hippocrates is by now quite abundant; following L.O. Brocker’s seminal article, “Die Methoden Galens in der Literarische Kritik”, RhM 40 (1885) 415–38, studies on the commentaries focusing on Galen’s exegetical method resumed with D. Manetti/A. Roselli, “Galeno commentatore di Ippocrate”, in Aufstieg und Niedergang der Römischen Welt 37.2 (Berlin: De Gruyter, 1994) 1529–635. For a bibliograhical selection, see at least H. von Staden, “‘A Woman Does Not Become Ambidextrous’: Galen and the Culture of the Scientific Commentary”, in R.K. Gibson/Chr. Shuttleworth Kraus (ed.), The Classical Commentary: Histories, Practices, Theory (Leiden: Brill, 2002) 109–39; L. Ferreri, “Alcune riflessioni sul concetto di ‘lectio difficilior’ nel mondo antico e nella filologia moderna”, Atti e Memorie dell’Accademia Toscana della Colombaria 56 (2005) 9–61; D. Manetti, “La terminologie du livre: à propos des emplois d’ὕφος et d’ἔδαφος dans deux passages de Galien”, REG

54

Amneris Roselli

A short paragraph on Galen’s intellectual background is a necessary introduction before moving on to discuss some of the most relevant issues in his work as ‘philologist’. Galen received an excellent philosophical and medical education, which he frequently mentions in his writings, wherein he proudly makes reference to his teachers and fellow students. His philosophical education took place in his homeland, Pergamum, under the guidance of teachers from the four main philosophical schools (Platonic, Stoic, Peripatetic, and Epicurean); his medical education began in Pergamum and Smyrnae and continued in Corinth and Alexandria, where he was instructed by illustrious physicians. Galen never mentions his grammatical education, most probably because it was of an unexceptional kind— namely, the same kind that was imparted to all boys born in good families like he was, being son and grandson of wealthy and appreciated architects and brought up in a lively city in Asia Minor, such as Pergamum in the second century ad. Nonetheless, clues to his grammatical and literary interests can be inferred from his writings, in which he proudly refers to the studies he chose to pursue during his lifetime; a powerful interest for language, lexicography, and Atticism emerges both from his remaining writings and from the titles of his lost works.5 Galen possessed a rich library—one that, we know now,6 was not exclusively of a medical character. He had personally copied some great classic texts from 119 (2006) 157–71; H. von Staden, “Staging the Past, Staging Oneself: Galen on Hellenistic Exegetical Traditions”, in C. Gill/T. Whitmarsh/J. Wilkins (ed.), Galen and the World of Knowledge (Cambridge: Cambridge University Press, 2009) 132–56; D. Manetti, “Galeno e la ekdosis di Ippocrate: una nota testuale”, in Harmonia: Scritti di filologia classica in onore di Angelo Casanova (Firenze: Firenze University Press, 2012) 475–81; D. Manetti, “Medicine and Exegesis”, in F. Montanari/S. Matthaios/A. Rengakos (ed.), Brill’s Companion to Ancient Greek Scholarship (Leiden: Brill, 2015) 1126–215, on pp. 1169–97. 5  See V. Boudon-Millot, Galien de Pergame: Un médecin grec à Rome (Paris: Les Belles Lettres, 2012) passim. 6  The recent discovery of the De indolentia—a short autobiographical, consolatory text written after the fire that destroyed many buidings in the Palatine area in 192—has unexpectedly increased our knowledge of Galen as philologist. The first part of the text contains a lot of information on Galen’s philological interest, in the years of his Roman sojourn, for other texts than the Hippocratic ones. De indolentia confirmed Galen’s interest for ‘grammatical’ matters and supplied a quantity of new information on his grammatical works. Philology emerges not merely as an instrument for reading Hippocrates, but as a core of authentic, individual interest, cultivated by Galen and absorbing a quantity of energy in the course of his long life. The editio princeps of De indolentia dates from 2008 (V. Boudon-Millot, “Un traité perdu de Galien miraculeusement retrouvé, le Sûr l’inutilité de se chagriner: texte grec et traduction française”, in V. Boudon-Millot/A. Guardasole/C. Magdelaine [ed.], La science médicale antique: nouveaux regards. Études réunies en l’honneur de J. Jouanna [Paris: Beauchesne, 2008] 72–123), and was rapidly followed by a new edition (Galien: Ne pas se chagriner [ed. V. Boudon-Millot/J. Jouanna; CUF; Paris: Les Belles Lettres, 2010]), complemented by a broad commentary. For an analysis of the many questions see C.K. Rothschild/T.W. Thompson (ed.), Galen’s ‘De indolentia’: Essays on a Newly Discovered Letter (Studien und Texte zu Antike und Christentum / Studies and Texts in Antiquity and Christianity 88; Tübingen: Mohr Siebeck, 2014).

Galen’s Practice of Textual Criticism

55

books preserved in Roman libraries; he had studied library catalogues, as well as identified and attributed previously unknown texts (De indolentia 16–17). His favourite authors, besides the medical ones, are mostly philosophers: Plato, Ari­ stotle, Theophrastus, and the Stoics—authors whose traces are disseminated in all his writings. The same can be maintained with regard to the comic and tragic poets, whom he read with a particular interest for their language. He did massive work as a lexicographer (De indolentia 24–28; De libris propriis 20). Galen’s reading of the ancient authors is characterized by his care in retrieving the authoritative editions of their books,7 or, alternatively, in copying and amending them, inserting for instance diacritics or punctuation—a practice that was becoming widespread during his time and to which he enthusiastically adheres (De indolentia 14).8 Galen evidently shared in the ‘antiquarian’ taste for ancient books (παλαιά, παλαιότατα ἀντίγραφα/βιβλία) that must have been widespread in the 2nd century (consider, for instance, Lucian of Samosata’s pamphlet Against an Ignorant Book Collector). Thanks to the network of his friends and the richness of Roman libraries,9 he pursued his aim to collect and to copy books, for personal use as well as for reference in compiling both grammatical writings (lexica and glossaries) and commentaries. Galen’s huge production on philosophical and grammatical subjects is lost. But the titles of his works remain, testifying to a tremendous output (De libris propriis 14–19). We especially regret the loss of the treatise on the method of exegesis (περὶ ἐξηγήσεως) that Galen mentions for the benefit of those who wish to complement his observations on the purpose of the commentaries, which he presents in the preface to the commentary on On Fractures.10 Despite the 7  The Attician edition of Plato mentioned by Galen in his Commentary on Plato’s Timaeus (Corpus Medicorum Graecorum, Suppl. I, 13, 3 Schröder, ἀττιῶν ἀντιγράφων is Ch. Daremberg’s amendment of the lesson ἀττικῶν in the manuscript), came to attention again after the discovery of De indolentia—which, in par. 13, mentions prestigious editions such as Atticus’. On the other hand, with regard to Plato, in the De indolentia Galen cites a “Plato of Panaetius”; see J.-B. Gourinat, “‘Le Platon de Panétius’. À propos d’un témoignage inédit de Galien”, PhilosAnt 8 (2008) 139–51; T. Dorandi, “‘Editori’ antichi di Platone”, Antiquorum Philosophia 4 (2010) 161–74. 8  De indolentia 14 “And what can I say about a stop or a comma which, as you know, are so important in obscure passages that someone who attends carefully to them has no need of an interpreter?” (Nutton’s translation). 9  De indolentia is a source of information on the books in the Roman libraries: cf. M.C. Nicholls, “Galen and Libraries in the Peri Alupias”, JRS 101 (2011) 123–42; A. Roselli, “Libri e biblioteche a Roma al tempo di Galeno: la testimonianza del de indolentia”, Galenos 4 (2010) 127–48; A. Stramaglia, “Libri perduti per sempre: Galeno, de indolentia 13; 16; 17–19”, RFIC 139 (2011) 118–47; E. Puglia, “La rovina dei libri di Anzio nel De indolentia di Galeno”, Segno e Testo 9 (2011) 53–62; F. Montana, “Gli Omeri di Aristarco in fumo: Gal. Indol. 13”, Prometheus 40 (2014) 262–70; M.C. Nicholls, “A Library at Antium?” in Rotschild/Thomsom, Galen’s ‘De indolentia’, 65–78. 10 Gal. In Hipp. Fract. 18 B 319 Kühn.

56

Amneris Roselli

­ umerous losses, we do not lack documents for knowing Galen as a scholar who n practiced grammatike; such documents are his extensive running commentaries on Hippocrates11, who was Galen’s reference author in terms of medical affiliation and intellectual assent.12

The Commentaries on Hippocrates In the 2nd century, a selected group of writings among those attributed to Hippocrates had become fundamental texts in a medical curriculum combining medical practice and theory; those treatises were read and quoted to teach bone surgery, the doctrine(s) of humors, the method of a patient’s clinical examination, and the method of prognosis. As Galen himself says, he commented on the most useful (χρησιμώτατα) and the “most authentic” (γνησιώτατα)13 treatises by Hippocrates, according to a recognized canon of Hippocratic works—a canon that, as far as we know, he ventured to modify in one single instance, attempting to demonstrate that Prorrhetic 1 was actually inauthentic because it was ridden with doctrinal and linguistic errors.14 The commentaries that survive in Greek or Arabic are those on the following treatises: On Fractures, On Joints and In the Surgery (three surgical treatises);15 Aphorisms;16 Prognostic;17 Regimen on Acute Diseases;18 Epidemics 1, 2 (remaining only in Arabic), 3, and 6 (lacking the final sections in Greek, complete in the Arabic version);19 Prorrhetic 1;20 Nature of Man;21 On Airs, Waters, and Places (remaining only in Arabic);22 and Oath (if the comment is authentic; fragments 11  For a general overview of his commentaries, see Gal. De libris propriis 9. 12  Philological observations are also present in other medical works by Galen (for instance, in De difficultate respirationis). 13 Gal. In Hipp. Epid. 3 17 A p. 484 Kühn (= CMG V 10,2,1 p. 3,16 Wenkebach). 14  See A. Roselli, “Galeno sull’autenticità del Prorretico”, in B. Holmes/K.-D. Fischer (ed.), The Frontiers of Ancient Science (Berlin: de Gryuter, 2015) 533–60. 15  The three commentaries lack a modern critical edition; they are accessible in Kühn’s edition, vols. 18 B 318–628; 18 A 300–767 and 18 B 629–925. 16  This commentary also lacks a critical edition; it can be read in Kühn’s edition, vols. 17 B 345–887 and 18 A 1–195. 17  Edited in vol. 18 B 1–317 Kühn = CMG V 9,2 197–378 Heeg. 18  Edited in vol. 15 418–919 Kühn = CMG V 9,1 117–366 Helmreich. 19  Edited by Kühn in vols. 17 A 1–302 (Epidemics 1); 17 A 480–791 (Epidemics 3); 17 A 793–1009 and 17 B 1–344 (Epidemics 6) and by E. Wenkebach, with the collaboration of F. Pfaff for the Arabic in CMG V 10,1 3–151 (Epidemics 1); CMG V 10,1 155–410 (Epidemics 2); CMG V 10,2,1 1–187 (Epidemics 3); CMG V 10,2,2 3–508 (Epidemics 6). The arabic text has been now edited by U. Vagelpohl in CMG Suppl. Or. V 1 (2014) and V 2 (2016). 20  Edited by Kühn in vol. 16 489–840 = CMG V 9,2 3–178 Diels. 21  Edited by Kühn in vol. 16 1–223 = CMG V 9,1 3–113 Mewaldt. 22  Unedited; editing in progress by G. Strohmaier for CMG.

Galen’s Practice of Textual Criticism

57

remaining in Hebrew).23 Commentaries on other treatises (On Sores, On Nutriment, On Wound in the Head) are lost. Galen presents his commentaries as the final written outcome of his practice of reading Hippocrates in the presence of disciples and friends (ἑταῖροι). Unfortunately, we are unable to assess the degree of correspondence between those oral commentaries and the written ones; it is certain, however, that over the years Galen updated the commentaries he had composed beforehand. Commenting occupied Galen during a long span of time and inevitably changed his exegetical perspective. He develops an increasing ‘philological’ interest during his second Roman sojourn, and, in his last commentaries, he increasingly makes references to the wording of the Hippocratic text, as well as to the variant readings he found in contemporary and ancient commentaries and in different editions. This philological attitude progressively intensifies also due to the fact that, in this last period, he is tackling particularly obscure and (he believes) highly corrupted texts—this is especially the case with Epidemics 6. In Rome, the reference editions of the Hippocratic works were probably those by Artemidoros Capiton and Dioscorides, two scholars of the Hadrianean age, unmentioned in other sources but often mentioned by Galen, who usually displays towards them an unsympathetic and sometimes aggressive attitude, being convinced that the two were shameless innovators, prone to excessively amending the texts transmitted in the most ancient copies and known to the commentators, especially the ancient ones.24 The words opening the Commentary on Hippocrates’ Epidemics 6—a methodological statement on what Galen is going to do in the commentary itself—can be a useful point of departure for a concise examination of Galen’s method and objectives in commenting on this treatise, as well as for incursions in his commentaries on other Hippocratic works. In fact, this passage contains remarkable information on textual criticism and lists some of the “philological” reasoning preceding and regulating Galen’s work as a commentator. I want to state, from the very beginning, that Galen is not an editor of Hippocratic texts; his scholarship exclusively emerges as an instrument for their evaluation and correct interpretation. Gal. In Hipp. Epid. 6 (proem, 17 A 793,1–795, 16 Kühn = p. 3,4–4,25 Wenkebach; the division in paragraphs is mine):

23  First edited by F. Rosenthal, Bulletin of the History of Medicine 40 (1966) 226–45. 24  See A. Roselli, “Galeno e la filologia del II secolo”, in E. Bona/C. Lévy/G. Magnaldi (ed.), Vestigia notitiai: Scritti in memoria di Michelangelo Giusta (Alessandria: Edizioni dell’Orso, 2012) 63–80; Ead., “Galeno e le edizioni ippocratiche di Artemidoro e Dioscoride”, in S. Fortuna/I. Garofalo/A. Lami/A. Roselli (ed.), Sulla tradizione indiretta dei testi medici greci: i commenti. Atti del IV seminario internazionale di Siena, Certosa di Pontignano, 3–4 giugno 2011 (Biblioteca di Galenos 5; Pisa Roma: Fabrizio Serra, 2012) 15–27; Manetti, “Medicine and Exegesis”, 1177–82.

58

Amneris Roselli

1. … this, more than other books by Hippocrates, has been offended (ἐλυμήναντο) by many among the interpreters, who have, this way or another, modified the lexis according to how each of them hoped to explain it (ἐξηγήσασθαι) plausibly (πιθανῶς). For this reason, I have been compelled to search for the most ancient copies (τὰ παλαιότατα τῶν ἀντιγράφων), as well as the commentaries (ὑπομνήματα) of those who have first explained it—among whom Zeuxis, Heraclides of Tarentum, Heraclides of Erythrae, and, before them, Bacchios and Glaukias. 2. If, then, after illustrating the ancient lesson (τὴν παλαιὰν γραφήν), they had said it was probable (εἰκός) that the lexis was wrong, and had accordingly speculated (ὑπονοεῖν) that Hippocrates’ lesson was this or that, I would have approved of them—on condition that, after the correction (μετὰ τὴν ἐπανόρθωσιν), I had seen that they were teaching something useful (χρήσιμον) and, simultaneously, consistent with the thought (γνώμης ἐχόμενον) of the ancient author. Since, in some cases, neither result is obtained, I established it was much better, preserving the ancient lesson (τὴν ἀρχαίαν γραφήν), to always make an effort to explain it—and when I was not able to do so, to amend it plausibly (πιθανὴν τὴν ἐπανόρθωσιν). 3. This is what Heraclides (of Tarentum) did in the second book on Epidemics (Epid. 2 2.20) in the case of the word (κατὰ τὴν λέξιν), in which it is written: πρὸς δὲ τὸ Ἀφροδίσιον αἱ οὐραὶ (ΟΥΡΑΙ) ἔβλεπον. Because according to the commentators, the word οὐραί is not convincing (ἀπιθάνως εἴρηται): “maybe, he says, it was written θύραι (ΘΥΡΑΙ) with theta (Θ)—but, since the middle stroke of the letter was ruined, the copyist (ὁ βιβλιο­γράφος) believed it was οὐραί (ΟΥΡΑΙ)”. It is hence possible that the loss of a thin fiber (of the papyrus) resulted into the loss of the letter—or that, from the beginning, it was written unclearly and vanished with time. 4. Among all those who have changed the ancient lessons (ὑπαλλαξάντων τὰς παλαιὰς γραφάς), Capiton and Dioscorides are, I believe, the ones who did it most impudently (τολμηρότατα). 5. Subsequently, I wondered whether it was better to mention all those who changed the lessons, or to mention only those who reasonably (εὐλόγως) did so, or, perhaps, not to mention anyone; and I found the best thing would be to mention everyone, if none of the readers were bothered by the length of the commentary. Then again, since many not only lament this type of commentaries but also those of average length, and only care for useful information (χρήσιμα), (I thought it is best) to provide an exegesis that is in-between (μέσην) these two types—and to declare this from the start, so that those who do not care for such material can avoid this commentary.

Taking the topics in this passage as a starting point, I shall now discuss in detail the cornerstones of Galen’s practice of textual criticism.

Galen’s Practice of Textual Criticism

59

The Practice of Textual Criticism Defining the Quality of the Text to be Commented On For the ancient readers as well, Epidemics 6 was a remarkably obscure text and, according to Galen, one particularly “offended” (ἐλυμήναντο) by the interventions of those commentators who attempted to make it understandable by modifying the ancient readings.25 In general, obscurity depends on brachylogy and incompleteness in the formulation of the argument, two deficiencies that render the texts especially liable to manipulation (this is the case of Epidemics 2 and 6). For this reason, such texts warrant a more careful approach, also with regard to the occurrence of variant readings, which can be detected by means of an accurate comparison with ancient editions and commentaries. Commenting on Hippocratic texts, Galen will need to take their ‘stylistic’ peculiarities into account, and it will be necessary for him to devote special attention to the correctness of such obscure texts. Accordingly, commenting on Epidemics 6, he pays more attention to the variants than in the other commentaries λυμαίνομαι. Galen’s level of attention to the quality of a text varies, depending on its style.

The Search for Ancient Witnesses (Both Ancient Copies and Ancient Commentaries) Being aware of dealing with a very obscure text as Epidemics 6, one that has undergone several acts of violence, Galen attempts to retrieve as many antigrapha of the text and commentaries as possible—especially the most ancient, and presumably less corrupted, ones. Introducing this Commentary (above p. 58), Galen resumes the argument he had used in the Commentary on In the Surgery (Off.)26—where he even lists the support materials of the most ancient books he (or, better, some ancient commentator) consulted: papyrus, parchment, or bark. 25  Galen usually employs the verb λυμαίνομαι to indicate damages to the body and bodily functions, and only rarely uses it in a philological context. The only other occurrence of the verb with reference to the quality of a text can be found in his Commentary on Epidemics 3 (17 A 633,8 Kühn = p. 100,3 Wenkebach), where, exactly like in the present case, Galen condemns ‘amendments’ of any kind (παντοίως) made by many of the most recent physicians (πολλοὶ τῶν νεωτέρων ἰατρῶν). This use of λυμαίνομαι is consistent with the use by the anonymous author of the Prolegomena in Aratum, who, introducing his commentary on the Phainomena, includes a broad typology of corruptors in the list of those who have “offended” the text; see: proleg. in Aratum, 33,10 Martin: “this poem has been offended by painters and astronomers and grammarians and geometers, each of them freely proposing their own drawings/lessons and interpretations” (ἐλύμηναν δὲ πολλοὶ τοῦτο τὸ ποίημα ζωγράφοι καὶ ἀστρονόμοι καὶ γραμματικοὶ καὶ γεωμέτραι, ἕκαστος αὐτῶν πρὸς τὸ βούλημα τὸ ἴδιον γραφὰς καὶ ἐξηγήσεις ἰδίας ποιούμενοι). 26 The Commentary on In the Surgery dates certainly from before the one on Epidemics 6, which was written after 192 AD.

60

Amneris Roselli

Gal. Comm. in Hipp. Off. 18 B p. 630,10–632,1 Kühn: 1. I shall now illustrate that which (is necessary) not because of the (brachylogical) nature of the book, but because of all those who have promptly altered, as they pleased (ἃ δὲ οὐκέτι τὸ βιβλίον, ἀλλ᾽ οἱ μεταγράφοντες [ἢ] ἑτοίμως εἰς ὅπερ ἂν αὐτοὶ βουληθῶσι),27 the lessons they have received from the most ancient (τῶν πρεσβυτέρων) (manuscripts/commentaries). In fact, some had made efforts to also retrieve very ancient books, written three hundred years beforehand—some on papyrus, others on animal hide, or bark, similar to those that are in use here in Pergamum. 2. My purpose was to learn all this from the early commentators, so as to find the genuine (τὰς γνησίας 28) among the most ancient and authoritative comment(ator)s (ἐκ τῶν πλείστων τε καὶ ἀξιοπιστοτάτων). 3. And the results of this enterprise exceeded my hopes: I found, in fact, that almost all the exemplaries and the commentaries by the commentators were mutually consistent; so that I have been surprised by the arrogance of those who have recently written commentaries or put together their own edition (ἰδίαν ἔκδοσιν) of all of Hippocrates’ works— among whom are Dioscorides and Artemidoros Capiton, who have proposed numerous innovations (καινοτομήσαντες) to the ancient lessons. 4. It seemed to me that, if I mentioned all the lessons, the commentary would be too long: I thought it preferable, then, to mention only the ancient ones, also adding some lessons that present small changes (βραχὺ μετακίνηνται),– and, among these, especially the lessons on which the early commentators of this book agree, and there are four of them: two, Zeuxis and Heraclides, who have written commentaries on all of Hippocrates’ works, while Bacchios Asclepiades did not comment on all of them.

The commentators mentioned in the proem to the Commentary on Epidemics 6 and in the Commentary on In the Surgery are Empiric and Herophilean physicians who date from the period across the 2nd and the 1st century b.c.e. Bacchios of Tanagra the Herophilean is the most ancient among them; Galen quotes him here and in two other commentaries (on Epidemics 3 and on Aphorisms), as well

27  See the passage from Origen Hom. 1 in Ps. 77, p. 352,14–15 Perrone, where he recommends not to rush to hasty corrections (προπετῶς): οὐ διὰ τοῦτο ἡμᾶς χρὴ τολμᾶν καὶ προπετῶς ἥκειν ἐπὶ τὴν διόρθωσιν. 28  The integration of the word γραφάς is justified by the recurrence of the term in this context 4 more times, but this is the only evidence of a connection between the adj. γνήσιος and γραφή.

Galen’s Practice of Textual Criticism

61

as in his Lexicon.29 References to the others—especially the Empirics Heraclides of Tarentum and Zeuxis—are much more frequent;30 those are clearly, to Galen, great and indisputable authorities. What Galen really means by the phrase ‘ancient copies’ (παλαιά, or παλαιότερα, or παλαιότατα ἀντίγραφα) is hard to establish. He refers to ‘ancient copies’ in about thirty passages of his commentaries. This happens most frequently in the commentaries on Prorrhetic 1 and Epidemics 6, both problematic works: the first, because of its doubtful authenticity (see above p. 56); the second, because of its obscurity. The concentration of references to ‘ancient copies’ in those two commentaries suggests that, while he was writing them, Galen actually had ancient manuscripts available for consultation. It must also be remarked, however, that the ‘ancient copies’ are mostly mentioned together with the (ancient) commentators and/or commentaries. Moreover, one should keep in mind that, as Galen reports, Heraclides of Tarentum had already claimed, in a commentary of his, to have searched for ancient copies in order to comment on a passage from Epidemics 2.2.6: “A supporter of this reading is Heraclides of Tarentum, one of the ancient commentators, who said he had found it in an ancient copy” (Vagelpohl’s translation).31 This passage illuminates the continuity of an exegetical practice grounded in the authority of the earlier commentators and suggests that Galen may be mostly referring to ancient commentaries also when referring to ‘ancient copies’. Whether Galen actually has ancient manuscripts at hand when he mentions them or whether, to the contrary, he refers to ancient manuscripts indirectly, filtered by the commentators to whom he has access, is a question that remains open.

Plausibility of the Text / Variants / Commentaries It should be noted that Galen’s concern with lesson authenticity is subordinated to his objective of offering, together with plausible lessons, a plausible interpretation of the text. In the choice among variants, as well as in the assessment of a unanimously transmitted text, the criterion of plausibility (πιθανῶς, πιθανός, ἀπιθάνως) must be followed, and the plausibility of the text, as well as of its 29  Besides the commentaries and the Lexicon, Galen quotes Bacchios in the treatises on pulse (De differentia pulsuum and De dignoscendis pulsibus). On Bacchios see H. von Staden, Herophilus: The Art of Medicine in Early Alexandria (Cambridge: Cambridge University Press, 1989) 484–500. 30  The fragments of Heraclides have been edited by K. Deichgräber, Die griechische Empirikerschule: Sammlung der Fragmente und Darstellung der Lehre (Berlin: Weidmannsche Buchhandlung, 1930) 172–202, and by A. Guardasole, Eraclide di Taranto. Frammenti (Testo critico, introduzione, traduzione e commentario a cura di A.G.; Napoli: D’Auria, 1997); the fragments of Zeuxis by Deichgräber, cit. 24–27; 209; 263. 31 Gal. In Hipp. Epid. 2, p. 221,41–222,1 Pfaff = p. 301,13 Vagelpohl.

62

Amneris Roselli

interpretation, is the ultimate aim of the exegete.32 Such a criterion may appear excessively subjective to our postlachmannian mentality. Nonetheless, plausibility, albeit inferior to certitude, is the legitimate objective of the interpreter.33 ‘Credibility’ is a standard: it must be an intrinsic characteristic of the text; it must be the aspiration of its interpreters; and it is the aim of Galen’s very exegetical activity. Problematic readings can be classified as corrupted or erroneous (the vocabulary for errors being essentially constituted by σφάλμα and ἁμάρτημα,34 but they are more often classified as ‘not convincing’ (ἀπίθανοι). A lack of plausibility elic­ its attention on the part of the interpreter and justifies the correction of a text. On the other hand, the search for a low level of difficulty (εὐπορεῖν) presents the risk of trivialization of the texts. The commentaries on Hippocrates contain a number of passages where, Galen suspects, trivialization has occurred. See, for instance, the Commentary on Epidemics 6 17 A 1005,3–12 Kühn (= p. 121,15–22 Wenkebach): I have said several times that in this book there is a huge number of variants (μυρίας εἶναι γραφάς)—and this is because, due to its obscurity, it is subject to modifications, this way or that way, according to how convincing the explanation will be considered (τὸ τῆς ἐξηγήσεως ἔσεσθαι πιθανόν). But I prefer ancient lessons, even when they seem unconvincing (κἂν ἀπίθανοι δοκῶσιν εἶναι) and present a higher degree of difficulty (ἀπορίαν). For the following reason one can be convinced that they were originally written like that: because, in spite of the difficulties in interpreting them (καίτοι γ᾽ ἀπορίας οὔσης περὶ τὴν ἐξήγησιν), the most ancient commentators agree about them. If they had dared modify them, they would have done so adopting a less difficult lexis (εἰς εὐπορωτέραν ἂν ὑπήλλαττον λέξιν), in order to make the explanation convincing (πιθανὴν γενέσθαι τὴν ἐξήγησιν).

32 See In Hipp. Epid. 6 17 B 9,8 Kühn (= p. 128, 19 Wenkebach), on the meaning of the word ἡσυχία: “this seems more plausible (πιθανότερον) from a certain perspective, and that from another one”; In Hipp. Epid. 6 17 A 908,12 (= p. 66, 12 Wenkebach) “because ... neither (way to divide the text with punctuation) is plausible (πιθανόν), many have set to modifying the text, like (they do) in all of the other places that appear obscure because of one word”. 33  See, for instance: In Hipp. Epid. 6 18 A 992,4 Kühn (= p. 113,25 Wenkebach) “if one dares to abandon the ancient reading (ἀποχωρεῖν τις τολμᾷ παλαιᾶς γραφῆς), one must do so according to a certain degree of plausibility (κατά τι πιθανόν); I just maintained that it is reasonable (εὔλογον) to speculate that here (the adjective) μακρός was written twice, and that one of the two occurences was omitted”; Galen here suggests the haplography of μακρός as a plausible explanation of the error and provides the criterion for correcting it. 34  The same vocabulary is recurrent in Origen, cf. Hom. in Ierem. 15.5.12–13 Klostermann ἐπισκεψάμενοι καὶ τὰς λοιπὰς ἐκδόσεις ἔγνωμεν γραφικὸν εἶναι ἁμάρτημα; and Hom. 1 in Psalm. 15 (p. 104,3–4 Perrone) (infra, n. 49).

Galen’s Practice of Textual Criticism

63

Here Galen speculates that lectiones faciliores may have intervened in the most recent copies in order to facilitate interpretation and maintains that a difficult reading can not result from an amendment. Moreover, in the Commentary on Prorrhetic 1 16 628,12–629,7 Kühn (= p. 69,22– 70,2 Diels), besides recognizing the eventuality of trivialization, Galen condemns the practice of dissimulating one’s “trivializing” intervention: All ancient manuscripts (ἁπάντων μὲν τῶν παλαιῶν ἀντιγράφων) display this reading, and all those who have commented (ἁπάντων δὲ τῶν ἐξηγησαμένων) on this book have accepted it. Nevertheless, some people born yesterday or the day before that35 change it into a lesson they find easier to explain (ἐξηγήσασθαι ῥᾷστον). I do not fault the eventuality that one who cannot find an explanation assumes (ὑπονοῆσαι) the lesson to be wrong and changes it (μεταγράψαι) following the meaning he believes to be correct; yet, to explain (the reading) as if, from the very beginning, it had been written just the way he dared change it ( οὕτως γεγραμμένον ἐξ ἀρχῆς ὡς αὐτὸς ἐτόλμησε μεταγράψαι), is something I believe should be chastised.

Galen opts for the ancient reading, although it may be more difficult to explain. However, being difficilior is not a merit for a variant; it is, instead, a tolerable fault that allows for the ancient reading to be saved. See In Hipp. Epid. 6 17 B 101, 8–9 Kühn (= p. 178,17 Wenkebach): “I prefer the ancient readings, although they may be harder to explain (ἀλλ᾽ ἐγὼ τὰς παλαιὰς αἱροῦμαι, κἂν χαλεπωτέραν ἔχωσιν τὴν ἐξήγησιν).” I would not suggest here that Galen chooses the lectio difficilior; what he prefers is the lectio antiquior. Recognizing that trivialization occurs is one thing; using the notion of ‘nontrivial lesson’ to choose among variants or to amend the transmitted lesson is something different.36

Objective of and Methodology for Amendment (ἐπανόρθωσις) When dealing with medical texts, the aim of amending is to reinstate a text that is medically useful (χρήσιμον). It is the one case when correcting is worth the effort. Commentators must illustrate the ancient readings and assess it; then, if not plausible, amend it, conjecturing (ὑπονοεῖν) what may actually be the origi­ nal lesson. In order for the amendment to be acceptable (ἀπεδεξάμην αὐτούς), it must fulfil the requirement of two different criteria: the first regards the value of 35  The reference is almost certainly to Artemidoros Capiton and Dioscorides. 36  An excellent investigation of the problem of the lectio difficilior in ancient philology, with broad documentation on Galen and the Christian authors, can be found in Ferreri, “Alcune riflessioni”, 9–61; the new Homelies by Origen provide two new, very explicit cases of choice between variants on the basis of the lectio difficilior (Hom. 1 in Ps. 77, p. 351,11–17 Perrone, ᾽Ασάφ / ᾽Ησαΐου see n. 46; Hom. 5 in Ps. 77 p. 411,10–13 Perrone; cf. also p. 15, n. 45 πίοισιν / πλείοισιν).

64

Amneris Roselli

the text from a medical perspective; the second is based on consistency with the author’s doctrine. In some cases, Galen also refers to a linguistic criterium—i.e., the use of a vocabulary and technical terminology coherent with the ones used by Hippocrates. All these criteria may be mutually combined and subsumed under the rule ‘explain the author by the author’—an Aristarchean formula which Galen (De dign. puls. 8 598,6 Kühn) regards as a “law of interpretation” (νόμος ἐξηγήσεως), and that eliminates the risk of dwelling in idle comments:37 and for me this is the law of interpretation: both to explain each author by himself; and not to rave unrestrainedly, uttering empty assumptions and undemonstrated propositions.38

These criteria are valid for the evaluation of the microunits of a text, such as words and short phrases, as well as for the evaluation of whole texts.

Causes of Corruption Throughout the commentaries, Galen has several occasions for speculating on the causes for textual corruption. Although he does not offer a systematic treatment of this topic, an extensive typology of causes for textual corruption can be easily inferred from the commentaries. Among the causes for errors one must surely list the negligence (ῥᾳθυμία) of the copyists (cf. Ιn Hipp. Epid. 2 2.13 (p. 230,20–27 Pfaff, see below): to copyists are attributed word omissions (for instance haplographies) as well as the addition of words at the margins, and the faulty reading of words unclearly written in their antigrapha.39 Moreover, Galen does not exclude the possibility of authorial mistakes—especially considering that not all the treatises he comments on had been prepared for publication by Hippocrates. A (highly corrupted) passage of the Commentary on In the Surgery lists the possible culprits as well as the circumstances wherein a mistake may have occurred, cf. Gal. In Hipp. Off. 18 B 778–9 Kühn: 37  The same rule is systematically applied by Origen; in the case of the commentary on Hosea, a comparison with other writings not only allows to preserve the transmitted text without amendments, but also provides a more profound interpretation (see Origène, Philocalie, 1–20 Sur les Ècritures [ed. M. Harl; Sources Chrétiennes 302; Paris: Éditions du Cerfs, 1983] 344); this method is defined as παρατήρησις or ἐξέτασις ὁμοίων ῥητῶν. 38  De dign. puls. 8 598,6 Kühn καὶ γάρ μοι καὶ νόμος οὗτος ἐξηγήσεως, ἕκαστον τῶν ἀνδρῶν ἐξ ἑαυτοῦ σαφηνίζεσθαι καὶ μὴ κεναῖς ὑπονοίαις καὶ φάσεσιν ἀναποδείκτοις ἀποληρεῖν, ὅ τι τις βούλεται. 39  Negligence on the part of copyists is indicated also by Origen as the main cause for errors, in Comm. in Matth. 15.14.87 Klostermann: πολλὴ γέγονεν ἡ τῶν ἀντιγράφων διαφορά, εἴτε ἀπὸ ῥᾳθυμίας τινῶν γραφέων, εἴτε ἀπὸ τόλμης τινῶν μοχθηρᾶς τῆς διορθώσεως τῶν γραφομένων, εἴτε καὶ ἀπὸ τῶν τὰ ἑαυτοῖς δοκοῦντα ἐν τῇ διορθώσει προστιθέντων ἢ ἀφαιρούντων (see n. 48).

Galen’s Practice of Textual Criticism

65

Some mistakes are the author’s (τοῦ γράψαντος) of the treatise, because he wrongly eliminated or added or changed one or two letters (παρὰ γράμματος ἑνὸς ἢ ­δυοῖν ἀφαίρεσιν ἢ πρόσθεσιν ἢ ὑπάλλαξιν ἁμαρτόντος); others were committed by the first copyist (ὑπὸ τοῦ πρώτου βιβλιογράφου), because the author himself had written something in obscure characters (ἀσαφέσι γράμμασι γεγραφότος αὐτοῦ τοῦ συν­θέν­τος [scripsi: συνθέτου Kühn]). *** Other errors are due to the fact that the papyrus broke, or to the fact that, due to some other circumstance, a letter got lost or confused. And maybe, in some cases, mistakes were committed by those who passed from the ancient writing to a writing system that is posterior to the age of the author, on the occasion of a μεταγραμματισμός; for all these reasons, a quantity of errors were accumulated in many of the copies.

A mention of μεταγραμματισμός as a cause for error is quite rare in ancient Greek texts. In another case (In Hipp. Epid. 6 17 B 111,7–16 Kühn = p. 185, 2–5 Wenkebach), though, Galen returns to this, urging to always keep in mind the possibility that an error may have intervened at the moment of passage (μετάθεσις) to an alphabet distinguishing between epsilon and eta and between omicron and omega, coming from an alphabet that did not have these distinctions.40 Not all mistakes, nonetheless, come from difficulties in copying. As mentioned above, one must also consider the commentators’ ease in modifying the readings they found difficult to explain or unclear.41 I choose to quote, for its subtlety, a passage from the Commentary on In the Surgery (18 B 729,16–730,6), where Galen focuses on an issue regarding not the general meaning of the text, but its exactness: so here as well, changing the case, or the word, they wrote ‘ἁπλοῦν, ἐγκύκλως [instead of ἔγκυκλον]’, understanding well enough (ὀρθῶς), but audaciously (τολμηρῶς) changing (the wording). In most cases, they even dared (ἐτόλμησαν) to change not having well understood (οὐδὲ καλῶς νοήσαντες);42 for this reason I mentioned (this reading) here, although beforehand and so far I had not mentioned an infinity of readings (of this kind). Because this also is apparent to me, that despite having well understood (καλῶς μὲν νοήσαντες) (the meaning of the passage), they wrote it

40 Gal. Ιn Hipp. Epid. 6 17 B 111,7–16 Kühn (p. 185.2–5 Wenkebach): “because of this, lots of attention must be paid to those readings which can be corrected replacing (μεταθέντας) the sound of the letter eta (η) with the sound of the letter epsilon (ε), or the other way round. The same must be done for omicron (ο) and omega (ω), because these two sounds were also graphically rendered by the same character (δι᾽ ἑνὸς χαρακτῆρος ἐγράφοντο). Galen thus explains the possibility of writing σκέψις instead of σκῆψις in Hipp. Epid. 6 3.23. 41  For the same observation in Origen, cf. supra n. 27. 42  A similar allegation of misunderstanding can be found in Origen; cf. Neuschäfer, “Origen” 132 and n. 211—and, naturally, in scholiastic literature.

66

Amneris Roselli

changing it into a clearer form (ἐπὶ τὸ σαφέστερον),43 making mistakes in secondary details, without understanding (μὴ νοήσαντες) how the author wanted (the text to be written).

Based on these premises, it is not surprising that Galen devotes so much space to the “impudence” (τόλμη) of the commentators who modify Hippocrates’ text following their own inclinations.

Ways to Amend Repeated criticism of those who alter the texts—Artemidoros and Dioscorides in the first place, but also several recent commentators—is occasionally expressed in harsh tones, and quite often through the verb τολμάω (dare) and its derivatives τολμηρός and τολμηρῶς, characterizing reckless, inconsiderate people or behaviour with no sense of their own limits.44 This style of expression45 in a philological-textual context is not unique to Galen: it can also be found in Latin authors (using audeo, ausus), as well as in Origen.46 43  See also Ιn Hipp. Epid. 6 17 B 98,2–7 Kühn (p. 176.16–21 Wenkebach): “what was the need, on the part of all exegetes, to modify this lesson in order to clarify it (ἐπὶ τὸ σαφέστερον)? And indeed, in those cases when they are in straits (to explain the text), they write differently in order to dispense with the difficulty (εὐπορίας ἕνεκεν). But I found no copy containing this reading, although I looked up several, exactly because I wanted to know the ancient readings, and those that were either accepted by everyone, or upon which there was a general disagreement”. 44  See also In Hipp. Epid. 6 18 A 992,4 Kühn (= p. 113,25 Wenkebach) “if one dares to abandon the ancient reading (ἀποχωρεῖν τις τολμᾷ παλαιᾶς γραφῆς), cited above, n. 33. 45  Galen employs the same vocabulary to attack those who have proposed anatomical theories, arguments, or descriptions that are totally unacceptable in his eyes, because contradicted by logic or experience: it is not, then, a vocabulary restricted to the philological domain. 46  On the centrality of this notion in Origen, see Neuschäfer, “Origen” 133–136 and notes 392ff. A passage from the Commentary on Matthew appears to be especially significant for a comparison with Galen: besides a reference to τόλμη, this passage also contains a compendium of causes for textual corruption that perfectly match the ones found in Galen’s commentaries, which emerge from the passages discussed so far. This had already been remarked by Neuschäfer, 390 and notes 187, 188; see Ιn Matth. commentarii 15.14.85–93 Klostermann: “now it is evident that great diversity exists between different manuscripts, because of the negligence of certain copyists (ἀπὸ ῥᾳθυμίας τινῶν γραφέων), or because of bad audacity on the part of some (ἀπὸ τόλμης τινῶν μοχθηρᾶς); because of those who have neglected the diorthosis ( τῆς διορθώσεως) of what has been written, or through the fault of those who have added to or removed from (προστιθέντων ἢ ἀφαιρούντων) the diorthosis whatever struck their minds” (see Neuschäfer, “Origen” 372, n. 24). An intentional alteration is also presumed in Hom. 2 in Ps. 15 (p. 104 Perrone): Origen finds that, in the Acts of the Apostles, the words ἠγαλλιάσατο ἡ δόξα μου are quoted as ἠγαλλιάσατο ἡ γλῶσσα μου, and he comments: “I take heart and say: the Apostles are not in contrast with the prophetic logos: perhaps a writing error (ἁμάρτημα γραφικόν) has intervened, that is, having someone not understood (μὴ νοήσαντος)

Galen’s Practice of Textual Criticism

67

Regarding Galen’s propensity for amendments, one can easily find passages where he takes a stance that demonstrates a remarkable openness—under certain conditions. For instance, in the Commentary on Epidemics 2 2.13 (p. 230,20–27 Pfaff = 333,7–13 Vagelpohl), he observes: When a text is incomprehensible because an error that occurred in the original (copy), I for one am not averse to changing, adding or removing a letter from it so that it becomes comprehensible. Even today we see many such errors and omissions occur because of the scribes’ lack of education and negligence; but I have no praise for people who change such things when they do so without knowing what the ancient reading was (Vagelpohl’s translation).

The effort to take into account the highest possible number of copies and commentaries must always precede any correcting intervention, steering away from erroneous amendments and repeated, gratuitous innovations. This method remains valid, although experience shows that some errors may be traced as far back as the first hand copy by the first copyist47 (whom one may imagine to be more attentive and less prone to errors) or even all the way back to the author. In a passage from the De difficultate respirationis, where he comments on two parallel chapters in Epidemics, Galen provides a clear and detailed reflection on the possible genesis of textual corruptions between the two extremes of carelessness (ῥᾳθυμία, ῥᾳδιουργία) and search for excessive precision (περιεργία) on the part of the copyist.48 While not a commentary, De difficultate respirationis tackles the same issues as the Commentaries on Epidemics. Here we find a comparative the expression ἠγαλλιάσατο ἡ δόξα μου, they have changed it into ἠγαλλιάσατο ἡ γλῶσσα μου; if someone does not grant that this is an error in the transcription, let them explain it, so as to clarify it (ἐπὶ τὸ σαφέστερον)”; also see Hom. 1 in Ps. 77 (p. 351 Perrone): “a writing error has occurred in the manuscripts of the Gospel (γέγονε δὲ περὶ τὰ ἀντίγραφα τοῦ εὐαγγελίου σφάλμα γραφικόν): so that Isaiah’s word be fulfilled—it says—‘I shall open my mouth in parables’. Probably, one of the first copyists (ἕνα τῶν ἀρχῆθεν γραφόντων), not knowing about the prophet Asaph, and having found so that Asaph’s word be fulfilled, must have supposed this to be an error (ἁμάρτημα) and dared (τετολμηκέναι), because of the rarity (ξενισμοῦ) of the prophet’s name, to change it from Asaph to Isaiah”; a remarkable case of trivialization! 47  Galen frequently ascribes errors to the first copyist (πρῶτος βιβλιογράφος), cf. e.g. In Hipp. nat. hom. 15 24,15 Kühn (= 15,12–13 Mewaldt) ὡς ἐνεδέχετο τὸν πρῶτον βιβλιογράφον ἁμαρτεῖν ἀντ᾽ἄλλου γράψαντ᾽ἄλλο; In Hipp. acut. 15 911,8 Kühn (= CMG V 9,1 362,9–11 Helmreich) τὰ δ᾽ ἐν ἀρχῇ τῆς ῥήσεως παραλελεῖφθαι ἁμαρτόντος τοῦ πρώτου βιβλιογράφου καὶ μετὰ ταῦτα μεινάσης τῆς ἁμαρτίας. 48  This category could also comprise the behavior of those who, not tolerating a solecism, ‘improve’ a text, and consequently lose its proper meaning: this is mentioned by Origen in the Commentary on Hosea (Philocalia 8.1.4–6: ὡς καὶ τολμᾶν τινὰς προφάσει διορθώσεως μετατιθέντας ἀλλοιοῦν τὸν ἐγκείμενον περὶ τὰ δοκοῦντα ἀνακολούθως γεγράφθαι ῥητὰ νοῦν).

68

Amneris Roselli

analysis of two passages, constituted by three or four pairs of adjectives that refer to respiration. Galen maintains that the correct text is the more extended one (Epid. 6 2.3), and he ascribes the omission in the shorter text (Epid. 2 3.7) not to the author but, instead, to an ancient copyist. The omission of a pair of adjectives may be unconscious or, to the contrary, conscious (in which case, it would be aimed at avoiding an apparent repetition in two pairs of adjectives which may be considered synonymous), but in this latter case, the intervention would be a case of excessive zeal on the part of the copyist, who did not limit his role to being a simple ‘servant’ (ὑπηρέτης) of the text (par. 3). Galen takes this occasion to remind us that an error interpolated in a text in the ancient phase of its transmission is preserved in the subsequent copies—either because it remains unnoticed, or because the ancient editors only highlighted the problem by means of a diacritic sign, without daring to modify the text.49 The final part of the passage—at this point unrelated to the initial issue—is about the practice of amendment and the criteria behind it. Galen only accepts interventions that modify—or add—very few letters, provided that the implemented changes preserve the truth from a medical perspective and are consistent with the author’s doctrine. This part presents an effective synthesis of the employed methodology and is worth reading through; cf. De diff. resp. 7 891,11 ff. Kühn: 1. We must begin the explanation of the rhesis in Epidemics 2 (2 3.7) that is given as: πνεῦμα· (1) σμικρόν, πυκνόν· (2) μέγα, ἀραιόν, (3) σμικρόν, ἀραιόν· (4) ἔξω μέγα, εἴσω μικρόν [...]. 2. In Epidemics 6 (6 2.3), this same rhesis is written in a more complete form; after the third pair of dyspneas, there is one more pair that has clearly been omitted in Book 2. This is not, in my opinion, an omission on the part of the author. In fact, it is clear that Hippocrates arrived at his number of dyspneas (i.e. four) not on the basis of experience or relying on memory, but through a process of logical division: so it is not likely that he overlooked one. Rather, it was one of the ancient copyists who made the mistake; afterwards, the book was published with this error which, it seems to me, has been perpetuated ever since. Some do not care about the books of the ancients and do not even notice if something is missing (in a text), or if it can be found in another book. Others do notice, but dare not fill the void. Ancient editors were not as ready to modify a reading which recurs in the same form in all the ancient manuscripts as nowadays; for them, it was sufficient to insert a sign (ἐπι-

49  The Αlexandrine practice, here only mentioned by Galen, is also recorded by Origen, in order to mark content discrepancies between the Greek tradition and the Hebrew one; cf. Ιn Matth. comm. 15.14.103–110 Klostermann: καὶ τινὰ μὲν ὠβελίσαμεν ἐν τῷ Ἑβραϊκῷ μὴ κείμενα (οὐ τολμήσαντες αὐτὰ πάντη περιελεῖν), τινὰ δὲ μετ’ ἀστερίσκων προσεθήκαμεν, ἵνα δῆλον ᾖ ὅτι μὴ κείμενα παρὰ τοῖς Ἑβδομήκοντα ἐκ τῶν λοιπῶν ἐκδόσεων συμφώνως τῷ Ἑβραϊκῷ προσεθήκαμεν.

Galen’s Practice of Textual Criticism

69

σημήνασθαι) that showed that, in that point, something was missing in the division (dihairesis) […]. 3. (p. 893 K) This is not the only place where a rhesis in Book 2 differs from that in Book 6. Further on, when Hippocrates compares [...], the second pair lacks completely in Book 2. This clearly shows that, also in this case, the copyist (γραφεύς) has been careless or negligent in the extreme. If he had (simply) omitted all the words, he would (merely) have been extremely careless, but if, on the other hand, he had thought they were equivalent to each other and not recognized the differences between them, he would have reached a peak of futility (περιεργία), revealing himself as one of those who want to correct Hippocrates’ writings rather than be of service (ὑπερέτης) to them. 4. This should be borne in mind in the many cases in which readings are obscure, and in which it seems plausible that something has been omitted or badly written. In fact, there is no reason not to render the phrasing complete and true by adding what is missing or correcting what is wrong. Just as it is hazardous to change the readings of the ancients, so it is the duty of a good exegete to solve these difficulties preserving the readings as they were written by means of (just) small additions and changes. 5. But I think that, later on, I will show in practice (δι᾽αὐτῶν τῶν ἔργων) what one ought to do. One should not proceed simplistically or haphazardly; one should, instead, show the truth of the matter (τήν τε τῶν πραγμάτων ἀλήθειαν), demonstrating, first and foremost, that this is what the ancient author thought (γνώμη), and that the modified reading (μετακοσμηθεῖσαν) agrees with the two (sc. both the truth of the matter and the ancient author’s thought).

Concluding Remarks Galen’s philology is a philology of medical texts. Accordingly, it abides by the standard of truth, and it is interested in the quality of a text in service of its correctness from a medical perspective. The most renowned editions of Hippocrates in Galen’s time must have been those by Artemidoros Capiton and Dioscorides, two scholars of the Hadrianean age. Galen frequently quotes them together, as if their editions shared the same variants; on other occasions, he quotes them separately. As a pair, the two are often accused of being shameless innovators, having modified the Hippocratic text without restraint (although, in a few cases, Galen approves their editorial choices). Especially in contrast with these two editors, Galen appears to be a (moderately) conservative philologist, respectful of the ancient lessons, and wary of modifying the transmitted text. Nonetheless, one cannot a priori exclude that Galen’s opinion may, in several cases, have been influenced by the text of the vulgata of his time—and, accordingly, that he

70

Amneris Roselli

may have been (wrongly) convinced that the recuperations of ancient lessons on the part of the two philologists were nothing but dangerous innovations on their part. An analysis of the method guiding Galen in his philological work, and the confidence he demonstrates in illustrating his textual choices to the reading public—a public, he presumed, able to understand them with the assistance of nothing more than a pedagogical repetition of known principles—reveals that he works in the context of a consolidated practice of reading ancient texts (Hippocrates in particular): his methodology is the brainchild of the most mature evolution of Alexandrian philology. Galen’s long practice in reading ancient texts (not only Hippocrates’) made him confident in the possibility of reconstructing the ancient wording, and, simultaneously, aware of the irremediableness of some corruptions intervening during the processes of copying.

Bibliography Boudon-Millot, V., Galien de Pergame: Un médecin grec à Rome (Paris: Les Belles Lettres, 2012). Boudon-Millot, V., “Un traité perdu de Galien miraculeusement retrouvé, le Sûr l’inutilité de se chagriner: texte grec et traduction française”, in V. Boudon-Millot/A. Guardasole/C. Magdelaine (ed.), La science médicale antique: nouveaux regards. Études réunies en l’honneur de J. Jouanna (Paris: Beauchesne, 2008) 72–123. Brocker, L.O., “Die Methoden Galens in der Literarische Kritik”, RhM 40 (1885) 415–38. Deichgräber, K., Die griechische Empirikerschule: Sammlung der Fragmente und Darstellung der Lehre (Berlin: Weidmannsche Buchhandlung, 1930). Dorandi, T., “‘Editori’ antichi di Platone”, Antiquorum Philosophia 4 (2010) 161–74. Ferreri, L., “Alcune riflessioni sul concetto di ‘lectio difficilior’ nel mondo antico e nella filologia moderna”, Atti e Memorie dell’Accademia Toscana della Colombaria 56 (2005) 9–61. Galen, Commentary on Plato’s Timaeus (Corpus Medicorum Graecorum, Suppl. I, 13, 3; Schröder). Galen, Galien: Ne pas se chagriner (ed. V. Boudon-Millot/J. Jouanna; CUF; Paris: Les Belles Lettres, 2010). Gourinat, J.-B., “‘Le Platon de Panétius’. À propos d’un témoignage inédit de Galien”, PhilosAnt 8 (2008) 139–51. Guardasole, A., Eraclide di Taranto: Frammenti (Testo critico, introduzione, traduzione e commentario a cura di A.G.; Napoli: D’Auria, 1997). Manetti, D., “Galeno e la ekdosis di Ippocrate: una nota testuale”, in Harmonia: Scritti di filologia classica in onore di Angelo Casanova (Firenze: Firenze University Press, 2012) 475–81. Manetti, D., “La terminologie du livre: à propos des emplois d’ὕφος et d’ἔδαφος dans deux passages de Galien”, REG 119 (2006) 157–71.

Galen’s Practice of Textual Criticism

71

Manetti, D., “Medicine and Exegesis”, in F. Montanari/S. Matthaios/A. Rengakos (ed.), Brill’s Companion to Ancient Greek Scholarship (Leiden: Brill, 2015) 1126–215. Manetti, D. & Roselli, A., “Galeno commentatore di Ippocrate”, in Aufstieg und Niedergang der Römischen Welt 37.2 (Berlin: De Gruyter, 1994) 1529–635. Martens, P.W., Origen and Scripture: The Contours of the Exegetical Life (Oxford: Oxford University Press, 2011). Montana, F., “Gli Omeri di Aristarco in fumo: Gal. Indol. 13”, Prometheus 40 (2014) 262–70. Neuschäfer, B., Origenes als Philologe (Schweizerische Beiträge zur Altertumswissenschaft 18/1–2; Basel: Friedrich Reinhard, 1987). Nicholls, M.C., “A Library at Antium?” in C.K. Rothschild/T.W. Thompson (ed.), Galen’s ‘De indolentia’: Essays on a Newly Discovered Letter (Studien und Texte zu Antike und Christentum / Studies and Texts in Antiquity and Christianity 88; Tübingen: Mohr Siebeck, 2014) 65–78. Nicholls, M.C., “Galen and Libraries in the Peri Alupias”, JRS 101 (2011) 123–42. Origenes, Die neuen Psalmenhomilien: Eine kritische Edition des Codex Monacensis Graecus 314 (ed. L. Perrone/M. Molin Pradel/E. Prinzivalli/A. Cacciari; Origenes Werke 13; Berlin: Walter de Gruyter, 2015). Origène, Philocalie, 1–20 Sur les Ècritures (ed. M. Harl; Sources Chrêtiennes 302; Paris: Éditions du Cerfs, 1983). Puglia, E., “La rovina dei libri di Anzio nel De indolentia di Galeno”, Segno e Testo 9 (2011) 53–62. Roselli, A., “Galeno e la filologia del II secolo”, in E. Bona/C. Lévy/G. Magnaldi (ed.), Vestigia notitiai: Scritti in memoria di Michelangelo Giusta (Alessandria: Edizioni dell’Orso, 2012) 63–80. Roselli, A., “Galeno e le edizioni ippocratiche di Artemidoro e Dioscoride”, in S. Fortuna/I. Garofalo/A. Lami/A. Roselli (ed.), Sulla tradizione indiretta dei testi medici greci: i commenti. Atti del IV seminario internazionale di Siena, Certosa di Pontignano, 3–4 giugno 2011 (Biblioteca di Galenos 5; Pisa Roma: Fabrizio Serra, 2012) 15–27. Roselli, A., “Galeno sull’autenticità del Prorretico”, in B. Holmes/K.-D. Fischer (ed.), The Frontiers of Ancient Science (Berlin: de Gryuter, 2015) 533–60. Roselli, A., “Libri e biblioteche a Roma al tempo di Galeno: la testimonianza del de indolentia”, Galenos 4 (2010) 127–48. Rosenthal, F., Bulletin of the History of Medicine 40 (1966) 226–45. Rothschild, C.K. & Thompson, T.W. (ed.), Galen’s ‘De indolentia’: Essays on a Newly Discovered Letter (Studien und Texte zu Antike und Christentum / Studies and Texts in Antiquity and Christianity 88; Tübingen: Mohr Siebeck, 2014). von Staden, H., “‘A Woman Does Not Become Ambidextrous’: Galen and the Culture of the Scientific Commentary”, in R.K. Gibson/Chr. Shuttleworth Kraus (ed.), The Classical Commentary: Histories, Practices, Theory (Leiden: Brill, 2002) 109–39. von Staden, H., Herophilus: The Art of Medicine in Early Alexandria (Cambridge: Cambridge University Press, 1989).

72

Amneris Roselli

von Staden, H., “Staging the Past, Staging Oneself: Galen on Hellenistic Exegetical Traditions”, in C. Gill/T. Whitmarsh/J. Wilkins (ed.), Galen and the World of Knowledge (Cambridge: Cambridge University Press, 2009) 132–56. Stramaglia, A., “Libri perduti per sempre: Galeno, De indolentia 13; 16; 17–19”, RFIC 139 (2011) 118–47.

Julio Trebolle Pre-Lucianic Readings of 3–4 Reigns in Marginal Notes of the Syrohexapla and in the Syriac Text of Jacob of Edessa The Antiochean Greek text of 3–4 Reigns has reached us through the Lucianic (L) manuscripts 19-82-93-108-127-700 Z 158 460, but Antiochean readings are also preserved in marginal notes of the Syrohexapla and in the text of 3 Reigns 1 of Jacob of Edessa (SyrJ).1 According to A. Rahlfs the Antiochean readings of these witnesses correspond to the Lucianic text of the 4th century c.e.2 Also, Richard J. Saley and Alison Salvesen seem to consider the Antiochean readings of 3 Reigns 1:1–49 in Jacob of Edessa’s text late and Lucianic.3 This paper attempts to show that those readings have a pre-Lucianic origin and go back to the earliest identifiable stage of the version of 3–4 Reigns.4 The proposal that marginal readings of the Syrohexapla and Jacob of Edessa’s readings go back to a pre-Lucianic or Old Greek (OG) layer may seem surprising and even, at first sight, erroneous, as apparently it goes against the historical development of the I thank Andrés Piquer Otero for the translation into English of the Spanish original. This paper has been produced in the framework of the public research project “Nueva Edición Políglota de Textos Bíblicos”, funded by the Spanish Ministerio de Ciencia e Innovación. 1  Mss 19-93-108-127 preserve a pure Antiochene text. Ms 82 does not have a pure Antiochene text, but attests sometimes readings of the Majority Greek text, especially in 4 Reigns. To these complete Lucianic mss, we should add ms 700 (r) which preserves the Antiochene text of 4 Reigns 1:19–4, 31; 10:3–11, 15; 12:4–17, 37; 18:9–19, 24 and the sixth century Palimpsest Z (Zuqninensis) which preserves 3 Reigns 1:11–14; 2:11–18; 2:32–3:2; 3:27–4:9; 4:28–6:16; 7:27– 8:33; 21:26–39. Besides those 7 mss, ms 460 is Lucianic in 4 Reigns. Finally ms 158 preserves many Antiochean readings in 3–4 Reigns. For Jacob of Edessa readings, see. E. Brooke/N. McLean/H.St.J. Thackeray, The Old Testament in Greek, Volume II. The Later Historical Books, Part II. I and II Kings (Cambridge: Cambridge University Press, 1930). 2  A. Rahlfs, Lucians Rezension der Königsbücher (Septuaginta-Studien III; Göttingen: Vandenhoeck & Ruprecht, 1911), 30–32. 3  R.J. Saley, The Samuel Manuscript of Jacob of Edessa: A Study in Its Underlying Textual Tradition (Leiden: Brill, 1998); A. Salvesen, The Books of Samuel in the Syriac Version of Jacob of Edessa (Leiden: Brill, 1999); id., “Jacob of Edessa and the Text of Scripture”, in L.V. Rutgers/P.W. van der Horst/H.W. Havelaar/L. Teugels (ed.), The Use of Sacred Books in the Ancient World (Leuven: Peeters, 1998), 235–46; R. Bas ter Haar (ed.), Jacob of Edessa and the Syriac Culture of His Day (Leiden: Brill, 2008). 4  A paper presented at the Annual SBL Meeting in Atlanta (November 2015) comes to the conclusion that, against Rahlfs’ opinion, “Lucianic readings inserted into the majority text attested by ms. 158 (g in Brooke-McLean) go back to the pre-Lucianic stratum of the Antiochean text.” Also, in a paper presented at the same SBL meeting Pablo Torijano analysed ms 460 which preserves an Antiochean text in 2 Kings, which has not been considered in the editions and studies by Brooke/McLean, Rahlfs and N. Fernández Marcos/J.R. Busto Saiz, El texto antioqueno de la Biblia griega II 1–2 Reyes (Madrid: CSIC, 1992).

74

Julio Trebolle

biblical text, which advances in the line of a growing incorporation of Hexaplaric readings into the LXX text, including its secondary versions. Since the Syrohexapla is the best witness for the Hexaplaric text of Kings, the possibility of its marginal readings reflecting an Old Greek text cannot but sound strange.

Textual History: Hebrew Texts—Primary and Secondary Versions— Witnesses in the Ancient Literature Studies in the history of the biblical text and the practice of textual criticism always begin with the Hebrew text and usually resort to the primary (Septuagint, Targums, Peshitta, and Vulgate) and secondary versions (Old Latin, Coptic, Ethiopic, and Syrohexapla), and more rarely to testimonies of ancient literature like Josephus and Rabbinic and Patristic sources, but only when necessary. Priority is given to the old over the new, and also to the texts in the original language or languages over those of the versions, be they primary or secondary. This also follows the notion of the history of the Bible being basically the history of an authoritative text which progressively imposes itself until almost erasing others which might have existed. Such is the case of the proto-Masoretic text in the Hebrew tradition, of the kaige and Hexaplaric texts in the Greek, and of the Vulgate in the Latin. Texts which might have existed before the authoritative or receptus text are marginalized and despoiled of authority. They also undergo a process of revision in order to adapt them to the new model of authoritative text. Thus, the Hebrew underneath LXX stopped being copied already in the last years of the Qumran period. The LXX version enjoyed in the beginning a certain status as authoritative, even inspired text, but given its differences with the proto-Masoretic text, it started losing authority and its text was gradually revised in order to adapt it to the new authoritative model. This process of revision affected especially those books which had circulated in several editions or textual forms in Hebrew, which would present more divergences when compared to the proto-Masoretic model. The books in question are those which in LXX constitute the “kaige group” (Barthélemy) or the “Theodition group” (Fernández Marcos.)5 Origen’s Hexaplaric recension culminated this process, by incorporating readings from the veritas hebraica represented by the “Three”, Theodotion, Aquila, and Symmachus. In Kings, it also rearranged the Greek text according to the Masoretic model. 5  D. Barthélemy, Les Devanciers d’Aquila (VT.S 10; Leiden, 1963), 47; N. Fernández Marcos, The Septuagint in Context: Introduction to the Greek Versions of the Bible (Leiden: Brill, 2000), 142–53; P.J. Gentry, “The Place of Theodotion-Job in the Textual History of the Septuagint”, in A. Salvesen (ed.), Origen’s Hexapla and Fragments (Tübingen: Mohr Siebeck, 1998) 199–230.

Pre-Lucianic Readings of 3–4 Reigns in Syriac

75

Regarding the secondary versions, the Old Latin (OL) of Kings, produced in the pre-Hexaplaric era, translated very literally a pre-Lucianic text, but the later textual tradition incorporated sections of text or single readings which come from a Greek kaige text and, at times, from Hexaplaric sources. Also, the Armenian and Georgian versions were produced from a pre-Lucianic text, but later revised in order to adapt their old text to the new Hexaplaric model. Thus, the Greek text of LXX and its versions constitute a mix of old texts with texts revised according to the proto-Masoretic tradition. The LXX books of 3–4 Reigns are a good example of this phenomenon. In section γγ, 3 Reigns 2:11–21:43, the B and Antiochean texts transmit an Old Greek text. On the other hand, in sections βγ and γδ, 3 Reigns 1:1–2:10 and 3 Reigns 22–4 Reigns respectively, the B text transmits the text of the kaige revision. In those sections, only the Antiochean text preserves a pre-Lucianic text quite close to the OG.6 This text is also attested by Josephus’ readings, the Old Latin, the pre-Hexaplaric layer of the Armenian version, by parallel readings from the book of Chronicles, and also by readings in the Ethiopic and Coptic versions which are based on an Old Greek text with the presence of pre-Lucianic variants.7 The Old Latin textual tradition of Kings presents many cases of double reading, one following the pre-Lucianic text and the other the kaige text.8 The value of the Antiochean text for reconstructing the OG is today generally acknowledged. As it is frequently repeated the problem consists in distinguishing the pre-Lucianic readings from those which correspond to the 4th century Antiochean text attested mainly by Theodoretus. Among many scholars prevails the idea sustained by Rahlfs and Brock, according to which the lion’s part of signature readings of the Antiochean text are to be attributed to the so-called “Lucianic recension”, produced in the 4th century. But Rahlfs’ study of 1911 could not take into account the existence of a kaige revision nor assess the value of the Qumran manuscripts when supporting the Septuagint version and, in particular, the Antiochean Greek in the books of Samuel-Kings. Also, Brock’s study focuses

6  F.M. Cross, “The History of the Biblical Text in the Light of Discoveries in the Judaean Desert”, HThR 57 (1964) 282–99; S.P. Brock, “Lucian redivivus: Some Reflections on Barthélemy’s Les Devanciers d’Aquila”, Studia Evangelica 5 (1968) 541–71; E. Tov, “Lucian and ProtoLucian: Toward a New Solution of the Problem”, RB 79 (1972) 101–13; E. Ulrich, “The Old Latin Translation of the LXX and the Hebrew Scrolls from Qumran”, in E. Tov (ed.), The Hebrew and Greek Texts of Samuel (1980 Proceedings IOSCS) (Jerusalem: Academon, 1980) 121–65. 7  H.S. Gehman, “The Old Ethiopic Version of I. Kings and Its Affinities”, JBL 50 (1931) 81–114, on p. 111; J.B. Payne, “The Sahidic Coptic Text of I Samuel”, JBL 72 (1953) 51–62, on pp. 59–60. 8  J. Trebolle, “Readings of the Old Latin (Beuron 91–95) reflecting ‘Additions’ of the Antiochean Text in III–IV Reges”, in A. Aejmelaeus/T. Kauhanen (ed.), The Legacy of Barthélemy: Fifty Years after Les Devanciers d’Aquila (Göttingen: Vandenhoeck & Ruprecht, 2017) 120–45.

76

Julio Trebolle

on 1 Samuel, whose text was not revised by the kaige group. His conclusions are not applicable in the same degree to the kaige sections of 2–4 Reigns.9 Generally speaking, scholars attribute to the late Lucianic text readings which have no explanation or parallel in the older pre-Lucianic or OG layer. The case of the adjective ἐπίποκος “covered with wool, woolly” is helpful. According to James K. Aitken,10 this derivative of the regular noun πόκος “wool, fleece”, “is only attested in the Lucianic version of 2 Kgdms 3:4. LSJ [Liddell-Scott-Jones Lexicon] (652) notes it as uncertain in the Septuagint owing to its late recension, and invites comparison with GDI [Sammlung der griechischen Dialekt-Inschriften] 3731.6, an inscription from Cos. They do not make clear whether the word is in that inscription or what the date is. It is in fact a 3rd-century b.c.e. inscription (cf. SEG [Supplementum Epigraphicum Graecum] 49:1102) and therefore older than the Septuagint, containing the phrase οιν] επιποκον τελε[αν ‘per]fect woolly [sheep’ (lines 6–7). The word is also partially reconstructed in another inscription from Cos in the expression ‘per]fect woolly [sheep’”. The adjective ἐπίποκος is indeed part of the OG in 4 Reigns 3:4 and is attested not only by the Antiochean manuscripts 82-700 46-52-236-313 44-107*-610*vid 92* 245, but by the best part of the manuscript tradition, including Codex Vaticanus, except for manuscripts d*ema(uid)orv. Rahlfs, in his edition, chooses the reading of those manuscripts, ἐπὶ πόκων, although he expresses his incertitude in the note “ἐπὶ πόκων: sic uel ἐπιπόκων ?”.11 Rahlfs seems to be guided by the presence of πόκος in Judges (6:37[2x].38[2x].39[2x].40), but there the term is translating the Hebrew ‫( נזה‬nizzah), whereas in 4 Reigns 3:4 the Hebrew word is ‫( צמר‬ṣemer). Rahlfs’ suspicion regarding the hapax ἐπίποκος reaches the point of making him choose a little-attested reading, ἐπὶ πόκων. A better understanding of 3rd century b.c.e. Greek through inscriptions or other sources of the period could shed some light on the antiquity and critical value of numerous readings in the Antiochean text. Lacking contemporary witnesses to the Greek version of Kings in the 2nd century b.c.e., it is in more recent testimonies, partial and marginal, where valuable pre-Lucianic elements of the Antiochean text may have been preserved. I have mentioned the testimonies of Flavius Josephus, the Old Latin, and the pre-Hexaplaric layer of the Armenian version, together with readings in the Coptic and Ethiopic texts. Those witnesses will assist in qualifying marginal readings in the Syrohexapla as pre-Lucianic ones, close to the OG text.

9  S.P. Brock, The Recensions of the LXX Version of I Samuel (Torino: Silvio Zamorani, 1996). 10  J.K. Aitken, No Stone Unturned: Greek Inscriptions and Septuagint Vocabulary (Winona Lake, IN: Eisenbrauns, 2014) 59. 11  A. Rahlfs, Septuaginta (Stuttgart: Württembergische Bibelanstalt Stuttgart, 19658) 697.

Pre-Lucianic Readings of 3–4 Reigns in Syriac

77

The Path of Textual Criticism: Marginal readings— Witnesses in the Ancient Literature—Secondary Versions— Greek Recensions—Old Greek—Hebrew Vorlage It is a well-known fact that the earlier layers of a text are preserved in isolated areas, removed from the centers of political, cultural and religious power, where the most updated and receptus text takes hold, in parallel with the language and styles more in agreement with the ruling tastes at the metropolis. Thus, Sardinian spoken in that island reflects an evolution of the Romance language less evolved than in other areas.12 In that island of Sardinia, and thanks to the quotations included in Lucifer of Cagliari’s work, long fragments of the Old Latin text of Kings have been preserved.13 Isolated readings of OL Kings have also been preserved in the marginal notes of Vulgate Bibles kept in the Northwestern Iberian Peninsula, a remote and isolated area where, after the arrival of Islam, Christians took refuge in, carrying with them OL texts originally from the north of Africa.14 Due to this marginal and late character, these witnesses preserve a fragmentary OL text, often corrupt and composed of a mix of diverse textual traditions. Nevertheless, those testimonies have preserved fossilized texts of very early and diverse origins. Thus, marginal readings of the codices Legionenses basically attest the text of the Old Latin version, but also transmit, generally under the siglae Al., readings which agree with the kaige text transmitted by the majority B text.15 That is why criticism should tell apart between readings of the OL textual tradition which reflect a kaige Greek text (OL2) from the properly OL readings which follow a preLucianic one and go back to the OG (OL1).16 A good example of this is the double translation of the Hebrew expression ‫ על הצבא‬in 1 Kgs 4:4: dux virtutis (OL2) = επι της δυναμεως and dux militiae (OL1) = επι της στρατιας (πατριας Β, reflected

12  L. Renzi, Introducción a la filología románica (Madrid: Gredos, 1982), 84 [Italian original: Introduzione alla filologia romanza, Bolonia: Il Mulino, 1976]. 13  G.F. Diercks, Luciferi Calaritani Opera quae supersunt. Ad fidem duorum codicum qui adhuc extant necnon adhibitis editionibus veteribus (Corpus Scriptorum Christianorum. Series Latina 8; Turnholti, 1978). 14  A. Moreno, Las glosas marginales de Vetus Latina en las Biblias Vulgatas Españolas. 1–2 Reyes (Madrid: Instituto de Filología del CSIC, 1992). 15  J. Trebolle, “The Textual History and the Text-Critical Value of the Old Latin Version in the Book of Judges”, in W. Kraus/S. Kreuzer/M. Meiser/M. Sigismund (ed.), Die Septuaginta— Text, Wirkung, Rezeption. 4. Internationale Fachtagung veranstaltet von Septuaginta Deutsch (LXX.D.), Wuppertal 19.–22. Juli 2012 (Tübingen: Mohr Siebeck, 2014) 53–72, on pp. 57–64. 16  J. Trebolle, “Textos ‘Kaige’ en la Vetus Latina de Reyes (2 Re 10,25–28)”, RB 89 (1982) 198–209; id., “From the Old Latin through the Old Greek to the Old Hebrew (2 Kgs 10,25– 28)”, Textus XI (1984) 17–36; P.-M. Bogaert, “Bulletin de la Bible latine (1955–75)”, Bulletin d’ancienne littérature chrétienne latine 5, Revue bénédictine de critique, d’histoire et de littérature religieuses 74–84 (1964–74) 162.

78

Julio Trebolle

by OL2 patrias and followed by Rahlfs’ edition).17 The version επι της δυναμεως is not attested in 3 Reigns 4:4, but agrees with one of the most acknowledged features of the kaige recension: the substitution of the original translation στρατια with the kaige version δυναμις, the same way that αρχιστρατηγος with αρχων (της) δυναμεως.18 The OL dux militiae and LXX αρχιστρατηγος in 2:46h seem to translate the Hebrew ‫ שר הצבא‬instead of MT ‫על הצבא‬. If the OL of Kings has been especially preserved in the Western limit—the finis terrae in Spain—Armenia and Georgia represent the Eastern border reached by the OG text of Kings, translated into Armenian and Georgian from an old Antiochean pre-Hexaplaric and pre-Lucianic text. These versions, Armenian and Georgian, were also revised with a Hexaplaric text coming also from Syria. As the Antiochean and later the Hexaplaric text expanded from Syria eastwards reaching Armenia and Georgia, the Greek Alexandrian text, represented by Codex Vaticanus (B) in particular, expanded southwards through the Ethiopic and Coptic versions. The Ethiopic has special value as it is a peculiarly careful translation, which generally speaking follows the B text and often is the only additional source to support that manuscript. It is free of the errors and frequent haplographies of Vaticanus, and thus it becomes “its best ‘corrector’… to the effect that agreement of Aeth with other groups as against B must mean that such a reading is correct”.19

17 A. Moreno, Las glosas marginales de Vetus Latina, 99–100. The edition by Brooke/ McLean/Thackeray has only the reading dux uirtutis. 18  J.D. Shenkel, Chronology and Recensional Development in the Greek Text of Kings (Cambridge, MA: Harvard University Press, 1968), 114 and 136. 19  J.A. Montgomery, The Books of Kings (ICC; Edinburgh: T. & T. Clark, 1951), 17.

Pre-Lucianic Readings of 3–4 Reigns in Syriac

79

These late and marginal witnesses will help us to identify valuable pre-Lucianic elements in the marginal readings of the Syrohexapla and in Jacob of Edessa in the following samples.

Readings of the Antiochean Text in Marginal Readings of the Syrohexapla Thirteen is the number of marginal readings of the Syrohexapla taken from the Greek text.20 Seven of them are marked as Lucianic with the letter ‫( ܠ‬4 Reg 9:9, 28; 10:24, 25; 11:1; 23:33; 23:35). All of the thirteen readings, with the only exception of 3 Reigns 3:25, are found in the kaige section γδ (3 Reg 22–4 Reg). This is already in itself a significant fact. The distinction between OG readings and kaige readings is key to recognizing the origin and value of the Antiochean readings collected in the marginal notes of the Syrohexapla. Their analysis will follow the order of appearance in the text. 3 Reg 3:25: Syh(mg) και το τεθνηκος ομοιως διελετε και δοτε αμφοτεραις, “And divide equally the dead (child) and give it to the two”.

This is a “plus” of the Antiochean text. Rahlfs is forced to recognize that this addition cannot be an invention of Lucian. It goes back to an earlier source as it is also found in Josephus (Ant VIII 31):21 κομισθηναι και το νεκρον και το ζων παιδιον μεταπεμπεται τινα των σωματοφηλακων και σπασαμενον εκελευσε την μαχαιραν αμφοτερα διχοτομησαι τα παιδια οπως εκατεραι λαβωσιν ανα ημισυ του τε ζωντος και του τετελευτηκοτος. According to Rahlfs the addition comes from a Jewish source, written or oral, from a time when the LXX was still in use among the Jews.22 The witness of Josephus proves that this is not a Lucianic addition but a pre-Lucianic reading dating back to the OG.

20 Rahlfs, Lucians Rezension der Königsbücher, 30. T.M. Law adds four more readings in 3 Reigns (1:8; 2:4, 16; 6:4), although signaling in a note: “These anonymous marginal readings have not been fully analysed in 3 Kgdms as it would have required us to go beyond the scope of this thesis”, T.M. Law, Origenes Orientalis: The Preservation of Origen’s Hexapla in the Syrohexapla of 3 Kingdoms (Göttingen: Vandenhoeck & Ruprecht, 2011), 25, n. 63. 21  “Lucian wird sie also nicht erst erfunden, sondern aus einer älteren Quelle geschöpft haben. Ob dies eine schriftliche Quelle war oder die mündliche Tradition der Juden, lässt sich allerdings nicht sagen”, Rahlfs, Lucians Rezension der Königsbücher, 284. 22  “An ein zufälliges Zusammentreffen kann man hier kaum denken. Auch gehört die eigentümliche Verballhornung des weisen Urteils Salomos offenbar der jüdischen Haggada an, und schon deshalb ist es wahrscheinlich, das der L-Zusatz aus einer älteren Zeit stammt, wo die Septuaginta noch von den Juden benutzt wurde”, Rahlfs, Lucians Rezension der Königsbücher, 102.

80

Julio Trebolle

3 Reg 22:10: Syh(mg) ενοπλοι εν οδω πυλης Σαμαρειας, “(They sat each on his throne,) armed, in the way of the gate of Samaria.”

This alleged “Lucianic” reading is present also in a marginal reading of the OL (91–95), (Sedebat unusquisque in throno suo) armati ad viam portae Samariae. This proves that the Antiochean εν οδω πυλης preserves the OG reading that translates the usual Hebrew expression ‫( דרך שער‬1 Sam 17:52; 2 Sam 15:2; 2 Kgs 11:19; 25:4). The B reading ενοπλοι εν ταις πυλαις Σαμαρειας belongs to the kaige text, although εν ταις πυλαις does not reproduce exactly MT ‫פתח שער‬, but seems to be representing the Hebrew ‫בשערי‬. The OG reading is recognized by the translation feature ‫ = איש‬εκαστος (L εκαθηντο εκαστος επι τον θρονον αυτου...) = unusquisque (OL), opposite to the kaige feature ‫ = איש‬ανηρ (B εκαθηντο ανηρ ...). 3 Reg 22:17: Syh(mg) ει κυριως αυτοι προς θεον αποστραφητω, “If they are legitimately with God, let come (now every one to his house in peace)”.

This is the pre-Lucianic (or OG) reading as opposed to the kaige reading ου κυριος τουτοις θεον reflecting MT ‫לא אדנים לאלה‬. The text edited by Rahlfs ου κυριος τουτοις omits εις θεον following only the witness of the manuscripts A V 246 245-318-342-707 and of the Syh(txt). Codex Vaticanus adds θεος as a loose element at the end of the clause. The majority text transmits a long reading: ου ­κυριος τουτοις εις θεον. The OL et ideo vellem, Domine, ut statueris his... seems to be a paraphrasis more than a translation. However his reflects B τουτοις (MT ‫ )לאלה‬against L προς θεον. The alternative reading Al. Si fidi vos estis in Deo (redeat unusquisque in domum suam in pace) reflects the Antiochean text ει κυριως αυτοι προς θεον. Possibly the OG text of the passage is to be considered lost. 4 Reg 3:4: Syh(mg) ην φερων φορον και επιστρεφον τω βασιλει Ισραηλ, “and he was bringing tribute and sending the king of Israel (one hundred thousand lambs...)”.

According to Rahlfs in this passage and also in 2 Kgs 17:3 the Hebrew expression for the payment of tribute is constructed with the verb ‫השיב‬, usually translated by επιστρεφειν; in both passages the Lucianic text introduced the version with φερειν, keeping in the first passage the old version in a typical Lucianic duplicate: και ην φερων φορον και επιστρεφων.23 However, the OL proves that the Antiochean text retains the OG reading φερων φορον, to which the kaige reading επιστρεφων was added. The codices Legionenses (Beuron 91–95) transmit two Latin readings. The reading preceded by Al. usually corresponds to the Greek kaige text. Such is the case here: Al. et Mosa rex Moab erat pecuarius, et praestabat regi Israel ex subiectione centum mil23 Rahlfs, Lucians Rezension der Königsbücher, 98.

Pre-Lucianic Readings of 3–4 Reigns in Syriac

81

lis = B και Μωσα βασιλευς Μωαβ ην νωκηδ και επεστρεψεν τω βασιλει Ισραηλ εν τη επαναστασει εκατον. The expression ex subiectione corresponds to εν τη επαναστασει, although it is not translating this Greek expression. This addition has no equivalent in MT. It is omitted by the pre-Lucianic text and also by the Armenian, Ethiopic and the OL. It appears with an obelus in the Syrohexapla. On the contrary, the reading et Mosa rex Moab ferebat tributum regi Israel centum millis translates a pre-Lucianic text. The typical duplicate of the later Lucianic text juxtaposes the kaige reading επιστρεφων to the OG φερων φορον. 4 Reg 6:8: Syh(mg) εις τον τοπον του φελμουνι ποιησωμεν ενεδρον, “to the place of felmouni we will ambush”.

This Antiochean reading has its counterpart in the OL (Beuron 91–95) in locum phalmunum obsessionem faciamus. This suggests that the L reading is pre-Lucianic, as opposed to the kaige reading represented by B (Rahlfs) εις τον τοπον τονδε τινα ελμωνι παρεμβαλω, “I will encamp at this certain place, Elmoni” (NETS).24 The verb ενεδρευειν corresponding to MT ‫ ארב‬appears in Judg 9:25.32.34.43; 16.2; 21:20 and in 1 Sam 15:5; 2 Sam 3:37 MT ‫בשלי‬. Likewise, the noun ενεδρον (‫ )ארב‬is frequently found in Judges. 4 Reg 7:2: Syh (mg) και εαν κυριος ποιηση, “And if the Lord does not produce (waterfalls in heaven)”.

The L reading και εαν (‫ )ואם‬is pre-Lucianic as evidenced by the Armenian and Ethiopic versions, as opposed to the B kaige variant ιδου (MT ‫)הנה‬. The primary versions, Peshitta, Targum, and Vulgate (si Dominus fecerit), also contain the reading transmitted by the Antiochean text. This is one of the rather frequent cases of agreement of Antiochean readings with one or more of the primary versions, as well as with secondary ones.25 4 Reg 9:28: Syh(mg) και ανηνεγκαν αυτον, “and (his servants) took him (to Jerusalem)”.

In this passage converge two readings that differ in the verb, αναφερειν (‫עלה‬, hiphil) or επιβιβαζειν (‫רכב‬, hiphil): και ανηνεγκαν αυτον οι παιδες αυτου L and 24 MT ‫אל מקום פלני אלמני תחנתי‬, BHS: inc; Vulg ponamus insidias cf. Syr Targ. Cf. A. Pietersma/B.G. Wright (ed.), A New English Translation of the Septuagint (Oxford: Oxford University Press, 2009). 25  J. Trebolle/P. Torijano, “Behavior of the Hebrew Medieval Manuscripts and the Vulgata, Aramaic and Syriac Versions vis-à-vis the Masoretic Text and the Greek Version in 1–2 Kings”, in E. Martín-Contreras/L. Miralles-Maciá (ed.), The Text of the Hebrew Bible: From the Rabbis to the Masoretes (Göttingen: Vandenhoeck & Ruprecht, 2014) 101–33.

82

Julio Trebolle

και επιβιβασαν αυτον οι παιδες αυτου επι το αρμα B (MT ‫)וירכבו אתו עבדיו‬. The B text merges both readings with the variant ηγαγον (‫בוא‬, hiphil): και επεβιβασαν αυτον οι παιδες αυτου επι το αρμα και ηγαγον αυτον. BHS takes επι το αρμα και ηγαγον αυτον as part of the OG and proposes to insert in MT ‫על המרכבה ויביאוהו‬. This Hebrew reading could have been part of a pre-masoretic text or of the masoretic textual family, as attested by the Vulgate super currum suum et tulerunt (although adding suum and wanting the pronoun corresponding to ‫ )–הו‬and also the Hexaplaric και ηγαγον αυτον (λ', et sursum duxerunt eum Syh, cf. BrookeMcLean).26 The OG text in 9:28 seems rather to be represented by the short L reading, reflected by the Armenian version and attested also by manuscripts A and 55. The short L reading has a parallel in a text preserved only by the preLucianic text in the plus of 2 Kgs 10:36: και ανεβιβασαν αυτον οι παιδες αυτου εν Ιερουσαλημ, reflected by the OL quem cum retulissent (mortuum) pueri eius in Hierusalem. The short L reading preserves the oldest attainable Greek form and reflects a Hebrew Vorlage, although the verb αναφερειν of L (‫עלה‬, hiphil) is less frequent than επιβιβαζειν or αναβιβαζειν (‫רכב‬, hiphil). 4 Reg 10:23: Syh(mg) και εξαποστειλατε αυτους και ειπον ουκ εισιν, “‘... and throw them out’. And they said, ‘There are not (but the servants of Baal)’”.

The Antiochean text is characterized by the plus attested by Syh(mg): και ιδετε ει εστιν μεθ’ υμων των δουλων κυριου και εξαποστειλατε αυτους και ειπον ουκ εισιν αλλ’ η οι δουλοι του Βααλ μονωτατοι (“‘…and see if there is with you someone of the servants of the Lord and throw them out’. And they said, ‘There are nothing but Baal’s servants’”). The B kaige text is characterized by the composite particle οτι αλλ’ η, which literally translates the two elements of MT ‫כי אם‬: και ιδετε ει εστιν (+ ωδε A Syh Arm * α' σ', MT ‫ )פה‬μεθ’ υμων των δουλων κυριου οτι αλλ’ η οι δουλοι του Βααλ μονωτατοι. The OG version translates the Hebrew ‫ כי אם‬simply as αλλ’ η. The Antiochean expression και εξαποστειλατε αυτους is also found in the manuscripts 247 CI f o 488 x 55 71 244 245 342 372 554 in a textual form that adds ευρισκομενους εκει, including the Hexaplaric addition εκει (MT ‫)פה‬. It also inserts the formula καθως ελαλησεν Ιου and repeats the terms των δουλων κυριου και ιδετε ει εστιν μεθ’ υμων των δουλων κυριου και εξαποστειλατε παντας τους δουλους κυριου ευρισκομενους εκει και εγενετο καθως ελαλησεν Ιου οτι ουκ ην εκει των δουλων κυριου. This text has its counterpart in the still more developed OL form: et eicite omnes seruos domini qui inuenti fuerint in tempulum Bahal et factum est sicut locutus est Ieu rex et cum nemo ­fuisset ibi de seruis domini. 26  This long reading is found in 2 Kgs 23:30, although without the words ‫—על המרכבה‬επι το αρμα: και επεβιβασαν (ανεβιβασαν L) αυτον οι παιδες αυτου... και ηγαγον αυτον (‫)ויבאהו‬ εις Ιερουσαλημ.

Pre-Lucianic Readings of 3–4 Reigns in Syriac

83

4 Reg 18:20 Syh(mg) πλην μη λογοις χειλεων και βουλη παραταξις γινεται εις πολεμον, “(What confidence is this...), if not in words of lips and so that an army leads to war in order of battle?”.

As opposed to the kaige B text πλην λογοι χειλεων βουλη και δυναμις εις πολεμον, “Mere words of lips are strategy and power for war”, the two transmitted textual forms of the OL follow the L text with some variations: fortasse uerbis labiorum cogitatio potestas autem in pugna est—Al. numquid in uerbis labiorum atque consilio uirtus est belli. The Armenian version seems also to follow the Antiochean text: quodsi dicas uerba labiorum et consilium quod et potentia opus sit (OTG).27 4 Reg 23:33 Syh(mg) και δεκα ταλαντα.

This reading is expressly attributed to the Lucianic text by the sign ‫ܠ‬. The reading δεκα is also attested by the Armenian text and by Hippolyte as well as by the Peshitta, as opposed to B εκατoν (MT ‫)מאה‬. Despite his reluctance to recognize pre-Lucianic elements in the Antiochean text, Rahlfs includes this case among those containing Vorlucianisches Gut. He does so, however, assuming that the Antiochean reading depends on the Peshitta, but the joint witness of the Armenian and the OL proves that this is a pre-Lucianic item. We pay finally special attention to the case of 4 Reigns 10:19. 4 Reg 10:19: Syh(mg) εξ αυτων.

This L reading is inseparable from the verb μη απολειφθητο εξ αυτων, “let not remain one of them”. The B text reads only the verb μη επισκεπητω (without εξ αυτων). The verb απολειπειν (or κατα-, υπολειπειν) corresponds to the Hebrew verb ‫שאר‬, while επισκεπειν is the kaige version of MT ‫פקד‬. Proof of it is the distribution of these verbs in this same chapter. The verb ‫ נשאר‬is found six times in vv. 11–21. Five times B and L have a compound of λειπειν (vv. 11 [2x]. 14.17.21). Twice in v. 21 B presents the readings απολειπεσθω and απολειφθη which also correspond to the Hebrew ‫נשאר‬, although there is not such correspondence in MT v. 21. The L text together with A Arm Aeth Syh follow MT in the omission of B λεγων… ου ζησεται. In v. 19 B επισκεπητω corresponds to MT ‫פקד‬, while L απολειφθητω represents the Hebrew ‫נשאר‬. Rahlfs signals the long insertion (einen längeren Zusatz) of LXX, which repeats what is said in v. 19, but with partly different vocabulary: v. 19, επισκεπητω, θυσια μεγαλη μοι, επισκεπη; v. 21, απολειπεσθω, θυσιαν μεγαλην ποιω, απολειφθη. According to Rahlfs the Lucianic text, as well as the Hexaplaric (i *) and the 27  J. Trebolle, “Reflejos de paralelismo hebreo en la Vetus Latina”, Sefarad 46 (1986) 463–72.

84

Julio Trebolle

Ethiopic text, have eliminated this addition transplanting the variant readings with slight modifications to v. 19: απολειπεσθω, θυσιαν μεγαλην εγω ποιω, απολειφθη, implying that the addition is prior to these changes.28 Rahlfs could not perceive the need to distinguish between OG and kaige texts. In v. 19 the B kaige text follows literally MT in the sentence order, forcing even the Greek syntax: παντες οι προφηται του Βααλ... καλεσατε προς με = MT ‫ קראו אלי‬...‫כל נביאי הבעל‬. The literal version ανηρ μη (επισκεπητω) = ‫ איש אל יפקד‬should also be noticed against the Greek μηδεις (απολειπεσθω). The OG and its Vorlage are tentatively reconstructed as follows:29 LXX Vorlage

Old Greek

L

B

καὶ νῦν καλέσατε (Aeth) πρός με πάντας τοὺς προφήτας τοῦ Βααλ καὶ τοὺς ἱερεῖς αὐτοῦ και29 πάντας τοὺς δούλους αὐτοῦ

19

19

MT καὶ νῦν

‫וְ ַע ָ ּ֣תה‬19

πάντες οἱ προφῆται τοῦ Βααλ

‫יאי‬ ֣ ֵ ‫ָכל־נְ ִב‬

πάντας τοὺς δούλους αὐτοῦ καὶ τοὺς ἱερεῖς αὐτοῦ καλέσατε πρός με ἀνὴρ μὴ ἐπισκεπήτω

‫ַה ַּ֡ב ַעל‬

‫ָּכל־ע ְֹב ָ ֣דיו‬ ֩‫וְ ָכל־ּכ ֲֹהנָ יו‬ ‫ִק ְר ֙אּו ֵא ֜ ַלי‬ ‫ִ ֣איׁש‬ ‫ַאל־יִ ָּפ ֵ ֗קד‬

28 Rahlfs, Lucians Rezension der Königsbücher, 207–8. 29  και: Pc Mss L Syr TargfMss Vulg και = (‫ו(כל‬. Cf. v. 21: και παντες … και παντες …

85

Pre-Lucianic Readings of 3–4 Reigns in Syriac

LXX Vorlage

‫ויהוא עשה‬

Old Greek

‫בעקבה‬ ‫למען האביד‬

καὶ Ιου ἐποίησεν ἐν πτερνισμῷ ἵνα ἀπολέσῃ

‫את עבדי‬ ‫הבעל‬ ‫ויאמר יהוא‬20 ‫קדשו‬ ‫עצרה‬ ‫לבעל‬

τοὺς δούλους τοῦ Βααλ 20 καὶ εἶπεν Ιου ἁγιάσατε θεραπειαν του Βααλ

‫ויקרא‬ ‫וישלח‬21 ‫יהוא‬ ‫בכל ישראל‬ ‫לאמר‬ ‫ועתה‬ ‫כל עבדי‬ ‫הבעל‬ ‫וכל‬ ‫כהניו‬ ‫וכל‬

καὶ ἐκήρυξε καὶ ἀπέστειλεν Ιου ἐν παντὶ Ισραηλ λέγων καὶ νῦν πάντες οἱ δοῦλοι τοῦ Βααλ καὶ πάντες οἱ ἱερεῖς αὐτοῦ καὶ πάντες οἱ 21

L

ἀπολειφθήτο ἐξ αῦτῶν ὅτι θυσίαν μεγάλην ἐγὼ ποιῶ τῷ Βααλ πᾶς ὃς ἐὰν ἀπολειφθῇ οὐ ζήσεται καὶ Ιου ἐποίησεν ἐν πτερνισμῷ ἵνα ἀπολέσῃ πάντας τοὺς δούλους τοῦ Βααλ 20 καὶ εἶπεν Ιου ἁγιάσατε θεραπείαν τοῦ Βααλ καὶ ἐκήρυξε30 θεραπείαν 21 καὶ ἀπέστειλεν Ιου ἐν παντὶ Ισραηλ > A L Arm Aeth Syh

30  εκηρυξε: singular, OG and Vulg.

B

MT

ἀνὴρ μὴ / μηδεὶς

ὅτι θυσία μεγάλη μοι τῷ Βααλ πᾶς ὃς ἐὰν ἐπισκεπῇ οὐ ζήσεται καὶ Ιου ἐποίησεν ἐν πτερνισμῷ ἵνα ἀπολέσῃ

‫ִּכי֩ ֙ ֶז ַבח‬ ‫ּגָ ֥דֹול‬ ‫ִ ֙לי‬ ‫ַל ַּ֔ב ַעל‬ ‫ּ֥כֹל ֲא ֶׁשר־‬ ‫יִ �ּפ ֵ ָ֖קד‬ ‫֣ל ֹא יִ ְֽח ֶי֑ה‬ ‫הּוא ָע ָ ׂ֣שה‬ ֙ ֵ‫וְ י‬ ‫ְב ָע ְק ָּ֔בה‬ ‫ְל ַ ֥מ ַען ַה ֲא ִ ֖ביד‬

τοὺς δούλους ‫ֶאת־ע ְֹב ֵ ֥די‬ τοῦ Βααλ ‫ַה ָ ּֽב ַעל׃‬ 20 καὶ εἶπεν Ιου ‫אמר יֵ ֗הּוא‬ ֶ ֹ ‫ וַ ּ֣י‬20 ἁγιάσατε ‫ַק ְּד ׁ֧שּו‬ ἱερείαν τῷ ‫ֲע ָצ ָ ֛רה‬ Βααλ ‫ַל ַ ּ֖ב ַעל‬ (+ καὶ θεραπειαν) ‫וַ ּיִ ְק ָ ֽראּו׃‬ καὶ ἐκήρυξαν 21 καὶ ‫ וַ ּיִ ְׁש ַל֤ח‬21 ἀπέστειλεν Ιου ‫הּוא‬ ֙ ֵ‫י‬ ἐν παντὶ ‫ְּב ָכל־יִ ְׂש ָר ֵ֔אל‬ Ισραηλ λέγων καὶ νῦν πάντες οἱ δοῦλοι τοῦ Βααλ καὶ πάντες οἱ ἱερεῖς αὐτοῦ καὶ πάντες οἱ

86

Julio Trebolle

LXX Vorlage

‫נביאיו‬ ‫איש‬ ‫אל יפקד‬ ‫כי זבח‬ ‫גדול עשה‬ ‫כל אשר‬ ‫נשאר‬ ‫לא יחיה‬ ‫ויבאו‬ ‫כל עבדי‬ ‫הבעל‬ ‫וכל‬ ‫כהניו‬ ‫וכל‬ ‫נביאיו‬ ‫לא נשאר‬ ‫איש‬ ‫אשר לא‬ ‫בא‬ ‫ויבאו‬ ‫בית‬ ‫הבעל‬ ‫וימלא‬ ‫הבית‬ ‫פה לפה‬

Old Greek προφῆται αὐτοῦ μηδεὶς ἀπολειπέσθω ὅτι θυσίαν μεγάλην ποιῶ ὃς ἂν ἀπολειφθῇ οὐ ζήσεται καὶ ἦλθον πάντες οἱ δοῦλοι τοῦ Βααλ καὶ πάντες οἱ ἱερεῖς αὐτοῦ καὶ πάντες οἱ προφῆται αὐτοῦ οὐ κατελείφθη ἀνήρ ὃς οὐ παρεγένετο καὶ εἰσῆλθον εἰς τὸν οἶκον τοῦ Βααλ καὶ ἐπλήσθη ὁ οἶκος στόμα εἰς στόμα

L

καὶ ἦλθον πάντες οἱ δοῦλοι τοῦ Βααλ > A L Arm Syh

οὐ κατελείφθη ἀνήρ ὃς οὐ παρεγένετο καὶ εἰσῆλθον εἰς τὸν οἶκον τοῦ Βααλ καὶ ἐπλήσθη ὁ οἶκος στόμα εἰς στόμα

B προφῆται αὐτοῦ μηδεὶς ἀπολειπέσθω ὅτι θυσίαν μεγάλην ποιῶ ὃς ἂν ἀπολειφθῇ οὐ ζήσεται καὶ ἦλθον πάντες οἱ δοῦλοι τοῦ Βααλ καὶ πάντες οἱ ἱερεῖς αὐτοῦ καὶ πάντες οἱ προφῆται αὐτοῦ οὐ κατελείφθη ἀνήρ ὃς οὐ παρεγένετο καὶ εἰσῆλθον εἰς τὸν οἶκον τοῦ Βααλ καὶ ἐπλήσθη ὁ οἶκος τοῦ Βααλ στόμα εἰς στόμα

MT

‫אּו‬ ֙ ֹ ‫וַ ּ֙יָב‬ ‫ָּכל־ע ְֹב ֵ ֣די‬ ‫ַה ַּ֔ב ַעל‬

‫וְ ֽל ֹא־נִ ְׁש ַ ֥אר‬ ‫ִ ֖איׁש‬ ‫א־בא‬ ֑ ָ ֹ ‫ׁשר ֽל‬ ֣ ֶ ‫ֲא‬ ‫אּו‬ ֙ ֹ ‫וַ ּ֙יָב‬ ‫ֵּב֣ית‬ ‫ַה ַּ֔ב ַעל‬ ‫וַ ּיִ ָּמ ֵ ֥לא ֵבית־‬ ‫ַה ַ ּ֖ב ַעל‬ ‫ֶ ּ֥פה ָל ֶ ֽפה׃‬

Pre-Lucianic Readings of 3–4 Reigns in Syriac

Old Greek

B: OG + kaige

MT

“And now all the prophets of Baal, all his worshipers and his priests summon to me; let a man not be inspected, for I have a great sacrifice for Baal; whoever is not inspected shall not live.” But Iou acted with trickery in order to destroy the worshipers of Baal. 20 And Iou said, “Sanctify a solemn festival for Baal.” And they proclaimed [it]. 21 And Iou sent in all Israel, saying, “And now, all the worshipers of Baal and all his priests and all his prophets, let not one remain, for I am making a great sacrifice; whoever remains shall not live.” And all the worshipers of Baal and all his priests and all his prophets came; not a man remained who did not come. And they entered into the house of Baal, and the house of Baal was filled mouth to mouth.

19

19

But Iou acted with trickery in order to destroy the worshipers of Baal. 20 And Iou said, “Sanctify a solemn service for Baal.” And he proclaimed [it]. 21 And Iou sent in all Israel, saying, “And now, all the worshipers of Baal and all his priests and all his prophets, let not one remain, for I am making a great sacrifice; whoever remains shall not live.” And all the worshipers of Baal and all his priests and all his prophets came; not a man remained who did not come. And they entered into the house of Baal, and the house of Baal was filled mouth to mouth. 19

“And now all the prophets of Baal, all his worshipers and all his priests call to me; let a man not be inspected, for I have a great sacrifice to Baal; whoever is not inspected shall not live.” But Jehu acted with trickery in order to destroy the worshipers of Baal. 20 And Jehu ordered, “Sanctify a solemn assembly for Baal.” So they proclaimed [it]. 21 And Jehu sent in all Israel;

and all the worshipers of Baal came,

not a man remained who did not come. And they entered into the house of Baal, and the house of Baal was filled mouth to mouth.

87

88

Julio Trebolle

Other Antiochean readings present in marginal readings of the Syrohexapla are: 4 Reg 9:5 κρυφιος; 9:9 και δωσω; 10:24 αυτου 1º; 10:25 συνετελεσαν; 11:1 οτι απεθανεν ο υιος αυτης; 15:11 και παντα οσα εποιησεν; 19:29 συνηγμενα; 23:35 κατα δυναμιν αυτου (B συντιμησιν). Some of these readings may also be pre-Lucianic.

Antiochean Readings of 3 Reigns 1 in the Text of Jacob of Edessa The revision of the Old Testament by Jacob, Bishop of Edessa, is a highly conflate work, ‘a curious eclectic or patchwork text’.31 Saley and Salvesen have studied the “distinctively Lucianic” readings found in the Samuel manuscript of Jacob of Edessa (SyrJ).32 In this manuscript 3 Reigns 1 is part of the second book of Samuel, as it happens also in the Antiochean text which places the end of 2 Reigns after 1 Kgs 2:10. The text of Jacob of Edessa for some reason does not contain Antiochean readings beyond 1:49. Saley distinguishes two types of readings: “distinctively Lucianic pluses” and “distinctively Lucianic substitutions”. I shall discuss each reading following this

17.

31  W. Wright, A Short History of Syriac Literature (London: Adam and Charles Black, 1894),

32  “The percentage of clearly Lucianic readings is by no means consistent throughout J. If we consider only the ‘L’ column in Table 2, for example, the percentages by grouping of passages are: 13 %, 6 %, 13 %, 5 %, 14 %, 30 %, 61 %. One is immediately struck by the higher ratio of these readings in the Kaige section than in the non-Kaige. Tempting as it might be to see a connection, none is forthcoming. There would appear to be no reason for this apart from either the texts available to Jacob, or his methodological caprice in using them,” R. J. Saley, “The Textual Vorlagen for Jacob of Edessa’s Revision of the Books of Samuel”, in R. Bas ther Haar (ed.), Jacob of Edessa and the Syriac Culture of his Day (Leiden: Brill, 2008) 113–26, on p. 121. “Where two different but mutually exclusive readings exist (for instance involving names or numbers), one of these often appears in a marginal note, the Peshitta reading in the text and the Greek in the margin or vice versa. There are around a hundred of such marginal notes in the manuscript, and although the hand in which they are written appears to differ from that of the main text, they may have been added shortly after the version was copied from the autograph or added later from the Peshitta or Syrohexapla. However, there are a few specifically Lucianic readings which may have been translated directly from a Greek text: for example, 2 Sam 10.18; 15.12; 21.8 have a Peshitta reading in the text and a Lucianic reading in the margin. Since the marginal notes are in accord with Jacob’s tendency of trying to include as much information and detail as possible, he may be responsible for most if not all of them,” A. Salvesen, The Books of Samuel in the Syriac Version of Jacob of Edessa, xiii. “In many places where JSam introduces material from the Greek tradition, the reading is common to the major Greek families, Egyptian, Hexaplaric, and Lucianic. Yet where his reading agrees with a particular family, it is often (though not exclusively) with the Lucianic tradition. The greatest number of these occur in the last chapter of the manuscript, 1 Kings 1”, A. Salvesen, “Jacob of Edessa’s Version of 1–2 Samuel: Its Method and Text-Critical Value”, in R. Bas ther Haar (ed.), Jacob of Edessa, 127–44, on p. 135–6; id., “Jacob of Edessa and the Text of Scripture”, 235–45.

Pre-Lucianic Readings of 3–4 Reigns in Syriac

89

classification, adding also readings that could be described as “distinctively Lucianic omissions”. Again the testimony of Josephus and of the secondary versions, particularly of the OL and Aeth, contribute to recognizing the pre-Lucianic character of the readings attested by Jacob of Edessa. Those “Lucianic additions” and “Lucianic substitutions” are neither Lucianic nor even additions or substitutions. They are mostly pre-Lucianic readings dating back to the oldest recognizable stratum of the LXX textual tradition. They may look like “additions” and “substitutions” because they contain more terms than the majority text. However, the majority text is the kaige text that “added” to the OG elements corresponding to the proto-Masoretic text. The Antiochean text frequently presents conflate readings made by the kaige reading and the preLucianic or OG reading. In textual criticism “addition” applies to the reading that adds something to the original text, not to the text of a recension. The Antiochean text purged of the kaige (also of the Hexaplaric and Lucianic) additions represents the pre-Lucianic or OG text. The same applies to the alleged “additions” of the OL and of its Greek Vorlage, the Antiochean text. These OL = L pluses are not additions to the B kaige text, but pre-Lucianic readings that represent or are very near to the OG text. On the contrary, the parallel readings of the majority text originated in the kaige revision. They were introduced into the OG textual tradition, often eliminating the OG readings. Some of these kaige readings entered into the Antiochean text, producing in this way the conflated readings typical of the late Lucianic text of the 4th century. This explains that in most cases they appear in a kaige section, as is the case of the Antiochean readings of Jacob of Edessa in 3 Reigns 1.

“Distinctively Lucianic Pluses” 1:8 και Σεμεει και Ρησει/Ρηει και υιοι δυνατοι B] και Σαμαιας και οι εταιροι αυτου οι οντες δυνατοι τω Δαυιδ L.

According to Saley, “The reading of GL and J, ‘and his companions,’ must result from a word division in the Vorlage of GL that was different from that of MT, i.e., ‫ ורעיו הגבורים‬rather than ‫ורעי והגבורים‬. The following οι οντες of GL may represent a subsequent addition to the Hebrew text, or it may simply be an interpretive translation. In either case, the aim was to smooth out the syntactic abruptness caused by the faulty word division.”33 Saley thinks that for the total passage J used only GL throughout: και Σαμαιας και οι εταιροι αυτου οι οντες δυνατοι τω Δαυειδ ουκ ησαν μετα Ορνια against B και Σεμεει και Ρησει και υιοι δυματοι του Δαυειδ ουκ ησαν οπισω Αδωνειου. But the L reading is pre-Lucianic versus the kaige variant transmitted by the majority text. 33 Saley, The Samuel Manuscript of Jacob of Edessa, 72–3.

90

Julio Trebolle

Rahlfs himself classified the L reading as vorlucianisches Gut: “L setzt hier ganz unverkennbar einen von MT abweichenden hebräische Wortlaut voraus: ‫שמעיה‬ ‫ ורעיו הגבורים‬statt ‫שמעי ורעי והגבורים‬. Lucian musste also, wenn er von LXX? ausgegangen wäre, die mit MT übereinstimmende LXX-Übersetzung nach einem von MT abweichenden hebräischen Texte korrigiert haben. Da jedoch der hebräische Text, wie uns die jüngeren griechischen Übersetzungen und Origenes lehren, schon im 2. und 3. Jahrh. n. Chr. sehr konstant war, ist eine solche hebräische Variante um 300 n. Chr. wenig wahscheinlich”.34 Again Josephus (Ant. VII 14.4) attests the pre-Lucianic nature of the Antiochean reading.35 1:9 Ζωελεθ B] Σελλαθ (Σελααθ o) L; ‫ ܠܐܬ‬SyrJ

The L reading involves also a name, Σελλαθ, which Rahlfs ranks among the L variants that cannot be explained as deriving from the majority text and transmit therefore pre-Lucianic material. Σελλαθ “setzt eine andere Aussprache von ‫זחלת‬ voraus”.36 1:11 ειπεν B] ηλθε L; ‫ ܘܐܬܐ‬SyrJ λεγων] και ειπεν L; ‫ ܘܐܡܪ‬SyrJ

Jacob reproduces the full L text: και ηλθε Ναθαν προς Βηρσαβεαι μητερα Σολομωντος και ειπεν. The two verbs are in accord only with L, “which appears to contain an ancient alternate to the other texts” (LXXB, και ειπεν... λεγων).37 The sequence και ηλθε... και ειπεν (‫ ויאמר‬...‫ )ויבא‬is common in the Hebrew narrative style (1:47).38 The spelling “Solomon” follows the pre-Lucianic form vs. “Salomon” of the majority text and also of the Peshitta and the Syrohexapla. 1:13 σου 1º B] + κατα κυριου του θεου L; ‫ ܒܡܪܝܐ ܐܠܗܟ‬SyrJ

34 Rahlfs, Lucians Rezension der Königsbücher, 185–6. 35  C.F. Burney, Notes on the Hebrew Text of the Books of Kings (Oxford, 1903), 4–5; Montgomery, The Books of Kings, 84; J. Gray, I & II Kings (2nd rev. edn; London: SCM Press, 1970), 79, n. a; “Schimi und Reï sind sonst als Parteigänger Salomos unbekannt; ist der Text verdorben?”, E. Würthwein, Die Bücher der Könige. 1. Könige 1–16 (Göttingen: Vandenhoeck & Ruprecht, 1977), 3, n. 1. 36 Rahlfs, Lucians Rezension der Königsbücher, 183. 37 Saley, The Samuel Manuscript of Jacob of Edessa, 74. 38  According to Rahlfs, in this case, as in several others “L’s Übersetzung gibt MT freier wieder, als die gewöhnliche griechische Übersetzung”, Rahlfs, Lucians Rezension der Königsbücher, 174.

Pre-Lucianic Readings of 3–4 Reigns in Syriac

91

Rahlfs may be right in considering this a Lucianic addition to the OG.39 1:20 συ κυριε μου βασιλευ οι οφθαλμοι παντος Ισραηλ προς σε Β] ει δια του κυριου μου του βασιλεως γεγονε το πραγμα τουτο οτι οι οφθαλμοι παντος του λαου προς ܵ σε L; ‫ܕܥܝܢܐ ܕܟܠܗ ܥܡܐ ܠܘܬܟ‬ ‫ ܐܢ ܡܢ ܡܪܐ ܕܝܠܝ ܡܠܟܐ ܗܘܐ ܣܘܥܪܢܐ ܗܢܐ ܡܛܠ‬SyrJ

This is the longest example encountered of “continuous Lucianic dependence”.40 According to Rahlfs this is a Lucianic addition from v. 27 replacing only ρημα by πραγμα. Rahlfs affirms that this addition displaced the previous sentence that concludes v. 19: και τον Σαλωμων τον δουλον σου ουκ εκαλεσεν. It also caused in v. 20 the loss of κυριε μου βασιλευ, since these terms are already present in the addition του κυριου μου του βασιλεως. But MT and the kaige mirroring expression in v. 20 ‫ =ואתה אדני המלך‬και συ κυριε μου βασιλευ could also be taken from v. 18. Furthermore the term λαου (‫ܥܡܐ‬, SyrJ) instead of the kaige reading Ισραηλ is surely a pre-Lucianic and OG element that can reflect a Hebrew original.41 According to M.J. Mulder the reading “in the eyes of all the people” solves “the exaggeration” found in the MT reading “the eyes of all Israel” with the omission of Judah.42 1:23 εισηλθεν B] + ναθαν L; ‫ ܢܐܬܐܢ‬SyrJ

The insertion of “Nathan” may be a late Lucianic element aiming to clarify an otherwise indefinite subject. 1:27 ουκ B] pr δια τι L; ‫ ܘܡܛܠ ܡܢܐ‬SyrJ

According to Saley, L “takes the word ‫ אם‬as the conjunction and then for semantic smoothness adds the subsequent words: ‘And if this thing came about through my lord the king, then why did you not make known…?’ To be sure this could have been triggered by the initial ‘and,’ though that could, just as well, have been secondary to this understanding of the particle.”43 1:40 εχορευον B] pr πας ο λαος L; ‫ ܘܟܠܗ ܥܡܐ‬SyrJ

39 Rahlfs, Lucians Rezension der Königsbücher, 174. 40 Saley, The Samuel Manuscript of Jacob of Edessa, 74. 41  J. Trebolle, “Textual Variants in Joshua-Kings Involving the Terms ‘People’ and ‘Israel’”, in K. De Troyer/T.M. Law/M. Liljeström (ed.), In the Footsteps of Sherlock Holmes. Studies in the Biblical Text in Honour of Anneli Aejmelaeus (Leuven: Peeters, 2014) 231–56. 42  M.J. Mulder, 1 Kings. Volume 1: 1 Kings 1–11 (Leuven: Peeters, 1998), 38. 43 Saley, The Samuel Manuscript of Jacob of Edessa, 76.

92

Julio Trebolle

This “addition” has to be placed in context. The Antiochean text conveys two juxtaposed readings: και πας ο λαος εχορευον εν χοροις και ευφραινομενοι ευφροσυνη μεγαλη ηυλουν εν αυλοις καὶ εχαιρον χαρα μεγαλη. The OL reproduces literally this doublet: et populus cantabat in canticis (et melodiis), et gaudebant gaudio magno organizantes in organis et iucundabantur in iucunditate magna. The first reading corresponds to the B kaige text και εχορευον εν χοροις και ευφραινομενοι ευφροσυνη μεγαλη, “dancing in choruses and rejoicing with great joy” (NETS). It translates a pil‘el form of the verb ‫)מחללים( חול‬, “dancing in the round”. The second reading ηυλουν εν αυλοις και εχαιρον χαρα μεγαλη is reflected by the OL organizantes in organis (Vulgate canentium tibiis) et iucundabantur in iucunditate magna. It belongs to the pre-Lucianic / OG text and corresponds to the piel form of ‫ חלל‬III, a denominative of ‫חליל‬, “flute”.44 1:41 κλητοι B] εσθιοντες και πινοντες μετ L; ‫ ܘܐܟܠܝܢ ܗܘܘ ܘܫܬܝܢ ܥܡܗ‬SyrJ

Jacob of Edessa combines “what are undoubtedly ancient alternative readings” supplying the conjunction “and” before the second reading: “the summoned ones who were with him” (‫ הקראים אשר אתו‬MT; οι κλητοι αυτου B kaige) and “the ones eating and drinking with him”.45 This second expression is found in MT 4:20 ‫ אכלים ושתים‬and in LXX 2:46a and LXXA (Hexaplaric) 4:19.24. 1:47 λεγοντες B] και εισεληλυθασι μονοι και ειπον L;‫ܘܥܠܘ ܠܘܬܗ ܠܚܘܕܝܗܘܢ ܘܐܡܪܘ‬ SyrJ.

The Antiochean addition might be a not very fortunate variant to the initial expression of the verse: ‫וגם באו עבדיו‬.46 1:49 και εξατεστησαν B 501 Aeth] + και ανεπηδησαν L; ‫ ܘܫܘܪܘ‬SyrJ(uid): ­αναπηδησαντες Jos (Ant., VII, 360).

Jacob of Edessa presents the conflate reading “and they leapt up, they arose”, following the L text which adds the conjunction “and”. Again Josephus proves the pre-Lucianic character of the Antiochean reading.

“Distinctively Lucianic Substitutions” 1:4 θαλπουσα τον βασιλεα] τω βασιλει συγκοιτος L; ‫ ܠܡܠܟܐ ܒܪܬ ܡܫܟܒܐ‬SyrJ: hab συγκοιμωμενη τω βασιλει Jos (Ant., VII.344). 44  M.J. Mulder, I Kings. Volume 1: 1 Kings 1–11 (Leuven: Peeters, 1998), 73. 45 Saley, The Samuel Manuscript of Jacob of Edessa, 81. 46 Burney, Notes on the Hebrew Text, 11–12.

Pre-Lucianic Readings of 3–4 Reigns in Syriac

93

According to Saley “The συγκοιμωμένη of Jos would seem to reflect the same tradition as GL. The compound form of the verb “to be” in J reflects the Greek imperfect of εἰμί, either directly or through SH [Syrohexapla], with the former seeming more likely. At any rate, J agrees with GL for the whole [και ην τω βασιλει συγκοιτος]”.47 Josephus attests the pre-Lucianic character of the Antiochean reading. 1:5.8.9.11.13.18.24.25.41.42.43 αδωνειας] ορνια L; ‫ ܐܘܪܢܝܐ ܐܘܪܘܢܝܐ‬SyrJ.

As stated by Saley, “The consistent use of this form by J makes this a striking example of his dependence upon the Lucianic tradition”. However, as he also says, “The spelling of the name in GL must be traced back to corruption in a Hebrew text”.48 Rahlfs classifies “Ornias” among the names of the Antiochean text that go back to Vorlucianisches Gut. Proof of this is found in 2 Samuel 24:18 ‫ ארניה‬MT, LXX Ορνα (OL). 1:6 και γε αυτος] ουτος L; ‫ ܗܢܐ‬SyrJ.

The L reading could well be the result of inner Greek corruption and its antiquity is questionable.49 L and Jacob ignore the particle γε, a feature that gives name to the kaige revision. Although ουτος may not be old, the omission of γε belongs to the OG. 1:7 εβοηθουν οπισω αδωνειου] αντελαμβανοντο αυτου L; ‫ ܘܡܤܝܥܝܢ ܗܘܘ ܠܗ‬SyrJ: secuti sunt Adonias et adiuuerunt eum Aeth.

J agrees completely with L in opposition to the other witnesses. According to Rahlfs the Antiochean “and they attended (Ornias)” is a freer translation of MT than that of the majority text: “and they were helping behind Adonias”.50 This literal version may belong to the kaige text. The Ethiopic contains a conflate reading. 1:21 εσομαι B] εσομεθα L; ‫ ܢܗܘܐ‬SyrJ.

The agreement of L with the Ethiopic version may indicate an ancient reading, but may also be an adaptation to the context. 47 Saley, The Samuel Manuscript of Jacob of Edessa, 91. 48 Saley, The Samuel Manuscript of Jacob of Edessa, 91; also according to Salvesen, “The latter reading from JSAM [ορνια] is a scribal error”, The Books of Samuel in the Syriac Version of Jacob of Edessa, 141. 49 Saley, The Samuel Manuscript of Jacob of Edessa, 91. 50 Rahlfs, Lucians Rezension der Königsbücher, 174.

94

Julio Trebolle

1:26 δουλον 2º B] υιον L; ‫ ܒܪܟ‬SyrJ.

According to Rahlfs Lucian introduced this change “weil diese Untertänigkeitsbezeichnung bei einem Sohne dem Griechen zu fremd klingen musste”.51 They are probably two alternative readings. “Solomon your servant” also occurs in 1:19, and “Solomon your son” in 1:13 and, in some witnesses, in 1:47.52

“Distinctively Lucianic Omissions” Saley does not take into account the readings in which Jacob of Edessa shares with the Antiochean text an omission in relation to the majority text. They are obviously less frequent than the additions and substitutions, but are in some sense more significant since the trend in textual transmission is to expand rather than to shorten. Again they are not Lucianic, nor are they omissions, because they go back to the pre-Lucianic or OG textual level. In these cases the kaige text adds to the pre-Lucianic or OG, rather than the Lucianic omitting from the kaige text. 1:12 δευρο] δη L SyrJ. Jacob of Edessa follows the L text in the omission of B δευρο (MT ‫)לכי‬.

The translation of ‫ לך‬or ‫ לכי‬as δευρο is a feature of the kaige text.53 The words δευρο (sing.) and δευτε (plural) are more frequent in the kaige sections of 3–4 Reigns than in any other biblical book. The OG version of ‫ לך‬/ ‫ לכי‬is πορευου. However, in v. 13 L has δευρο, which shows that crossings and contaminations between different texts frequently occur. 1:28 ενωπιον του βασιλεως] om L SyrJ.

The kaige text follows MT in the repetition ‫= ותבא לפני המלך ותעמד לפני המלך‬ και εισηλθεν ενωπιον του βασιλεως και εστη ενωπιον αυτου. Modern translations like the RSV overlook the repetition with a synonymous expression: “So she came into the king’s presence, and stood before the king”. The Antiochean text followed by Jacob of Edessa read και εισηλθεν και εστη ενωπιον του βασιλεως: “she came and stood before the king”. BHS proposes to correct the second ‫לפני‬ ‫ המלך‬by ‫ לפניו‬following LXX ενωπιον αυτου. The short reading is surely OG. Rahlfs states that the majority text reduces the repetition substituting του βασιλεως (2º) by αυτου. He affirms that Lucian reworked the text beyond what the 51 Rahlfs, Lucians Rezension der Königsbücher, 174. 52 Saley, The Samuel Manuscript of Jacob of Edessa, 93. 53  H. Ávalos, “δεῦρο—δεῦτε and the Imperatives of ‫הלך‬. New Criteria for the ‘Kaige’ Recension of Reigns”, Estudios Bíblicos 49 (1989) 165–76.

Pre-Lucianic Readings of 3–4 Reigns in Syriac

95

majority text had already done. Leaving aside his prejudice against the existence of a Lucian before Lucian, his statement remains valid: “ενωπιον του βασιλεως hat L nur einmal, an der naturgemässen Stelle”.54 1:44 ο βασιλευς B 243 509 56 55 Aetha] om L Aeths SyrJ: post αυτου AMN rell Arm Syh.

The short reading of L Aeths SyrJ may be preferable.

Conclusion In summary and conclusion, the marginal notes of the Syrohexapla as well as the text of Jacob of Edessa in 3 Reigns 1 (SyrJ) belong to the sections where the majority text conveys the kaige text. External and internal evidence proves that in most cases these readings are pre-Lucianic, older than the kaige readings of the majority text. Pre-Lucianic sources as Josephus, the OL and the Ethiopic version contain identical or similar variants. Moreover while the kaige text tends to a verbatim reproduction of the proto-MT, the Antiochean text preserves characteristics of the OG translation based on a divergent Hebrew text. Any textual variant, addition or substitution, is such in relation to the original or to the oldest recognizable text. However the study of the Greek text of Kings is generally based on the B text as edited by Rahlfs. But in the kaige sections this text, far from being OG, is a kaige text which adapted the OG to the proto-MT model. The terminology commonly used is therefore inadequate. A plus of L to the B text is generally labelled as an “addition” to the LXX original. On the contrary, the B readings parallel to the L readings are to be considered additions to a pre-Lucianic or OG text. This is the case of the typical “Lucianic” duplicates. The Antiochean textual tradition incorporated kaige additions to a previous preLucianic or OG text. Such is the case of 4 Reigns 3:4: Syh(mg) ην φερων φορον (OG) + και επιστρεφον (kaige) τω βασιλει Ισραηλ. Finally, the question arises: how to incorporate in a text-critical edition of 3–4 Reigns data of Syriac witnesses (the marginal readings of the Syrohexapla and readings of Jacob of Edessa, but also of the Peshitta) that, despite their late and complex nature, represent textual elements of the Greek pre-Lucianic Antiochean tradition. As affirmed by Rahlfs, they attest the spreading of the late Antiochean or Lucianic text also among the Syriac-speaking population. But they have also a special value as a peculiar witness of pre-Lucianic / OG readings.

54 Rahlfs, Lucians Rezension der Königsbücher, 179.

96

Julio Trebolle

Bibliography Aitken, J.K., No Stone Unturned: Greek Inscriptions and Septuagint Vocabulary (Winona Lake, IN: Eisenbrauns, 2014). Ávalos, H., “δεῦρο—δεῦτε and the Imperatives of ‫הלך‬. New Criteria for the ‘Kaige’ Recension of Reigns”, Estudios Bíblicos 49 (1989) 165–76. Barthélemy, D., Les Devanciers d’Aquila (VT.S 10; Leiden, 1963). Bas ter Haar, R. (ed.), Jacob of Edessa and the Syriac Culture of His Day (Leiden: Brill, 2008). Bogaert, P.-M., “Bulletin de la Bible latine (1955–75)”, Bulletin d’ancienne littérature chrétienne latine 5, Revue bénédictine de critique, d’histoire et de littérature religieuses 74–84 (1964–74) 162. Brock, S.P., “Lucian redivivus: Some Reflections on Barthélemy’s Les Devanciers d’Aquila”, Studia Evangelica 5 (1968) 541–71. Brock, S.P., The Recensions of the LXX Version of I Samuel (Torino: Silvio Zamorani, 1996). Brooke E., & McLean, N. & Thackeray, H.St.J., The Old Testament in Greek, Volume II. The Later Historical Books, Part II. I and II Kings (Cambridge: Cambridge University Press, 1930). Burney, C.F., Notes on the Hebrew Text of the Books of Kings (Oxford, 1903). Cross, F.M., “The History of the Biblical Text in the Light of Discoveries in the Judaean Desert”, HThR 57 (1964) 282–99. Diercks, G.F., Luciferi Calaritani Opera quae supersunt: Ad fidem duorum codicum qui adhuc extant necnon adhibitis editionibus veteribus (Corpus Scriptorum Christianorum. Series Latina 8; Turnholti, 1978). Fernández Marcos, N., The Septuagint in Context: Introduction to the Greek Versions of the Bible (Leiden: Brill, 2000). Fernández Marcos, N. & Busto Saiz, J.R., El texto antioqueno de la Biblia griega II 1–2 Reyes (Madrid: CSIC, 1992). Gehman, H.S., “The Old Ethiopic Version of I. Kings and Its Affinities”, JBL 50 (1931) 81–114. Gentry, P.J., “The Place of Theodotion-Job in the Textual History of the Septuagint”, in A. Salvesen (ed.), Origen’s Hexapla and Fragments (Tübingen: Mohr Siebeck, 1998) 199–230. Gray, J., I & II Kings (2nd rev. edn; London: SCM Press, 1970). Law, T.M., Origenes Orientalis: The Preservation of Origen’s Hexaplar in the Syrohexaplar of 3 Kingdoms (Göttingen: Vandenhoeck & Ruprecht, 2011). Montgomery, J.A., The Books of Kings (ICC; ed. Henry S. Gehman; Edinburgh: T & T Clark, 1951). Moreno, A., Las glosas marginales de Vetus Latina en las Biblias Vulgatas Españolas. 1–2 Reyes (Madrid: Instituto de Filología del CSIC, 1992). Mulder, M.J., 1 Kings. Volume 1: 1 Kings 1–11 (Leuven: Peeters, 1998). Payne, J.B., “The Sahidic Coptic Text of I Samuel”, JBL 72 (1953) 51–62. Pietersma, A. & Wright, B.G. (ed.), A New English Translation of the Septuagint (Oxford: Oxford University Press, 2009). Rahlfs, A., Lucians Rezension der Königsbücher (Septuaginta-Studien III; Göttingen: Vandenhoeck & Ruprecht, 1911).

Pre-Lucianic Readings of 3–4 Reigns in Syriac

97

Rahlfs, A., Septuaginta (Stuttgart: Württembergische Bibelanstalt Stuttgart, 19658). Renzi, L., Introducción a la filología románica (Madrid: Gredos, 1982) [Italian original: Introduzione alla filologia romanza, Bolonia: Il Mulino, 1976]. Saley, R.J., The Samuel Manuscript of Jacob of Edessa: A Study in Its Underlying Textual Tradition (Leiden: Brill, 1998). Saley, R. J., “The Textual Vorlagen for Jacob of Edessa’s Revision of the Books of Samuel”, in R. Bas ther Haar (ed.), Jacob of Edessa and the Syriac Culture of his Day (Leiden: Brill, 2008) 113–26. Salvesen, A., The Books of Samuel in the Syriac Version of Jacob of Edessa (Leiden: Brill, 1999). Salvesen, A., “Jacob of Edessa and the Text of Scripture”, in L.V. Rutgers/P.W. van der Horst/ H.W. Havelaar/L. Teugels (ed.), The Use of Sacred Books in the Ancient World (Leuven: Peeters, 1998), 235–46. Salvesen, A., “Jacob of Edessa’s Version of 1–2 Samuel: Its Method and Text-Critical Value”, in R. Bas ther Haar (ed.), Jacob of Edessa, 127–144. Shenkel, J.D., Chronology and Recensional Development in the Greek Text of Kings (Cambridge, MA: Harvard University Press, 1968). Tov, E., “Lucian and Proto-Lucian: Toward a New Solution of the Problem”, RB 79 (1972) 101–13. Trebolle, J., “From the Old Latin through the Old Greek to the Old Hebrew (2 Kgs 10,25–28)”, Textus XI (1984) 17–36. Trebolle, J., “Readings of the Old Latin (Beuron 91–95) reflecting ‘Additions’ of the Antiochean Text in III–IV Reges”, in A. Aejmelaeus/T. Kauhanen (ed.), The Legacy of Barthélemy: Fifty Years after Les Devanciers d’Aquila (Göttingen: Vandenhoeck & Ruprecht, 2017) 120–45. Trebolle, J., “Reflejos de paralelismo hebreo en la Vetus Latina”, Sefarad 46 (1986) 463–72. Trebolle, J., “Textos ‘Kaige’ en la Vetus Latina de Reyes (2 Re 10,25–28)”, RB 89 (1982) 198–209. Trebolle, J., “Textual Variants in Joshua-Kings Involving the Terms ‘People’ and ‘Israel’”, in K. De Troyer/T.M. Law/M. Liljeström (ed.), In the Footsteps of Sherlock Holmes. Studies in the Biblical Text in Honour of Anneli Aejmelaeus (Leuven: Peeters, 2014) 231–56. Trebolle, J., “The Textual History and the Text-Critical Value of the Old Latin Version in the Book of Judges”, in W. Kraus/S. Kreuzer/M. Meiser/M. Sigismund (ed.), Die Septuaginta— Text, Wirkung, Rezeption. 4. Internationale Fachtagung veranstaltet von Septuaginta Deutsch (LXX.D.), Wuppertal 19.–22. Juli 2012 (Tübingen: Mohr Siebeck, 2014) 53–72. Trebolle, J. & Torijano, P., “Behavior of the Hebrew Medieval Manuscripts and the Vulgata, Aramaic and Syriac Versions vis-à-vis the Masoretic Text and the Greek Version in 1–2 Kings”, in E. Martín-Contreras/L. Miralles-Maciá (ed.), The Text of the Hebrew Bible: From the Rabbis to the Masoretes (Göttingen: Vandenhoeck & Ruprecht, 2014) 101–33. Ulrich, E. “The Old Latin Translation of the LXX and the Hebrew Scrolls from Qumran”, in E. Tov (ed.), The Hebrew and Greek Texts of Samuel (1980 Proceedings IOSCS) (Jerusalem: Academon, 1980) 121–65. Wright, W., A Short History of Syriac Literature (London: Adam and Charles Black, 1894). Würthwein, E., Die Bücher der Könige. 1. Könige 1–16 (Göttingen: Vandenhoeck & Ruprecht, 1977).

Kristin De Troyer The Scribe of the Marginal Notes of Manuscript 344 (Ra 344; BM v)1 Manuscript 344 is a manuscript from Mount Athos, from the Pantokrator Monastery, where it is known as manuscript no 24. The siglum in Brooke-McLean is “v”. It contains the Octateuch and is dated to the 10th century.2 The manuscript contains hexaplaric notes, that is both notes from the three revisors, as well as notes from the fifth column, written by (mainly) a second hand.3 These hexaplaric notes have been collated by Brooke-McLean in the second apparatus of their edition.4 In this contribution, I will focus on the hexaplaric notes as found in the margins of the text of the Book of Joshua in manuscript 344. These at times complex hexaplaric notes give us an insight into how a scribe read a text and how he composed marginal notes. Note that I have organized the examples in such a way that the hexaplaric notes become more complex with each case. Example 1: 1:7: ανδριζου ] + οι λ̥ σφοδρα Mvz5

1  With thanks to Felix Albrecht from the Septuaginta Unternehmen, Göttingen, for providing me with the images. 2  A. Rahlfs, Verzeichnis der griechischen Handschriften des Alten Testaments, für das Septuaginta-Unternehmen aufgestellt (MSU 2; Berlin: Weidmannsche Buchhandlung, 1914), 22. 3  Ibidem. I disagree with Olivier Munnich, who writes: “... et présente dans ses marges un nombre considérable de notes provenant de la même main que le texte scriptuaire”. See O. Munnich, “Les Hexaples d’Origène à la lumière de la tradition manuscrite de la Bible grecque”, in O. Munnich/A. le Boulluec, Origiana Sexta: Origène et la Bible / Origen and the Bible. Actes du Colloquium Origenianum Sextum Chantilly, 30 août—3 septembre 1993 (BETL 118; Louvain: Peeters/Leuven University Press, 1995) 167–85, on pp. 172–3. With thanks to Peter Gentry for sending me a copy of the article, after which I remembered that I had the book in my Septuagint section of my library, marked with a note which read: “incorporate into article on ms 344.” 4  A.E. Brooke/N. McLean/H.St.J. Thackeray, The Old Testament in Greek According to the Text of Codex Vaticanus, Supplemented from Other Uncial Manuscripts, with a Critical Apparatus Containing the Variants of the Chief Ancient Authorities for the Text of the Septuagint. Volume 2 (Cambridge Library Collection; Cambridge: CUP, 2009), 675–783 (reprint of A.E. Brooke/N. McLean/H.St.J. Thackeray, The Old Testament in Greek According to the Text of Codex Vaticanus, Supplemented from Other Uncial Manuscripts, with a Critical Apparatus Containing the Variants of the Chief Ancient Authorities for the Text of the Septuagint. Part IV: Joshua, Judges and Ruth [Cambridge: CUP, 1917]). 5  The information is taken from the second apparatus of the Cambridge edition. Differences between the apparatus and the manuscript are noted in the text of this contribution.

100

Kristin De Troyer

The first example is an easy example: a marginal note in manuscript 344, as well as in Mz,6 indicates that the remaining of the Early Jewish Revisors (EJRs), οἱ λοιποί, have a plus after the imperative: namely σφόδρα. When looking at the plus, it becomes clear that the EJR have attempted to bring the text of the Old Greek closer to the Hebrew text as current in their days. The MT in 1:7 reads: ‫רק חזק‬ ‫“( ואמץ מאד‬only be strong and very courageous”7). The Old Greek has translated the imperatives (“be strong and manly”8), but not the adverb (“very”). The BHS notes in its apparatus that the Old Greek lacks this adverb.9 From the Cambridge apparatus, it can be gleaned that Origen took the additional word over from the EJR and added it to his text, which he consequently marked with an asterisk. The latter is visible in the Syrohexapla. The hexaplaric reading can be found in many of the hexaplaric witnesses. Whereas most of the scholarly discussion focuses on the presence of the word “the law” (“be strong and courageous, being careful to act in accordance with all the law that my servant Moses commanded you”) and its absence from the Old Greek (“… to observe and act as Moses my servant commanded you”),10 there is considerably less discussion regarding the presence or absence of the adverb. This “plus”, however, is also discussed by Van der Meer. He writes as follows: “Likewise the two additional particles ‫ רק‬and ‫ מאד‬in verse 7, which modify the two imperatives, stress the Janus-aspect of the beginning of verse 7: the particle ‫רק‬, only, however, introduces a restriction, the particle ‫מאד‬, very much, stresses the previous statement.”11 Van der Meer considers these particles to be typical DtrNadditions, which indicate “the shift in the presentation of Joshua from an obedient and successful military leader, with whom Yhwh apparently maintains a direct contact, towards a devout observer of the torah”.12 In other words, the adverb ‫ מאד‬already belongs to a rather old layer of the Hebrew text. Consequently, Van der Meer considers the absence of this word (and the others) as “unexpected”13 and studies its absence in the Old Greek. Van der Meer first points to the “liter-

6  Codex M is the famous Codex Coislinianus from the Bibliothèque Nationale of Paris, which contains scholia and hexaplaric notes. “z” in the Cambridge edition is manuscript nr 85 in the Göttingen system, also a hexaplaric manuscript. See A. Rahlfs, Verzeichnis der griechi­ schen Handschriften. 7  Translations into English are taken from the NRSV. 8  Translations into English are taken from NETS. 9  R. Meyer, Josua et judices (BHS 4; Stuttgart: Württembergische Bibelanstallt, 1972), 1. 10  See M. Rösel, “The Septuagint-Version of the Book of Joshua”, SJOT 16/1 (2002) 5–23, on pp. 21–2; Michael van der Meer, Formation and Reformulation: The Redaction of the Book of Joshua in the Light of the Oldest Textual Witnesses (VTSup 102; Leiden: Brill, 2004), 210–222. See example 2. 11  Ibidem, 217. 12 Ibidem. 13  Ibidem, 218.

The Scribe of the Marginal Notes of Manuscript 344 (Ra 344; BM v)

101

ary initiative on the part of the Greek translator”14 with regard to the translation of the first particle ‫רק‬, indicating that for the translator there was no separation between verse 6 and 7, “but rather stresses the logical continuity between the two segments beginning with the imperatives to remain courageous, verses 6 and 7: ἶσχυε καὶ άνδρίζου … ἴσχυε οὖν καὶ ἀνδρίζου”.15 For the second particle, ‫מאד‬, Van der Meer states that “this minus does not point to a secondary interpolation (either from a glossator [scholarly consensus], or a second editor [Tov]), but rather stems from the same harmonising interpretation by the Greek translator of the Hebrew text.”16 The problem with the view of Van der Meer is that he does not take into account the textual history, in this case the data from the EJRs. Whereas Van der Meer notes that the plus results from the “same harmonising interpretation”, the EJRs did not recognize this interpretation and felt it necessary to add the adverb back into the text. The question now is: did the scribe of the marginalia of manuscript 344 also not recognize this harmonising interpretation of the Old Greek, or did he simply note that the EJR had a different text, with the addition of the adverb? So far, the best option is to follow the latter suggestion: the scribe of the marginal notes of manuscript 344 noted that the EJR had a minor plus, which was identified as σφόδρα. Example 2: 1:7: καθοτι ] pr οι λ̥ κατα παντα τον νομον Mvz

The second example is part of a larger debate about the relation between the Hebrew and Greek text of the Book of Joshua. The Hebrew text reads: “and all the law that my servant Moses commanded you”. The OG reads: “according to all that my servant Moses commanded you”. The most important variant between the MT and the OG is the expression “all the law”. According to Michael van der Meer, the translator took verse 7 as the continuation of verse 6 and understood the commands to be Moses’ instruction to Joshua to remain courageous and fearless, rather than all the torah or a particular book of the torah. Thus, the translator felt the word torah to be problematic and inconsistent with his interpretation of the text and passed over ‫כל־התורה‬.17 The Old Greek translator shortened the text, as he did with the adverb. The EJR, however, clearly did not think that the Old Greek text was representing the Hebrew text, and thus added the reference to “all the law”. The reading was taken over in the hexaplaric tradition, with the asterisk indicating that the OG was made to align with the Hebrew text. The scribe of the marginal notes of manuscript 344, as well

14 Ibidem. 15  Ibidem, 219. 16 Ibidem. 17  Ibidem, 218–22.

102

Kristin De Troyer

as the scribes of M and z, noted the reading from the EJR and added it to the margins of manuscript 344. What is remarkable, however, is that the scribe who composed the marginal note in manuscript 344 not only jotted down the words “all the law” in his marginal note, but also the preposition “according”: κατα παντα τον νομον. The scribe of the marginal note thus not only offered the additional reading of the EJR, but also offered the alternative beginning of the phrase in the EJR! In other words, the scribe of the marginal note offered a plus and a variant reading. This full reading also proves to be the variant reading in the hexaplaric tradition. The scribe thus stands in the hexaplaric tradition, or at least seems to compare the text of manuscript 344 with a hexaplaric manuscript. Example 3: 1:5: ουκ 1°—υμων ] α' ου στηλωθησεται ανηρ εις προσωπον σου v

A marginal note in 344 indicates that the phrase οὐκ ἀντιστήσεται ἆνθρωπος κατενώπιον ὑμῶν is replaced in Aquila with οὐ στηλωθήσεται ἀνήρ εἰς πρόσωπόν σου. The verb was further used in 7:13; 23:9, where it is said that the people continued to resist, to stand against, their enemies—no notes of Aquila are registered for the use of the verb in these verses. The note indicates that the scribe of the marginal note either was able to differentiate between different Jewish Revisors or that he found this note in his hexaplaric manuscript or tradition. Example 4: 1:7: πρασσης ] θ' πορευη Mvz

A marginal note in manuscript 344, as well as in M and z, indicates that the finite verb πράσσῃς is replaced in Theodotion with πορεύῃ. The MT reads: “so that you may prosper wherever you walk”, which Theodotion captured perfectly with the use of the verb πορεύομαι, to go, to proceed. This verb also is used in Josh 1:9, in the beautiful reassurance that “the Lord is with you wherever you go.” With the introduction of πορεύομαι in 1:7, the Theodotionic revised Greek text, uses the same verb in 1:7 as in 1:9 and thus reflects the MT, where also in 1:7 and in 1:9 the verb ‫ הלך‬is used. The reading of Theodotion was, however, not taken over in the Hexapla, as the use of the Old Greek verb πράσσω may have been more appropriate in this specific verse: it is about “understanding all the things that you do”. The Old Greek has stressed the understanding of all things. In this statement, the use of ­πορεύομαι is not appropriate, since in all the cases where πορεύομαι is used in the Book of Joshua, it is to indicate movement—that is, people, rivers, or boundaries going from one point to another. The OG thus rendered the Hebrew phrase appropriately. Theodotion, however, opted for a verb reflecting better the Hebrew text and thereby imitating the double use of the verb in 1:7–9. The scribe of the marginal note to 1:7 knew the reading of Theodotion and jotted down the variant

The Scribe of the Marginal Notes of Manuscript 344 (Ra 344; BM v)

103

in the margins. Most likely he had found this note of a Theodotionic variant in his hexaplaric manuscript. In all the above-mentioned cases, the scribe of the marginal notes has taken his information from a hexaplaric manuscript or the hexaplaric tradition, offering the hexaplaric readings. The scribe has been rather precise, with indications of precisely what is found in one or more of the EJR and which one of the EJR was the source of the variant. Example 5: 2:22: ημερας ] + οι λ̥ ου  εως επεστρεψαν οι διωκοντες v

The marginal note consists of four different parts: First, there is the squiggle line referring back to the place in the text of the manuscript to indicate where the variant occurs. Then it reads οι λ̥, the remaining (of the EJR), followed by ου, and then on the next line an asterisk with the variant reading, ἕως ἐπέστρεψαν οἱ διώκοντες. With regard to the squiggle line: the marginal squiggle line is mirrored in the text with another squiggle line, which is positioned directly after ἡμέρας. Then the note indicates that the variant is in all the EJR. The variant is an actual plus, namely ἕως ἐπέστρεψαν οἱ διώκοντες “until the pursuers returned”. The latter phrase is not present in the OG, but is attested in Hebrew in the MT. The variant is clearly an attempt to align the OG with the MT and is well attested in the hexaplaric tradition. The marginal note also indicates that the plus is not marked with an asterisk: ου . There is however an asterisk in the hexaplaric tradition; more specifically, in the recovered readings of the Syrohexapla manuscript from Masius, as far as one can see.18 By remarking that there is no asterisk the note in the margin of manuscript 344 gives away interesting information about the scribe of the marginal note. It could either indicate that the scribe of the marginal notes knew that the variant was an alignment towards the Hebrew text and thus needed an asterisk or that the scribe had multiple manuscripts at his disposition: a main model manuscript which he had been following so far, and which in this case, lacked the asterisk and another manuscript with hexaplaric notes. The first option would identify our scribe as someone who is well aware of the reasons why an asterisk is added to a text and its note. The second option presumes that the scribe had multiple manuscripts at his disposition and that he was following one model manuscript, 18  “sub * uid SM”, see Brooke/McLean/Thackeray, The Old Testament in Greek, 681.

104

Kristin De Troyer

but had information from another manuscript; otherwise, how would he have known that there was supposed to be an asterisk? Example 6: 3:10: εν τουτω ] pr οι λ̥ και ειπεν ις̅ Mvz : ο' θ' εν τουτω v

In this case there are two marginal notes. On the left of the text is a note in the usual hand of the scribe of the marginal notes, as we have encountered him so far. It refers to the text in the manuscript, which unlike the OG (ἐν τούτῷ), reads ἐν τῷ νῦν and indicates that before ἐν τῷ νῦν there is in all the EJR an additional phrase that reads: καὶ εἶπεν ιησους. “And Joshua said”, which is the translation of the phrase found in the MT. The latter phrase (in the MT) is actually not necessary, as the beginning of 3:9 already reads: “And Joshua said”. The repetition of these introductory words in 3:10 may indicate that 3:9 is maybe a rather late insertion in the text as it encourages the Israelites to “Draw near and hear the words of the Lord.” What is important for this contribution is that the reading “And Joshua said” is marked with an asterisk in Masius’ readings of the Syrohexapla and that it can be found in the hexaplaric witness tradition. Also important is the position of the referring squiggle: it is put before another squiggle, which is a reference to yet another variant. It is also not put on top of the word, as the scribe normally does, but in front of it. Compare the two squiggles in 1:7, the squiggle referring to each variant reading is on top or at the end of the word:

The position of the squiggle in 3:10 may indicate that the scribe of the marginal notes (from here onwards: “MN scribe”, Marginal Notes scribe) that we have been studying so far, has offered his notes after the scribe of the text (from here onwards: “MS scribe”, ManuScript scribe) himself has jotted down his own marginal note in 3:10, drawing the attention to a variant that he had found—a variant to which we now turn. To the right of the text, there is a second marginal note. The script of the second marginal note strongly resembles the hand of the scribe of the text of manuscript 344—the note to the right of the text may indeed be the scribe of the text himself, the MS scribe.19 This second marginal note reads that 19  The difference between the hands has led me to conclude that there are at least two scribes of marginal notes: the scribe of the text himself and another scribe; precisely this difference has also made me disagree with Munnich, see note 3.

The Scribe of the Marginal Notes of Manuscript 344 (Ra 344; BM v)

105

the reading ἐν τῷ νῦν has a variant, namely ἐν τούτῷ in the “the seventy” (ο')20 and in Theodotion. Munnich correctly remarks: “… le copiste a constaté que son edition de la Septante ne correspondait pas à celle d’Origène.”21 Indeed, the copyist noted that the text of his manuscript was different from the Old Greek text. This note also indicates that the scribe of the note and, in this case the text of manuscript 344 himself, was either copying from a manuscript, which had hexaplaric notes22 or that he was following a main manuscript for his text and comparing it with another manuscript which had marginal notes. The scribe clearly did not have the intention to replace the text of his model manuscript with the text of, in this case, the Old Greek. To the contrary, he copied the text he was supposed to copy but added a variant in the margins. This observation can lead to two conclusions: either the MS scribe followed a manuscript that had marginal notes or the MS scribe had two manuscripts in front of him—one was his model text and another was one with marginal notes. The MS scribe thus seems to be in precisely the same position as was the later scribe of the marginal notes, the MN scribe! Example 7: 1:1: υπουργω ] ο' υπουργω v

The marginal note attached to 1:1 indicates that the reading ὑπουργῷ stems from the Old Greek text. The reading ὑπουργῷ is in the text and then repeated in the margin. The variant is preceded by an omicron indicating that this is a reading, which is from the Seventy/the Old Greek. Indeed, this reading is found also in Codex Vaticanus et al. It thus seems that in manuscript 344, the ο' refers to the Old Greek text. But, which text is meant precisely? More examples are needed to answer that question. Example 8: 8:29: εσπερας ] pr ο'  καιρου της v

A marginal note in manuscript 344 indicates that in the fifth column, ο’, the reading ἑσπέραs preceded by καιροῦ τῆς and that this reading is sub asterisk. What can we learn about the scribe and his notes from this note? The reading καιροῦ τῆς is also found in the margins of Codex M, with no critical signs however. In the Syro-hexapla, the reading is marked with an asterisk. Manuscript 344 20  Whereas the siglum ο’ is by almost everyone identified as the text of the “Seventy”, I think it actually could be rendered with Old Greek. Especially in the edition of the Book of Esther by R. Hanhart, the latter symbol is used to indicate the Old Greek in contrast to the socalled Lucianic text. 21  Munnich, “Les Hexaples d’Origène”, 173. 22  This is the conclusion of Munnich: “C’est la prevue que celles-si se trouvaient notes dans l’exemplaire de la recension origénienne auquel il se raportait.” See Munnich, “Les Hexaples d’Origène”, 174.

106

Kristin De Troyer

has the reading and the asterisk. The reading “the time (of)” ‫ עת‬is found in the MT. The addition “the time of ” to the Old Greek “the evening” is thus certainly a correction towards the MT. The fact that the marginal note incorporates an asterisk indicates that the MN scribe had a hexaplaric text in front of him, i.e. a text where the Old Greek has already been provided with the critical signs used in the Hexapla. In my opinion, the siglum ο' in manuscript 344 demonstrates that the MN scribe had a hexaplaric text in front of him—a text where the Old Greek is already provided with the critical signs.23 The siglum ο' thus refers clearly to the text of the fifth column of the Hexapla. From example 5, however, we have learned that this hexaplaric manuscript which the MN scribe is following was, however, not his main model, but a second manuscript which he consulted alongside his model manuscript. Example 9: 8:29: βοθρον ] + ο' προς την πυλην της πολεως v

The next marginal note to 8:29 indicates that in o’ the reading βόθρον has a plus that reads πρὸς τὴν πύλην τῆς πόλεως.24 What is not in Brooke-McLean, but clearly in the margins of the manuscript, is that this variant is noted with an asterisk.25 The text of the OG reads: “and they threw it (= his body, added KDT) down the pit.” The Hebrew text reads differently: they did not throw the body in the pit, but “they threw it down at the entrance of the gate of city.” In other words, the MT makes sure that no dead corpse is buried within the city.26 The same 23  Similarly, albeit only admitting to one manuscript, Munnich: “De tells exemples établissent, semble-t-il, que le texte Septante present dans la synopse était bien la recension critique d’Origène, avec ses complements conformes à l’hébreu et ses signes critiques.” See Munnich, “Les Hexaples d’Origène”, 177. 24  This is one of the examples where an o' reading is quoted without any reference to EJR variants. See also example 7, 8, and 10. Munnich stresses that most o' readings are accompanied by readings from the EJR and draws the conclusion from that fact that the author of the marginal notes (which is in his opinion also the author of the manuscript) consulted one manuscript, which contained not only the variant readings of the EJR, but also “un texte suivi”, a model text: “En outré, puisque dans tous les cas—ou Presque—où il cite la Septante origénienne, il rapporte le choix des autres réviseurs, cela montre que son exemplaire de reference—ou son archetype—ne comportait pas seulement, pour Aquila, Symmaque et Théodotion, des notes sporadiques, mais un text suivi,” see Munnich, “Les Hexaples d’Origène”, 174. As I have noted many more marginal notes where only o' readings are offered—at least in the book of Joshua—I think Munnich’s conclusion is a bit too strong; these single readings may need further study to buttress my idea that the scribe(s) consulted two manuscripts. 25  This is an example where BM needs to be corrected! See footnote 5. 26  According to Michael van der Meer, the Old Greek reading “into a pit” “maybe attributed to the reasoning of the Old Greek translator: since Ai was burnt to the ground (8:28), there could have been no city-gate left to bury the king, so what remained for the king’s corpse was nothing more than an anonymous pit” (see Van der Meer, Formation and Reformulation, 470). Van der Meer opposes Leah Mazor who ascribes the difference “to an exchange with ‫( פחת‬βό-

The Scribe of the Marginal Notes of Manuscript 344 (Ra 344; BM v)

107

expression is also used in 20:24 where further details are given about the place where a slayer is allowed to explain his case and negotiate entrance in a city. The latter verse is notably absent from the OG. Again, the appearance of ο' and an asterisk together only makes sense if the MN scribe had a manuscript in front of him which contained the already annotated fifth column of the Hexapla. Example 10: 1:1: μωσυη 1° ] [οι ο' χω M :] + ο' χω  δουλου κς̅ v

The MN scribe indicates that the fifth column has the plus δούλου κς̅ in the text but not the asterisk, hence χω from χώρις, without. Again, the scribe clearly seems to know that the plus is sub asterisk in the fifth column. Again, the scribe seems to consult for his notes a manuscript with hexaplaric notes but has as his main model a manuscript that is different from the manuscript with the hexaplaric notes. Example 11: 2:20: τουτω ] + οι λ̥ ω ωρκισας ημας v

There are more elements visible in the manuscript than are mentioned in BM apparatus.

First there is the squiggle, referring back to the place in the text where the ­variant belongs. The squiggle is followed by οἱ λοίποι, the remaining. Then there is an element, which may be a lemnisk.27 The next line of the marginal note, ­mentions the fifth column and the remaining of the EJR, followed by the reading ὧ ὧρκισας ἡμᾶς “that you made us swear”. With regard to the lemnisk in the first line, ­Gentry quotes Aristarchus and concludes that the lemnisk-sign “marks misplaced texts and variants”.28 The remaining of the EJR thus had a variant, either a textual variant to the one offered or a variant in a different place. Both the fifth column and the EJR have the variant reading, which is a plus: ὧ ὧρκισας ἡμᾶς “that you made θρος)” (Ibidem). In my opinion, the MT redactor used the possibility of metathesis to elaborate the location: from ‫ פחת‬to ‫ פתח‬and further to ‫אל־פתח שער העיר‬. 27  See P.J. Gentry, “The Aristarchian Signs in the Textual Tradition of LXX Ecclesiastes”,. in K. De Troyer/T.M. Law/M. Liljeström (ed.), In the Footsteps of Sherlock Holmes. Studies in the Biblical Text in Honour of Anneli Aejmelaeus (CBET 72; Louvain: Peeters, 2014) 463–78, on pp. 472ff. 28  Ibidem, 473.

108

Kristin De Troyer

us swear”. The difference between the fifth column and the MT is not the variant itself but the word to which the variant, i.e. the plus, is added. The fifth column has the plus attached to the demonstrative pronoun: (τῷ ὅρκῳ σου) τούτῳ but the MT does not have the demonstrative pronoun ‫הזה‬.29 The MN scribe has thus again been very precise. He has jotted down the fifth column reading but also that the EJR have noted this variant in a different place. The EJR have added the plus to τῷ ὅρκῳ σου τούτῳ from 2:19. Indeed, in the BM critical apparatus to 2:19 we find the following note: 2:19: τουτω ] + οι λ̥  ω ωρκισας ημας v, where the OG has the plus, “the oath that you made us swear”, which reading is absent from the MT in the parallel place, but present in 2:20.30 It needs to be said here that the MN scribe seems to have also altered his annotation system. The MN scribe is here using in 2:20 as well as in 2:19 a lambda omicron, but without the article οἱ in front of the lambda omicron. This variation, however, can be found in the rest of the marginal notes in the manuscript and is thus not exceptional. Example 12: 18:9: ιησουν Bhq E] + εν σηλω ad-gjnsv(txt)wza2

My last example is introduced to again complicate matters. Manuscript 344 has in 18:9 a reading in the text that is a plus in comparison to the OG. The plus is clearly a hexaplaric plus, but there is no note in the margins! In other words, manuscript 344 has at times a hexaplaric reading in the text without being aware of it. Alternatively, the MN scribe who made the marginal notes did not notice this reading or there were no notes in the hexaplaric manuscript which he also consulted.

Some Conclusions –  There are in manuscript 344 marginal notes from the scribe of the manu-

script himself (MS scribe), as well as from a scribe of the marginal notes (MN scribe), and rarely subsequent MN scribes. –  The MS scribe of manuscript 344 seems to have copied faithfully from a model manuscript and occasionally jotted down variants, which he had found in a hexaplaric manuscript. –  The MN scribe follows a similar pattern. He had two manuscripts: one which he consulted as his main model and one which he consulted for marginal, hexaplaric notes.

29  Interestingly, the demonstrative pronoun is also present in the Syriac version! 30  Unfortunately, I was not able to consult the page with the 2:19 reading. It would be interesting to see whether the MN scribe had also used a lemnisk there!

The Scribe of the Marginal Notes of Manuscript 344 (Ra 344; BM v)

109

–  The siglum ο' stands for the fifth column in which there was an Old Greek text as substratum that was revised and annotated by Origen. –  There is some variation as to how the MN scribe referred to the remaining of the Early Jewish Revisors, as well as to the fifth column of the Hexapla.

Bibliography Brooke, A.E. & McLean, N. & Thackeray, H.St.J., The Old Testament in Greek According to the Text of Codex Vaticanus, Supplemented from Other Uncial Manuscripts, with a Critical Apparatus Containing the Variants of the Chief Ancient Authorities for the Text of the Septuagint. Volume 2 (Cambridge Library Collection; Cambridge: CUP, 2009), (reprint of A.E. Brooke/N. McLean/H.St.J. Thackeray, The Old Testament in Greek According to the Text of Codex Vaticanus, Supplemented from Other Uncial Manuscripts, with a Critical Apparatus Containing the Variants of the Chief Ancient Authorities for the Text of the Septuagint. Part IV: Joshua, Judges and Ruth [Cambridge: CUP, 1917]). Gentry, P.J., “The Aristarchian Signs in the Textual Tradition of LXX Ecclesiastes”, in K. De Troyer/T.M. Law/M. Liljeström (ed.), In the Footsteps of Sherlock Holmes: Studies in the Biblical Text in Honour of Anneli Aejmelaeus (CBET 72; Louvain: Peeters, 2014) 463–78. Meyer, R., Josua et judices (BHS 4; Stuttgart: Württembergische Bibelanstallt, 1972). Munnich, O. “Les Hexaples d’Origène à la lumière de la tradition manuscrite de la Bible grecque”, in O. Munnich/A. le Boulluec, Origiana Sexta: Origène et la Bible / Origen and the Bible. Actes du Colloquium Origenianum Sextum Chantilly, 30 août—3 septembre 1993 (BETL 118; Louvain: Peeters/Leuven University Press, 1995) 167–85. Rahlfs, A. Verzeichnis der griechischen Handschriften des Alten Testaments, für das SeptuagintaUnternehmen aufgestellt (MSU 2; Berlin: Weidmannsche Buchhandlung, 1914). Rösel, M., “The Septuagint-Version of the Book of Joshua”, SJOT 16/1 (2002) 5–23. van der Meer, M., Formation and Reformulation. The Redaction of the Book of Joshua in the Light of the Oldest Textual Witnesses (VTSup 102; Leiden: Brill, 2004).

Part II Textual History of the Hebrew Bible

Peter J. Gentry and John D. Meade1 MasPsa and the Early History of the Hebrew Psalter Introduction The constant and ceaseless quest of the human mind to impose meaningful organisation on the input provided is known in cognitive psychology as the Gestalt effect and is a fundamental feature of human life. No observer is objective. Let us remember that the raw data are not self-interpreting. Each human brings to the data a metanarrative as the mind seeks to assemble all the data and order them into a cohesive and meaningful whole.2 Naturally there will be competing explanations to explore and we have a marvellous opportunity at the Symposium in Tbilisi to consider which larger picture is able to encompass the most data in the simplest way and offer explanatory power in other areas of knowledge that are related. This expression of epistemological foundations is never more important than in dialogue over the early history of the Hebrew Text and in particular, the focus of our research—an early text known as MasPsa and the role it may or may not play in the history of the Hebrew Psalter.

MasPsa The text known as Masada Psalms fragment “a” is part of a scroll of the Book of Psalms 25.5 cm in height. It was discovered November 20, 1963 in a room designated Casemate 1039. There is evidence that the room was inhabited by one of the families of the Zealots, but “the circumstances of the discovery and the objects found there gave the impression that articles from various rooms were thrown into disorder into this one, and heaped up there.”3 MasPsa consists of two

1  We would like to express our gratitude and thanks to Paul Sanders and especially Raymond de Hoop for constructive and detailed criticism of an earlier draft of this work. Nonetheless, the opinions expressed as well as the errors must remain ours. 2  Langer calls this “premature cognitive commitments”. See E.J. Langer, Mindfulness (Boston: Da Capo, 1989, 2014). 3 Shemaryahu Talmon with Carol Newsom and Yigael Yadin, Masada VI: The Yigael Yadin Excavations 1963–1965 Final Reports (Jerusalem: Israel Exploration Society, 1999), 76. Nahman Avigad believes the Romans threw the spoils of coins and documents on the heap of stones in this room after the Zealots removed the wood they needed from the roof and the ballista stones fell in. See E. Netzer, “The Last Days and Hours at Masada by Nahman Avigad”, in

114

Peter J. Gentry and John D. Meade

fragments that can be conjoined on the basis of matching up the contours of the edges and also of the text in its three columns. Yadin designated the script or writing “late Herodian formal style” and proposed a date of the first half of the first century c.e., 20–30 years before t he fal l of Masada.4 Close attention to palaeography in the editio princeps by Talmon led him to classify the script simply as “formal Herodian” and date the fragment to the end of the last century b.c.e. (30–1 b.c.e.). In t he 2007 Festschrift for Florentino García Martínez, Armin Lange concurred with a date of 30–1 b.c.e.5 Emanuel Tov in his Third Edition of Textual Criticism of the Hebrew Bible also altered his dates for the Masada Psalms fragments to 50–30 b.c.e.,6 but as of a 2013 publication, Peter Flint nowhere discusses the proposed re-dating by Talmon.7 Eibert Tigchelaar challenged the dating of MasEzek by Talmon as “formal Herodian” and evaluated the script as “late Herodian formal style”.8 Nonetheless, Talmon compares the script of MasEzek to MasLeva,b and MasDeut and does not compare it to the script of MasPsa.9 Tigchelaar also acknowledges that paleographical experts vary widely in dating some manuscripts and calls for a re-evaluation of the later Herodian and post-Herodian hands.10 An earlier date would have some significance for Flint’s approach to the Qumran Psalters (discussed infra). The consonantal text is almost identical to that of the later MT, but the manuscript has a special layout and in Talmon’s view is almost certain to have been brought to Masada from Jerusalem by the Zealots.11 E. Tov argues that the nonbiblical texts discovered at Masada were brought there by fugitives from Qumran.12 This is based largely on similarities between non-biblical texts found at both sites. While the biblical corpora at Masada and Qumran are similar in faR. Ngo (ed.), Masada: The Dead Sea Desert’s Fortress (Washington, DC: Biblical Archaeology Society, 2014), 11. 4  Y. Yadin, “The Excavations of Masada 1963/64. Preliminary Report”, Israel Exploration Journal 15 (1965) 103–4. 5  A. Lange, “ ‘Nobody Dared to Add to Them, to Take from Them, or to Make Changes’ (Josephus, Ag. Ap. 1.42): The Textual Standarization of Jewish Scriptures in Light of the Dead Sea Scrolls”, in A. Hilhorst/É. Puech/E. Tigchelaar (ed.), Flores Florentino: Dead Sea Scrolls and Other Early Jewish Studies in Honour of Florentino García Martínez (Supplements for the Journal for the Study of Judaism 122; Leiden: Brill, 2007), 110, n. 14. 6  E. Tov, Textual Criticism of the Hebrew Bible (3rd revised and expanded edn; Minneapolis: Fortress, 2012), 29. 7  P.W. Flint, “The Dead Sea Psalms Scrolls: Psalms Manuscripts, Editions, and the Oxford Hebrew Bible”, in S. Gillingham (ed.), Jewish and Christian Approaches: Conflict and Convergence (Oxford: Oxford University Press, 2013) 11–34. 8  E. Tigchelaar, “Notes on the Ezekiel Scroll from Masada (MasEzek)”, Revue de Qumran 22/86 (2005) 273–5. 9 Talmon, Masada VI, 60. 10  Tigchelaar, “Notes on the Ezekiel Scroll from Masada (MasEzek)”, 274. 11 Talmon, Masada VI, 24. 12  E. Tov, “A Qumran Origin for the Masada Non-Biblical Texts?” DSD (2000) 57–73.

MasPsa and the Early History of the Hebrew Psalter

115

vouring Torah, Prophets, and Psalms, this argument is not significant for determining the source of the biblical texts. What has not been adequately explored are the differences between MasPsa and biblical texts found at Qumran. Tov argues that though the biblical texts from Masada may have come from Qumran, the character of the manuscripts and the text make it probable that they were from Jerusalem as the home of the later mediaeval MT.13 Possibly MasPsa came from the Jerusalem Temple, while the non-biblical texts were brought to Masada by fugitives from other sites. We cannot be sure based on evidence available at the present time. We shall analyse in turn (1) the character of the text, (2) the layout of the manuscript, and (3) its putative provenance. Few variants obtain between MasPsa and the MT as represented by the Aleppo Codex and the Leningrad Codex. Aside from a couple of instances of spelling words ‫ ָמ ֵלא‬instead of ‫ ָח ֵסר‬and also a couple occurrences of the reverse situation, six variants are listed by Talmon in increasing measure of significance. The reading of MT is presented left of the bracket; the reading of MasPsa is after it to the right: 83:14a 81:9b 83:8a 83:7a 83:10a/b 83:12a

‫ֹלהי‬ ַ ‫אלהים ] ֱא‬ ‫ע־לי‬ ִ ‫ם־ת ְש ַמ‬ ִ ‫אם תשמע קולי ] ִא‬ ‫גבל עמון ועמלק ]ּגְ ָבל וְ ַעֹּמון וַ ֲע ָמ ֵלק‬ ‫אלהי אדום ] ָא ֳה ֵלי ֱאֹדום‬ ‫יס ָרא‬ ְ ‫ ְּכ ִס‬init 10b] trahit ad fin 10a ‫ שיתמו‬sup lin MasPsa

The reading in 81:9b is probably a harmonisation with v. 12. The variant in 83:8a entails a waw in a list. The difference between “gods” and “tents” in 83:7a is doubtless due to metathesis. The problem in 83:10 is a difference in stichometry, to be discussed later. The case of 83:12a is a prima manus correction for a parablepsis based on homoioteleuton. None of these are significant variants. Emanuel Tov has rightly classified the manuscript as an example of the proto-MT (Tov, 2008), which belongs to what he terms “the first circle” of texts, i.e. master scrolls preserved in the setting of the Temple. We can note that MasPsa has 29–30 lines per column, which aligns with deluxe scroll standards, while 11QPsa has only 16–17 lines and 4QPsb also has only 16–17 lines. The number of lines per column sets MasPsa apart from the Qumran manuscripts and indicates that it was a master scroll or copied from one and corrected. These are the most complete examples from Qumran Psalms, exhibiting both top and bottom margins. Analysis and attention will be focused on the layout of the text.

13  Tov, “A Qumran Origin”, 61, 71.

116

Peter J. Gentry and John D. Meade

Colometry in MasPsa and Pausal Forms/ Accents in the Masoretic Text One full column of text is extant as well as two half columns of text to the left and right side of the fully preserved column. The description by Talmon is as follows: Lines are invariably divided into hemistichs, forming two half-columns separated by a margin of between 1–2.3 cm, depending on the amount of text which the scribe accommodated in the first half-line. Whereas the beginnings of lines, viz. of the first stichs, are justified, the beginnings of the second half-lines are irregular. A hemistich contains between 2–5 words, contingent on the length of words and the sense unit: e.g. ‫( בחשכה יתהלכו‬col. II, l. 8b); ‫( גבל ועמון ועמלק‬l. 20b); ‫פלׁשת עם יׁשבי צור‬ (l. 21a); ‫( היו זרוע לבני לוט סלה‬l. 22a). The shortest fully preserved hemistich, ‫( מואב והגרים‬col. II, 20a), contains ten letters and one inter-word space, viz. 11 spaces altogether; the longest, ‫( וכזבח וכצלמנע [כ]ל [נ]סיכמו‬col. II, l. 25a) 20 letters and three spaces, viz. 23 spaces in all. The shortest line, ‫כקׁש לפני רוח כאׁש תבער יער‬ (col. II, l. 27), has 20 letters and four inter-word spaces, i.e. a total of 24 spaces; the longest, ‫( וכזבח וכצלמנע [כ]ל [נ]סיכמו אׁשר אמרו נירׁשה לנו‬col. II, l.25), 35 letters and seven inter-word spaces, viz. 42 spaces in all. (83)14

Talmon immediately follows this description by the statement that “in the masoretic [sic] tradition this system of half-lines, termed ‫אריח על גבי אריח‬, half-brick on top of half-brick, is essentially based on verses composed of sense units based upon parallelismus membrorum.”15 His analysis is premature. While a number of aspects of the manuscript (e.g. gewil material prepared for writing) correspond to the later rabbinic regulations for the proper preparation of biblical texts, there is also development to be noted between MasPsa and Masoretic manuscripts.16 For the sake of clarity in analysis and description, let us define the terminology used in describing lineation both in the manuscript and in poetry. Hebrew poetry is generally based on couplets—occasionally triplets—containing two (or three, even four) cola. ‘Hemistich’ is the term used by Talmon for colon in his description of colometry or stichometry. Each column of MasPsa normally contains two cola of text that are divided by a space. The two cola of text correspond generally to the standard couplet of poems in Hebrew. For Psalms 81, 82, 84, and 85, the superscription occupies the right hand colon while the left hand colon is vacat. 14  As a minor point, we may note that the shortest fully preserved hemistich is in fact col. III, l. 22a (= 84:13a) containing ‫יהוה צבאות‬, i.e. nine letters and one inter-word space. 15 Talmon, Masada VI, 83. 16  Cf. b. Meg. 16b and Soferim 12:10. Tov notes that the exact form in stichometry required by the Talmud is not found at Qumran. See E. Tov, Scribal Practices and Approaches Reflected in the Texts Found in the Judean Desert (Leiden: Brill, 2004), 174.

MasPsa and the Early History of the Hebrew Psalter

117

In every case the superscription is either longer or shorter than an average colon. For 83, 84, and 85, an entire empty line is also left between the end of the previous psalm and the beginning of the next. In 83, the text of the psalm begins in the left hand colon while the superscription is in the right hand of the same line. Here the superscription is the length of an average colon. Psalms 81–85 contain a number of triplets (e.g. 81:6, 8, 11). Each member of the triplet that makes up the verse is treated as a colon in the manuscript. This better coincides with the definition of a line of Hebrew poetry based on syntax by M. O’Connor in Hebrew Verse Structure than the treatment in our later Masoretic manuscripts.17 Comparison between the layout of MasPsa with the layout of the text, the pausal forms and the accents in the Aleppo Codex as representative of the Masoretic Text (MT-A) is instructive. First, in only one instance in MasPsa the scribe placed an interval or space in the middle of a half-line or hemistich (in the terminology of Talmon): col. II l. 24 = 83:11b. In this hemistich, ‫ לאדמה‬marks the end of a colon and ‫ שיתמו‬begins the next colon according to the accents and layout in MT-A. After a close look at the photograph, we would disagree with the reconstruction of Talmon. There is a normal inter-word space between ‫ שיתמו‬and ‫נדיבימו‬. Therefore ‫ שיתמו‬belongs with 83:12a and not with 11b. Appropriate space between hemistichs exists between ‫ לאדמה‬and ‫שיתמו‬. Second, in terms of pausal forms, comparison reveals that the end of cola in MasPsa correspond to 37 pausal forms (i.e. all of the pausal forms in MT for the delimited stretch of text), 138 neutral forms, and one contextual form (designated in the first column of the table as P, N, and C respectively). Third, comparison in terms of accents in MT with intervals or spaces in Mas­ Psa requires discussion. Paul Sanders has shown in his analysis of colometry in the Aleppo Codex that in the three Books of Poetry, only the following disjunctive accents mark the end of a colon (numbered according to disjunctive value): silluq [1], ʕole weyored [2], ’atnah� [3], reviaʕ gadol [4]*, s�innor [7]*

According to Sanders, the accents indicated by an asterisk only mark the end of a colon if they are preceded by a weaker disjunctive accent.18 In MasPsa the end of a half-line always corresponds to a disjunctive accent in MT (63× = silluq, 5× = ‘ole weyored, 54× = ’atnah�, 9× = rebia‘, 2× = s�innor, 3× = deh�i, and 1× = pazer). All cases of silluq [1], ‘ole weyored [2], ’atnah� [3] mark a 17  M.P. O’Connor, Hebrew Verse Structure (2nd edn; Winona Lake, IN: Eisenbrauns, 1980, 1997). See the discussion below of the layout in the Aleppo Codex. 18  P. Sanders, “The Colometric Layout of Psalms 1 to 14 in the Aleppo Codex”, in M. Korpel/J. Oesch (ed.) Studies in Scriptural Unit Division (Pericope 3; Assen: Van Gorcum, 2002) 226–57.

118

Peter J. Gentry and John D. Meade

colon end. Both cases of s�innor are preceded by a weaker disjunctive accent and so match MT in marking a colon end. The instances of rebia‘ and deh�i call for comment. In Ps 81:6a, 11a, and 82:5a the hemistich is ended by a rebia‘ gadol. Thus accentuation in MT matches colometry in MasPsa. In Ps 83:7a, 18a; 84:2a, the hemistich is ended by means of a so-called defective rebia‘ mugraš, because the gereš is lacking. This is important since in general, when we find such a defective rebia‘ mugraš, the ’atnah� is missing and the defective rebi‘a mugraš has the function of ’atnah�. This implies that in these cases accentuation in MT matches the colometry in MasPsa. In Ps 84:3c, the rebi‘a is preceded by an ’atnah� and is a plene rebi‘a mugraš, which as a rule does not mark the end of a colon. In Ps 84:11a we have rebi‘a qat�an preceding ‘ole weyored. Normally rebi‘a qat�an does not mark a colon end, but rather ‘ole weyored in this situation. If the restored text and its stichometry is correct, it is not in line with accentuation in MT. In v. 11b we are dealing with a rebi‘a gadol which has in this particular position (no preceding disjunctive, like mehuppak and ‘azla legarmeh; cf. Ps 81:6a, 11a; 82:5a) the function of a precursor for the combination deh�i—’atnah�. The stichometry suggested by the restored text would be somewhat odd. A probable reconstruction, not completely in line with the accentuation, could be: ‫ּתֹופף‬ ֵ ‫ָב ַ֗ח ְר ִתי ִ ֭ה ְס‬ ‫י־ר ַשע׃‬ ֽ ֶ ‫ִ֝מ ּ֗דּור ְב ָא ֳה ֵל‬

‫ִ ֤כי ֽטֹוב־י֥ ֹום ַב ֲח ֵצ ֶ ֗ריָך ֵ֫מ ָ ֥א ֶלף‬ ‫ֹלהי‬ ֑ ַ ‫ְב ֵב֣ית ֱא‬

For a short stich or colon, see e.g. Ps 84:2b, 13a. In Ps 84:4a a colon end has pazer and in 84:4d, 7a, and 9a a colon end corresponds to deh�i in MT. In all cases the poetry entails a triplet. This explains why colon ends are marked by the disjunctive accents s�innor [7], deh�i [9], and pazer [10]. We can consider colometry in MasPsa equal to accentuation in MT-A. A similar situation to Ps 84:3c and 11b is Ps 84:4e where ’atnah� is found with ‫צ ָב ֑אֹות‬.ְ This contradiction to MT-A also entails a triplet. Finally we must mention 83:18a where the word ‫אבדּו‬ ֽ ֵ ֹ ‫ וְ י‬with silluq is penultimate and the final word in the colon of MasPsa is ‫ ְ �וֽיֵ ְד ֗עּו‬with rebi‘a gadol. This based entirely upon reconstruction of MasPsa. If the reconstruction of Talmon is correct, the colometry is contrary to MT-A, but preferred by the editors of BHS. In sum, only five cola (Ps 83:18a; 84:3c, 11a, 11b; 84:4e) of 138 in MasPsa (in case reconstruction is correct), i.e. 3.6%, contradict the tradition in MT. If Ps 83:18a is discounted, the percentage of difference is 2.9%. The physical layout of the text of the Masada Psalms scroll, then, uses spacing as terminal markers19 to 19  We are using the term “terminal markers” according to the definition and usage of E.J. Revell, The Pausal System: Divisions in the Hebrew Biblical Text as Marked by Voweling and Stress Position (R. de Hoop/P. Sanders (ed.); Sheffield: Phoenix Press, 2015).

‫‪119‬‬

‫‪MasPsa and the Early History of the Hebrew Psalter‬‬

‫‪divide the text into cola (grammatical and sense units) identical to the MT tradi‬‬‫‪tion over 1000 years later. This, in itself, is astonishing. At the same time, clearly‬‬ ‫‪the tradents of MasPsa construed the text in a slightly different way, and there is‬‬ ‫‪change from the time of MasPsa to that of MT.‬‬

‫‪H = Hemistich‬‬ ‫‪Accents at‬‬ ‫‪H End‬‬

‫‪MasPsa‬‬

‫‪MT-A‬‬

‫‪81‬‬

‫‪Column I‬‬ ‫‪H1: silluq‬‬ ‫‪H2: vacat‬‬ ‫�‪H1: ’atnah‬‬ ‫‪H2: silluq‬‬ ‫�‪H1: ’atnah‬‬ ‫‪H2: silluq‬‬ ‫�‪H1: ’atnah‬‬ ‫‪H2: silluq‬‬ ‫�‪H1: ’atnah‬‬ ‫‪H2: silluq‬‬ ‫‪H1: rebi‘a‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫‪H2: ‘ole‬‬ ‫‪weyored‬‬ ‫�‪H1: ’atnah‬‬ ‫‪H2: silluq‬‬ ‫�‪H1: ’atnah‬‬ ‫‪H2: silluq‬‬ ‫�‪H1: ’atnah‬‬ ‫‪H2: silluq‬‬ ‫‪H1: rebi‘a‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬

‫[למנצח על הגתית לאסף]‬

‫ל־הגִ ִ֬תית ְל ָא ָ ֽסף׃‬ ‫ַל ְמנַ ֵ֬צ ַח׀ ַ ֽע ַ‬

‫‪1a N‬‬

‫[הרנינו לאלהים עוזנו]‬ ‫[הריעו לאלהי י]עקב‬ ‫[שאו־זמרה ותנו תף]‬ ‫[כנור נעים] עם [נ]בל‬ ‫[תקעו בחדש שופר]‬ ‫[בכסה ליום] חגנו‬ ‫[כי חק לישראל הוא]‬ ‫[משפט לא]להי יעקב‬ ‫[עדות ביהוסף שמו]‬ ‫[בצאתו על א]רץ מצרים‬ ‫[שפת לא ידעתי אשמע]‬ ‫[הסירותי] מסבל שכמו‬ ‫[כפיו מדוד תעברנה]‬ ‫[בצ]רה ק[ר]את וא[ח]לצכה‬

‫עּוז֑נּו‬ ‫אֹלהים ֵ‬ ‫ַ ֭ה ְרנִ ינּו ֵל ִ ֣‬ ‫אֹלהי יַ ֲע ֽקֹב׃‬ ‫ָ֝ה ִ ֗ריעּו ֵל ֵ ֥‬ ‫תף‬ ‫ּותנּו־ ֑ ֹ‬ ‫אּו־ז ְ֭מ ָרה ְ‬ ‫ְ ֽש ִ‬ ‫ם־נ ֶֽבל׃‬ ‫ִכּנ֖ ֹור נָ ִ ֣עים ִע ָ‬ ‫ׁשֹופר‬ ‫ִת ְק ֣עּו ַב ֣חֹ ֶדׁש ָ ֑‬ ‫ַ֝ב ֵ֗כ ֶסה ְלי֣ ֹום ַח ֵגֽנּו׃‬ ‫ִ ֤כי ֣חֹק ְליִ ְש ָר ֵ ֣אל ֑הּוא‬ ‫אֹלהי יַ ֲע ֽקֹב׃‬ ‫ִ֝מ ְש ֗ ָפט ֵל ֵ ֥‬ ‫ֹוסף ָש ֗מֹו‬ ‫יה ֵ ֤‬ ‫ֵע֤דּות׀ ִ ֽב ֘‬ ‫ל־א ֶרץ ִמ ְצ ָ ֑ריִם‬ ‫ְ ֭ב ֵצאתֹו ַע ֶ ֣‬ ‫ְש ַ ֖פת לֹא־יָ ַ ֣ד ְע ִתי ֶא ְש ָ ֽמע׃‬ ‫ֹותי ִמ ֵ ֣ס ֶבל ִש ְכ ֑מֹו‬ ‫ֲה ִס ֣יר ִ‬ ‫ַ֝כ ֗ ָפיו ִמ ּ֥דּוד ַת ֲע ֽבֹ ְרנָ ה׃‬ ‫את וָ ֲא ַ֫ח ְל ֶ ֥צךָ‬ ‫ַב ָצ ָ ֥רה ָק ָ ֗ר ָ‬

‫‪2a N‬‬ ‫‪2b N‬‬ ‫‪3a N‬‬ ‫‪3b P‬‬ ‫‪4a N‬‬ ‫‪4b N‬‬ ‫‪5a N‬‬ ‫‪5b N‬‬ ‫‪6a N‬‬ ‫‪6b P‬‬ ‫‪6c P‬‬ ‫‪7a N‬‬ ‫‪7b N‬‬ ‫‪8a N‬‬

‫[אענך בסתר רעם]‬ ‫[אבח]נך על מי מריבה סלה‬ ‫[שמע עמי ואעידה בך]‬ ‫[י]שראל אם תשמע קולי‬ ‫[לא יהיה בך אל זר]‬ ‫[ולא] תשתחוה לאל נכר‬ ‫[אנכי יהוה אלהיך]‬ ‫[המע]לך מארץ מצרים‬ ‫[הרחב פיך ואמלאהו]‬ ‫[ולא] שמע עמי לקולי‬ ‫[וישראל לא אבה לי]‬ ‫[ואש]לחהו בשרירות לבם‬

‫ֶ֭א ֶענְ ָך ְב ֵ ֣ס ֶתר ַ ֑ר ַעם‬ ‫יבה ֶ ֽס ָלה׃‬ ‫ל־מי ְמ ִר ָ ֣‬ ‫ֶא ְב ָ ֽחנְ ָ֨ך ַע ֵ ֖‬ ‫ְש ַ ֣מע ַ ֭ע ִמי וְ ָא ִ ֣ע ָידה ָ ֑בְך‬ ‫ע־לי׃‬ ‫ם־ת ְש ַ ֽמ ִ ֽ‬ ‫ִי ְ֝ש ָר ֵ֗אל ִא ִ ֽ‬ ‫ֽל ֹא־יִ ְה ֶי ֣ה ְ ֭בָך ֵ ֣אל ָז֑ר‬ ‫וְ ֥ל ֹא ִ֝ת ְש ַת ֲח ֶ ֗וה ְל ֵ ֣אל נֵ ָ ֽכר׃‬ ‫ֹלהיָך‬ ‫ָאנ ִֹ֨כי׀ יְ ֘הָו֤ה ֱא ֶ֗‬ ‫ַ ֽ֭ה ַמ ַע ְלָך ֵמ ֶ ֣א ֶרץ ִמ ְצ ָ ֑ריִם‬ ‫ב־פיָך וַ ֲא ַמ ְל ֵ ֽאהּו׃‬ ‫ַה ְר ֶח ֗ ִ֝‬ ‫קֹולי‬ ‫א־ש ַ ֣מע ַע ִ ֣מי ְל ִ ֑‬ ‫וְ ל ֹ ָ‬ ‫א־א ָבה ִ ֽלי׃‬ ‫וְ ֝יִ ְש ָר ֵ֗אל ל ֹ ָ ֥‬ ‫ׁשל ֵחהּו ִב ְש ִר ֣ירּות ִל ָ ֑בם‬ ‫ָ ֭ו ֲא ְ‬

‫‪8b N‬‬ ‫‪8c P‬‬ ‫‪9a P‬‬ ‫‪9b N‬‬ ‫‪10a N‬‬ ‫‪10b N‬‬ ‫‪11a N‬‬ ‫‪11b P‬‬ ‫‪11c N‬‬ ‫‪12a N‬‬ ‫‪12b N‬‬ ‫‪13a N‬‬

‫‪120‬‬

‫‪Peter J. Gentry and John D. Meade‬‬

‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬

‫[ילכו במועצותיהם]‬ ‫[לו] עמי שמע לי‬ ‫[ישראל בדרכי יהלכו]‬ ‫[כמעט] אויביהם אכניע‬ ‫[ועל צריהם אשיב ידי]‬ ‫[משנא]י יהוה יכחשו לו‬

‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫‪H2: vacat‬‬

‫[ויהי עתם] לעולם‬ ‫ויאכילהו מחלב חטה‬ ‫[ומצור דב]ש אשביעך‬

‫יהם׃‬ ‫צֹות ֶ ֽ‬ ‫מֹוע ֵ‬ ‫ֵי ְ֝ל ֗כּו ְ ֽב ֲ‬ ‫֗לּו ַ ֭ע ִמי ש ֵ ֹ֣מ ַ ֽע ִ ֑לי‬ ‫ִי ְ֝ש ָר ֵ֗אל ִב ְד ָר ַ ֥כי יְ ַה ֵלֽכּו׃‬ ‫יהם ַא ְכ ִנ ַ֑יע‬ ‫אֹויְב ֶ ֣‬ ‫ֵ‬ ‫ִ ֭כ ְמ ַעט‬ ‫יהם ָא ִ ֥שיב יָ ִ ֽדי׃‬ ‫וְ ַ ֥על ָ֝צ ֵר ֶ֗‬ ‫ׁשּו־לֹו‬ ‫֑‬ ‫ְמ ַשנְ ֵ ֣אי ְי֭הוָ ה יְ ַכ ֲח‬

‫‪13b N‬‬ ‫‪14a N‬‬ ‫‪14b P‬‬ ‫‪15a N‬‬ ‫‪15b N‬‬ ‫‪16a N‬‬

‫‪Column II‬‬

‫‪Accents at‬‬ ‫‪H End‬‬

‫עֹולם׃‬ ‫יהי ִע ָ ֣תם ְל ָ ֽ‬ ‫וִ ִ ֖‬ ‫וַ ֭ ֽיַ ֲא ִכ ֵילהּו ֵמ ֵ ֣ח ֶלב ִח ָ ֑טה‬ ‫יעָך׃‬ ‫ּו ִ֝מ ּ֗צּור ְד ַ ֣בׁש ַא ְש ִב ֶ ֽ‬

‫‪MasPsa‬‬

‫‪MT-A‬‬ ‫‪Column II Continued‬‬

‫‪H1: ‘ole‬‬ ‫‪weyored‬‬ ‫‪H2: vacat‬‬ ‫�‪H1: ’atnah‬‬ ‫‪H2: silluq‬‬ ‫�‪H1: ’atnah‬‬ ‫‪H2: silluq‬‬ ‫�‪H1: ’atnah‬‬ ‫‪H2: silluq‬‬ ‫�‪H1: ’atnah‬‬ ‫‪H2: silluq‬‬ ‫‪H1: rebi‘a‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬

‫‪16b N‬‬ ‫‪17a N‬‬ ‫‪17b P‬‬

‫‪82‬‬

‫מזמו[ר] לאסף‬

‫ִמזְ ֹ֗מור ְל ָ֫א ָ ֥סף‬

‫‪1a N‬‬

‫אלהים נצב בעדת אל‬ ‫בקרב אלהים ישפט‬ ‫עד מתי תׁשפטו עול‬ ‫ופני רׁשעים תשאו סלה‬ ‫ׁשפטו דל ויתום‬ ‫עני ורׁש הצדי]קו[‬ ‫פלטו דל ואביון‬ ‫מיד רשעים הצילו‬ ‫לא ידעו ולא יבינו‬ ‫בחשכה יתהלכו‬ ‫ימוטו כל מוסדי ארץ‬ ‫אני אמרתי אלהים אתם‬ ‫ובני עליון כלכם‬ ‫אכן כאדם תמותון‬ ‫וכאחד השרים תפלו‬ ‫קומה אלהים שפטה הארץ‬ ‫כי אתה תנחל בכל הגוים‬

‫ת־אל‬ ‫ֹלהים נִ ָ ּ֥צב ַּב ֲע ַד ֵ ֑‬ ‫ֱ ֽא ִ֗‬ ‫ֹלהים יִ ְׁש ֽ ֹּפט׃‬ ‫ּב� ֶ ְ֖ק ֶרב ֱא ִ ֣‬ ‫טּו־עוֶ ל‬ ‫ד־מ ַ ֥תי ִּת ְׁש ְּפ ָ ֑‬ ‫ַע ָ‬ ‫אּו־ס ָלה׃‬ ‫ּופ ֵנ֥י ְ֝ר ָׁש ֗ ִעים ִּת ְׂש ֶ ֽ‬ ‫ְ‬ ‫יָֹתום‬ ‫טּו־דל וְ ֑‬ ‫ִׁש ְפ ַ ֥‬ ‫ָע ִנ֖י וָ ָ ֣רׁש ַה ְצ ִ ּֽדיקּו׃‬ ‫טּו־דל וְ ֶא ְבֹי֑ ון‬ ‫ַּפ ְּל ַ ֥‬ ‫ִמ ַּי֖ד ְר ָׁש ִ ֣עים ַה ִ ּֽצילּו׃‬ ‫֤ל ֹא ָי ְ�ֽד ֨עּו׀ וְ ֥ל ֹא ִ֗יָבינּו‬ ‫יִת ַה ָּל֑כּו‬ ‫ַּב ֲח ֵׁש ָ ֥כה ְ‬ ‫וס ֵדי ָ ֽא ֶרץ׃‬ ‫ל־ֹמ ְ‬ ‫֝י ִֹּ֗מוטּו ָּכ ֥‬ ‫ֹלהים ַא ֶ ּ֑תם‬ ‫י־א ַמ ְר ִּתי ֱא ִ ֣‬ ‫ֲ ֽאנִ ָ֭‬ ‫ּוב ֵנ֖י ֶע ְלֹי֣ ון ֻּכ ְּל ֶ ֽכם׃‬ ‫ְ‬ ‫מּותּון‬ ‫ָ֭א ֵכן ְּכ ָא ָ ֣דם ְּת ֑‬ ‫ּוכ ַא ַ ֖חד ַה ָּׂש ִ ֣רים ִּת ֽ ֹּפלּו׃‬ ‫ְ‬ ‫ֹלהים ָׁש ְפ ָ ֣טה ָה ָ ֑א ֶרץ‬ ‫קּומה ֱ֭א ִ‬ ‫ָ֣‬ ‫ל־הֹּגויִ ֽם׃‬ ‫י־א ָ ּ֥תה ִ֝תנְ ַ֗חל ְּב ָכ ַ‬ ‫ִ ּֽכ ַ‬

‫‪1b N‬‬ ‫‪1c N‬‬ ‫‪2a P‬‬ ‫‪2b P‬‬ ‫‪3a N‬‬ ‫‪3b N‬‬ ‫‪4a N‬‬ ‫‪4b N‬‬ ‫‪5a N‬‬ ‫‪5b P‬‬ ‫‪5c P‬‬ ‫‪6a N‬‬ ‫‪6b N‬‬ ‫‪7a N‬‬ ‫‪7b P‬‬ ‫‪8a P‬‬ ‫‪8b N‬‬

‫‪121‬‬ ‫‪Accents at‬‬ ‫‪H End‬‬

‫‪MasPsa and the Early History of the Hebrew Psalter‬‬

‫‪MasPsa‬‬

‫‪MT-A‬‬ ‫‪Column II Continued‬‬

‫‪linea vacat‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫‪H2: rebi‘a‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬

‫ִ ׁ֖שיר ִמזְ ֹ֣מור ְל ָא ָ ֽסף׃‬ ‫שיר מזמור לאסף‬ ‫י־לְ֑ך‬ ‫ל־ּד ִמ ָ‬ ‫ֹלהים ַא ֳ‬ ‫ֱא ִ ֥‬ ‫אלהים אל דמי לך‬ ‫ל־ּת ְׁש ֣קֹט ֵ ֽאל׃‬ ‫ל־ּת ֱח ַ ֖רׁש וְ ַא ִ‬ ‫ַא ֶ‬ ‫אל תחרש ואל תׁשקט אל‬ ‫ויְביָך יֶ ֱה ָמי֑ ּון‬ ‫י־ה ֵּנ֣ה ֭א ֶ‬ ‫ִ ּֽכ ִ‬ ‫כי הנה אויביך יהמיון‬ ‫ו ְ֝מ ַׂשנְ ֶ֗איָך ָנ ְׂ֣שאּו ֽר ֹאׁש׃‬ ‫ומשנאיך נשאו ראש‬ ‫ל־ע ְּמָך יַ ֲע ִ ֣רימּו ֹ֑סוד‬ ‫ַ ֽע ַ ֭‬ ‫על עמך [יע]ר[ימו ס]וד‬ ‫פּונֽיָך׃‬ ‫ל־צ ֶ‬ ‫יִתיָ ֲע ֗צּו ַע ְ‬ ‫ְ֝ו ְ‬ ‫ויתיעצו על צפוניך‬ ‫ָא ְמ ֗רּו ְ ֭לכּו וְ נַ ְכ ִח ֵיד֣ם ִמֹּג֑ וי‬ ‫אמרו לכו ונכח]יד[ם מגוי‬ ‫וְ ֽל ֹא־יִ ּזָ ֵכ֖ר ֵ ֽׁשם־יִ ְׂש ָר ֵ ֣אל ֹֽעוד׃‬ ‫ולא יזכר שם ישראל עוד‬ ‫ֹנוע ֣צּו ֵל֣ב יַ ְח ָ ּ֑דו‬ ‫ִ ּ֤כי ֲ‬ ‫כי נועצו לב יח]דו[‬ ‫ָ֝ע ֗ ֶליָך ְּב ִ ֣רית יִ ְכ ֽר ֹתּו׃‬ ‫עליך ברית יכרתו‬ ‫אלים‬ ‫ָא ֳה ֵל֣י ֱ֭אֹדום וְ יִ ְׁש ְמ ֵע ֗ ִ‬ ‫אלהי אדום וישמעא[לים]‬ ‫ֹמואב וְ ַהגְ ִ ֽרים׃‬ ‫ָ֥‬ ‫מואב והגרים‬ ‫ּגְ ָב֣ל ְ֭ו ַעֹּמון וַ ֲע ָמ ֵל֑ק‬ ‫גבל עמון ועמלק‬ ‫ְ֝פ ֗ ֶל ֶׁשת ִעם־ ֹ֥י ְׁש ֵבי ֹֽצור׃‬ ‫פלשת עם ישבי צור‬ ‫ם־אּׁשּור נִ ְלָו֣ה ִע ָ ּ֑מם‬ ‫ּגַ ַ֭‬ ‫גם אשור נלוה עמם‬ ‫י־ֹלוט ֶ ֽס ָלה׃‬ ‫ָ ֤הי֥ ּו זְ ֹ֖ר ַוע ִל ְבנֵ ֣‬ ‫היו זרוע לבני לוט סלה‬ ‫ה־ל ֶ ֥הם ְּכ ִמ ְד ָי�֑ן‬ ‫ֲע ֵ ֽׂש ָ‬ ‫עשה להם כמדין כסיסרא‬ ‫יֹׁשון׃‬ ‫יס ָ ֥רא ְ֝כ ִ֗יָבין ְּב ַנ ַ֣חל ִק ֽ‬ ‫ְ ּֽכ ִ ֽס ְ‬ ‫כיבין בנחל קיׁשון‬ ‫ין־ּד ֹאר‬ ‫נִ ְׁש ְמ ֥דּו ְ ֽב ֵע ֑‬ ‫נשמדו בעין דאר‬ ‫ָ ֥היּו ֗֝ד ֹ ֶמן ָל ֲא ָד ָ ֽמה׃‬ ‫היו דמן לאדמה‬ ‫יתֹמו ְ֭נ ִד ֵיבֹמו ְּכע ֵ ֹ֣רב וְ ִכזְ ֵ ֑אב‬ ‫ִׁש ֵ ֣‬ ‫שיתמו נדיבימו כערב וכזאב‬ ‫יכֹמו׃‬ ‫וכזבח וכצלמנע [כ]ל [נ]סיכמו ּוֽ ְכ ֶז ַ֥בח ו ְ֝כ ַצ ְל ֻמ ָּ֗נע ָּכל־נְ ִס ֵ ֽ‬ ‫ׁשר ָ֭א ְמרּו ִנ ֲ֣יר ָׁשה ָּל֑נּו‬ ‫ֲא ֶ ֣‬ ‫אשר אמרו נירׁשה לנו‬ ‫ֹלהים׃‬ ‫ֵ֝֗את נְ ֹ֣אות ֱא ִ ֽ‬ ‫את נאות אלהים‬ ‫יתֹמו ַכּגַ ְל ַּג֑ל‬ ‫ֹלהי ִׁש ֵ ֥‬ ‫ֱ ֽא ַ֗‬ ‫אלהים ׁשיתמו כגלגל‬ ‫י־ר ַּוח׃‬ ‫ְ֝כ ַ ֗קׁש ִל ְפנֵ ֽ‬ ‫כקׁש לפני רוח‬ ‫ר־י ַ֑ער‬ ‫ְּכ ֵ ֥אׁש ִּת ְב ַע ָ‬ ‫כאש תבער יער‬ ‫ו ְ֝כ ֶל ָה ָ֗בה ְּת ַל ֵ ֥הט ָה ִ ֽרים׃‬ ‫וכלהבה תלהט הרים‬ ‫ֵ ֭כן ִּת ְר ְּד ֵפ֣ם ְּב ַס ֲע ֶ ֑רָך‬ ‫כן תרדפם בסערך‬ ‫סּופ ְתָך֥ ְת ַב ֲה ֵ ֽלם׃‬ ‫ּוב ָ‬ ‫ְ‬ ‫ובסופתך תבהלם‬ ‫יהם ָק ֹ֑לון‬ ‫ַמ ֵּל֣א ְפנֵ ֶ ֣‬ ‫מלא פניהם קלון‬

‫‪83‬‬ ‫‪1a N‬‬ ‫‪2a P‬‬ ‫‪2b N‬‬ ‫‪3a N‬‬ ‫‪3b N‬‬ ‫‪4a N‬‬ ‫‪4b N‬‬ ‫‪5a N‬‬ ‫‪5b N‬‬ ‫‪6a N‬‬ ‫‪6b P‬‬ ‫‪7a N‬‬ ‫‪7b N‬‬ ‫‪8a N‬‬ ‫‪8b N‬‬ ‫‪9a N‬‬ ‫‪9b P‬‬ ‫‪10a N‬‬ ‫‪10b N‬‬ ‫‪11a N‬‬ ‫‪11b N‬‬ ‫‪12a N‬‬ ‫‪12b N‬‬ ‫‪13a N‬‬ ‫‪13b N‬‬ ‫‪14a N‬‬ ‫‪14b N‬‬ ‫‪15a P‬‬ ‫‪15b N‬‬ ‫‪16a P‬‬ ‫‪16b N‬‬ ‫‪17a N‬‬

‫‪122‬‬

‫‪Peter J. Gentry and John D. Meade‬‬

‫‪Accents at‬‬ ‫‪H End‬‬

‫‪MasPsa‬‬

‫‪MT-A‬‬ ‫‪Column III‬‬

‫‪H1: silluq‬‬ ‫‪H2: rebi‘a‬‬ ‫‪H1: silluq +‬‬

‫וידעו‬

‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬

‫[ויבקשו שמך יהוה]‬ ‫[יבשו ויבהלו עדי עד]‬ ‫ויח[פרו ויאבדו וידעו]‬

‫הוה׃‬ ‫וִ ַיב ְק ׁ֖שּו ִש ְמָך֣ יְ ָ ֽ‬ ‫י־עד‬ ‫יֵ ֖בֹׁשּו וְ ָיִב ֲה ֥לּו ֲע ֵד ֗ ַ‬ ‫אבדּו׃‬ ‫ְ �וֽיַ ְח ְפ ֥רּו וְ י ֹ ֵ ֽ‬

‫‪17b N‬‬ ‫‪18a N‬‬ ‫‪18b P‬‬

‫[כי אתה שמך יהוה לבדך]‬ ‫עלי[ון על כל הארץ]‬

‫הו֣ה ְל ַב ֶ ֑דָך‬ ‫י־א ָ֬תה ִש ְמָך֣ יְ ָ‬ ‫ְ �וֽיֵ ְד ֗עּו ִ ֽכ ַ‬ ‫ל־ה ָ ֽא ֶרץ׃‬ ‫ל־כ ָ‬ ‫֝ ֶע ְלי֗ ֹון ַע ָ‬

‫‪19a P‬‬ ‫‪19b P‬‬ ‫‪84‬‬

‫‪linea vacat ‬‬ ‫‪H1: silluq‬‬ ‫‪H2: vacat‬‬ ‫‪H1: rebi‘a‬‬ ‫‪H2: silluq‬‬ ‫‪H1: s�innor‬‬ ‫‪H2: ‘ole‬‬ ‫‪weyored‬‬ ‫‪H1: rebi‘a‬‬ ‫‪H2: silluq‬‬ ‫‪H1: pazer‬‬ ‫‪H2: s�innor‬‬ ‫‪H1: ‘ole‬‬ ‫‪weyored‬‬ ‫‪H2: deh�i‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫‪H2: deh�i‬‬ ‫�‪H1: ’atnah‬‬ ‫‪H2: silluq‬‬ ‫�‪H1: ’atnah‬‬ ‫‪H2: silluq‬‬ ‫‪H1: deh�i‬‬ ‫�‪H2: ’atnah‬‬ ‫‪H1: silluq‬‬ ‫�‪H2: ’atnah‬‬

‫‪83‬‬

‫למנצח [על הגתית לבני קרח‬ ‫מזמור]‬

‫ל־הגִ ִ ֑תית ִל ְבנֵ י־ ֥קֹ ַרח‬ ‫ַל ְמנַ ֵ ֥צ ַח ַ ֽע ַ‬ ‫ִמזְ ֽמֹור׃‬

‫מה ידידות מ[שכנותיך]‬ ‫[יהוה צבאות]‬ ‫נכספה וגם כ[לתה נפשי]‬ ‫[לחצרות יהוה]‬

‫נֹותיָך‬ ‫ַמה־יְ ִד ֥ידֹות ִמ ְש ְכ ֶ֗‬ ‫הו֥ה ְצ ָב ֽאֹות׃‬ ‫יְ ָ‬ ‫ם־כ ְל ָ֨תה׀ נַ ְפ ִש֮י‬ ‫נִ ְכ ְס ֬ ָפה וְ גַ ָ‬ ‫ְל ַח ְצ ֪רֹות יְ ֫הָו֥ה‬

‫‪2a N‬‬ ‫‪2b N‬‬ ‫‪3a N‬‬ ‫‪3b N‬‬

‫לבי ובשרי יר[ננו]‬ ‫[אל אל חי]‬ ‫גם צפור מצא[ה בית]‬ ‫[ודרור קן לה]‬ ‫אשר שתה א[פרחיה]‬

‫ּוב ָש ִ ֑רי ְי ַ֝רנְ נ֗ ּו‬ ‫ִל ִ ֥בי ְ‬ ‫ל־חי׃‬ ‫ֶ ֣אל ֵ ֽא ָ ֽ‬ ‫ם־צ ּ֨פֹור׀ ָ ֪מ ְצ ָאה ַ֡ביִת‬ ‫גַ ִ‬ ‫ְּוד ֤רֹור׀ ֵ ֥קן ָל ּ֮ה‬ ‫יה‬ ‫ר־ש ָתה ֶא ְפ ֫ר ֹ ֶ ֥ח ָ‬ ‫ֲא ֶש ָ ֪‬

‫‪3c N‬‬ ‫‪3d P‬‬ ‫‪4a N‬‬ ‫‪4b N‬‬ ‫‪4c N‬‬

‫[את מזבחותיך]‬ ‫יהוה צבאות [מלכי ואלהי]‬ ‫[אשרי יושבי ביתך]‬ ‫עוד יהללוך ס[לה]‬ ‫[אשרי אדם עוז לו בך]‬ ‫מסלות בלבבם‬ ‫[עברי בעמק הבכא]‬ ‫מעין ישיתוהו‬ ‫[גם ברכות יעטה מורה]‬ ‫ילכו מחיל אל ח[יל]‬ ‫[יראה אל אלהים בציון]‬ ‫יהוה אלהים צב[אות]‬ ‫[שמעה תפלתי]‬ ‫ה[אזינה] אלהי יע[קב סלה]‬ ‫[מגננו ראה אלהים]‬

‫חֹותיָך‬ ‫ת־מזְ ְב ֶ‬ ‫ֶ ֽא ִ ֭‬ ‫אֹלהי׃‬ ‫הו֣ה ְצ ָב ֑אֹות ַ֝מ ְל ִ֗כי וֵ ָ ֽ‬ ‫יְ ָ‬ ‫יתָך‬ ‫יֹוש ֵב֣י ֵב ֶ ֑‬ ‫ַ ֭א ְש ֵרי ְ‬ ‫֝֗עֹוד יְ ַֽה ְל ֥לּוָך ֶ ֽס ָלה׃‬ ‫ֹוז־לֹו ָ ֑בְך‬ ‫ַא ְש ֵ ֣רי ָ ֭א ָדם ֽע ֥‬ ‫ְ֝מ ִס ּ֗לֹות ִב ְל ָב ָ ֽבם׃‬ ‫ע ְֹב ֵ ֤רי׀ ְב ֵע ֶ֣מק ַ ֭ה ָב ָכא‬ ‫יתּוהּו‬ ‫ַמ ְע ָי�֣ן יְ ִש ֑‬ ‫מֹורה׃‬ ‫ם־ב ָר ֗כֹות יַ ְע ֶ ֥טה ֶ ֽ‬ ‫גַ ְ֝‬ ‫ל־חיִ ל‬ ‫ֵי ְ֭לכּו ֵמ ַ ֣חיִ ל ֶא ָ ֑‬ ‫ֹלהים ְב ִצּיֽ ֹון׃‬ ‫ל־א ִ ֣‬ ‫יֵ ָר ֶ ֖אה ֶא ֱ‬ ‫ֹלהים ְ ֭צ ָבאֹות‬ ‫יְ ֘הָו֤ה ֱא ִ ֣‬ ‫ִש ְמ ָ ֣עה ְת ִפ ָל ִ ֑תי‬ ‫ֹלהי יַ ֲע ֣קֹב ֶ ֽס ָלה׃‬ ‫ַה ֲא ִ֨זינָ ה ֱא ֵ ֖‬ ‫ֹלהים‬ ‫ָ ֭מגִ נֵ נּו ְר ֵ ֣אה ֱא ִ ֑‬

‫‪1N‬‬

‫‪4d N‬‬ ‫‪4e P‬‬ ‫‪5a P‬‬ ‫‪5b P‬‬ ‫‪6a P‬‬ ‫‪6b N‬‬ ‫‪7a N‬‬ ‫‪7b N‬‬ ‫‪7c N‬‬ ‫‪8a P‬‬ ‫‪8b N‬‬ ‫‪9a N‬‬ ‫‪9b N‬‬ ‫‪9c P‬‬ ‫‪10a N‬‬

MasPsa and the Early History of the Hebrew Psalter

10b P 11a N 11b N 11c C 11d N 12a N 12b N 12c N 13a N 13b P

‫יחָך׃‬ ֽ ֶ ‫וְ ַ֝ה ֵ֗בט ְפ ֵנ֣י ְמ ִש‬ ‫ִ ֤כי ֽטֹוב־י֥ ֹום ַב ֲח ֵצ ֶ ֗ריָך‬ ‫ֵ֫מ ָ ֥א ֶלף ָב ַ֗ח ְר ִתי‬ ‫ֹלהי‬ ֑ ַ ‫ּתֹופף ְב ֵב֣ית ֱא‬ ֵ ‫ִ ֭ה ְס‬ ‫י־ר ַשע׃‬ ֽ ֶ ‫ִ֝מ ּ֗דּור ְב ָא ֳה ֵל‬ ‫הו֪ה ֱא ֹ֫ל ִ ֥הים‬ ָ ְ‫ּומגֵ ֮ן י‬ ָ ‫ִ ֤כי ֶ֨ש ֶמׁש׀‬

]‫וה[בט פני משיחך‬ ]‫[כי טוב יום בחצריך‬ ]‫[מאלף בחרתי‬ ]‫[הסתופף בבית אלהי‬ ]‫מד[ו]ר באה[לי רשע‬ ]‫[כי שמש ומגן יהוה אלהים‬

‫הו֑ה‬ ָ ְ‫יִתן י‬ ֣ ֵ ‫ֵ ֣חן ְו ָ֭כבֹוד‬ ]‫חן וכבוד יתן יה[וה‬ ‫ע־טֹוב ַ ֽלה ְֹל ִ ֥כים ְב ָת ִ ֽמים׃‬ ֗֝ ַ‫יִמנ‬ ְ ‫[לא ימנע טוב להלכים בתמים] ֥ל ֹא‬ ‫הו֥ה ְצ ָב ֑אֹות‬ ָ ְ‫י‬ ‫יהוה צבאות‬ ‫ַ ֽא ְש ֵ ֥רי ָ֝א ָ ֗דם ב ֵ ֹ֥ט ַח ָ ֽבְך׃‬ ]‫[אשרי אדם בטח בך‬

85

123 H1: silluq H2: rebi‘a H1: rebi‘a H2: ’atnah� H1: silluq H2: ‘ole weyored H1: ’atnah� H2: silluq H1: ’atnah� H2: silluq linea vacat

1N 2a P 2b N 3a P 3b P 4a P 4b P 5a N 5b N 6a N 6b N

‫ַל ְמנַ ֵ֬צ ַח׀ ִל ְבנֵ י־ ֬קֹ ַרח ִמזְ ֽמֹור׃‬

]‫למנצח לבני קרח מזמ[ור‬

‫הו֣ה ַא ְר ֶצָ֑ך‬ ָ ְ‫ית י‬ ָ ‫ָר ִ ֣צ‬ ‫ַ֝ ֗ש ְב ָת ְש ִ ֯ ֣בות יַ ֲע ֽקֹב׃‬ ‫את ֲעֹו֣ ן ַע ֶ ֑מָך‬ ָ ‫ָנ ָ֭ש‬ ‫אתם ֶ ֽס ָלה׃‬ ֣ ָ ‫ל־ח ָט‬ ַ ‫ית ָכ‬ ָ ‫ִכ ִ ֖ס‬ ‫ל־ע ְב ָר ֶ ֑תָך‬ ֶ ‫ָא ַ ֥ס ְפ ָת ָכ‬ ‫ֹות ֵמ ֲח ֥רֹון ַא ֶ ֽפָך׃‬ ָ ‫ֱ֝ה ִש ֗יב‬ ‫ֹלהי יִ ְש ֵ ֑ענּו‬ ֣ ֵ ‫ּובנּו ֱא‬ ֵ ‫ׁ֭ש‬ ‫וְ ָה ֵ ֖פר ַ ֽכ ַע ְסָך֣ ִע ָ ֽמנּו׃‬ ‫ף־בנּו‬ ֑ ָ ַ‫עֹולם ֶת ֱאנ‬ ֥ ָ ‫ַה ְל‬ ‫ִת ְמ ֥ש ְֹך ַ֝א ְפ ָ֗ך ְל ֣ד ֹר וָ ֽד ֹר׃‬

‫רצית יהוה ארצך‬ ]‫[שבת שבות יעקב‬ ‫נשאת עון עמך‬ ]‫[כסית כל חטאתם סלה‬ ‫אספת כל עברתך‬ ]‫[השיבות מחרון אפך‬ ‫שובנו אלהי ישענו‬ ]‫[והפר כעסך עמנו‬ ‫הלעולם תאנף בנו‬ ]‫[תמשך אפך לדר ודר‬

H1: silluq H2: vacat H1: ’atnah� H2: silluq H1: ’atnah� H2: silluq H1: ’atnah� H2: silluq H1: ’atnah� H2: silluq H1: ’atnah� H2: silluq

Fourth, comparison between MasPsa and the Aleppo Codex (A) in terms of the layout of the text requires evaluation. Here we are comparing the layout in a codex transmitting the same textual tradition in a scroll approximately 1000 years later in time. A first look indicates an almost identical layout. The Masada Psalms scroll has one column of text divided into bicola by intervals or spaces. The layout of the Aleppo Codex entails two columns of text per page (at least in Psalms). Each column of a page of the Aleppo Codex corresponds to a single column in MasPsa, and usually lines of text are divided into two parts by an interval or space. Paul Sanders produced a significant study of the colometry of Psalms 1–14 in the Aleppo Codex.20 He classifies the colometry evinced there in 13 categories. At least 50% of the time, each line contains a couplet (two cola) separated by a space. For the remaining 50%, two, three, or even four cola per line are written in different patterns, occasionally with no interval or space at all. He concludes that 20 Ibid.

‫‪124‬‬

‫‪Peter J. Gentry and John D. Meade‬‬

‫‪this is due mainly to the fact that the page size demanded by the codex format‬‬ ‫‪required narrower columns of text.‬‬ ‫‪We shall carefully compare the approach to stichometry in A against that in‬‬ ‫‪MasPsa. A chart displays the differences between the layout in A and MasPsa.‬‬ ‫‪Each line in the chart shows only one colon or half-line in the column of MasPsa.‬‬ ‫‪The column to the right shows how the two cola of the Masada Psalms manu‬‬‫‪script are arranged as one line in the Aleppo Codex. Verse numbers follow MT.‬‬ ‫‪The number in the most right column is the category from Sanders for this pat‬‬‫‪tern of colometry in the Aleppo Codex, elaborated infra.‬‬

‫‪Stichometry of MasPsa versus the Aleppo Codex‬‬ ‫)‪Aleppo Codex (A‬‬

‫‪MasPsa‬‬

‫‪81‬‬

‫‪Column I‬‬

‫ל־הגִ ִ֬תית ְל ָא ָ ֽסף‬ ‫‪ַ 13‬ל ְמנַ ֵ֬צ ַח׀ ַ ֽע ַ‬ ‫אֹלהי יַ ֲע ֽקֹב׃‬ ‫עּוז֑נּו ָ֝ה ִ ֗ריעּו ֵל ֵ ֥‬ ‫אֹלהים ֵ‬ ‫‪֭ ַ 1‬ה ְרנִ ינּו ֵל ִ ֣‬ ‫תף‬ ‫ּותנּו־ ֑ ֹ‬ ‫אּו־ז ְ֭מ ָרה ְ‬ ‫‪ֽ ְ 1‬ש ִ‬

‫ם־נ ֶֽבל׃‬ ‫ִכּנ֖ ֹור נָ ִ ֣עים ִע ָ‬

‫ׁשֹופר‬ ‫‪ִ 1‬ת ְק ֣עּו ַב ֣חֹ ֶדׁש ָ ֑‬

‫ַ֝ב ֵ֗כ ֶסה ְלי֣ ֹום ַח ֵגֽנּו׃‬

‫‪֤ ִ 1‬כי ֣חֹק ְליִ ְש ָר ֵ ֣אל ֑הּוא‬

‫אֹלהי יַ ֲע ֽקֹב׃‬ ‫ִ֝מ ְש ֗ ָפט ֵל ֵ ֥‬

‫ֹוסף ָש ֗מֹו‬ ‫יה ֵ ֤‬ ‫‪ֵ 4‬ע֤דּות׀ ִ ֽב ֘‬

‫ל־א ֶרץ‬ ‫ְ ֭ב ֵצאתֹו ַע ֶ ֣‬

‫ֹותי‬ ‫‪ִ 9‬מ ְצ ָ ֑ריִם ְש ַ ֖פת לֹא־יָ ַ ֣ד ְע ִתי ֶא ְש ָ ֽמע׃ ֲה ִס ֣יר ִ‬ ‫ַ֝כ ֗ ָפיו ִמ ּ֥דּוד ַת ֲע ֽבֹ ְרנָ ה׃‬ ‫‪ִ 1‬מ ֵ ֣ס ֶבל ִש ְכ ֑מֹו‬ ‫את וָ ֲא ַ֫ח ְל ֶ ֥צךָ ֶ֭א ֶענְ ָך ְב ֵ ֣ס ֶתר‬ ‫? ַב ָצ ָ ֥רה ָק ָ ֗ר ָ‬ ‫ְמ ִר ָיב֣ה ֶ ֽס ָלה׃‬ ‫ל־מי‬ ‫נְָך ַע ֵ ֖‬ ‫? ַ ֑ר ַעם ֶא ְב ָ ֽח ֨‬ ‫ִי ְ֝ש ָר ֵ֗אל‬ ‫‪ְ 4‬ש ַ ֣מע ַ ֭ע ִמי וְ ָא ִ ֣ע ָידה ָ ֑בְך‬ ‫ע־לי׃ ֽל ֹא־יִ ְה ֶי ֣ה ְ ֭בָך ֵ ֣אל ָז֑ר‬ ‫ם־ת ְש ַ ֽמ ִ ֽ‬ ‫‪ִ 1‬א ִ ֽ‬ ‫ָאנ ִֹ֨כי׀ יְ ֘הָו֤ה‬ ‫‪ 4‬וְ ֥ל ֹא ִ֝ת ְש ַת ֲח ֶ ֗וה ְל ֵ ֣אל נֵ ָ ֽכר׃‬ ‫ֹלהיָך ַ ֽ֭ה ַמ ַע ְלָך ֵמ ֶ ֣א ֶרץ ִמ ְצ ָ ֑ריִם ַה ְר ֶחב‬ ‫‪ֱ 4‬א ֶ֗‬ ‫‪ִ֝ ֗ 1‬פיָך וַ ֲא ַמ ְל ֵ ֽאהּו׃‬

‫קֹולי‬ ‫א־ש ַ ֣מע ַע ִ ֣מי ְל ִ ֑‬ ‫וְ ל ֹ ָ‬

‫[למנצח על הגתית לאסף]‬ ‫[הרנינו לאלהים עוזנו]‬ ‫[הריעו לאלהי י]עקב‬ ‫[שאו־זמרה ותנו תף]‬ ‫[כנור נעים] עם [נ]בל‬ ‫[תקעו בחדש שופר]‬ ‫[בכסה ליום] חגנו‬ ‫[כי חק לישראל הוא]‬ ‫[משפט לא]להי יעקב‬ ‫[עדות ביהוסף שמו]‬ ‫[בצאתו על א]רץ מצרים‬ ‫[שפת לא ידעתי אשמע]‬ ‫[הסירותי] מסבל שכמו‬ ‫[כפיו מדוד תעברנה]‬ ‫[בצ]רה ק[ר]את וא[ח]לצכה‬ ‫[אענך בסתר רעם]‬ ‫[אבח]נך על מי מריבה סלה‬ ‫[שמע עמי ואעידה בך]‬ ‫[י]שראל אם תשמע קולי‬ ‫[לא יהיה בך אל זר]‬ ‫[ולא] תשתחוה לאל נכר‬ ‫[אנכי יהוה אלהיך]‬ ‫[המע]לך מארץ מצרים‬ ‫[הרחב פיך ואמלאהו]‬

‫‪1a‬‬ ‫‪2a‬‬ ‫‪2b‬‬ ‫‪3a‬‬ ‫‪3b‬‬ ‫‪4a‬‬ ‫‪4b‬‬ ‫‪5a‬‬ ‫‪5b‬‬ ‫‪6a‬‬ ‫‪6b‬‬ ‫‪6c‬‬ ‫‪7a‬‬ ‫‪7b‬‬ ‫‪8a‬‬ ‫‪8b‬‬ ‫‪8c‬‬ ‫‪9a‬‬ ‫‪9b‬‬ ‫‪10a‬‬ ‫‪10b‬‬ ‫‪11a‬‬ ‫‪11b‬‬ ‫‪11c‬‬

‫‪125‬‬

‫‪MasPsa and the Early History of the Hebrew Psalter‬‬

‫ׁשל ֵחהּו‬ ‫ָ ֭ו ֲא ְ‬ ‫א־א ָבה ִ ֽלי׃‬ ‫‪ 4‬וְ ֝יִ ְש ָר ֵ֗אל ל ֹ ָ ֥‬ ‫יהם׃‬ ‫צֹות ֶ ֽ‬ ‫מֹוע ֵ‬ ‫‪ִ 4‬ב ְש ִר ֣ירּות ִל ָ ֑בם ֵי ְ֝ל ֗כּו ְ ֽב ֲ‬ ‫‪֗ 1‬לּו ַ ֭ע ִמי ש ֵ ֹ֣מ ַ ֽע ִ ֑לי‬

‫ִי ְ֝ש ָר ֵ֗אל ִב ְד ָר ַ ֥כי יְ ַה ֵלֽכּו׃‬

‫יהם ַא ְכ ִנ ַ֑יע וְ ַ ֥על ָ֝צ ֵר ֶ֗יהם‬ ‫אֹויְב ֶ ֣‬ ‫ֵ‬ ‫‪֭ ִ 4‬כ ְמ ַעט‬ ‫‪ָ 1‬א ִ ֥שיב ִ ֽיָדי׃‬

‫ְמ ַשנְ ֵא֣י ְי֭הוָ ה ַיְכ ֲחׁשּו־‬

‫עֹולם׃‬ ‫יהי ִע ָ ֣תם ְל ָ ֽ‬ ‫‪֑ 11‬לֹו וִ ִ ֖‬ ‫‪ֵ 1‬מ ֵ ֣ח ֶלב ִח ָ ֑טה‬

‫֭יַא ִכ ֵילהּו‬ ‫וַ ֽ ֲ‬

‫יעָך׃‬ ‫ּו ִ֝מ ּ֗צּור ְד ַבׁ֣ש ַא ְש ִב ֶ ֽ‬

‫[ולא] שמע עמי לקולי‬ ‫[וישראל לא אבה לי]‬ ‫[ואש]לחהו בשרירות לבם‬ ‫[ילכו במועצותיהם]‬ ‫[לו] עמי שמע לי‬ ‫[ישראל בדרכי יהלכו]‬ ‫[כמעט] אויביהם אכניע‬ ‫[ועל צריהם אשיב ידי]‬ ‫[משנא]י יהוה יכחשו לו‬

‫‪12a‬‬ ‫‪12b‬‬ ‫‪13a‬‬ ‫‪13b‬‬ ‫‪14a‬‬ ‫‪14b‬‬ ‫‪15a‬‬ ‫‪15b‬‬ ‫‪16a‬‬

‫[ויהי עתם] לעולם‬ ‫ויאכילהו מחלב חטה‬ ‫[ומצור דב]ש אשביעך‬

‫‪16b‬‬ ‫‪17a‬‬ ‫‪17b‬‬

‫‪Column II‬‬

‫‪linea vacat‬‬

‫‪linea vacat‬‬

‫‪Aleppo Codex ‬‬

‫‪MasPsa ‬‬ ‫‪Column II Continued‬‬

‫מזמו[ר] לאסף‬

‫‪ִ 13‬מזְ ֹ֗מור ְל ָ֫א ָ ֥סף‬ ‫ּב� ֶ ְ֖ק ֶרב‬ ‫ת־אל‬ ‫ֹלהים נִ ָ ּ֥צב ַּב ֲע ַד ֵ ֑‬ ‫‪ֽ ֱ 4‬א ִ֗‬ ‫טּו־עוֶ ל‬ ‫ד־מ ַ ֥תי ִּת ְׁש ְּפ ָ ֑‬ ‫ֹלהים יִ ְׁש ֽ ֹּפט׃ ַע ָ‬ ‫‪ֱ 1‬א ִ ֣‬ ‫טּו־דל וְ יָ ֹ֑תום‬ ‫אּו־ס ָלה׃ ִׁש ְפ ַ ֥‬ ‫ּופ ֵנ֥י ְ֝ר ָׁש ֗ ִעים ִּת ְׂש ֶ ֽ‬ ‫‪ְ 1‬‬ ‫‪ָ 1‬ע ִנ֖י וָ ָ ֣רׁש ַה ְצ ִ ּֽדיקּו׃‬ ‫‪ִ 1‬מ ַּי֖ד ְר ָׁש ִ ֣עים ַה ִ ּֽצילּו׃‬ ‫יִת ַה ָּל֑כּו‬ ‫‪ַּ 1‬ב ֲח ֵׁש ָ ֥כה ְ‬

‫טּו־דל וְ ֶא ְבֹי֑ ון‬ ‫ַּפ ְּל ַ ֥‬ ‫֤ל ֹא ָי ְ�ֽד ֨עּו׀ וְ ֥ל ֹא ִ֗יָבינּו‬ ‫וס ֵדי ָ ֽא ֶרץ׃‬ ‫ל־ֹמ ְ‬ ‫֝י ִֹּ֗מוטּו ָּכ ֥‬

‫ֹלהים ַא ֶ ּ֑תם‬ ‫י־א ַמ ְר ִּתי ֱא ִ ֣‬ ‫‪ֽ ֲ 1‬אנִ ָ֭‬ ‫מּותּון‬ ‫‪ָ֭ 4‬א ֵכן ְּכ ָא ָ ֣דם ְּת ֑‬

‫ּוב ֵנ֖י ֶע ְלֹי֣ ון ֻּכ ְּל ֶ ֽכם׃‬ ‫ְ‬ ‫ּוכ ַא ַ ֖חד‬ ‫ְ‬

‫ֹלהים‬ ‫קּומה ֱ֭א ִ‬ ‫ָ֣‬ ‫‪ַ 1‬ה ָּׂש ִ ֣רים ִּת ֽ ֹּפלּו׃‬ ‫ל־הֹּגויִ ֽם׃‬ ‫י־א ָ ּ֥תה ִ֝תנְ ַ֗חל ְּב ָכ ַ‬ ‫‪ָׁ 1‬ש ְפ ָ ֣טה ָה ָ ֑א ֶרץ ִ ּֽכ ַ‬

‫‪linea vacat‬‬

‫אלהים נצב בעדת אל‬ ‫בקרב אלהים ישפט‬ ‫עד מתי תׁשפטו עול‬ ‫ופני רׁשעים תשאו סלה‬ ‫ ‬ ‫ׁשפטו דל ויתום‬ ‫עני ורׁש הצדי]קו[‬ ‫פלטו דל ואביון‬ ‫מיד רשעים הצילו‬ ‫לא ידעו ולא יבינו‬ ‫בחשכה יתהלכו‬ ‫ימוטו כל מוסדי ארץ‬ ‫אני אמרתי אלהים אתם‬ ‫ובני עליון כלכם‬ ‫אכן כאדם תמותון‬ ‫וכאחד השרים תפלו‬ ‫קומה אלהים שפטה הארץ‬ ‫כי אתה תנחל בכל הגוים‬

‫‪82‬‬ ‫‪1a‬‬ ‫‪1b‬‬ ‫‪1c‬‬ ‫‪2a‬‬ ‫‪2b‬‬ ‫‪3a‬‬ ‫‪3b‬‬ ‫‪4a‬‬ ‫‪4b‬‬ ‫‪5a‬‬ ‫‪5b‬‬ ‫‪5c‬‬ ‫‪6a‬‬ ‫‪6b‬‬ ‫‪7a‬‬ ‫‪7b‬‬ ‫‪8a‬‬ ‫‪8b‬‬

‫‪126‬‬

‫‪Peter J. Gentry and John D. Meade‬‬

‫‪Aleppo Codex‬‬

‫‪MasPsa‬‬ ‫‪Column II Continued‬‬

‫‪linea vacat‬‬

‫‪ׁ֖ ִ 13‬שיר ִמזְ ֹ֣מור ְל ָא ָ ֽסף׃‬ ‫י־לְ֑ך‬ ‫ל־ּד ִמ ָ‬ ‫ֹלהים ַא ֳ‬ ‫‪ֱ 4‬א ִ ֥‬

‫שיר מזמור לאסף‬ ‫ל־ּת ֱח ַ ֖רׁש וְ ַאל־‬ ‫ַא ֶ‬

‫ויְביָך יֶ ֱה ָמי֑ ּון‬ ‫י־ה ֵּנ֣ה ֭א ֶ‬ ‫ִ ּֽכ ִ‬ ‫‪ִּ 1‬ת ְׁש ֣קֹט ֵ ֽאל׃‬ ‫ל־ע ְּמָך‬ ‫ַ ֽע ַ ֭‬ ‫‪ 4‬ו ְ֝מ ַׂשנְ ֶ֗איָך ָנ ְׂ֣שאּו ֽר ֹאׁש׃‬ ‫פּונֽיָך׃‬ ‫ל־צ ֶ‬ ‫יִתיָ ֲע ֗צּו ַע ְ‬ ‫ְ֝ו ְ‬ ‫‪ 1‬יַ ֲע ִ ֣רימּו ֹ֑סוד‬ ‫‪ָ 4‬א ְמ ֗רּו ְ ֭לכּו וְ נַ ְכ ִח ֵיד֣ם ִמֹּג֑ וי וְ ֽל ֹא־יִּזָ ֵכ֖ר‬ ‫ֹנוע ֣צּו ֵל֣ב ְיַח ָ ּ֑דו‬ ‫ם־יִׂש ָר ֵא֣ל ֹֽעוד׃ ִּכ֤י ֲ‬ ‫‪ֽׁ ֵ 1‬ש ְ‬ ‫‪ָ֝ 4‬ע ֗ ֶליָך ְּב ִ ֣רית יִ ְכ ֽר ֹתּו׃‬

‫ָא ֳה ֵל֣י ֱ֭אֹדום‬

‫ֹמואב וְ ַהגְ ִ ֽרים׃‬ ‫ָ֥‬ ‫אלים‬ ‫‪ 1‬וְ יִ ְׁש ְמ ֵע ֗ ִ‬ ‫ְ֝פ ֗ ֶל ֶׁשת ִעם־ ֹ֥י ְׁש ֵבי ֹֽצור׃‬ ‫? ּגְ ָב֣ל ְ֭ו ַעֹּמון וַ ֲע ָמ ֵל֑ק‬ ‫ָ ֤הי֥ ּו זְ ֹ֖ר ַוע‬ ‫ם־אּׁשּור נִ ְלָו֣ה ִע ָ ּ֑מם‬ ‫‪ּ 4‬גַ ַ֭‬ ‫ה־ל ֶ ֥הם ְּכ ִמ ְדָי�֑ן‬ ‫ֲע ֵ ֽׂש ָ‬ ‫נֵי־ֹלוט ֶ ֽס ָלה׃‬ ‫‪ִ 1‬ל ְב ֣‬ ‫יֹׁשון׃ נִ ְׁש ְמ ֥דּו ְ ֽב ֵעין־‬ ‫יס ָ ֥רא ְ֝כ ִ֗יָבין ְּב ַנ ַ֣חל ִק ֽ‬ ‫‪ּֽ ְ 4‬כ ִ ֽס ְ‬ ‫יתֹמו ְ֭נ ִד ֵיבֹמו‬ ‫‪ּ֑ 4‬ד ֹאר ָ ֥היּו ֗֝ד ֹ ֶמן ָל ֲא ָד ָ ֽמה׃ ִׁש ֵ ֣‬ ‫יכֹמו׃‬ ‫‪ְּ 1‬כע ֵ ֹ֣רב וְ ִכזְ ֵ ֑אב ּוֽ ְכ ֶז ַ֥בח ו ְ֝כ ַצ ְל ֻמ ָּ֗נע ָּכל־נְ ִס ֵ ֽ‬ ‫ֹלהים׃‬ ‫ׁשר ָ֭א ְמרּו ִנ ֲ֣יר ָׁשה ָּל֑נּו ֵ֝֗את נְ ֹ֣אות ֱא ִ ֽ‬ ‫‪ֲ 1‬א ֶ ֣‬ ‫נֵי־ר ַּוח׃‬ ‫יתֹמו ַכּגַ ְל ַּג֑ל ְ֝כ ַ ֗קׁש ִל ְפ ֽ‬ ‫ֹֽלהי ִׁש ֵ ֥‬ ‫‪ֱ 1‬א ַ֗‬ ‫ר־י ַ֑ער‬ ‫‪ְּ 1‬כ ֵ ֥אׁש ִּת ְב ַע ָ‬ ‫‪֭ ֵ 1‬כן ִּת ְר ְּד ֵפ֣ם ְּב ַס ֲע ֶ ֑רָך‬ ‫יהם ָק ֹ֑לון‬ ‫‪ַ 1‬מ ֵּל֣א ְפנֵ ֶ ֣‬

‫‪linea vacat‬‬

‫ו ְ֝כ ֶל ָה ָ֗בה ְּת ַל ֵ ֥הט ָה ִ ֽרים׃‬ ‫סּופ ְתָך֥ ְת ַב ֲה ֵ ֽלם׃‬ ‫ּוב ָ‬ ‫ְ‬ ‫הוה׃‬ ‫וִ ַיב ְק ׁ֖שּו ִש ְמָך֣ יְ ָ ֽ‬

‫‪linea vacat‬‬

‫אלהים אל דמי לך‬ ‫אל תחרש ואל תׁשקט אל‬ ‫כי הנה אויביך יהמיון‬ ‫ומשנאיך נשאו ראש‬ ‫על עמך [יע]ר[ימו ס]וד‬ ‫ויתיעצו על צפוניך‬ ‫אמרו לכו ונכח]יד[ם מגוי‬ ‫ולא יזכר שם ישראל עוד‬ ‫כי נועצו לב יח]דו[‬ ‫עליך ברית יכרתו‬ ‫אלהי אדום וישמעא[לים]‬ ‫מואב והגרים‬ ‫גבל עמון ועמלק‬ ‫פלשת עם ישבי צור‬ ‫גם אשור נלוה עמם‬ ‫היו זרוע לבני לוט סלה‬ ‫עשה להם כמדין כסיסרא‬ ‫כיבין בנחל קיׁשון‬ ‫נשמדו בעין דאר‬ ‫היו דמן לאדמה‬ ‫שיתמו נדיבמו כערב וכזאב‬ ‫וכזבח וכצלמנע [כ]ל [נ]סיכמו‬ ‫אשר אמרו נירׁשה לנו‬ ‫את נאות אלהים‬ ‫אלהים ׁשיתמו כגלגל‬ ‫כקׁש לפני רוח‬ ‫כאש תבער יער‬ ‫וכלהבה תלהט הרים‬ ‫כן תרדפם בסערך‬ ‫ובסופתך תבהלם‬ ‫מלא פניהם קלון‬

‫‪83‬‬ ‫‪1a‬‬ ‫‪1b‬‬ ‫‪2a‬‬ ‫‪2b‬‬ ‫‪3a‬‬ ‫‪3b‬‬ ‫‪4a‬‬ ‫‪4b‬‬ ‫‪5a‬‬ ‫‪5b‬‬ ‫‪6a‬‬ ‫‪6b‬‬ ‫‪7a‬‬ ‫‪7b‬‬ ‫‪8a‬‬ ‫‪8b‬‬ ‫‪9a‬‬ ‫‪9b‬‬ ‫‪10a‬‬ ‫‪10b‬‬ ‫‪11a‬‬ ‫‪11b‬‬ ‫‪12a‬‬ ‫‪12b‬‬ ‫‪13a‬‬ ‫‪13b‬‬ ‫‪14a‬‬ ‫‪14b‬‬ ‫‪15a‬‬ ‫‪15b‬‬ ‫‪16a‬‬ ‫‪16b‬‬ ‫‪17a‬‬

‫‪127‬‬

‫‪MasPsa and the Early History of the Hebrew Psalter‬‬

‫‪Aleppo Codex‬‬

‫‪MasPsa‬‬ ‫‪Column III‬‬

‫אבדּו׃‬ ‫ֽיַח ְפ ֥רּו וְ י ֹ ֵ ֽ‬ ‫י־עד ְ �ו ְ‬ ‫‪ 1‬יֵ ֖בֹׁשּו וְ ָיִב ֲה ֥לּו ֲע ֵד ֗ ַ‬ ‫י־א ָ֬תה ִש ְמָך֣‬ ‫‪� ְ 3‬ו ְֽיֵד ֗עּו ִ ֽכ ַ‬ ‫ל־ה ָ ֽא ֶרץ׃‬ ‫ל־כ ָ‬ ‫‪ֶ ֝ 13‬ע ְלי֗ ֹון ַע ָ‬

‫יְהו֣ה ְל ַב ֶ ֑דָך‬ ‫ָ‬

‫[ויבקשו שמך יהוה]‬ ‫[יבשו ויבהלו עדי עד]‬ ‫ויח[פרו ויאבדו וידעו]‬ ‫[כי אתה שמך יהוה לבדך]‬ ‫עלי[ון על כל הארץ]‬

‫‪83‬‬ ‫‪17b‬‬ ‫‪18a‬‬ ‫‪18b‬‬ ‫‪19a‬‬ ‫‪19b‬‬ ‫‪84‬‬

‫‪13‬‬ ‫‪3‬‬ ‫‪3‬‬ ‫‪4‬‬ ‫‪4‬‬ ‫‪4‬‬ ‫‪1‬‬ ‫‪4‬‬ ‫‪4‬‬ ‫‪4‬‬ ‫‪1‬‬ ‫‪3‬‬ ‫‪4‬‬ ‫‪4‬‬ ‫‪4‬‬ ‫‪1‬‬

‫למנצח [על הגתית לבני קרח‬ ‫ל־הגִ ִ ֑תית ִל ְבנֵ י־ ֥קֹ ַרח ִמזְ ֽמֹור׃‬ ‫ַל ְמנַ ֵ ֥צ ַח ַ ֽע ַ‬ ‫מזמור]‬ ‫מה ידידות מ[שכנותיך]‬ ‫הו֥ה ְצ ָב ֽאֹות׃‬ ‫נֹותיָך יְ ָ‬ ‫ַמה־יְ ִד ֥ידֹות ִמ ְש ְכ ֶ֗‬ ‫[יהוה צבאות]‬ ‫נַ ְפ ִש֮י ְל ַח ְצ ֪רֹות יְ ֫הָו֥ה נכספה וגם כ[לתה נפשי]‬ ‫ם־כ ְל ָ֨תה׀‬ ‫נִ ְכ ְס ֬ ָפה וְ גַ ָ‬ ‫[לחצרות יהוה]‬ ‫לבי ובשרי יר[ננו]‬ ‫ם־צ ּ֨פֹור‬ ‫ל־חי׃ גַ ִ‬ ‫ּוב ָש ִ ֑רי ְי ַ֝רנְ נ֗ ּו ֶ ֣אל ֵ ֽא ָ ֽ‬ ‫ִל ִ ֥בי ְ‬ ‫[אל אל חי]׃‬ ‫גם צפור מצא[ה בית]‬ ‫ר־ש ָתה‬ ‫ֲא ֶש ָ ֪‬ ‫׀ ָ ֪מ ְצ ָאה ַ֡ביִת ְּוד ֤רֹור׀ ֵ ֥קן ָל ּ֮ה‬ ‫[ודרור קן לה]‬ ‫אשר שתה א[פרחיה]‬ ‫הו֣ה ְצ ָב ֑אֹות‬ ‫חֹותיָך יְ ָ‬ ‫ת־מזְ ְב ֶ‬ ‫יה ֶ ֽא ִ ֭‬ ‫ֶא ְפ ֫ר ֹ ֶ ֥ח ָ‬ ‫[את מזבחותיך]‬ ‫יהוה צבאות [מלכי ואלהי]‬ ‫יתָך‬ ‫יֹוש ֵב֣י ֵב ֶ ֑‬ ‫ַ ֭א ְש ֵרי ְ‬ ‫אֹלהי׃‬ ‫ַ֝מ ְל ִ֗כי וֵ ָ ֽ‬ ‫[אשרי יושבי ביתך]‬ ‫עוד יהללוך ס[לה]‬ ‫ַא ְש ֵ ֣רי ָ ֭א ָדם ֽעֹוז־‬ ‫֝֗עֹוד יְ ַֽה ְל ֥לּוָך ֶ ֽס ָלה׃‬ ‫[אשרי אדם עוז לו בך]‬ ‫֥לֹו ָ ֑בְך ְ֝מ ִס ּ֗לֹות ִב ְל ָב ָ ֽבם׃ ע ְֹב ֵ ֤רי׀ ְב ֵע ֶ֣מק ַ ֭ה ָב ָכא מסלות בלבבם‬ ‫[עברי בעמק הבכא]‬ ‫מעין ישיתוהו‬ ‫מֹורה׃‬ ‫ם־ב ָר ֗כֹות יַ ְע ֶ ֥טה ֶ ֽ‬ ‫יתּוהּו גַ ְ֝‬ ‫ַמ ְע ָי�֣ן יְ ִש ֑‬ ‫[גם ברכות יעטה מורה]‬ ‫ֹלהים ְב ִצּיֽ ֹון׃ ילכו מחיל אל ח[יל]‬ ‫ל־א ִ ֣‬ ‫ל־חיִ ל יֵ ָר ֶ ֖אה ֶא ֱ‬ ‫ֵי ְ֭לכּו ֵמ ַ ֣חיִ ל ֶא ָ ֑‬ ‫[יראה אל אלהים בציון]‬ ‫יהוה אלהים צב[אות]‬ ‫ֹלהים ְ ֭צ ָבאֹות ִש ְמ ָע֣ה ְת ִפ ָל ִ ֑תי‬ ‫יְ ֘הָו֤ה ֱא ִ ֣‬ ‫[שמעה תפלתי]‬ ‫ה[אזינה] אלהי יע[קב סלה]‬ ‫ָ ֭מגִ נֵ נּו ְר ֵ ֣אה‬ ‫ֹלהי יַ ֲע ֣קֹב ֶ ֽס ָלה׃‬ ‫ַה ֲא ִ֨זינָ ה ֱא ֵ ֖‬ ‫[מגננו ראה אלהים]‬ ‫ִ ֤כי ֽטֹוב־י֥ ֹום‬ ‫יחָך׃‬ ‫ֹלהים וְ ַ֝ה ֵ֗בט ְפ ֵנ֣י ְמ ִש ֶ ֽ‬ ‫ֱא ִ ֑‬ ‫וה[בט פני משיחך]‬ ‫[כי טוב יום בחצריך]‬ ‫ּתֹופף‬ ‫ָב ַ֗ח ְר ִתי ִ ֭ה ְס ֵ‬ ‫ַב ֲח ֵצ ֶ ֗ריָך ֵ֫מ ָ ֥א ֶלף‬ ‫[מאלף בחרתי]‬ ‫[הסתופף בבית אלהי]‬ ‫י־ר ַשע׃‬ ‫ִ֝מ ּ֗דּור ְב ָא ֳה ֵל ֶ ֽ‬ ‫ֹלהי‬ ‫ְב ֵב֣ית ֱא ַ ֑‬

‫‪1‬‬ ‫‪2a‬‬ ‫‪2b‬‬ ‫‪3a‬‬ ‫‪3b‬‬ ‫‪3c‬‬ ‫‪3d‬‬ ‫‪4a‬‬ ‫‪4b‬‬ ‫‪4c‬‬ ‫‪4d‬‬ ‫‪4e‬‬ ‫‪5a‬‬ ‫‪5b‬‬ ‫‪6a‬‬ ‫‪6b‬‬ ‫‪7a‬‬ ‫‪7b‬‬ ‫‪7c‬‬ ‫‪8a‬‬ ‫‪8b‬‬ ‫‪9a‬‬ ‫‪9b‬‬ ‫‪9c‬‬ ‫‪10a‬‬ ‫‪10b‬‬ ‫‪11a‬‬ ‫‪11b‬‬ ‫‪11c‬‬

128 11d 12a 12b 12c 13a 13b

Peter J. Gentry and John D. Meade

]‫מד[ו]ר באה[לי רשע‬ ]‫הו֑ה [כי שמש ומגן יהוה אלהים‬ ָ ְ‫יִתן י‬ ֣ ֵ ‫הו֪ה ֱא ֹ֫ל ִ ֥הים ֵ ֣חן ְו ָ֭כבֹוד‬ ָ ְ‫ּומגֵ ֮ן י‬ ָ ‫ ִ ֤כי ֶ֨ש ֶמׁש׀‬2 ]‫חן וכבוד יתן יה[וה‬ ]‫[לא ימנע טוב להלכים בתמים‬ ‫ַ ֽלה ְֹל ִ ֥כים ְב ָת ִ ֽמים׃‬ ‫ע־טֹוב‬ ֗֝ ַ‫יִמנ‬ ְ ‫ ֥ל ֹא‬3 ‫יהוה צבאות‬ ‫ַ ֽא ְש ֵ ֥רי ָ֝א ָ ֗דם ב ֵ ֹ֥ט ַח ָ ֽבְך׃‬ ‫הו֥ה ְצ ָב ֑אֹות‬ ָ ְ‫ י‬1 ]‫[אשרי אדם בטח בך‬

85 1

]‫למנצח לבני קרח מזמ[ור‬

2a 2b 3a 3b 4a 4b 5a 5b 6a 6b

‫רצית יהוה ארצך‬ ]‫[שבת שבות יעקב‬ ‫נשאת עון עמך‬ ]‫[כסית כל חטאתם סלה‬ ‫אספת כל עברתך‬ ]‫[השיבות מחרון אפך‬ ‫שובנו אלהי ישענו‬ ]‫[והפר כעסך עמנו‬ ‫הלעולם תאנף בנו‬ ]‫[תמשך אפך לדר ודר‬

‫ ַל ְמנַ ֵ֬צ ַח׀ ִל ְבנֵ י־ ֬קֹ ַרח ִמזְ ֽמֹור׃‬13 linea vacat ‫הו֣ה ַא ְר ֶצָ֑ך ַ֝ ֗ש ְב ָת ְש ִ ֯ ֣בות יַ ֲע ֽקֹב׃‬ ָ ְ‫ית י‬ ָ ‫ ָר ִ ֣צ‬1 ‫אתם ֶ ֽס ָלה׃‬ ֣ ָ ‫ל־ח ָט‬ ַ ‫ית ָכ‬ ָ ‫ִכ ִ ֖ס‬

‫את ֲעֹו֣ ן ַע ֶ ֑מָך‬ ָ ‫ ָנ ָ֭ש‬1

‫ֹות ֵמ ֲח ֥רֹון ַא ֶ ֽפָך׃‬ ָ ‫ֱ֝ה ִש ֗יב‬

‫ל־ע ְב ָר ֶ ֑תָך‬ ֶ ‫ ָא ַ ֥ס ְפ ָת ָכ‬1

‫וְ ָה ֵ ֖פר ַ ֽכ ַע ְסָך֣ ִע ָ ֽמנּו׃‬

‫ֹלהי יִ ְש ֵע֑נּו‬ ֣ ֵ ‫ּובנּו ֱא‬ ֵ ‫ ׁ֭ש‬1

‫ִת ְמ ֥ש ְֹך ַ֝א ְפ ָ֗ך ְל ֣ד ֹר וָ ֽד ֹר׃‬

‫ף־בנּו‬ ֑ ָ ַ‫עֹולם ֶת ֱאנ‬ ֥ ָ ‫ ַה ְל‬1

Most lines are classified as either category 1 or 4 defined by Sanders as follows:

Category 1 The lines of Category 1 show one blank space. The last word before the blank space (on the right) is the last word of a colon. The first word after the blank space (on the left) is the first word of the following colon. The last word on the line is the last word of that colon and touches the left margin. The first word on the line (on the right) is commonly the first word of the first colon, but in some cases one or more words of the first colon have been written on the preceding line.21 or

zzzzzzzzzzzzzzz yyyyyyyyyyyyyyy (yyyyy) zzzzzzzzzzzzzzz yyyyyyyyyyyyyyy

21  P. Sanders, “The Colometric Layout”, 231–2.

MasPsa and the Early History of the Hebrew Psalter

129

Category 4 The lines of Category 4 show one blank space. The word immediately preceding the blank space (on the right) is the last word of a colon as delimited by the accents. The word immediately following the blank space (on the left) is the first word of the following colon. However, the last word on the line bears a conjunctive or a minor disjunctive accent that does not suggest the end of a colon. The first word on the line (on the right) is commonly the first word of the first colon, but one or more words of this colon may have been written on the preceding line.22

or

4. zzzzzzzz yyyyyyyyyyyyyyyyyyyyyy (zzzzzzzz) (yyyyy) zzzzzzzz yyyyyyyyyyyyyyyyyyyyyy (zzzzzzzz)

In Category 13, lines are left blank on the left. Two situations are outside the categories given by Sanders for his analysis of Psalms 1–14. In Ps 81:8a, 8b, 8c, the scribe appears to have arranged bits of three cola across a couple of lines as follows: yyyyyyyyyyy zzzzzzzzzzzz

xxxxxxxxxx zzzzzyyyyy

To illustrate the approach of the scribe of the Aleppo Codex, the last word of 81:6b in MasPsa is at the beginning of the next line in A. The stichometry in A and MasPsa is not synchronised until 7b. The last word of 8b is at the beginning of the next line in A. The bound phrase in 8c is broken by a major space and split in two in A to re-synchronise (read across line 8c for ‫יבה‬ ֣ ָ ‫ְמ ִר‬ ‫ ֵ ֖מי‬in A). Similar examples exist, e.g. in Ps 81:11a and 13a. Finally, note that the scribe of the Aleppo Codex has parts of 6b–6c–7a all on one line with no spaces. It is not possible to compare the two texts in every line since, e.g. 83:18b involves reconstruction in MasPsa and we do not know if the word ‫ וידעו‬belonged to 18b or 19a as in A. The differences between MasPsa and A are due not only to the narrow page of the codex format as Sanders showed, but also to different approaches to arranging tricola and places where there is simply a real difference in stichometry. These data show that in the intervening centuries between MasPsa and A, with the move from scroll to codex and the vowels and accents now represented in writing, the presentation as two hemistichs divided by a space on one line is only a 22  Ibid., 234–6.

130

Peter J. Gentry and John D. Meade

memory. Note particularly 81:6a where the scribe of A puts a space between the bound and free members of a bound phrase. Unlike MasPsa, where spacing is the key to dividing the text into sense units, the pausal forms and accents are the key terminal markers in A and the spacing is only a memory of a past scribal technique. Unfortunately only a handful of Hebrew biblical manuscripts are known between the III–VII centuries c.e. Two separated fragments of Exodus belong to the same manuscript, MS London-Ashkar. These manuscripts are not listed by Tov in the most recent revision of Textual Criticism of the Hebrew Bible in 2012. Paul Sanders in a recent article has shown that the Ashkar-Gilson Hebrew Manuscript at Duke University was part of a model scroll that had an important influence on later manuscripts like the Aleppo and Leningrad Codices.23 Although these are not manuscripts of the Psalms, they may provide a plausible bridge between MasPsa and the early Masoretic manuscripts. More research is needed to cast light on the evolution from the first century b.c.e. to the tenth century c.e. in light of these texts.

Hebrew Manuscripts from III–VII Centuries c.e. Genesis Cambridge T-S NS 3.21 and 4.3 Exodus Duke University, Ashkar-Gilson Hebrew Manuscript #2 + MS London, formerly of Jew’s College Exodus Oxford Bodleian Lib. Ms. Heb. D.89 (P) i Leviticus EGLev Numbers Berlin, Staatliche Museum, P 10598 Kings Oxford, Ashmolean Museum, Pap. 47–48 Job Oxford, Ashmolean Museum, Ant. Pap. 49–50

We can now focus attention on the question of other texts that show the special layout of MasPsa and also ask the question: how far back can this tradition be traced? An essay by Emanuel Tov published in 1996 was the first attention given to texts from the Judean Desert with a special layout for poetical units.24 Tov

23  P. Sanders, “The Ashkar-Gilson Manuscript: Remnant of a Proto-Masoretic Model Scroll of the Torah”, Journal of Hebrew Scriptures 14 Article 7 (2014) 1–25. http://www.jhsonline.org/ Articles/article_201.pdf accessed on December 14, 2014. 24  E. Tov, “Special Layout of Poetical Units in the Texts from the Judean Desert”, in J. Dyk (ed.), Give Ear to My Words: Psalms and other Poetry in and around the Hebrew Bible: Essays in Honour of Professor N. A. van Uchelen (Amsterdam: Societas Hebraica Amstelodamensis, 1996) 115–28.

MasPsa and the Early History of the Hebrew Psalter

131

improved upon his own research in his magisterial volume on scribal practices appearing in 2004.25

MSS With 2 Hemistichs Per Line and a Space Between Hemistichs: 4QpaleoDeutr (not dated; Deut 32, reconstructed) 1QPsa (not dated; Ps 119 only, other Psalms in prose, reconstructed) 4QPsc (50–68 c.e.; Pss 16–53) 8QPs (1–100 c.e.; only Pss 17–18 preserved) 11QPsb (30–1 b.c.e.; Ps 119 only, other Psalms in prose) 5/6HevPs (50–68 c.e.; Pss 7–16, 18, 22–25, 29–31) MasPsa (30–1 b.c.e.; fragments of Pss 81–85) 4QProva (50 b.c.e.–30 c.e.; Prov 1–2) 2QSir (50–1 b.c.e.; Sir 6; Reconstructed) MasSir (10 b.c.e.–50 c.e.; Sir 39–44)

Three of the ten texts are doubtful as they are fragmentary and involve reconstruction. Tov lists twenty other texts laid out in colometrically, but either they have no spaces between hemistichs or display a different approach in layout. He also lists thirty-one texts that could display a stichometric layout for poetry but do not. We have checked all of the manuscripts in the photographs. Tov rightly notes that MasPsa is a deluxe manuscript among all of those exhibiting stichometry. The one manuscript at Qumran which most closely resembles MasPsa is 4QPsb although it has only one hemistich per line in the column of text and it only has 16–17 lines per column compared to 29–30 lines in MasPsa. The stichometry and text of 4QPsb are not as close to MT as MasPsa and so belongs to the “second circle of texts”, i.e. biblical texts circulating apart from the setting of the Temple. How far back can the tradition exhibited by MasPsa be traced? Perhaps related to this question is the provenance of MasPsa. Division of poetic lines dates to at least the first millennium b.c.e. As a number of scholars have noted, there is a funerary inscription in Aramaic from the fifth century b.c.e. where the genre is poetry (KAI 269). The inscription has four lines corresponding to the four couplets of poetry, but there are no spaces to mark the hemistichs.26 25  E. Tov, Scribal Practices and Approaches Reflected in the Texts Found in the Judean Desert (Leiden: Brill, 2004), 166–78. 26  J.C.L. Gibson, Textbook of Syrian Semitic Inscriptions. Volume II. Aramaic Inscriptions (Oxford: Clarendon, 1975), 120–1 and Figure 13. For an actual photograph, see H. Donner/ W. Röllig, Kanaanäische und Aramäische Inschriften (3 volumes; Wiesbaden: Otto Harrassowitz, 1962–1964), pl. 34.

132

Peter J. Gentry and John D. Meade

There is also interesting evidence from the Septuagint. The John Rylands Papyrus 458 from the second century b.c.e. preserves fragments of Deut 22–28. Spaces between words function as terminal markers equivalent to pausal forms and accents in MT, although there are a few differences from MT.27 Similar evidence is provided by the spacing in the Fouad Papyrus from approximately 50 c.e., which presents Deut 32 in one hemistich per line.28 1QDeutb is an example in Hebrew from Qumran where the spacing corresponds to the vowels and accents added later in MT.29 The great uncials of the Septuagint of the fourth century offer surprising data. Before the development of the codex, the books of the Bible were copied first one book per scroll and later in blocks where several books of the same genre were transmitted in a single scroll or smaller codex. The evidence for this is supplied in Rahlfs’ Verzeichnis where he lists all the manuscripts in the blocks in which they were transmitted. (This section is entitled “Übersicht über das handschriftliche Material für die einzelnen Teile des A.T.”).30 There are seven blocks as follows: 1) Octateuchus 2) Reg., Par., Esdr. 3) Est., Idt., Tob. 4) Mac. I–IV 5) Ps.Od. 6) Libri sapientiales (Prov. Eccl. Cant. Iob. Sap. Sir. Ps. Sal.) 7) XVI prophetae

In the fourth century c.e. when scribes attempted to produce a pandect Bible, they had to assemble a group of seven codices or rolls, roughly speaking. The colophons present early evidence for this. The colophon to Esther in Codex Sinaiticus speaks of a book that contained Kingdoms through Esther, i.e. the historical literature. Here Esther was part of the second block and not part of the third with Judith and Tobit. The columns and layout in Codex Sinaiticus further support this. The historical books and prophets before the poetical books are written in 27  E.J. Revell, “The Oldest Evidence for the Hebrew Accent System”, Bulletin of the John Rylands Library 54 (1971) 214–22. See also E.J. Revell, “The Occurrence of Pausal Forms”, Journal of Semitic Studies 57/2 (2012) 213–30; idem, “Minor Pausal Forms and their Function”, Vetus Testamentum 65/3 (2015) 457–65; idem, “Pausal Forms Marked with Conjunctive Accents”, Journal of Semitic Studies 61/1 (2016) 67–84. 28 See P. Sanders, The Provenance of Deuteronomy 32 (Oudtestamentische Studiën 37; Leiden: Brill, 1996), 121–2 and passim. 29  Ibid, 214. 30  A. Rahlfs, Verzeichnis der griechischen Handschriften des Alten Testament (Mitteilungen des Septuaginta-Unternehmens der Königlichen Gesellschaft der Wissenschaften 2; Berlin: Weidmannsche Buchhandlung, 1914), 373–439.

MasPsa and the Early History of the Hebrew Psalter

133

four columns as are the gospels which follow. Nonetheless the poetical books are written in two columns. This is due to the fact that it was normal practice for scribes to follow the physical layout of the source text in spite of the fact that it was common practice to reproduce poetic text in longer lines than the short columns of prose text. We compared the stichometry of MasPsa with that of Codex Vaticanus and Codex Sinaiticus. A chart details the comparison hemistich for hemistich. The first chart gives an illustration using Psalm 81.

Illustration 81

MasPsa

Codex Sinaiticus 

Column I 1a

]‫ [למנצח על הגתית לאסף‬ειϲ το τελοϲ ϋπερ των ληνων ψαλμοϲ τω

2a ]‫[הרנינו לאלהים עוזנו‬ 2b ‫[הריעו לאלהי י]עקב‬ 3a ]‫[שאו־זמרה ותנו תף‬ 3b ‫[כנור נעים] עם [נ]בל‬ 4a ]‫[תקעו בחדש שופר‬ 4b ‫[בכסה ליום] חגנו‬ 5a ]‫[כי חק לישראל הוא‬ 5b ‫[משפט לא]להי יעקב‬ 6a ]‫[עדות ביהוסף שמו‬ 6b ‫[בצאתו על א]רץ מצרים‬ 6c ]‫[שפת לא ידעתי אשמע‬ 7a ‫[הסירותי] מסבל שכמו‬ 7b ]‫[כפיו מדוד תעברנה‬ 8a ‫[בצ]רה ק[ר]את וא[ח]לצכה‬ 8b ]‫[אענך בסתר רעם‬ 8c ‫[אבח]נך על מי מריבה סלה‬ 9a 9b 10a

]‫[שמע עמי ואעידה בך‬ ‫[י]שראל אם תשמע קולי‬ ]‫[לא יהיה בך אל זר‬

10b 11a

‫[ולא] תשתחוה לאל נכר‬ ]‫[אנכי יהוה אלהיך‬

αϲαφʼ αγαλλιαϲθε τω θω τω βοηθω ημων αλαλαξατε τω θω ϊακωβ λαβετε ψαλμον και δοτε τυμπανον ψαλτηριον τερπνον μετα κιθαραϲ ϲαλπιϲατε εν νεομηνια ϲαλπιγγι εν ευϲημω ημερα εορτηϲ ημων οτι προϲταγμα τω ϊϲραηλ εϲτιν και κριμα τω θω ϊακωβ μαρτυριον εν τω ϊωϲηφ εθετο αυτον εν τω εξελθεν αυτον εκ γηϲ αιγυπτου γλωϲϲαν ην ουκ εγνω ηκουϲεν απεϲτηϲεν απο αρϲεων τον νωτον αυτου αι χειρεϲ αυτου εν τω κοφινω εδουλευϲαν εν θλιψει επεκαλεϲω μαι και ερυϲαμην ϲε επηκουϲα ϲου εν αποκρυφω καταιγιδοϲ εδοκιμαϲα ϲε επι ϋδατοϲ αντιλογιαϲ διαψαλμα ακουϲον λαοϲ μου και λαληϲω ϲοι ϊϲραηλ και διαμαρτυρομαι ϲοι εαν ακουϲηϲ μου ουκ εϲται εν ϲοι θϲ προϲφατοϲ ουδε προϲκυνηϲειϲ θω αλλοτριω εγω γαρ ειμι κϲ ο θϲ ϲου ο αναγαγων

134 11b 11c 12a 12b 13a

13b 14a 14b

Peter J. Gentry and John D. Meade

‫[המע]לך מארץ מצרים‬ ]‫[הרחב פיך ואמלאהו‬ ‫[ולא] שמע עמי לקולי‬ ]‫[וישראל לא אבה לי‬ ‫[ואש]לחהו בשרירות לבם‬

ϲε εκ γηϲ αιγυπτου 2H » 1 πλατυνον το ϲτομα ϲου και πληρωϲω αυτο και ουκ ηκουϲεν ο λαοϲ μου τηϲ φωνηϲ μου και ιϲραηλ ου προϲεϲχεν μοι και εξαπεϲτεϲτειλα αυτουϲ κατα τα επιτηδευματα των καρδιων αυτων ]‫ [ילכו במועצותיהם‬πορευϲονται εν τοιϲ επιτηδευμαϲιν αυτων ‫ [לו] עמי שמע לי‬ει ο λαοϲ ┬ ηκουϲεν μου ]‫ [ישראל בדרכי יהלכו‬ϊϲραηλ ταιϲ οδοιϲ μου ει επορευθη

Detailed Comparison of Stichometry in MasPsa with Codices Vaticanus and Sinaiticus

81

MasPsa

Cod. Sinaiticus 

Cod. Vaticanus 

Column I 1a

]‫[למנצח על הגתית לאסף‬

2a

linea vacat

]‫[הרנינו לאלהים עוזנו‬

H1: H2: H1:

=S =S =S

=B =B =B

2b 3a 3b 4a 4b 5a 5b 6a 6b 6c 7a 7b 8a 8b 8c 9a

‫[הריעו לאלהי י]עקב‬ ]‫[שאו־זמרה ותנו תף‬ ‫[כנור נעים] עם [נ]בל‬ ]‫[תקעו בחדש שופר‬ ‫[בכסה ליום] חגנו‬ ]‫[כי חק לישראל הוא‬ ‫[משפט לא]להי יעקב‬ ]‫[עדות ביהוסף שמו‬ ‫[בצאתו על א]רץ מצרים‬ ]‫[שפת לא ידעתי אשמע‬ ‫[הסירותי] מסבל שכמו‬ ]‫[כפיו מדוד תעברנה‬ ‫[בצ]רה ק[ר]את וא[ח]לצכה‬ ]‫[אענך בסתר רעם‬ ‫[אבח]נך על מי מריבה סלה‬ ]‫[שמע עמי ואעידה בך‬

H2: H1: H2: H1: H2: H1: H2: H1: H2: H1: H2: H1: H2: H1: H2: H1:

=S =S =S =S =S =S =S =S =S =S =S =S =S =S = S (‫ סלה‬l. separatum) =S

=B =B =B =B =B =B =B =B =B =B =B =B =B =B B=S 9a + 9b l. unum

‫‪135‬‬

‫‪MasPsa and the Early History of the Hebrew Psalter‬‬

‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪= B31‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬

‫‪9b + 10a linea unum‬‬

‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬

‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬

‫[י]שראל אם תשמע קולי‬ ‫[לא יהיה בך אל זר]‬ ‫[ולא] תשתחוה לאל נכר‬ ‫[אנכי יהוה אלהיך]‬ ‫[המע]לך מארץ מצרים‬ ‫[הרחב פיך ואמלאהו]‬ ‫[ולא] שמע עמי לקולי‬ ‫[וישראל לא אבה לי]‬ ‫[ואש]לחהו בשרירות לבם‬ ‫[ילכו במועצותיהם]‬ ‫[לו] עמי שמע לי‬ ‫[ישראל בדרכי יהלכו]‬ ‫[כמעט] אויביהם אכניע‬ ‫[ועל צריהם אשיב ידי]‬ ‫[משנא]י יהוה יכחשו לו‬

‫‪9b‬‬ ‫‪10a‬‬ ‫‪10b‬‬ ‫‪11a‬‬ ‫‪11b‬‬ ‫‪11c‬‬ ‫‪12a‬‬ ‫‪12b‬‬ ‫‪13a‬‬ ‫‪13b‬‬ ‫‪14a‬‬ ‫‪14b‬‬ ‫‪15a‬‬ ‫‪15b‬‬ ‫‪16a‬‬

‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪non vacat‬‬

‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪non vacat‬‬

‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬

‫[ויהי עתם] לעולם‬ ‫ויאכילהו מחלב חטה‬ ‫[ומצור דב]ש אשביעך‬

‫‪16b‬‬ ‫‪17a‬‬ ‫‪17b‬‬

‫‪=B‬‬

‫‪=S‬‬ ‫‪non vacat‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫)‪ l. separatum‬סלה( ‪= S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪6a + 6b linea unum‬‬

‫‪=B‬‬

‫‪=S‬‬

‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬

‫‪=S‬‬ ‫‪11a + 11b linea unum‬‬

‫‪Column II‬‬

‫‪linea vacat‬‬

‫‪Column II Continued‬‬

‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪B=S‬‬

‫מזמו[ר] לאסף‬

‫‪linea vacat‬‬

‫אלהים נצב בעדת אל‬ ‫בקרב אלהים יׁשפט‬ ‫עד מתי תׁשפטו עול‬ ‫ופני רׁשעים תׂשאו סלה‬ ‫ׁשפטו דל ויתום‬ ‫עני ורׁש הצדי]קו[‬ ‫פלטו דל ואביון‬ ‫מיד רׁשעים הצילו‬ ‫לא ידעו ולא יבינו‬ ‫בחׁשכה יתהלכו‬ ‫ימוטו כל מוסדי ארץ‬ ‫אני אמרתי אלהים אתם‬ ‫ובני עליון כלכם‬ ‫אכן כאדם תמותון‬

‫‪82‬‬ ‫‪1a‬‬ ‫‪1b‬‬ ‫‪1c‬‬ ‫‪2a‬‬ ‫‪2b‬‬ ‫‪3a‬‬ ‫‪3b‬‬ ‫‪4a‬‬ ‫‪4b‬‬ ‫‪5a‬‬ ‫‪5b‬‬ ‫‪5c‬‬ ‫‪6a‬‬ ‫‪6b‬‬ ‫‪7a‬‬

‫‪ at the beginning of 14b is at the end of 14a; nonetheless there is a colon‬ישראל ‪31  The word‬‬ ‫‪in B before it and a colon at the end of 14b to mark the correct stichometry.‬‬

‫‪Peter J. Gentry and John D. Meade‬‬

‫‪136‬‬

‫וכאחד הׂשרים תפלו‬ ‫קומה אלהים ׁשפטה הארץ‬ ‫כי אתה תנחל בכל הגוים‬

‫‪7b‬‬ ‫‪8a‬‬ ‫‪8b‬‬

‫‪Column II Continued‬‬

‫‪83‬‬

‫ׁשיר מזמור לאסף‬ ‫אלהים אל דמי לך‬ ‫אל תחרׁש ואל תׁשקט אל‬ ‫כי הנה אויביך יהמיון‬ ‫ומׂשנאיך נׂשאו ראׁש‬ ‫על עמך [יע]ר[ימו ס]וד‬ ‫ויתיעצו על צפוניך‬ ‫אמרו לכו ונכח]יד[ם מגוי‬ ‫ולא יזכר ׁשם יׂשראל עוד‬ ‫כי נועצו לב יח]דו[‬ ‫עליך ברית יכרתו‬ ‫אלהי אדום וישמעא[לים]‬ ‫מואב והגרים‬ ‫גבל עמון ועמלק‬ ‫פלׁשת עם יׁשבי צור‬ ‫גם אׁשור נלוה עמם‬ ‫היו זרוע לבני לוט סלה‬

‫‪1a‬‬ ‫‪2a‬‬ ‫‪2b‬‬ ‫‪3a‬‬ ‫‪3b‬‬ ‫‪4a‬‬ ‫‪4b‬‬ ‫‪5a‬‬ ‫‪5b‬‬ ‫‪6a‬‬ ‫‪6b‬‬ ‫‪7a‬‬ ‫‪7b‬‬ ‫‪8a‬‬ ‫‪8b‬‬ ‫‪9a‬‬ ‫‪9b‬‬ ‫‪10a‬‬ ‫‪10b‬‬ ‫‪11a‬‬ ‫‪11b‬‬ ‫‪12a‬‬ ‫‪12b‬‬

‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬

‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬

‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬

‫‪non vacat‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬

‫‪non vacat‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬

‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬

‫‪linea vacat‬‬

‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪B=S‬‬

‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪12a + 12b linea unum‬‬

‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬

‫‪B=S‬‬

‫‪13a + 13b linea unum‬‬

‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬

‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬

‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬

‫עׂשה להם כמדין כסיסרא‬ ‫כיבין בנחל קיׁשון‬ ‫נׁשמדו בעין דאר‬ ‫היו דמן לאדמה ׁשיתמו‬ ‫נדיבמו כערב וכזאב‬ ‫וכזבח וכצלמנע [כ]ל‬ ‫[נ]סיכמו‬ ‫אׁשר אמרו נירׁשה לנו‬ ‫את נאות אלהים‬ ‫אלהים ׁשיתמו כגלגל‬ ‫כקׁש לפני רוח‬ ‫כאׁש תבער יער‬ ‫וכלהבה תלהט הרים‬ ‫כן תרדפם בסערך‬ ‫ובסופתך תבהלם‬ ‫מלא פניהם קלון‬

‫‪13a‬‬ ‫‪13b‬‬ ‫‪14a‬‬ ‫‪14b‬‬ ‫‪15a‬‬ ‫‪15b‬‬ ‫‪16a‬‬ ‫‪16b‬‬ ‫‪17a‬‬

‫‪137‬‬

‫‪MasPsa and the Early History of the Hebrew Psalter‬‬

‫‪Column III‬‬ ‫‪=S‬‬ ‫‪=S‬‬

‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬

‫‪=B‬‬ ‫‪=B‬‬ ‫‪B=S‬‬ ‫‪B=S‬‬ ‫‪B=S‬‬ ‫‪non vacat‬‬ ‫‪=B‬‬ ‫‪non vacat‬‬ ‫‪B=S‬‬

‫‪non vacat‬‬ ‫‪2a + 2b linea unum‬‬

‫‪B=S‬‬

‫‪3a + 3b linea unum‬‬

‫‪B=S‬‬

‫‪3c + 3d linea unum‬‬

‫‪B=S‬‬

‫‪=S‬‬ ‫‪4b + 4c linea unum‬‬

‫‪B=S‬‬

‫‪4d + 4e linea unum‬‬

‫‪=B‬‬ ‫‪B=S‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪B=S‬‬

‫‪=S‬‬ ‫)‪ l. separatum‬סלה( ‪= S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪7a + 7b linea unum‬‬

‫‪=B‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪B=S‬‬

‫‪=S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪9a + 9b linea unum‬‬

‫‪B=S‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪B=S‬‬ ‫‪B=S‬‬ ‫‪B=S‬‬ ‫‪=B‬‬ ‫‪=B‬‬ ‫‪B=S‬‬

‫)‪ l. separatum‬סלה( ‪= S‬‬ ‫‪=S‬‬ ‫‪=S‬‬ ‫‪ trahit ad 11a‬מאלף‬ ‫‪≠S‬‬ ‫‪ trahit ad 11c‬בחרתי‬ ‫‪=S‬‬ ‫‪ trahit ad 12b‬אלהים‬ ‫‪ trahit ad 12c‬יה[וה‬

‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬ ‫‪H2:‬‬ ‫‪H1:‬‬

‫‪ trahit ad 19a‬וידעו‬ ‫‪ trahit ad 19b‬לבדך‬ ‫‪non vacat‬‬

‫[ויבקשו שמך יהוה]‬ ‫[יבשו ויבהלו עדי עד]‬ ‫ויח[פרו ויאבדו וידעו]‬ ‫[כי אתה שמך יהוה לבדך]‬ ‫עלי[ון על כל הארץ]‬ ‫‪linea vacat‬‬

‫למנצח [על הגתית לבני קרח‬ ‫מזמור]‬ ‫‪linea vacat‬‬

‫מה ידידות מ[שכנותיך]‬ ‫[יהוה צבאות]‬ ‫נכספה וגם כ[לתה נפשי]‬ ‫[לחצרות יהוה]‬ ‫לבי ובשרי יר[ננו]‬ ‫[אל אל חי]‬ ‫גם צפור מצא[ה בית]‬ ‫[ודרור קן לה]‬ ‫אשר שתה א[פרחיה]‬ ‫[את מזבחותיך]‬ ‫יהוה צבאות [מלכי ואלהי]‬ ‫[אשרי יושבי ביתך]‬ ‫עוד יהללוך ס[לה]‬ ‫[אשרי אדם עוז לו בך]‬ ‫מסלות בלבבם‬ ‫[עברי בעמק הבכא]‬ ‫מעין ישיתוהו‬ ‫[גם ברכות יעטה מורה]‬ ‫ילכו מחיל אל ח[יל]‬ ‫[יראה אל אלהים בציון]‬ ‫יהוה אלהים צב[אות]‬ ‫[שמעה תפלתי]‬ ‫ה[אזינה] אלהי יע[קב סלה]‬ ‫[מגננו ראה אלהים]‬ ‫וה[בט פני משיחך]‬ ‫[כי טוב יום בחצריך]‬ ‫[מאלף בחרתי]‬ ‫[הסתופף בבית אלהי]‬ ‫מד[ו]ר באה[לי רשע]‬ ‫[כי שמש ומגן יהוה אלהים]‬ ‫חן וכבוד יתן יה[וה]‬

‫‪83‬‬ ‫‪17b‬‬ ‫‪18a‬‬ ‫‪18b‬‬ ‫‪19a‬‬ ‫‪19b‬‬ ‫‪84‬‬ ‫‪1a‬‬

‫‪2a‬‬ ‫‪2b‬‬ ‫‪3a‬‬ ‫‪3b‬‬ ‫‪3c‬‬ ‫‪3d‬‬ ‫‪4a‬‬ ‫‪4b‬‬ ‫‪4c‬‬ ‫‪4d‬‬ ‫‪4e‬‬ ‫‪5a‬‬ ‫‪5b‬‬ ‫‪6a‬‬ ‫‪6b‬‬ ‫‪7a‬‬ ‫‪7b‬‬ ‫‪7c‬‬ ‫‪8a‬‬ ‫‪8b‬‬ ‫‪9a‬‬ ‫‪9b‬‬ ‫‪9c‬‬ ‫‪10a‬‬ ‫‪10b‬‬ ‫‪11a‬‬ ‫‪11b‬‬ ‫‪11c‬‬ ‫‪11d‬‬ ‫‪12a‬‬ ‫‪12b‬‬

138

Peter J. Gentry and John D. Meade

]‫[לא ימנע טוב להלכים‬ ]‫[בתמים‬ ‫יהוה צבאות‬ ]‫[אשרי אדם בטח בך‬

12c 12c 13a 13b 85 1

]‫למנצח לבני קרח מזמ[ור‬

2a 2b 3a 3b 4a 4b 5a 5b 6a 6b

‫רצית יהוה ארצך‬ ]‫[שבת שבות יעקב‬ ‫נשאת עון עמך‬ ]‫[כסית כל חטאתם סלה‬ ‫אספת כל עברתך‬ ]‫[השיבות מחרון אפך‬ ‫שובנו אלהי ישענו‬ ]‫[והפר כעסך עמנו‬ ‫הלעולם תאנף בנו‬ ]‫[תמשך אפך לדר ודר‬

linea vacat: H1 + H2

H2: H1: H2: H1: H2: H1: H2: H1: H2: H1: H2: H1: H2: H1: H2:

=S 13a + 13b linea unum

=B B=S

non vacat =S non vacat =S =S =S = S (‫ סלה‬l. separatum) =S =S =S =S =S =S

non vacat =B non vacat =B =B =B B=S =B =B =B =B =B =B

The result is interesting. Of approximately 145 hemistichs, the stichometry of B differs in 8 and combines 26 lines into 13 while that of S differs in 9 and combines 26 lines into 13. B has διάψαλμα on the same line in one instance as in the case of MasPsa. Everywhere else in both B and S διάψαλμα is given a line by itself due to the fact that there is not sufficient room to have it on the line to which it belongs. In sum, both B and S agree with the stichometry of MasPsa in over 75% of the lines, and B is a bit closer than S. Even the combined lines match the ends of lines in MasPsa so that the agreement could be claimed to be 99%. And although Talmon argues strongly for the stichometry of MT which places Sisera with 83:10b, he does not note that both B and S place it with 10a as in MasPsa.32 Thus five hundred years after the original translation, B and S have largely preserved the terminal markers of a Hebrew parent text like MasPsa. The evidence suggests that the original translation into Greek followed the stichometric layout in poetic texts and even in the earliest texts had spaces corresponding to the reading tradition, and that this tradition was preserved for some time. The correspondence with Codex Alexandrinus, a century later, is not as strong.

32  Talmon, Masada VI, 88–9.

MasPsa and the Early History of the Hebrew Psalter

139

Conclusions The connection between MasPsa and MT-A is strong. (1) The wording of the text between the two manuscripts was nearly identical with few significant variants. (2) The breaks between couplets (i.e. the space between the A and B cola) and the ends of the couplets were shown to agree closely with the Masoretic terminal markers (pausal forms and accents). Although the Masoretes employed a system of spaces within each column of text, development had occurred between the time of MasPsa and the Masoretes to the extent that the spaces in the Aleppo codex appear only to be a memory of the earlier stichometry. Unfortunately there are no known Hebrew Psalms manuscripts from the III–VII centuries which might illuminate the exact points of development. (3) The stichometries of MasPsa and the codices of Sinaiticus and Vaticanus agreed in the couplet and line breaks to an even higher degree than MT showing significant continuity from MasPsa to at least the fourth century c.e. This demonst rat es t hat MasPsa represents a much older tradition than just the end of the first century b.c.e. or beginning of t he first century c.e. Emanuel Tov has already identified MasPsa as a witness to the proto-MT on the basis of its text and physical layout and has concluded that it was part of the inner circle of texts, that is, those texts which are believed to be connected to the Temple.33 We have shown that MasPsa not only has a near identical text to the Aleppo codex but also has divisions according to terminal markers in the later manuscript, further demonstrating that MasPsa and Aleppo can be traced either to a common source or that Aleppo came from an intermediate manuscript which was both derived and developed from MasPsa. In 2010 Eugene Ulrich stated, “despite suggestions to the contrary, the future still awaits demonstration that the texts preserved in the medieval MT transmit the texts guarded by the priests in the Jerusalem temple as opposed to other popular or ‘vulgar’ texts that were less well preserved by less well qualified people. Nor has a line of succession—from temple priests to Pharisees to rabbis—been

33  E. Tov, “The Text of the Hebrew/Aramaic and Greek Bible Used in the Ancient Synagogues”, in Hebrew Bible, Greek Bible, and Qumran (Tübingen: Mohr Siebeck, 2008) 171–88, on p. 177. For ancient testimonia to texts in the temple see (1) passages from the OT (Exod 25:16; 40:20; Deut 10:12; 31:9; 31:24–26; Josh 24:25–26 et al.). The OT attests the same practice of storing up documents in the shrine as other ancient peoples. For the Hittites, cf. G. Beckman, Hittite Diplomatic Texts (2nd edn; Atlanta: Scholars Press, 1999). For the general practice in the ancient Near East, cf. K. Van Der Toorn, Scribal Culture and the Making of the Hebrew Bible (Cambridge, MA: Harvard University Press, 2007), 6. (2) The following passages from the Second Temple Period: CD A 7:14–18; Letter of Aristeas 30, 32, 39, 41, 46; 2 Maccabees 2:13–15; Jewish War 7.148–150 and thirteen other places in Josephus.

140

Peter J. Gentry and John D. Meade

convincingly shown.”34 Further work should be undertaken to determine the exact linkages between priests and Pharisees at Masada and how the Pharisaic texts come into the possession of the rabbis. We cannot provide full demonstration for this succession here, but our study has increased the probability of such a linkage between the text of Psalms of the Second Temple and the text of Psalms of the Masoretic text.35

Implications for the Work of Peter Flint Peter Flint has devoted enormous research on all the fragments of the Psalms at Qumran. His work is well known and need not be rehearsed here in detail. He seeks to develop a trajectory from 4QPsa—an extremely fragmentary text—to 11QPsa and from there to our MT Psalter of 150 psalms. Although the text of the Psalms at Qumran does not differ greatly from MT, the arrangement and number does; additional Psalms are included. Is Flint’s reconstruction persuasive? First, he does not engage the convincing research of Roger Beckwith in 1996 and Jonathan Ben-Dov in 2008 whose extensive studies on astronomy and calendrical issues demonstrate that the Qumran Community added psalms and arranged them differently due to their solar calendar and the liturgical system based thereon.36 Second, nowhere in his many contributions can we see any treatment of the superscription to LXX Psalm 151: Οὗτος ὁ ψαλμὸς ἰδιόγραφος εἰς Δαυιδ καὶ ἔξωθεν τοῦ ἀριθμοῦ· ὅτε ἐμονομάχησεν τῷ Γολιαδ. This psalm is autobiographical regarding David and outside the number. When he fought Goliath (Translation ours).

34  E. Ulrich, “Methodological Reflections, on Determining Scriptural Status in First Century Judaism”, in M.L. Grossman (ed.), Rediscovering the Dead Sea Scrolls: An Assessment of Old and New Approaches and Methods (Grand Rapids: Eerdmans, 2010) 145–61, on p. 155. 35  We did not have space to analyze MasPsb, but most scholars, including Peter Flint, have analysed it and have concluded that it represents the end of the book of Psalms and there was no Psalm 151. If this conclusion is granted, then Masada also possessed a text which would indicate that the final shape of the Psalter agreed with the later Masoretic Psalter. Cf. P.W. Flint, The Dead Sea Psalms Scrolls and the Book Of Psalms, 140–1. 36 R.T. Beckwith, Calendar and Chronology, Jewish and Christian, 141–66; J. Ben-Dov, Head of All Years, 49–52.

MasPsa and the Early History of the Hebrew Psalter

141

The superscription specifically states that Psalm 151 is “outside the number”. Therefore the translator of the Greek Psalter was aware of 150 as a fixed number at the time of translation. The number 150 is also considered proven from MasPsb (a text which we have not had time to analyse) by most scholars, including Flint.37 We do not, of course, have an exact date for the translation of the Hebrew Psalms into Greek. Surely it was translated after the Torah, but almost certainly before the Prologue to Greek Ben Sira. As Jannes Smith notes in an excellent contribution to the forthcoming Brill publication Textual History of the Bible, Munnich, Olofsson, and Schaper argue for a date in the second century b.c.e.38 The problem is compounded by the proposal of Ulrich and Rüsen-Weinhold that our current Greek Psalter is a revision of the original. While intriguing, this theory about a phantom translation of a phantom parent text remains unverified.39 Comparison with the book of Daniel is invalid since of approximately 90 extant manuscripts of Daniel only 3 attest the OG (88, 967, Syh) as opposed to that of Theodotion, whereas of approximately 1,300 known manuscripts of the Psalter all attest the same text.40 There is no manuscript evidence for Ulrich’s proposed OG.

Bibliography Barthélemy, D., Critique Textuelle de L’Ancien Testament (1. Josué, Juges, Ruth, Samuel, Rois, Chroniques, Esdras, Néhémie, Esther. Rapport final du Comité pour l’analyse textuelle de l’Ancien Testament hébreu institué par l’Alliance Biblique Universelle, établi en coopération avec A.R. Hulst, N. Lohfink, W.D. McHardy, H.P. Rüger, coéditeur, J.A. Sanders, coéditeur; Orbis Biblicus et Orientalis 50/1; Göttingen: Vandenhoeck & Ruprecht, 1982). Beckman, G., Hittite Diplomatic Texts (2nd edn; SBL Writings from the Ancient World Series 7; Atlanta: Scholars Press, 1999). Beckwith, R.T., Calendar and Chronology, Jewish and Christian: Biblical, Intertestamental and Patristic Studies (Leiden: Brill, 1996). Ben-Dov, J., Head of All Years: Astronomy and Calendars at Qumran in their Ancient Context (Studies on the Texts of the Desert of Judah 78; Leiden: Brill, 2008). Bouzard, Jr., W.C., “The Date of the Psalms Scroll from the Cave of Letters (5/6HEVPS) Recon�sidered”, Dead Sea Discoveries 10/3 (2003) 319–37.

37  P.W. Flint, The Dead Sea Psalms Scrolls and the Book Of Psalms, 140–141. 38  J. Smith, “10.3.1. Septuagint [Psalms > Primary Translations]”, in A. Lange/E. Tov (ed.), Textual History of the Bible: The Hebrew Bible Volume 1C, Writings (Leiden: Brill, 2017), 82–88. 39  There is not space here to rehearse the convincing evidence provided by Smith that it is unlikely. 40  This number is based on current research of the Septuaginta-Unternehmen der Akademie der Wissenschaften zu Göttingen, 2015.

142

Peter J. Gentry and John D. Meade

Dávid, N. & Lange, A. & de Troyer, K. & Tzoref, S., The Hebrew Bible in Light of the Dead Sea Scrolls (Forschungen zur Religion und Literatur des Alten und Neuen Testaments 239; Göttingen: Vandenhoeck & Ruprecht, 2012). de Hoop, R. & Sanders, P. (ed.), Have a Break (Forthcoming 2015). Donner H. & Röllig. W., Kanaanäische und Aramäische Inschriften (3 volumes; Wiesbaden: Otto Harrassowitz, 1962–1964 [Rev edn, 2005]). Dunand, F., Papyrus Grecs Bibliques (Papyrus F. INV. 266) Volumina de la Genèse et du Deutéronome: Texte et Planches (Extrait des Études de Papyrologie 9; Cairo: Imprimerie de l’Institut Français d’Archéologie Orientale, 1966). Dunand, F., Papyrus Grecs Bibliques (Papyrus F. INV. 266) Volumina de la Genèse et du Deutéronome: Introduction (Recherches d’Archéologie, de Philologie et d’Histoire 27; Cairo: Imprimerie de l’Institut Français d’Archéologie Orientale, 1966). Flint, P.W., The Dead Sea Psalms Scrolls and the Book of Psalms (Studies on the Texts of the Desert of Judah 17; Leiden: Brill, 1997). Flint, P.W., “The Contribution of the Cave 4 Psalms Scrolls To the Psalms Debate”, Dead Sea Discoveries 5/3 (1998) 320–33. Flint, P.W., “The Book of Psalms in the Light of the Dead Sea Scrolls”, Vetus Testamentum 48/4 (1998) 453–72. Flint, P.W., “Five Surprises in the Qumran Psalms Scrolls”, in A. Hilhorst/É. Puech/E. Tigchelaar (ed.), Flores Florentino: Dead Sea Scrolls and Other Early Jewish Studies in Honour of Florentino García Martínez (Supplements to the Journal for the Study of Judaism 122; Leiden: Brill, 2007) 183–96. Flint, P.W., “The Dead Sea Psalms Scrolls: Psalms Manuscripts, Editions, and the Oxford Hebrew Bible”, in S. Gillingham (ed.), Jewish and Christian Approaches: Conflict and Convergence (Oxford: Oxford University Press, 2013) 11–34. Gentry, P.J., “The Text of the Old Testament”, Journal of the Evangelical Theological Society 52/1 (2009) 19–45. Gibson, J.C.L., Textbook of Syrian Semitic Inscriptions. Volume II. Aramaic Inscriptions (Oxford: Clarendon, 1975). Grossman, M.L. (ed.), Rediscovering the Dead Sea Scrolls: An Assessment of Old and New Approaches and Methods (Grand Rapids: Eerdmans, 2010). Hallo, W.W., “The Concept of Canonicity in Cuneiform and Biblical Literature: A Comparative Reappraisal”, in The World’s Oldest Literature: Studies in Sumerian Belles-Lettres (Leiden: Brill, 2010) 699–716. Hallo, W.W., “Assyriology and the Canon”, in The World’s Oldest Literature: Studies in Sumerian Belles-Lettres (Leiden: Brill, 2010) 85–91. Kitchen, K.A. & Lawrence, P.J.N., Treaty, Law, and Covenant in the Ancient Near East (3 Vols; Wiesbaden: Harrassowitz, 2012). Law, T.M., When God Spoke Greek: The Septuagint and the Making of the Christian Bible (Oxford: Oxford University Press, 2013). Lim, T.H., The Formation of the Jewish Canon (New Haven: Yale University Press, 2013).

MasPsa and the Early History of the Hebrew Psalter

143

Millard, A., “Text and Comment”, in G.A. Tuttle (ed.), Biblical and Near Eastern Studies: Essays in Honor of William Sanford LaSor (Grand Rapids: Eerdmans, 1978) 245–52. O’Connor, M.P., Hebrew Verse Structure (2nd edn; Winona Lake, IN: Eisenbrauns, 1980, 1997). Pelletier, A., La Lettre d’Aristée a Philocrate (Sources Chrétiennes 89; Paris: Les Éditions du Cerf, 1962). Rahlfs, A. (ed.), Psalmi Cum Odis (Septuaginta Vetus Testamentum Graecum Vol. 10; 3rd edn; Göttingen: Vandenhoeck & Ruprecht, 1931, 1967, 1979). Revell, E.J., “The Oldest Evidence for the Hebrew Accent System”, Bulletin of the John Rylands Library 54 (1971) 214–22. Revell, E.J., “The Occurrence of Pausal Forms”, Journal of Semitic Studies 57/2 (2012) 213–30. Revell, E.J., “Minor Pausal Forms and their Function”, Vetus Testamentum 65/3 (2015) 457–65. Revell, E.J., The Pausal System: Divisions in the Hebrew Biblical Text as Marked by Voweling and Stress Position (R. de Hoop/P. Sanders (ed.); Sheffield: Phoenix Press, 2015). Revell, E.J., “Pausal Forms Marked with Conjunctive Accents”, Journal of Semitic Studies 61/1 (2016) 67–84. Sanders, P., “The Ashkar-Gilson Manuscript: Remnant of a Proto-Masoretic Model Scroll of the Torah”, Journal of Hebrew Scriptures 14 (2014) 1–25 (jhs.2014.v14.a7). Sanders, P., “The Colometric Layout of Psalms 1 to 14 in the Aleppo Codex”, in M. Korpel/J. Oesch (ed.), Studies in Scriptural Unit Division (Pericope 3; Assen: Van Gorcum, 2002) 226–57. Schuller, E., “Review of The Dead Sea Psalms Scrolls and the Book of Psalms, by Peter W. Flint”, Hebrew Studies, 41 (2000) 297–303. Smith, J., “10.3.1. Septuagint [Psalms > Primary Translations]”, in A. Lange/E. Tov (ed.), Textual History of the Bible (Leiden: Brill, forthcoming). Smith, J., “The Meaning and Function of Ἀλληλουϊα in the Old Greek Psalter”, in M.K.H. Peeters (ed.), XII Congress of the International Organization for Septuagint and Cognate Studies, Leiden, 2004 (SCS 54; Atlanta: SBL, 2006) 141–51. Smith, J., “The Text-Critical Significance of Oxyrhynchus Papyrus 5101 (Ra 2227) for the Old Greek Psalter”, Journal of Septuagint and Cognate Studies 45 (2012) 5–22. Smith, J., Translated Hallelujahs: A Linguistic and Exegetical Commentary on Select Septuagint Psalms (Leuven: Peeters, 2011). Talmon, S., “Review of The Dead Sea Psalms Scrolls and the Book of Psalms, by Peter W. Flint”, JBL 118/3 (1999) 545–7. Talmon, S. & Yadin, Y., Masada VI: The Yigael Yadin Excavations 1963–1965 (Jerusalem: Israel Exploration Society, 1999). Tigchelaar, E., “Notes on the Ezekiel Scroll from Masada (MasEzek)”, Revue de Qumran 22/86 (2005) 269–75. Tov, E., “A Qumran Origin for the Masada Non-Biblical Texts?” DSD (2000) 57–73. Tov, E., “The Background of the Stichometric Arrangements of Poetry in the Judean Desert Scrolls”, in Textual Criticism of the Hebrew Bible, Qumran, Septuagint, Collected Essays, Volume 3 (Supplements to Vetus Testamentum 167; Leiden: Brill, 2015) 325–36.

144

Peter J. Gentry and John D. Meade

Tov, E., “The Biblical Texts from the Judean Desert—An Overview and Analysis”, in Hebrew Bible, Greek Bible and Qumran: Collected Essays by Emanuel Tov (Tübingen: Mohr-Siebeck, 2008). Tov, E., Scribal Practices and Approaches Reflected in the Texts Found in the Judean Desert (Leiden: Brill, 2004). Tov, E., Textual Criticism of the Hebrew Bible (3rd revised and expanded edn; Minneapolis: Fortress Press, 2012). Ulrich, E., “Methodological Reflections on Determining Scriptural Status in First Century Judaism”, in M.L. Grossman (ed.), Rediscovering the Dead Sea Scrolls: An Assessment of Old and New Approaches and Methods (Grand Rapids: Eerdmans, 2010) 145–61. VanderKam, J.C., The Dead Sea Scrolls and the Bible (Grand Rapids: Eerdmans, 2012). Van der Toorn, K., Scribal Culture and the Making of the Hebrew Bible (Cambridge, MA: Harvard University Press, 2007). van Liere, F., An Introduction to the Medieval Bible (Cambridge: Cambridge University Press, 2014). von Weissenberg, H. & Pakkala, J. & Marttila, M. (ed.), Changes in Scripture: Rewriting and Interpreting Authoritative Traditions in the Second Temple Period (Berlin: de Gruyter, 2011). Williams, P.J. “The Bible, the Septuagint, and the Apocrypha: A Consideration of Their Singularity”, in G. Khan & D. Lipton (ed.), Studies on the Text and Versions of the Hebrew Bible in Honour of Robert Gordon (Leiden: Brill, 2012). Yeivin, I. Introduction to the Tiberian Masorah (Transl. and edited by E.J. Revell; Masoretic Studies 5; Missoula: Scholars Press, 1980).

DJD Volumes Tov, E., Revised Lists of the Texts from the Judean Desert (Leiden: Brill, 2010). Barthélemy, D./J.T. Milik (ed.), Qumran Cave 1 (Discoveries in the Judaean Desert I; Oxford: Clarendon Press, 1955). Baillet, M. & Milik, J.T. & de Vaux, R. (ed.), Les ‘Petites Grottes’ de Qumran (Texte) (Discoveries in the Judaean Desert III; Oxford: Clarendon Press, 1962). Baillet, M. & Milik, J.T. & de Vaux, R. (ed.), Les ‘Petites Grottes’ de Qumran (Planches) (Discoveries in the Judaean Desert III; Oxford: Clarendon Press, 1962). Skehan, P.W. & Ulrich, E. & Sanderson, J.E. (ed.), Qumrân Cave 4: IV. Palaeo-Hebrew and Greek Biblical Manuscripts (Discoveries in the Judaean Desert IX; Oxford: Clarendon Press, 1992). Sanders, J.A. (ed.), The Psalms Scroll of Qumrân Cave 11 (11QPsa) (Discoveries in the Judaean Desert IV; Oxford: Clarendon Press, 1965). Ulrich, E. & Cross, F.M. & White Crawford, S. & Duncan, J.A. & Skehan, P.W. & Tov, E. & Trebolle Barrera, J. (ed.), Qumran Cave 4: IX. Deuteronomy, Joshua, Judges, Kings (Discoveries in the Judaean Desert XIV; Oxford: Clarendon Press, 1995).

MasPsa and the Early History of the Hebrew Psalter

145

Ulrich, E. (ed.), Qumran Cave 4: XVI. Psalms to Chronicles (Discoveries in the Judaean Desert XVI; Oxford: Clarendon Press, 2000). Puech, É. (ed.), Qumrân Grotte 4: XVIII. Textes Hébreux (4Q521–4Q528, 4Q576–4Q579) (Discoveries in the Judaean Desert XXV; Oxford: Clarendon Press, 1998). Charlesworth, J. et al. (ed.), Miscellaneous Texts from the Judaean Desert (Discoveries in the Judaean Desert XXXVIII; Oxford: Clarendon Press, 2001).

Emanuel Tov The Possible Revision of Hebrew Texts According to MT

Background The topic of this study is the revision of ancient sources according to the Masoretic text (MT). From our vantage point such a revision sounds like a logical concept since the developments of the past 2000 years have made us accustomed to the idea that MT is the central Scripture text. However, this was not always the case, and the assumption of revisional tendencies cannot automatically be assumed across the board. Each source needs to be analyzed separately, and what is true for the realm of one Greek translation does not necessarily apply to other Greek translations, or to Aramaic translations, and definitely not for Hebrew texts. For example, adaptation of biblical quotations to MT in rabbinic sources is less frequent than one might think. Penkower claims that quotations deviating from MT1 in the manuscripts of the rabbinic literature or the traditional Jewish commentators were not often changed in manuscripts,2 while they were in printed editions.3 The discussion will first lead us to the revision of ancient translations, especially the Septuagint (LXX), and then to the main topic, revision of Hebrew manuscripts.

1  See the Appendix in M. Ben Yashar et al., The Bible in Rabbinic Interpretation: Rabbinic Derashot on Prophets and Writings in Talmudic and Midrashic Literature, Vol. 1, Hosea (Ramat Gan: Bar-Ilan University Press, 2003), 747–52, and subsequent volumes; J.S. Penkower, Masorah and Text Criticism in the Early Modern Mediterranean: Moses ibn Zabara and Menahem de Lonzano (Jerusalem: Magnes, 2014) and the literature mentioned in these works. The most detailed description of the critical approach toward the biblical quotations in rabbinic literature is found in M.H. Goshen-Gottstein/S. Talmon, The Hebrew University Bible: The Book of Ezekiel (Jerusalem: Magnes, 2004), xxxi–xxxvi. 2  E-mail message 30.12.2014. 3  See C. Rabin/S. Talmon/E. Tov, The Hebrew University Bible, The Book of Jeremiah (Jerusalem: Magnes, 1997), xxx: “It is a well-known fact that in most Midrash collections ‘deviant’ quotations were corrected by editors and printers, who may have made use of one or more manuscripts, but did not record them in an apparatus”. The changes made by editors were illustrated in detail by S. Esh, “Variant Readings in Mediaeval Hebrew Commentaries: R. Samuel Ben Meir (Rashbam)”, Textus 5 (1966) 84–92 and J.S. Penkower, “The Text of the Bible Used by Rashi as Reflected in His Biblical Commentaries”, in A. Grossman/S. Japhet (ed.), Rashi: The Man and His Work (Heb.; Jerusalem: The Zalman Shazar Center for Jewish History, 2008) 1.99–122.

148

Emanuel Tov

Revision of Ancient Scripture Translations to MT The evaluation of the ancient translations is complicated because neither details identical with MT nor scribal corrections towards the text now known as MT were necessarily meant as corrections towards that text. The assumption that the wording of a translation has been changed in the direction of the proto-MT is often not without a residue of doubt. The most frequent revisional activity involved changing the original wording of a translation towards the proto-MT when the original text of that translation deviated from it. Thus the Old Greek translation, based on a text that deviated slightly or much from the proto-MT, was later adapted to the proto-MT when that became increasingly accepted in the Jewish communities as a binding text. Greek translations. Two types of possible approximation to the proto-MT may be distinguished, (a) a precise representation of the proto-MT in a new translation and (b) the change of isolated elements in LXX manuscripts. In the first scenario no revisional activity was involved. a. A precise representation of the proto-MT. A precise translation of the protoMT need not necessarily reflect a revision to that text, since it could represent a literal translation of Hebrew Scripture made once the proto-MT became the standard source text for Hebrew translations. We need to ask ourselves whether there existed at all early Greek translations that were novel translations of the proto-MT that were not based on the Old Greek (OG). In my view such translations are not evidenced. There are indications that translations like kaige-Theodotion (Th) and Aquila are not independent, but adapted the OG to new norms. It can be proven that these new translations share with the OG a layer of idiosyncratic renderings, which they changed in the direction of the proto-MT. Such a new translation should therefore be considered a revision of the OG based on the proto-MT. Criteria have been developed in order to characterize such a common layer of idiosyncratic renderings.4 Several early revisers reworked the OG in different localities and at different times, viz. kaige-Th, Aquila, and Symmachus as well as some anonymous revisers.5 The mentioned three translations have much in common, and this common stratum was probably caused by Aquila and Symmachus basing themselves on 4  For the theoretical background, see E. Tov, Textual Criticism of the Hebrew Bible (3rd, revised and expanded edn; Minneapolis: Fortress Press, 2012), 141 (henceforth: TCHB). The common exclusive features of the OG and the new translations have been analyzed in the case of the kaige-Th revision by D. Barthélemy, Les devanciers d’Aquila (VTSup 10; Leiden: Brill, 1963), 91–102 and in the case of an internal revision in the LXX of Jeremiah in my study The Septuagint Translation of Jeremiah and Baruch: A Discussion of an Early Revision of Jeremiah 29–52 and Baruch 1:1–3:8 (HSM 8; Missoula, MT: Scholars Press, 1976), 19–40. 5  One such anonymous reviser of the second part or all of Jeremiah is described in my study mentioned in the previous note.

The Possible Revision of Hebrew Texts According to MT

149

the earlier version of kaige-Th. These two translators thus did not change the OG directly, but they revised kaige-Th.6 b. Change of isolated elements in the uncial manuscripts of the LXX. It is assumed, but usually cannot be proven without a residue of doubt, that early scribes and readers adapted OG manuscripts to MT7 when that Hebrew source became gradually accepted. In all these cases there is no physical evidence in the form of visible corrections in the manuscripts, and identity with MT of the Greek readings compared with other Greek sources is the main reason for assuming the revisional nature of an early witness. In these cases it is assumed that the rendering agreeing with MT replaced earlier renderings differing from it. Thus it is often claimed that the uncial manuscripts A and B had been altered occasionally in the direction of MT. In the Göttingen editions of the Pentateuch these revisional elements are not described in the introductions, but they are analyzed in a series of monographs by Wevers. Thus for codex A in Numbers8 and for codex B in Deuteronomy (e.g. Deut 11:19),9 Wevers records many idiosyncratic readings agreeing with MT. Likewise, manuscripts Q and V in Jeremiah reflect a pre-Hexaplaric revision.10 By the same token, on the basis of a comparison with such early sources as 4QLXXLeva, 4QLXXpapLevb, and 4QLXXNum differing more from MT than the uncial manuscripts,11 it is often assumed that the text of the uncials reflects several revisional elements.12 Similar revisions are included in several early sources, as compared with the later uncials: P.Fouad 266b of Deuteronomy 17–33 (middle of the 1st century b.c.e.)13 = Ra 848 (“Deut I”); P.Fouad 266c (847) of Deuteronomy 6  This view was first suggested by Barthélemy in his revolutionary study Devanciers, 246– 65 and confirmed by others. 7  In the discussion below “MT” stands for the proto-Masoretic text. 8  J.W. Wevers, Text History of the Greek Numbers (AKWG, Phil.-Hist. Kl. 3, 125; MSU XII; Göttingen: Vandenhoeck & Ruprecht, 1982), 73–6. 9  J.W. Wevers, Text History of the Greek Deuteronomy (AKWG, Phil.-Hist. Kl. 3, 106; MSU XVI; Göttingen: Vandenhoeck & Ruprecht, 1978), 48–51. 10  See J. Ziegler, Ieremias, Baruch, Threni, Epistula Ieremiae (Septuaginta, Vetus Testamentum Graecum auctoritate societas litterarum gottingensis editum, XV; Göttingen: Vandenhoeck & Ruprecht), 63–6. 11  Furthermore, the translation vocabulary of the Qumran texts has been adapted in the uncials to that of the later LXX books. 12  See my studies “The Greek Biblical Texts from the Judean Desert”, in S. McKendrick/ O.A. O’Sullivan (ed.), The Bible as Book: The Transmission of the Greek Text (London: British Library and Oak Knoll Press/The Scriptorium: Center for Christian Antiquities, 2003) 97–122. Revised version: Hebrew Bible, Greek Bible, and Qumran: Collected Essays (TSAJ 121; Tübingen: Mohr Siebeck, 2008), 339–64; “The Use of the Earliest Greek Scripture Fragments in Text Editions”, forthcoming. 13  Publication: F. Dunand, Papyrus grecs bibliques (Papyrus F. Inv. 266): Volumina de la Genèse et du Deutéronome, Texte et planches (Extrait des études de papyrologie IX; Cairo: Institut francais d’archéologie orientale, 1966); eadem, Papyrus grecs bibliques (Papyrus F. Inv. 266):

150

Emanuel Tov

10–11, 31–33 (50–1 b.c.e.)14 = Ra 847 (“Deut II”); 7QpapLXXExod of Exodus 28 (1st century b.c.e.) = Ra 805;15 hands A and B of 8HevXIIgr (end of the 1st century b.c.e.) = Ra 943; P.Oxy. 77.5101 of Psalms (1st or 2nd century c.e.)16 = Ra 2227. These texts contain many revisional elements as recognized by their agreements with MT and additionally most of them represent the tetragrammaton in Hebrew characters. The latter feature is also reflected in P.Oxy. 50.3522 of Job 42 (1st century b.c.e.), containing the tetragrammaton in the paleo-Hebrew script. In my view, this feature is a sign of the revisional character of the Greek sources. By the same token, the uncial manuscripts of the LXX as compared with the pre-Hexaplaric Chester-Beatty-Scheide papyrus 967 dating to the beginning of the 3rd century c.e. often reflect revised elements, for example in the sequence of the chapters in Ezekiel. In this detail, papyrus 967, differing from the MT and the majority of the Greek textual witnesses, may well represent an earlier sequence.17 Likewise, P. Katz detected a group of Philonic manuscripts (manuscripts U and F, sometimes joined by other sources) that represent an approximation of Philo’s text to MT. In his view this textual tradition probably represents an unknown early revision, similar to Aquila.18 Katz finds that this unknown recension has much in common with the recensions that Rahlfs detected in Ruth (an unknown ­ olumina de la Genèse et du Deutéronome, Introduction (Recherches d’archéologie, de philoloV gie et d’histoire XXVII; Cairo: Institut francais d’archéologie orientale, 1966). 14  Publication: Z. Aly, Three Rolls of the Early Septuagint: Genesis and Deuteronomy (PTA 27; Bonn: Habelt, 1980). 15  The evidence is limited. See the analysis of J.W. Wevers, “Pre-Origen Recensional Activity in the Greek Exodus”, in D. Fraenkel et al. (ed.), Studien zur Septuaginta: Robert Hanhart zu Ehren (MSU XX; Göttingen: Vandenhoeck & Ruprecht, 1990) 64–73. On the other hand, E. Ulrich, The Dead Sea Scrolls and the Developmental Composition of the Bible (VTSup 169; Leiden: Brill, 2015), 156 classifies this text as “independent scribal commonplace”. 16  D. Colomo & W.B. Henry, “5101. LXX, Psalms xxvi 9–14, xliv 4–8, xlvii 13–15, xlviii 6–21, xlix 2–16, lxiii 6–lxiv 5”, in: A. Benaissa (ed.), The Oxyrhynchus Papyri (Graeco-Roman Memoirs 98; London: Egypt Exploration Society, 2011) 77.1–11. For an analysis, see J. Smith, “The Text-Critical Significance of Oxyrhynchus Papyrus 5101 (Ra 2227) for the Old Greek Psalter”, BIOSCS 45 (2012) 5–22. 17  See J. Lust, “Major Divergences between LXX and MT in Ezekiel”, in A. Schenker (ed.), The Earliest Text of the Hebrew Bible: The Relationship between the Masoretic Text and the Hebrew Base of the Septuagint Reconsidered (SBLSCS 52; Atlanta: Scholars Press, 2003) 83–92; idem, “The Ezekiel Text”, in Y.A.P. Goldman et al. (ed.), Sôfer Mahîr: Essays in Honour of Adrian Schenker Offered by Editors of Biblia Hebraica Quinta (VTSup 110; Leiden: Brill, 2006) 153–67. 18  P. Katz, The Aberrant Text of Bible Quotations in Some Philonic Writings and Its Place in the Textual History of the Greek Bible (Cambridge: University Press, 1950), 95–103. P. Kahle, The Cairo Genizah (2nd edn; Oxford: Blackwell, 1959), 247–9, who does not accept Katz’s conclusions, believes that manuscripts U and F of Philo reflect the text as known to Philo himself and which was similar to 8HevXIIgr dated to the same period. Although this assumption is remotely possible, it does not explain the difference between the various groups of Philo manuscripts. According to Katz, the manuscripts containing the aberrant LXX quotations also reflect a better Philo text.

The Possible Revision of Hebrew Texts According to MT

151

recension R together with the Catena group)19 and with the manuscript groups b c q r in Exodus-Judges. Targumim. The Vorlagen of the known Targumim were presumably close to MT, but in antiquity they were made even closer to that text since these translations were conceived of as the “official Jewish translations”. This assumption is strengthened by the fact that the oldest known Targum, that of Job from Qumran (11QtgJob), differed much more from MT than the ones that are known from the later manuscripts.20 Better known is the situation in the next centuries. From an early period onwards the text of the Targumim was often juxtaposed in the manuscripts with MT, verse after verse. This proximity brought the two closer together and there are indeed proven cases of changes in the text of the Targumim towards MT.21

Revising Ancient Hebrew Manuscripts in the Direction of MT The manifold corrections of ancient Greek sources toward MT have influenced scholars, sometimes unconsciously, in their assumption that also Hebrew sources were corrected to MT, especially when scribal corrections are still visible in the manuscripts. Some remarks to this effect are made in the DJD editions,22 while others are embedded in several monographs.23 These scribal changes may take several forms, mainly as additions between the lines or in the linear text or as deletions indicated with various techniques. Rarely elements have been added in 19  A. Rahlfs, Studie über den griechischen Text des Buches Ruth (MSU III,2; NGWG Phil.Hist. Kl.; Göttingen: Weidmannsche Buchhandlung,1922), 47–164. 20  See R. Weiss, “Recensional Variations between the Aramaic Translation to Job from Qumran Cave 11 and the Massoretic Text”, Shnaton 1 (Heb. with Eng. summ.; Jerusalem, 1975) 123–7; id., The Aramaic Targum of Job (Heb. with Eng. summ.; Tel Aviv: Tel Aviv University, 1979). On the other hand, according to Ego, the tradition of 11QtgJob does not stand in a direct line of development with the later Targumim: B. Ego, “Targumim, Overview Article”, in A. Lange and E. Tov (ed.), Textual History of the Bible, The Hebrew Bible, Vol. 1A, Overview Articles (Leiden: Brill, 2016), § 1.3.3.1.2. 21  For examples, see W.F. Smelik, “Targum, Former Prophets”, ibid., § 3–5.1.3.4. 22  E.g., J.E. Sanderson in her edition of 4QExodc in Qumran Cave 4.VII: Genesis to Numbers (ed. E. Ulrich/F.M. Cross; DJD XII; Oxford: Clarendon, 1994), 102. Furthermore, the notations “= MT” attached to the recording of the corrections in DJD may have contributed to the confusion: Formal notations such as my remark on 4QJera in Jer 17:16 ‫ = ואני לא‬MT (DJD XVII, 153) may have been taken as denoting a correction toward MT. 23  Thus E.Y. Kutscher, The Language and Linguistic Background of the Isaiah Scroll (1 Q Isa) (STDJ 6; Leiden: Brill, 1974), 554 (1QIsaa); A. Lange, Handbuch der Textfunde vom Toten Meer, I: Die Handschriften biblischer Bücher von Qumran und den anderen Fundorten (Tübingen: Mohr Siebeck, 2009), 55 (MurGen-Exod-Num), 188 (4QJoshb), 300 (4QJera); D. Barthélemy, Studies in the Text of the Old Testament: An Introduction to the Hebrew Old Testament Text Project (Winona Lake, IN: Eisenbrauns, 2012), 387 (MurXII), 404–405 (1QIsab).

152

Emanuel Tov

the margins between the columns. However, several arguments may be invoked against this linkage between the mentioned revision of ancient translations and the correction of Hebrew manuscripts. (1) The evidence for the revision of the ancient translations differs much from that invoked for the Hebrew Judean Desert scrolls. The translational revisions are embedded in the texts themselves from which they can be extrapolated, while the presumed revisions of Dead Sea Scrolls are visible in the form of scribal interventions. Seemingly such scribal evidence is more convincing than the presumed translational revisions, but in actual fact, it is less stable. For the translational revisions may be inferred with relative confidence from the evidence, while the intention behind the scribal corrections in Hebrew manuscripts as approximations towards MT is less clear. The agreement between the corrections and MT is a fact, but it is not known whether the changes have been inserted on the basis of MT as a standard text or as corrections of scribal errors toward the scroll’s Vorlage that happened to agree with MT.24 (2) The assumption of correction to MT is based on the further assumption that the changes in the scrolls reflect a gradual move toward MT as an accepted text until its complete acceptance as the sole Bible text. That process presumably was successfully completed when MT succeeded to oust all the other texts after a Kulturkampf taking place at the end of the 1st century c.e. Several argument s have been suggested in support of such a theory of stabilization, and the main argument runs as follows. After several centuries of textual plurality, a period of uniformity and stability can be discerned within Judaism at the end of the 1st century c.e. This sit uat ion is usual ly expl ained as reflect ing a conscious effort t o stabilize the Scripture text, involving the creation of a standard text for Palestine as a whole. However, textual stability (that is, when all sources reflect the same text) should not be confused with stabilization, that is, an organized attempt to create a stable text, since that stability may have been caused by a number of factors. Stabilization involves a conscious process, since it reflects an attempt to impose a text on a group or region. An alternative explanation of the evidence could be the assumption of historical coincidence. For example, one could claim—as I do25—that after the destruction of the Temple and with the splitting off from Judaism of the Samaritans and Christians, MT was the only text surviving within Judaism as opposed to textual plurality in the previous period. As a result, in my view, there was no movement from pluriformity to uniformity but, in the words 24  These two possibilities have been carefully weighed by J.T. Milik in his analysis of MurXII in P. Benoit/ G.M. Crowfoot/J.T. Milik/R. de Vaux, Les grottes de Murabba‘ât (DJD II; Oxford: Clarendon, 1961), 183. 25  Tov, TCHB, 174–180; idem, “The Myth of the Stabilization of the Text of Hebrew Scripture”, in E. Martín Contreras/L. Miralles-Maciá (ed.), The Text of the Hebrew Bible: From the Rabbis to Masoretes (JAJSup 13; Göttingen: Vandenhoeck & Ruprecht, 2014) 37–45.

The Possible Revision of Hebrew Texts According to MT

153

of van der Woude, “there was a basically uniform tradition besides a pluriform tradition in Palestine Judaism in the last centuries BC”.26 (3) A statistical analysis of the totality of the Judean Desert manuscripts is sometimes invoked in support of the assumption of a process of stabilization. Grosso modo, the argument runs as follows: Also before the scrolls were found, scholars were aware of textual multiplicity involving MT, the Samaritan Pentateuch (SP), the LXX and papyrus Nash of the Decalogue and Shema‘ in the preChristian centuries, but in the 1st century c.e. we only meet the text of MT. It was often thought that this uniform textual tradition was obtained through a process of elimination of the other texts. After the discovery of the Dead Sea scrolls the quantities of the ancient copies were used to illustrate this process. It was claimed that the presence of proto-Masoretic scrolls in the late Judean Desert sites from the period of Bar Kochba (135 c.e.) show that at that time MT had ousted the other texts. More precisely, according to Lange, proto-Masoretic manuscripts are found in larger quantities from the second half of the last century b.c.e. onwards in Qumran and elsewhere.27 However, this assumption is based on a very small number (6) of scrolls from Qumran.28 Further, these texts are very fragmentary, and with a content of around 100 fragmentary words, their textual profile cannot be established well.29 Because of these problems, of the six texts mentioned by Lange, only the 1st century c.e. scroll 4QGenb (50–100 c.e.) may be considered truly proto-Masoretic,

26  A.S. van der Woude, “Pluriformity and Uniformity: Reflections on the Transmission of the Text of the Old Testament”, in J.N. Brenner/F. García Martínez (ed.), Sacred History and Sacred Texts in Early Judaism: A Symposium in Honour of A.S. van der Woude (Kampen: Kok Pharos, 1992) 151–69, on p. 163. 27  A. Lange, “‘They Confirmed the Reading’ (y. Ta‘an. 4:68a): The Textual Standardization of Jewish Scriptures in the Second Temple Period”, in A. Lange et al. (ed.), From Qumran to Aleppo: A Discussion with Emanuel Tov about the Textual History of Jewish Scriptures in Honor of His 65th Birthday (FRLANT 230; Göttingen: Vandenhoeck & Ruprecht, 2009) 29–80. Lange points to a religious factor as the background of the diffusion of these texts: “The chronological distribution of the manuscripts of biblical books … shows, that the proto-Masoretic text was created in the second half of the first century b.c.e. as part of a concentrated effort to preserve the cultural heritage of Judea. Early proto-Masoretic manuscripts like 4QJera demonstrate that for this purpose existing manuscript traditions were used at least partly” (Lange, “They Confirmed”, 56). The same view is presented in idem, “The Textual Plurality of Jewish Scriptures in the Second Temple Period in Light of the Dead Sea Scrolls”, in N. Dávid/A. Lange (ed.), ­Qumran and the Bible: Studying the Jewish and Christian Scriptures in Light of the Dead Sea Scrolls (CBET 57; Leuven: Peeters, 2010) 43–96. 28 4QXIIe (75–50 b.c.e.); 4QEzeka (50 b.c.e.); 4QDeut e (50–25 b.c.e.); 2QRut ha (30–1 b.c.e.); 4QDeutg (1–50 c.e.); 4QGenb (50–100 c.e.). The l at t er t ext is analyzed bel ow. 29  Furthermore, 4QXIIe is not proto-Masoretic as it reflects a completely different orthography system (‫רוש‬, ‫ )ידעתמה‬as well as four corrections and one major variant, in 90 more or less complete words.

154

Emanuel Tov

but the Qumran provenance of this scroll has been greatly doubted.30 Further, the assumption of the rising presence of proto-MT scrolls in Qumran in the second half of the 1st century b.c.e. is in jeopardy since the peak in the presumed presence of proto-Masoretic scrolls is shared with the enlarged production of all scrolls in the second part of the 1st century b.c.e.31 As a result, a process of stabilization cannot be established. We now reach the difficult question of whether we are able to distinguish between an assumed approximation of a scroll towards MT and scribal correction in accord with the scroll’s Vorlage. This question will be analyzed when paying attention to three different groups of texts: (a) proto-MT scrolls closely resembling the medieval MT, (b) MT-like scrolls containing a larger number of variants than in group (a), and (c) scrolls that differ much from MT. It seems that the evidence of groups (b) and (c) shows that in all three groups a revision in accord with MT is unlikely. a. A few proto-MT scrolls that closely resemble the medieval MT contain a number of corrections towards MT. It may be claimed that the corrections in these scrolls were made by the original scribe or a later one in order to achieve full agreement with MT. The group of true proto-Masoretic texts is small and fragmentary,32 and even smaller is the number of the texts that contain corrections agreeing with MT.33 The four texts mentioned here are meant to be exhaustive for this group. 1. 4QGenb (50–100 c.e.). With 358 partially preserved words, this scroll reflects only one variant in spelling (‫ למארת‬1:15 [= v. 14, 16 MT]; MT ‫)למאורת‬, and none in content. Further, the addition of the erroneously left out gimel in 1:16 ‫ הגדולים‬reflects a clear scribal error. 2. MasLevb (30 b.c.e.–30 c.e.). The 96 fragmentary lines of MasLevb contain no variants, but there is one orthographical difference. Four apparent mistakes were corrected towards the text that is now MT:

31.

30  This scroll has not been found in controlled excavations. See James R. Davila, DJD XII,

31  Lange, “They Confirmed”, notes: “It seems that this time was a zenith of scribal culture” (p. 53). This assumption is also confirmed by the statistics in my study “Some Thoughts about the Diffusion of Biblical Manuscripts in Antiquity”, in S. Metso et al. (ed.), Transmission of Traditions and Production of Texts (STDJ 92; Leiden: Brill, 2010) 151–72. Revised version: Textual Criticism of the Hebrew Bible, Qumran, Septuagint: Collected Writings, Volume 3 (VTSup 167; Leiden: Brill, 2015) 60–81. 32  This group consists of all the texts from the Judean Desert from sites other than Qumran, and in addition the late scroll 4QGenb for which see n. 30. 33  A few somewhat substantial texts contain very few corrections or none at all, such as MasPsa (one correction in II 24).

The Possible Revision of Hebrew Texts According to MT

155

10:17 + ‫( הוא‬supralinear) = MT 11:10 ‫הםה‬, with final mem = MT ‫המה‬. This variant probably corrects an original ‫ הם‬to ‫הםה‬. 11:32 ‫ = יטמא‬MT 11:35 ‫ = וכירים‬MT

3. MasEzek (50–1 b.c.e.). The 120 fragmentary lines of MasEzek contain 8 orthographical variants (Ezek 35:12; 36:17, 18, 21, 26, 30; 37:6, 10) and five variants in content (35:14; 36:21; 37:7, 7, 17).34 This scroll usually agrees with MT against the LXX,35 and also contains 5 corrections: 35:13 ‫דבריכם‬, ‫ = עלי‬MT 36:25 ‫ = טמאותיכם‬MT 36:30 ‫ = ותנובת‬MT 37:4 ‫ = הנבא‬MT

The nature of two corrections is unclear: the supralinear addition of ‫ בית‬in 36:22 and of an unspecified letter in 37:11, both against MT.36 4. MurXII (c. 115 c.e.). This long scroll, with 3,60537 more or less identifiable words differs from MT in only 41 instances,38 23 in spelling and 18 in small details. The closeness of MurXII to MT is visible in such unusual spellings as ‫אילכה‬ for ‫( אלכה‬Mic 1:8) and ‫ אבי‬for ‫( אביא‬Mic 1:15).39 Ten corrective additions, above or under the line, revise MurXII towards the text now included in MT:40 Orthography (2):

–  Am 9:8 addition of yod in ‫ = השמיד‬MT –  Mic 7:4 addition of waw in ‫ = מבוכתם‬MT

Linguistic correction (1) –  Joel 3:2 ‫ = ההםה‬MT

34  See S. Talmon in idem & Y. Yadin, Masada VI, The Yigael Yadin Excavations 1963–1965, Final Reports, Hebrew Fragments from Masada (Jerusalem: Israel Exploration Society/The Hebrew University of Jerusalem, 1999) 68–72. 35  See Talmon, ibid. 36  According to Talmon, 68, the corrections were made in order to bring the scroll in line with MT. If this were the case, the last two instances do not follow this pattern. 37 Lange, Handbuch, 346 counts 3803 partially preserved words. 38 Barthélemy, Studies, 386–7 counts 42 variants (compared with BHK3). 39  For further details on these spellings, see Barthélemy, Studies, 387–8. 40 Barthélemy, Studies, 387.

156

Emanuel Tov

Supralinear correction of scribal omissions (6): –  Am 7:13 ‫ = עוד‬MT –  Am 8:11 ‫ = והשלחתי רעב בארץ‬MT. The text makes no sense without the added words. –  Obad 2 ‫ = בגוים‬MT –  Mic 4:7 ‫ = את‬MT –  Nah 1:11 ‫ = על‬MT –  Hag 2:3 ‫ = אתו‬MT

Elements omitted (1):

–  Obad 13 waw of “‫ ואל‬apparently” scratched out41 = MT—not visible on the plate.

The number of corrections in this scroll is seemingly impressive,42 but when compared with other scrolls it is not large. In this long scroll we find an average of one scribal intervention in 54 lines, as opposed to an average of one intervention in 4–10 lines in many other scrolls. The data recorded in a chart of scribal intervention in the Dead Sea Scrolls43 shows that the rate of intervention in MurXII points to a group of carefully transmitted texts. Most corrections pertain to quantitative differences from MT regarding elements that had been omitted erroneously by the first hand, with one possible erasure (Obad 13). On the other hand, the 41 variants44 pertain to qualitative differences in content and orthography. Therefore, in this carefully written scroll, the ten corrections seem to represent rare mistakes that were corrected according to the scroll’s Vorlage. It is not logical to assume that only quantitative differences were corrected while qualitative variants, that is “differences”, were left in the text. b. MT-like45 scrolls contain a larger number of variants than the scrolls in group (a) as well as some corrections in the direction of the text that is now MT.46 It cannot be claimed that these corrections were made in conformity with MT since most deviations from MT in these texts were not corrected to MT. I do accept the view that all revisions are inconsistent, but to invoke inconsistency in this case would be exaggerated. The evidence for this group is not exhaustive.

41  Thus J.T. Milik, DJD II, 183. 42  There may be a few additional corrections in columns that are not well readable. 43  E. Tov, Scribal Practices and Approaches Reflected in the Texts Found in the Judean Desert (STDJ 54; Leiden: Brill, 2004), 331–5. 44  The instances are listed in DJD II, 183–4, 205. 45  For the term, see Tov, TCHB, 31. Lange, “They Confirmed”, 54 names these scrolls semiMasoretic (see also Lange, Handbuch, 16), while for Barthélemy, Studies, 383–409 they are preMasoretic. 46 In Scribal Practices, 332–5, I list 19 scrolls of this group.

The Possible Revision of Hebrew Texts According to MT

157

1. 1QIsab (50–25 b.c.e.). The base t ext of 1QIsa b was corrected only eight times towards a text that is now MT.47 These corrections involve letters or words added to the text and words deleted, probably by the scribe himself.48 Orthography (2): 41:7 ‫ = טוב‬MT. the prima manu word ‫ טב‬is not attested elsewhere in Scripture. 66:24 ‫ = האנשים‬MT. the p. m. word ‫ הנשים‬is probably a phonetic variant.

Erroneously omitted letters (4) 55:13 ‫ ספד‬corrected to ‫ = סרפד‬MT. MT ‫ סרפד‬is a hapax, so that a mistake is likely. 56:6 ‫ = בני הנכר‬MT. The p.m. reading may well reflect a variant, occurring also in Isa 60:10; 61:5; 62:8. For the corrected form, see ‫ בן הנכר‬in 56:3. 58:11 ‫ = וכמוצא‬MT. The p.m. reading is a clear error. 61:1 ‫ = שלחני‬MT. The p.m. reading is a mistake caused by the context that is replete with words starting with a lamed (no less than five consecutive words immediately following and one occurring two words before the present word).

Erroneously omitted words (2) 55:10 ‫ = כי‬MT added above the line before ‫כאשר‬. 65:24 ‫ = והיה‬MT added above the line before ‫טרם‬.

The eight presumed corrections (= MT) form a minute group as compared with a very large group of divergences from MT in this scroll that have not been corrected: 161 differences in orthography,49 many in sense divisions,50 and 183 in textual details.51 It is incomprehensible that with so many variants in the scroll, only eight details would have been singled out for correction towards MT. Rather, these instances of scribal errors were corrected to the Vorlage of the scroll that 47  The evidence is presented by D. Barthélemy, Critique textuelle de l’Ancien Testament, 3. Ézéchiel, Daniel et les 12 Prophètes (OBO 50/3; Fribourg/Göttingen: Éditions Universitaires/ Vandenhoeck & Ruprecht, 1992), cii–cxvi, translated posthumously in his Studies, 389–409 (405). Barthélemy speaks about nine instances, but one instance is not an approximation to MT: In 55:7 a word was erased before ‫ ;יעזב‬this is clearly a scribal error of a word erroneously written and immediately recognized by the scribe as faulty. 48  Thus E. Ulrich and P.W. Flint, Qumran Cave 1.II: The Isaiah Scrolls, Part 1: Plates and Transcriptions, Part 2: Introductions, Commentary, and Textual Variants (DJD XXXII; Oxford: Clarendon, 2010), II:200: “All the corrections appear to be penned by the original scribe; no later hands are detectable in the manuscript”. 49  See Ulrich/Flint, Isaiah, 200–4. 50  See ibid., 205–8. 51  See ibid., 208.

158

Emanuel Tov

was very close to the medieval text. Some of the instances listed above are clear instances of scribal errors.52 2. 4QJera (225–175 b.c.e.). DJD XV53 lists the following 31 corrections, all but four (12:4; 17:18; 18:19; 22:16) supralinear, subdivided into four groups. The uncorrected text often forms a very unlikely text (8:12; 17:10, 18; 18:19, 23). –  Orthography (1): Jer 9:14 –  Linguistic corrections (7): 2 phonetic corrections (14:6; 17:17), 5 different forms (4 supralinear corrections [9:11; 10:11; 17:21, 24]; 1 erased letter together with a supralinear letter [17:11]). –  Additions of details (19): a very large addition of 7:30–8:3 between the lines, in the margin, and under the text; smaller pluses (9) in 8:12; 13:5; 17:14, 16, 19, 19; 18:23; 22:12, 14; erasures (4) together with supralinear additions in 17:18, 19 and 22:3, 12; additions of single letters (5): 12:5; 17:10, 16, 19, 21. –  Erasures (4): 12:4; 17:18; 18:19; 22:16.

The 31 scribal interventions of 4QJera in 130 partially preserved lines present an average of one intervention in every four lines. Together with 1QIsaa this scroll is thus among the scrolls displaying the greatest amount of scribal intervention.54 This situation was created by careless copying visible in the large number of presumed scribal errors55 among these interventions.56 Beyond the corrected elements, the scroll differs from MT in 19 details in orthography and in 16 other details, all of which have not been corrected.57 Almost all the aforementioned corrections make 4QJera agree with MT in a text that is already close to MT. However, the fact that some of the corrections revise the base text away from MT (9:11; 17:11; 22:3) together with the fact that a larger number of deviations from MT have not been corrected suggest that the apparent corrections to MT actually correct scribal errors, probably in agreement with the scribe’s Vorlage.

52  The amount of scribal intervention in this scroll is that of an average biblical scroll: Based on a calculation of 564 fragmentarily preserved lines presented in DJD XXXII, we reckon with an average of one scribal intervention per 63 lines. 53  E. Tov, “4QJera”, in E. Ulrich et al., Qumran Cave 4.X: The Prophets (DJD XV; Oxford: Clarendon, 1997), 153. 54  See the comparative data in Tov, Scribal Practices, 331–5. 55  Tov, “Textual Base”, 309–10. 56  It is difficult to determine who inserted the corrections, the original scribe or a later one, because most corrections are small and often crammed in between the lines. However, the very long addition in 7:30–8:3 was added by a scribe other than the original one. See my analysis in DJD XV, 152. 57  See the apparatus in DJD XV. Lange, Handbuch, 299 calculates 74 variants.

The Possible Revision of Hebrew Texts According to MT

159

3. 4QGeng (75–50 b.c.e.). This scroll with 132 preserved words contains 4 variants in matters of content and 9 in orthography. It is closer to MT than to SP and the LXX. It also contains two corrections towards the text now included in MT: 1:14 ‫ אלהים‬added above the line between ‫ יהי‬and ‫ויאמר‬ 2:6–7 or 2:19 ‫וייצר‬

c. Scrolls that differ much from MT, and hence contain a much larger number of variants than groups (a) and (b). Since the corrections pertain only to a very small minority of the variants (mainly scribal errors) it cannot be claimed that they were made on the basis of MT. Merely two representative examples are provided. 1. 4QCantb (25–1 b.c.e.). In t his scrol l t hat differs much from MT58 only ‫עת‬, erroneously copied from 2:12b in v. 12a, was crossed out with a line, thus creating a text like MT. 2. 11QPsa (1–50 c.e.). In addit ion t o t he many variant s in t his scrol l , t here are 21 corrections, 8 in scribal errors (= MT),59 8 in matters of spelling (6 = MT,60 1 similar to MT,61 1 different from MT62), 3 in linguistic details (= MT),63 and 2 in content (= MT).64 The scribe of this scroll also erased letters or words in 28 scribal errors.65 The resulting text in these erasures almost always agrees with MT. Two tetragrammata canceled with cancellation dots in Ps 138:1 (col. XXI 2) and Ps 145:1 (col. XVI 7) seemingly point to corrections towards MT, but these instances should be viewed in light of other data. Both Psalm 138 and 145 make abundant use of the tetragrammaton, making it easy to understand that the scribe of this scroll was a little overzealous in the use of that word which was subsequently erased (= MT). 11QPsa contains so many variants vis-à-vis MT that it would not be logical that someone would have corrected the scroll towards MT in the few instances mentioned above. One correction does not agree with MT (XVI 15) making it likely that the scroll was corrected towards its Vorlage. This assumption is strengthened by the fact that the non-canonical segments of this scroll were corrected in the exact same way.66

58  In 32 details or groups of details as well as two very large omissions of verses. 59  Col. III 9; XII 13; XVI 10, 10; XX 17; XXIII 12; XXV 11, 12. 60  Col. VII 13; XVII 6; XX 9; XXIII 6; XXV 5, 9. For the details, see Tov, “Textual Base”, 312. 61  Col. XXI 6. 62  Col. VI 15. 63  Col. III 2; VIII 6; XVII 6. 64  Col. XIII 5, 6. 65  See the list in DJD IV, 14. 66  For example, XIX 8; XXIV 12, 13; XXVII 2; XXVIII 3, 11.

160

Emanuel Tov

Conclusions Correction towards MT has been substantiated well in the Greek Bible versions, in complete versions as well as in individual readings, but it has not been proven for Hebrew texts. The assumption of correcting Hebrew scrolls according to MT is not impossible, but is not borne out by the evidence relating to three groups of texts: (a) proto-MT scrolls closely resembling the medieval MT, (b) MT-like scrolls containing a larger number of variants than in group (a), (c) scrolls that differ much from MT. 1. A major problem for the assumption of revision according to MT is the correlation between the deviations from MT that were revised and those that were not revised. In group (b) the unrevised deviations are more numerous than the assumed corrections, while in group (c) they form an absolute majority. It is inconceivable that only a small segment of the deviations from MT was revised. Above I claimed that it would be exaggerated to invoke inconsistency in this case. 2. Most corrections agreeing with MT seem to be corrections of single scribal errors.67 It is particularly significant that the great majority of the corrections pertain to corrective pluses of elements left out. It is inconceivable that these are corrections to MT, since a process of correction to MT would have pertained to all textual phenomena. It is therefore noteworthy that differences in content and orthography have usually not been corrected. 3. If it were true that the early scroll 4QJera (225–175 b.c.e.) reflect s correct ions to MT, the corrective tendency would have started at a much earlier time than is generally assumed. Since there is no evidence other than this scroll for such an early revision towards MT, the corrections in this scroll, even by a different hand, probably display correction towards the scroll’s Vorlage and not towards MT. Correction towards MT, though theoretically possible, has not been substantiated in the scrolls that differ much from MT. The only group of texts for which this possibility cannot be refuted is in the scrolls of group (a) that from the outset are close to MT such as the proto-Masoretic scrolls from the Judean Desert. However, in the case of these scrolls, such as MurXII, it does not seem likely that this scroll that was already proto-Masoretic would have been corrected even more in the direction of MT. More likely, the scroll’s scribal errors were corrected according to its Vorlage.

67  Thus Tov, Scribal Practices, 223–5.

The Possible Revision of Hebrew Texts According to MT

161

Bibliography Aly, Z., Three Rolls of the Early Septuagint: Genesis and Deuteronomy (PTA 27; Bonn: Habelt, 1980). Barthélemy, D., Critique textuelle de l’Ancien Testament, 3. Ézéchiel, Daniel et les 12 Prophètes (OBO 50/3; Fribourg/Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 1992). Barthélemy, D., Les devanciers d’Aquila (VTSup 10; Leiden: Brill, 1963). Barthélemy, D., Studies in the Text of the Old Testament: An Introduction to the Hebrew Old Testament Text Project (Winona Lake, IN: Eisenbrauns, 2012). Benoit, P. & Crowfoot, G.M. & Milik, J.T. & de Vaux, R., Les grottes de Murabba‘ât (DJD II; Oxford: Clarendon, 1961). Ben Yashar, M. et al., The Bible in Rabbinic Interpretation: Rabbinic Derashot on Prophets and Writings in Talmudic and Midrashic Literature, Vol. 1, Hosea (Ramat Gan: Bar-Ilan University Press, 2003). Colomo D. & Henry, W.B., “5101. LXX, Psalms xxvi 9–14, xliv 4–8, xlvii 13–15, xlviii 6–21, xlix 2–16, lxiii 6–lxiv 5”, in A. Benaissa (ed.), The Oxyrhynchus Papyri (Graeco-Roman Memoirs 98; London: Egypt Exploration Society, 2011) 77.1–11. Dunand, F., Papyrus grecs bibliques (Papyrus F. Inv. 266): Volumina de la Genèse et du Deutéronome, Texte et planches (Extrait des études de papyrologie IX; Cairo: Institut francais d’archéologie orientale, 1966). Dunand, F., Papyrus grecs bibliques (Papyrus F. Inv. 266): Volumina de la Genèse et du Deutéronome, Introduction (Recherches d’archéologie, de philologie et d’histoire XXVII; Cairo: Institut francais d’archéologie orientale, 1966). Ego, B., “Targumim, Overview Article”, in A. Lange (ed.), Textual History of the Bible, vol. 1, § 1.3.3.1.2, forthcoming. Esh, S., “Variant Readings in Mediaeval Hebrew Commentaries: R. Samuel Ben Meir (Rashbam)”, Textus 5 (1966) 84–92. Goshen-Gottstein, M.H. & Talmon, S., The Hebrew University Bible: The Book of Ezekiel (Jerusalem: Magnes, 2004). Kahle, P., The Cairo Genizah (2nd edn; Oxford: Blackwell, 1959). Katz, P., The Aberrant Text of Bible Quotations in Some Philonic Writings and Its Place in the Textual History of the Greek Bible (Cambridge: University Press, 1950). Kutscher, E.Y., The Language and Linguistic Background of the Isaiah Scroll (1 Q Isa) (STDJ 6; Leiden: Brill, 1974). Lange, A., Handbuch der Textfunde vom Toten Meer, I: Die Handschriften biblischer Bücher von Qumran und den anderen Fundorten (Tübingen: Mohr Siebeck, 2009). Lange, A., “The Textual Plurality of Jewish Scriptures in the Second Temple Period in Light of the Dead Sea Scrolls”, in N. Dávid/A. Lange (ed.), Qumran and the Bible: Studying the Jewish and Christian Scriptures in Light of the Dead Sea Scrolls (CBET 57; Leuven: Peeters, 2010) 43–96. Lange, A., “‘They Confirmed the Reading’ (y. Ta‘an. 4:68a): The Textual Standardization of Jewish Scriptures in the Second Temple Period”, in A. Lange et al. (ed.), From Qumran to Aleppo:

162

Emanuel Tov

A Discussion with Emanuel Tov about the Textual History of Jewish Scriptures in Honor of His 65th Birthday (FRLANT 230; Göttingen: Vandenhoeck & Ruprecht, 2009) 29–80. Lust, J., “Major Divergences between LXX and MT in Ezekiel”, in A. Schenker (ed.), The Earliest Text of the Hebrew Bible: The Relationship between the Masoretic Text and the Hebrew Base of the Septuagint Reconsidered (SBLSCS 52; Atlanta: Scholars Press, 2003) 83–92. Lust, J., “The Ezekiel Text”, in Y.A.P. Goldman et al. (ed.), Sôfer Mahîr: Essays in Honour of Adrian Schenker Offered by Editors of Biblia Hebraica Quinta (VTSup 110; Leiden: Brill, 2006) 153–67. Penkower, J.S., Masorah and Text Criticism in the Early Modern Mediterranean: Moses ibn Zabara and Menahem de Lonzano (Jerusalem: Magnes, 2014). Penkower, J.S., “The Text of the Bible Used by Rashi as Reflected in His Biblical Commentaries”, in A. Grossman/S. Japhet (ed.), Rashi: The Man and His Work (Heb.; Jerusalem: The Zalman Shazar Center for Jewish History, 2008) 1.99–122. Rabin, C. & Talmon, S. & Tov, E., The Hebrew University Bible, The Book of Jeremiah (Jerusalem: Magnes, 1997). Rahlfs, A., Studie über den griechischen Text des Buches Ruth (MSU III,2; NGWG Phil.-Hist. Kl.; Göttingen: Weidmannsche Buchhandlung,1922). Sanderson J.E., “4QExodc”, in E. Ulrich/F.M. Cross (ed.), Qumran Cave 4.VII: Genesis to Numbers (DJD XII; Oxford: Clarendon, 1994) Smith, J., “The Text-Critical Significance of Oxyrhynchus Papyrus 5101 (Ra 2227) for the Old Greek Psalter”, BIOSCS 45 (2012) 5–22. Talmon S. & Yadin, Y., Masada VI, The Yigael Yadin Excavations 1963–1965, Final Reports, Hebrew Fragments from Masada (Jerusalem: Israel Exploration Society/The Hebrew University of Jerusalem, 1999) 68–72. Tov, E., “4QJera”, in E. Ulrich et al. (ed.), Qumran Cave 4.X: The Prophets (DJD XV; Oxford: Clarendon, 1997). Tov, E., “The Use of the Earliest Greek Scripture Fragments in Text Editions”, forthcoming. Tov, E., Scribal Practices and Approaches Reflected in the Texts Found in the Judean Desert (STDJ 54; Leiden: Brill, 2004). Tov, E., “Some Thoughts about the Diffusion of Biblical Manuscripts in Antiquity”, in S. Metso et al. (ed.), Transmission of Traditions and Production of Texts (STDJ 92; Leiden: Brill, 2010) 151–72. Revised version: Textual Criticism of the Hebrew Bible, Qumran, Septuagint: Collected Writings, Volume 3 (VTSup 167; Leiden: Brill, 2015) 60–81. Tov, E., Textual Criticism of the Hebrew Bible (3rd, revised and expanded edn; Minneapolis: Fortress Press, 2012). Tov, E., “The Greek Biblical Texts from the Judean Desert”, in S. McKendrick/O.A. O’Sullivan (ed.), The Bible as Book: The Transmission of the Greek Text (London: British Library and Oak Knoll Press/The Scriptorium: Center for Christian Antiquities, 2003) 97–122. Revised version: Hebrew Bible, Greek Bible, and Qumran: Collected Essays (TSAJ 121; Tübingen: Mohr Siebeck, 2008) 339–64.

The Possible Revision of Hebrew Texts According to MT

163

Tov, E., “The Myth of the Stabilization of the Text of Hebrew Scripture”, in E. Martín Contreras/L. Miralles-Maciá (ed.), The Text of the Hebrew Bible: From the Rabbis to Masoretes (JAJSup 13; Göttingen: Vandenhoeck & Ruprecht, 2014) 37–45. Tov, E., The Septuagint Translation of Jeremiah and Baruch: A Discussion of an Early Revision of Jeremiah 29–52 and Baruch 1:1–3:8 (HSM 8; Missoula, MT: Scholars Press, 1976). Ulrich, E., The Dead Sea Scrolls and the Developmental Composition of the Bible (VTSup 169; Leiden, 2015). Ulrich E. & Flint, P.W., Qumran Cave 1.II: The Isaiah Scrolls, Part 1: Plates and Transcriptions, Part 2: Introductions, Commentary, and Textual Variants (DJD XXXII; Oxford: Clarendon, 2010). Weiss, R., “Recensional Variations between the Aramaic Translation to Job from Qumran Cave 11 and the Massoretic Text”, Shnaton 1 (Heb. with Eng. summ.; Jerusalem, 1975) 123–7. Weiss, R., The Aramaic Targum of Job (Heb. with Eng. summ.; Tel Aviv: Tel Aviv University, 1979). Wevers, J.W., “Pre-Origen Recensional Activity in the Greek Exodus”, in D. Fraenkel et al. (ed.), Studien zur Septuaginta: Robert Hanhart zu Ehren (MSU XX; Göttingen: Vandenhoeck & Ruprecht, 1990) 64–73. Wevers, J.W., Text History of the Greek Deuteronomy (AKWG, Phil.-Hist. Kl. 3, 106; MSU XVI; Göttingen: Vandenhoeck & Ruprecht, 1978). Wevers, J.W., Text History of the Greek Numbers (AKWG, Phil.-Hist. Kl. 3, 125; MSU XII; Göttingen: Vandenhoeck & Ruprecht, 1982). van der Woude, A.S., “Pluriformity and Uniformity: Reflections on the Transmission of the Text of the Old Testament”, in J.N. Brenner/F. García Martínez (ed.), Sacred History and Sacred Texts in Early Judaism: A Symposium in Honour of A.S. van der Woude (Kampen: Kok Pharos, 1992), 151–69. Ziegler, J., Ieremias, Baruch, Threni, Epistula Ieremiae (Septuaginta, Vetus Testamentum Graecum auctoritate societas litterarum gottingensis editum, XV; Göttingen: Vandenhoeck & Ruprecht).

Anneli Aejmelaeus Rewriting David and Goliath? Almost a quarter of a century ago, I dealt with the story of David and Goliath in my inaugural lecture at the University of Göttingen, using it as an example of a case in which the Septuagint reveals to us a form of the text that is considerably shorter and earlier than the MT.1 The idea that something in the MT could be secondary, although represented by some scholars more than a century earlier, was not popular in those days. I do not know if I convinced anyone in my audience. Afterwards the wife of a colleague came to me and expressed her opinion that the motif of a small shepherd boy who slays the bad giant, present in the longer text, is so essential to the story that it cannot have been added later. I shall come back to the motif of the shepherd boy, which is important, although I do not think it proves the originality of the longer version. The question whether the story was shortened or expanded was however the main question at that time, and many colleagues in the field had difficulty, as many still do, in accepting that the MT could contain such extensive and late changes. At that time, already four decades after Qumran, new attitudes to the MT and to textual study of the Hebrew Bible were only beginning to dawn on the majority of scholars. Today, more than six decades after the Qumran discoveries and well after the completion of the publication of these findings, the situation is somewhat, but not totally, different. Take for instance the most recent commentaries on the Books of Samuel. Walter Dietrich, writing for the Biblischer Kommentar series, has a very complicated explanation for the growth of the story, in which the Septuagint plays no role at all.2 According to Dietrich, the shorter text is a shortened and simplified form of the complicated long story. By contrast, the commentary of Graeme Auld represents a totally different view: the shorter text of the Septuagint, or more correctly the Vorlage of the Septuagint translation, reveals an earlier phase in the development of the text.3 It is clear that the witness of the Septuagint does not explain everything, but it cannot simply be put aside: the evidence needs to be taken into account. The two forms of the text that are extant represent two different stages in its development—not the first and the second edition, but

1  See “Übersetzung als Schlüssel zum Original”, in A. Aejmelaeus, On the Trail of the Septuagint Translators: Collected Essays (Contributions to Biblical Exegesis & Theology 50; Leuven: Peeters, 2007) 143–56. 2 W. Dietrich, Samuel (Biblischer Kommentar: Altes Testament; VIII/24 [16,1–17,58], VIII/25 [17,1—18,30], and VIII/26 [18,1—19,24]; Neukirchener: Neukirchen, 2012–2013). 3  A.G. Auld, I–II Samuel: A Commentary (The Old Testament Library; Louisville, KY: Westminster John Knox, 2011).

166

Anneli Aejmelaeus

more probably snapshots of two fairly late phases in the gradual evolution of the book. Thus, the story of David and Goliath continues to divide scholars into different camps, as it has done long before. It may be interesting to note that Julius Wellhausen, one of the early proponents of the priority of the shorter text, was later hesitant about the matter and gave different views in different editions of his Die Composition des Hexateuchs und der historischen Bücher des A.T. (2nd edn 1889; 3rd edn 1899). His study on the text of the Books of Samuel (1871)—in which he deals with the matter most extensively—speaks for the priority of the shorter text.4 In this paper, it is however not my aim to present the arguments for the priority of the shorter form of the story. Others have done that convincingly.5 Instead, I shall concentrate on the question of how the longer version came about: Where did the additions come from? Could the emergence of the longer form of the story perhaps be explained through the strategy and technique of “rewriting”? The idea of “rewritten scripture” is fairly new in scholarship, having been lively debated during the last decade, although the phenomenon was introduced by Geza Vermes already in the 1960s.6 By now, “rewriting” has become a permanent topic of our discussions on the development of sacred texts and sacred literature in the Second Temple period. My question is, does it apply to David and Goliath?

The Evidence Concerning the two Different Versions The story of David and Goliath and its aftermath comprises the whole two chapters of 1 Sam 17 and 18, the longer version exceeding the shorter one by ca. 40%. My aim is not to discuss the whole text in great detail, but it is good to have an overall view of it, so that we know what we are talking about. The basic facts about the Greek evidence are the following: (1) There are two more extensive blocks of text and numerous smaller additions (from a few lines to single words) that are not present in the original translation of the Septuagint.7 (2) In the ­textual 4 J. Wellhausen, Der Text der Bücher Samuelis (Göttingen: Vandenhoeck & Ruprecht, 1871). 5  See, for instance, the contributions of Emanuel Tov and Johan Lust in D. Barthélemy/ D.W. Gooding/J. Lust/E. Tov, The Story of David and Goliath (OBO 73, 1986), and E. Tov, “The Composition of 1 Samuel 16–18 in the Light of the Septuagint”, in The Greek and Hebrew Bible: Collected Essays on the Septuagint (VTS 72; Leiden: Brill, 1999) 333–62. 6  G. Vermes, Scripture and Tradition in Judaism: Haggadic Studies (Studia Post-Biblica 4; Leiden: Brill, 1961). 7  The two longer passages not present in the Septuagint are: 1 Sam 17:12–31 and 17:55– 18:5; whole sentences are lacking in vv. 17:37, 38, 41, 42, 48, 50, 51; 18:6, 8, 10–11, 12, 17–19, 21, 26, 27, 28, 29–30.

Rewriting David and Goliath?

167

transmission of the Septuagint, we happen to have a group of 12 manuscripts, including Vaticanus (B V 119-527-799 121 29 71 244 245 460 707), which do not have any of these additions. (3) The majority of the Greek manuscripts have secondarily complemented them, whereas not all manuscripts have all of them. It seems that the smaller complements sometimes escaped the attention of the scribes. (4) The translation in these secondary complements is clearly different from the mode of translation in 1 Samuel otherwise and the various manuscript groups show also different formulations in certain details of the additions. (5) In some details, the Greek complements even seem to have had a Vorlage different from the MT (for 17:12 see below). (6) On the other hand, the Greek text also shows a few details that are not present in the MT (17:32 ‫ אדני‬instead of ‫ ;אדם‬for 17:36 and 43, see below). The most important of such variants and pluses will be briefly discussed below. Let us have a brief review of the contents of the two different versions of the text we are discussing. The different versions of the story of David and Goliath deal with the beginning of David’s career in the service of Saul. How did he come to Saul’s court? The preceding ch. 16 already tells that Saul’s servants recommend David, the son of Jesse from Bethlehem, to Saul as “a skillful musician, a mighty man of valor, a warrior, one prudent in speech, and a handsome man” (‫י ֵ ֹ֣ד ַע ַ֠נּגֵ ן וְ גִ ּ֙בֹור ַ֜חיִ ל וְ ִ ֧איׁש‬ ‫ּת ַאר‬ ֹ ֑ ‫ ִמ ְל ָח ָ ֛מה ּונְ ֥בֹון ָּד ָ ֖בר וְ ִ ֣איׁש‬16:18). David is brought to Saul, who is immediately attracted by him and makes him his armour-bearer. The motif of David playing the harp must be a later reworking of this passage; it is most peculiar that a musician would need to be recommended as “a mighty man of valor, a warrior, one prudent in speech, and a handsome man” and additionally to become an armourbearer. The oldest story of David’s arrival at Saul’s court was probably about the grown-up warrior David, and this story connects with 1 Sam 14:52, which says: 1 Sam 14:52 ‫ּבֹור‬ ֙ ִ‫ל־איׁש ּג‬ ֤ ִ ‫יְמי ָׁש ֑אּול וְ ָר ָ֙אה ָׁש ֜אּול ָּכ‬ ֣ ֵ ‫ל־ּפ ִל ְׁש ִּ֔תים ּ֖כֹל‬ ְ ‫וַ ְּת ִ ֤הי ַה ִּמ ְל ָח ָמ ֙ה ֲחז� ָ ָ֣קה ַע‬

‫ן־חיִ ל וַ ּיַ ַא ְס ֵ ֖פהּו ֵא ָ ֽליו‬ ַ֔ ‫ל־ּב‬ ֶ ‫וְ ָכ‬

The war against the Philistines was severe all the days of Saul, and when Saul saw any mighty man, and any valiant man, then he took them to himself.

So he took the valiant man, David, to be his armour-bearer. This is the starting point for the shorter story of the fight with Goliath: David is in Saul’s service and standing on his side when Goliath makes his challenge. As such, the shorter story connects to the preceding text without difficulty and does not seem to have any major discrepancies or to be lacking anything. By the way, speaking of the oldest form of the story, the slaying of Goliath was most probably altogether secondarily attributed to David. According to 2 Sam 21:19 it was Elhanan from Bethlehem who killed Goliath—probably the same

168

Anneli Aejmelaeus

“Goliath the Gittite, the shaft of whose spear was like a weaver’s beam” (‫ֵ ֚את ּגָ ְל ָי֣ת‬ ‫יתֹו ִּכ ְמנ֖ ֹור א ְֹר ִ ֽגים‬ ֔ ִ‫)הּגִ ִּ֔תי וְ ֵע֣ץ ֲחנ‬. ַ 8 But for some reason, the story was connected with David, whose career thus gets a glorious beginning—all the more glorious when this “mighty man of valor” is by and by depicted as an inexperienced youth. The curious detail in 1 Sam 17:54 that David brings the head of Goliath to Jerusalem, which has not yet been conquered, may perhaps stem from the original connection. The first and longest expansion of the MT (1 Sam 17:12–31) begins with an introduction of David and his father Jesse. This first joint between the old and new material is confusing, as the characters have been mentioned before. It is presupposed that David is not staying permanently with Saul, but time and again goes back to his father to tend his sheep, and this is also expressed in v. 15, which thus functions as a link to the shorter version. The father happens to send David to see his brothers on the battle field (vv. 17–19), and so he arrives, hears Goliath challenge the Israelites (vv. 23–24), makes queries about the situation, and argues with his elder brother (vv. 28–29) before volunteering to fight with the giant. The connection with the shorter story is not very skillful: the expansion actually has the character of a flashback and could have been built in as such, but now Goliath is said to have repeated his challenge during forty days (v. 16) in order to give David time to appear. This unnecessary feature is obviously meant as a further link to the shorter story, but it really makes the story absurd: for forty days, morning and evening, Goliath makes his challenge, and every time the Israelites are as surprised and scared. The actual fight does not show big differences. The longer version adds a bit more action describing how the two parties approach each other (vv. 41 and 48), which perhaps creates excitement and emphasizes the contrast between the giant and David, who is much smaller and quicker in his movements. David uses the weapons of a shepherd and says that he used to tend the flocks of his father. Most noteworthy in the common middle section are the dialogues of David, first with Saul (vv. 32–37) and then with Goliath (vv. 43–47). David’s first answer to Goliath is found only in the Septuagint (“No, but worse than a dog!”), but I am sure it is original to the story and was removed from the MT on purpose—perhaps because its rough tone was not considered suitable for the shepherd boy David. The other plus of the Septuagint (v. 36) is probably also original in the shorter version; it may have been removed because the same formulation occurs in v. 26 of the longer version (“slay him and remove the reproach from Israel, for who is this uncircumcised that…”).9 The theological motive of both the shorter and 8  1 Chr 20:5 is a variant form of 2 Sam 21:19, which secondarily removes the discrepancy. 9  In both cases, the plus of the Septuagint connects with details of the preceding story, and thus proves to be part of the shorter version (that is, the common content). At 1 Sam 16:18 David is described as a man “skilful in speech”, and the Septuagint plus in v. 43 is an example

Rewriting David and Goliath?

169

the longer version are given here in the often quoted vv. 46–47: “…that all the earth may know that there is a God in Israel (cf. 1 Kings 18:36), and that all this assembly may know that the Lord does not deliver by sword or by spear; for the battle is the Lord’s and He will give you into our hands.”10 This is not a speech of a young boy seeking adventure. The theology of these verses would certainly deserve another paper.11 After the slaying of Goliath, the shorter story continues almost immediately with the song of the women (1 Sam 18:6): “Saul has slain his thousands, and David his ten thousands.” The song actually presupposes that David has already before his fight with Goliath spent some time as Saul’s armour-bearer and taken part in the war against the Philistines. From here on, Saul becomes more and more suspicious against David (already in the shorter version). Not long before, he was strongly attracted by David (16:21–22) and now he wishes to get rid of him, making him a commander of a thousand (in the shorter version 18:13). For the same purpose, he offers his daughter Michal to him in marriage on the brideprice of a hundred Philistine foreskins (18:25). As Saul’s aversion grows, so does David’s success. At the end, everyone loves David (18:28). By contrast, the second large expansion, 1 Sam 17:55–18:6, creates more distance between the fight with Goliath and Saul’s growing displeasure, but it is also very puzzling, as David appears to be unknown to Saul and his court. “Whose son is this young man?” he asks and, since no one knows it, he has to ask David himself (v. 58). This is generally seen as the most serious contradiction caused by the additions. The scribe who added this part obviously had no idea about David’s earlier connection to Saul. He must have totally ignored the story of David becoming Saul’s armour-bearer at 16:18–22 and the mention of Jesse in that connection. Further additions in ch. 18 deal with Saul’s growing suspicion. David’s constant success and Jonathan’s love for him give Saul more reason to see a threat in David. He even attempts to kill him (18:10–11), which actually comes a bit too early in the story (the original at 1 Sam 19:9–10 is much more appropriate)! He promises his elder daughter Merab to David in marriage, on the condition that he continues “to fight the Lord’s battles” (cf. 1 Sam 25:28), only hoping that the Philistines would take care of him. When the time comes, Merab is however given to someone else.

of that. At 1 Sam 17:10 Goliath reproaches the ranks of Israel, and in v. 36 David expresses his aim to remove that reproach. 10  Cf. 2 Chr 20:15; 1QM (1Q33) XI 1–2. 11  The motif of trusting in God instead of weapons has its roots deep in the Hebrew Bible: Ex 14:14; Hos 1:7; Isa 31:1; Zech 4:6; Ps 20:8; 44:7–8. On the other hand, the divine epithet “the living God” (1 Sam 17:26, 36) is late: it is found otherwise only at Deut 5:26 and in Jer 10:10; 23:36 in additions of the MT.

170

Anneli Aejmelaeus

Where is the Origin of the Additions? There are mainly two models that have been used to explain the origins of the longer form of the story. The first one presupposes an alternative, independent story from which the supplementary parts were taken. The second model presupposes that the additions did not have an independent existence but were written into the text of 1 Samuel as a kind of “recomposition” or “rewriting”. The theory of an independent story can be traced back to the name of Julius Wellhausen,12 and has been represented by most scholars who are for the priority of the shorter story. The theory of “recomposition” was more recently presented by A. Graeme Auld and Craig Y.S. Ho.13 The additional considerations concerning the strategy of “rewriting” are my own.

Independent Alternative Account Let us start with the theory of an independent alternative account of the fight between David and Goliath. According to Julius Wellhausen the story was complemented from a “Flugblatt” (a “fly-sheet” or “flyer”), which seems to refer to a written source. This is a kind of source-critical solution, in the spirit of the 19th century. Wellhausen thought that the alternative story was created on the basis of 17:34 (present in the shorter text) in which David says that he “had been tending his father’s sheep”. According to v. 40 he also uses a shepherd’s bag to carry the stones. Wellhausen points out that David speaks in the past tense, when he refers to his experience with lions and bears while tending the sheep. In the shorter text, being a shepherd was clearly in the past, David having become Saul’s armourbearer. In the so-called alternative account, however, David is still a shepherd boy. Verse 17:15 that allows David to commute between Bethlehem and Saul’s court was probably created to smooth out the discrepancy. Wellhausen’s source-critical solution has been followed with small variations by many commentators. Of the more recent commentaries, the same solution is represented by Kyle McCarter (Anchor Bible, 1980).14 He presents the translation of the additional parts of the MT separately from the primary, shorter narrative and argues that the pieces fit together so well that they must form an independent alternative account. However, according to McCarter, this alternative story is perhaps not quoted in its entirety in those parts in which there is an overlap with the primary narrative. McCarter thinks that the alternative account was combined with the primary narrative fairly late—he says in the 4th century b.c.e.,

12  J. Wellhausen, Der Text der Bücher Samuelis, 105. 13  A.G. Auld/C.Y.S. Ho, “The Making of David and Goliath”, JSOT 56 (1992) 19–39. 14  K. McCarter, I Samuel (Anchor Bible, 1980), 399–409.

Rewriting David and Goliath?

171

which some of us might consider fairly early—and he reasons that it may even have circulated before that independently for some time. This theory was further elaborated by Johan Lust in the volume The Story of David and Goliath (OBO 73, 1986),15 which consists of contributions by four scholars, Dominique Barthélemy, David W. Gooding, Johan Lust, and Emanuel Tov, all with differing views on the problem. Of the four scholars, Lust and Tov represent the priority of the shorter text. According to Lust, the additions in ch. 17 (that is, verses 12–31 plus 55–58 plus one verse 18:2) contain “relics of a narrative that once was the opening story of the history of David”.16 Indeed, the introduction of the characters that opens the first long addition shows similarity with the introductions of Samuel’s and Saul’s fathers in the beginning of the respective sections (1 Sam 1 and 9). Lust also argues that it must have been originally an introduction of David’s father (v. 12)—without however noticing that there are a few manuscripts that might support that view. Instead of ‫ן־איׁש ֶא ְפ ָר ִ֜תי ַה ֶּ֗זה‬ ִ֙ ‫וְ ָדוִ ד֩ ֶּב‬ ‫הּודה‬ ֔ ָ ְ‫מ ֵ ּ֥בית ֶ ֙ל ֶח ֙ם י‬, ִ the Hexaplaric group O (= 247–376) has και ην ανθρωπος εκ Βεεθλεεμ Ιουδα, which presupposes the Hebrew text ‫והיה איש מבית לחם יהודה‬. To begin a story with the conjunction and a proper name, as in the MT, is not normal in classical Hebrew. Imperfect consecutive ‫ ויהי‬is what is most often found at the beginning. But if it happened to be ‫( והיה‬cf. 17:48), it might have been possible to confuse it with ‫—ודוד‬at least in a late script, like that of 4QSama, this seems to be possible. This would mean that ‫ ֶּבן‬and ‫ ֶא ְפ ָר ִ֜תי ַה ֶּ֗זה‬were added to change the introduction of the father to that of David,17 whereas the O group is following an earlier form, represented possibly by, for instance, Theodotion. As I already mentioned, it seems that on a few points the Greek translations of the additional parts presuppose a Hebrew text different from the MT, so that it would not be extraordinary to presuppose that the beginning was originally different. This is not however as weighty an argument as Lust seems to think. The changed formulation of the beginning does not prove that the story had an independent existence. What is extraordinary in Lust’s explanation is that he finds the alternative account of the MT additions to be the older one of the two versions. In this older story, he sees features of a “romantic epic” or a “fairy tale”, whereas the shorter story is the later one and is called a “heroic epic”. Lust suggests that the alternative story, the “romantic epic”, was first the only one given in ch. 17 and that it was at some stage replaced by the “heroic epic”—this is the form that was translated into 15  J. Lust, “The Story of David and Goliath in Hebrew and Greek” and “Second Thoughts on David and Goliath”, in D. Barthélemy/D.W. Gooding/J. Lust/E. Tov, The Story of David and Goliath (OBO 73, 1986) 5–18 and 87–91. 16  Lust, ibid. 13. 17  By many, ‫ הזה‬is considered to be grammatically suspect (see McCarter, I Samuel, 301). The odd expression might be explained by its late origin.

172

Anneli Aejmelaeus

the Septuagint—and later on, another redactor added the removed text again and thus combined both stories. With this very complicated solution Lust, in a way, actually gives the priority to both versions. A solution like this really calls for Occam’s razor! Emanuel Tov, who was one of the four scholars who published the volume The Story of David and Goliath, also represents a variation of the theory of an independent alternative story.18 In his contribution, he shows that the Greek additions cannot have been part of the original Greek translation. A few years later, he published another extensive article in which he goes through the whole text again in great detail.19 He seems however to be mainly occupied by the argumentation for the priority of the shorter text—and the impossibility shortening the text—so that he does not so much deal with the question of the origins of the additions. Like so many before him, Tov presupposes that the alternative account once existed as a written source, from which the redactor excerpted details that he liked and considered worth preserving, possibly motivated by the idea that “God can bring victory to his people even through initially unimportant figures.”20 This idea is however already present in the shorter version. There is one more scholar to be mentioned. Gene Ulrich has frequently used the story of David and Goliath as an example of the pluriformity of the biblical text in which the MT and the Septuagint witness “variant literary editions”, the Septuagint representing an earlier form of the text and the MT an intentionally expanded, revised edition.21 As a matter of fact, the term “rewriting” has also been applied by Ulrich in this connection. In a recent article of his, the story of David and Goliath is given as evidence for “the Rewriting That Produced Revised Editions”.22 Ulrich’s main point of interest is the evidence provided by our story for his theory explaining the development of the biblical text, so that he does not deal more extensively with the question where the expansions of the MT came from. He refers either to alternate David traditions that were preserved in the pluses or to the insertion of “components of a second version of the story quite different in content, details, and style”.23 18  E. Tov, “The Nature of the Differences between MT and the LXX”, in The Story of David and Goliath (OBO 73, 1986) 19–46. 19  Tov, “The Composition of 1 Samuel 16–18”. 20  Tov, ibid. 355. 21  E. Ulrich, “Crossing the Borders from ‘Pre-Scripture’ to Scripture (Rewritten) to ‘Rewritten Scripture’”, in J. Zsengellér (ed.), Rewritten Bible after Fifty Years: Texts, Terms, or Techniques? A Last Dialogue with Geza Vermes (JSJS 166; Leiden: Brill, 2014) 83–104. See also idem, The Dead Sea Scrolls and the Origins of the Bible (Studies in the Dead Sea Scrolls and Related Literature; Grand Rapids, MI: Eerdmans/Leiden: Brill, 1999), esp. pp. 38, 41, 72. 22  Ulrich, “Crossing the Borders from ’Pre-Scripture’ to Scripture (Rewritten) to ’Rewritten Scripture’”, 87. 23 Ulrich, The Dead Sea Scrolls and the Origins of the Bible, 38.

Rewriting David and Goliath?

173

“Recomposition” or “Rewriting” We now come to the second model, the “recomposition” theory, according to which the additions were written as complements to the shorter story within the First Book of Samuel. The first serious challenge, in my knowledge, to the “alternative account” theory, represented by the mentioned prominent scholars, was in the article “The Making of David and Goliath” by Graeme Auld and Craig Ho (1992).24 They do not see “enough evidence for the existence of another similar, once-independent story” that would be fragmentarily represented in the additions of the MT. Rather, they suggest that the longer story was “a literary creation by a redactor out of existing material in 1 Samuel”.25 They refer to the observation of previous scholars that parallels and contrasts with similar stories are a frequent phenomenon in the Books of Samuel and draw the conclusion: “A book that is full of parallel structures might have attracted further parallel details to enhance the literary effect.”26 For the practice of incorporating interpretations into the original text of various genres in the Hebrew Bible, they make a very general reference to Michael Fishbane’s Biblical Interpretation in Ancient Israel (1985).27 In their analysis of chs. 17–18, Auld and Ho discuss correspondences between the stories of Saul and the narrative items added in the longer version of our text. They begin with the introduction of “a man and his son” (17:12), which is parallel to the introduction of Saul and his father (9:1–2), but in what follows they see a contrast between the two young men: Saul was the most handsome and the tallest man in Israel, expected to become something special, whereas no one expects anything from the smallest of Jesse’s sons. Another parallel is seen in the errand that the father gives to his son (9:3 and 17:17–18); in this case, the contrast is that Saul does not succeed in fulfilling his task, as he does not find the lost asses, but David accomplishes his task in a splendid manner. Both young men ask questions (9:7 and 17:26), which characterize Saul as “a brainless man” but David as someone actively “showing initiative”. Both young men are brought into contact with the leader of the nation by their small errands. Here the contrast is that Saul is offered the kingship, whereas David is offered a royal status as the king’s son-in-law, which is however meant as a trap. Both of them 24  Auld/Ho, “The Making of David and Goliath”. Graeme Auld represents the same view in his recent commentary I–II Samuel: A Commentary (The Old Testament Library; Louisville, KY: Westminster John Knox: 2011), but since he does not develop it further there, I am referring to Auld and Ho. 25  Auld/Ho, ibid. 24. 26  Auld/Ho, ibid. 24–25. Reference is made to J.P. Fokkelman, Narrative Art and Poetry in the Books of Samuel: A Full Interpretation Based on Stylistic and Structural Analysis. Volume II: The Crossing Fates (1 Sam. 13–31 & 2 Sam. 1) (Assen: Van Gorcum, 1986). 27  M. Fishbane, Biblical Interpretation in Ancient Israel (Oxford: Clarendon, 1985).

174

Anneli Aejmelaeus

humbly belittle themselves, Saul referring to his family—which is the smallest in Benjamin, the smallest tribe of Israel (9:21)—and David asking, “Who am I and who is my family that I should become the king’s son-in-law?” (18:18). A parallel is seen also in the meal that Saul takes part in (9:23–24), in that it is interpreted as a covenant meal, whereas there is a covenant—without a meal—between David and Jonathan (18:3). Here, at the latest, it becomes evident that some of the parallels listed by Auld and Ho are fairly artificial, concentrating mainly on formal details. The sacrificial meal in ch. 9 actually parallels more suitably with the story of David’s anointing in 16:13, which in fact contains other parallels, above all that David is anointed by Samuel, just like Saul was (10:1 and 16:13). Nevertheless, I believe that Auld and Ho were on the right track, although their theory needs some fine-tuning.

Defining “Rewriting” as Exegesis and Gap-Filling That it is a question of a “recomposition of an original story”, as Auld and Ho suggest, or “rewriting”, is a much more powerful solution than the theory of an independent alternative account. There are clearly parallels, not just with the stories about Saul, but also with many other texts. However, I see a weakness in the theory of Auld and Ho, in that demonstrating parallels or even contrasts between the two first kings is certainly not enough to motivate such literary activity. The parallels are not the goal—but rather the means for attaining the goal, which is to highlight and complement certain features of the story. The overall tendency is of course that David grows in glory, whereas Saul, the rejected king, is presented in ever darker tones. The motivation behind the “rewriting” must however have been of the kind typical for “rewriting” in general, that is, exegesis and eisegesis, explanation of problematic items in the narrative and gap-filling. When discussing this kind of rewriting within the later parts of the Hebrew Bible, Geza Vermes speaks of “a midrashic process”.28 I think, this term “midrashic process” applies very well to the story of David and Goliath. According to most exegetes, the shorter version of the story of David and Goliath is consistent in itself, whereas the additions bring along contradictions. However, where do we find a composite text totally without contradictions? The method of literary criticism is based on these contradictions! On the other hand, a smooth story like the shorter version is by no means without problems ­calling 28  G. Vermes, “Bible and Midrash: Early Old Testament Exegesis”, in P.R. Ackroyd/C.F. Evans (ed.), The Cambridge History of the Bible (1970) 199–231: “Post-biblical midrash is to be distinguished from the biblical only by an external factor, canonization.” M. Fishbane, Biblical Interpretation in Ancient Israel (1985), uses the term “aggadic exegesis”.

Rewriting David and Goliath?

175

for exegesis. The story of the beginning of David’s career is of course a very important narrative, one that inspired the interpreters to give answers to all the questions that it might raise. (1) An important key is the motif of the shepherd boy, which already plays a role in the shorter narrative and even more clearly in the long addition in ch. 17. None of the scholars I have referred to discuss this motif in greater detail. I shall deal with this motif in a moment. (2) There are also gaps that need to be filled in concerning Saul’s growing aversion against David. For instance, the shorter story does not give the full picture of the reasons for the change in Saul’s attitude. In the longer version, David’s increasing success arouses Saul’s jealousy and the covenant with Jonathan is clearly a cause of rage for Saul (as we can read in 20:30–31).29 It is Saul’s dynasty that is threatened by David (which is expressed in the plus at the end of 18:8). However, (3) we can also observe that not all the additions would have been absolutely necessary from the point of view of David’s career and need not have been added at the same time or out of the same motivation. For instance, concerning Saul’s daughters, there might have been the question why the elder daughter was not offered first, which reminds us of Jacob’s marriage with Leah before Rachel. There were however older stories about David’s marriage with Michal only, so that Saul’s offer of Merab had to end without a marriage. The marriage to the king’s daughter is often seen as the reward promised for slaying the giant (17:25). It is however more plausible that the idea of the reward came from the older stories about David’s marriage to Michal.30 Of the mentioned three aspects, I am only able to develop here the first-­ mentioned.

The Shepherd Motif as the Key Let us look a bit more closely at the shepherd motif, which I think is the key. The story of David and Goliath was developed in several steps in the direction that David becomes younger and younger and is finally just a small shepherd boy. In the shorter text, as Wellhausen noted, David says that he “had been shepherding” his father’s sheep—but no longer is.31 The shorter text, however, already includes a feature that makes David appear as an inexperienced youth: the verses that describe how Saul wants to equip David with his own armour 29  David refers to a covenant at 20:8, to which the longer text provides a point of reference. 30  Also the verses that serve to adjust the additions to the older story (17:15 David commuting between Saul and his father and its pair 18:2 making an end to it—which disturb the connection between the surrounding verses—and 17:16 according to which Goliath makes his challenge during forty days) may have been added at a later stage. 31 Wellhausen, Der Text der Bücher Samuelis, 105.

176

Anneli Aejmelaeus

and weapons (17:38–39). This looks like a later insertion in the shorter text, because Saul already sent David off and gave him his blessing in the previous verse. In the long expansion (17:12–31), this development is taken still further, to the point that David is just a small boy. Whether a shepherd boy or warrior, whether long or short version, it is of course very clever of David to slay the giant from afar with the sling and the stone and not to come to close combat with him. However, in order to give the glory to the God of Israel, who rescues the helpless and gives victory over powerful enemies, David had to become smaller, although this creates a problem for the story-teller: How could it happen that a small shepherd boy was on the battle field when Goliath made his challenge? To solve this problem the editor of the longer version used an old narrative motif. The shepherd boy David was, of course, sent on an errand by his father. There is however a still deeper reason for letting David appear as a shepherd. The shepherd motif appears in prominent passages concerning David’s career. In 2 Sam 7, David receives the promise of an eternal dynasty, and in this connection the Lord says to David through Nathan: 2 Sam 7:8 ‫ל־ע ִ ּ֖מי ַעל־יִ ְׂש ָר ֵ ֽאל‬ ַ ‫ן־הּנָ ֶ ֔וה ֵמ ַא ַ ֖חר ַה ּ֑צ ֹאן ִ ֽל ְהי֣ ֹות נָ ֔ ִגיד ַע‬ ַ ‫ֲא ִנ֤י ְל ַק ְח ִּ֙ת ֙יָך ִמ‬ I took you from the pasture, from following the sheep, that you should be ruler over my people Israel.

Also 2 Sam 5, which gives an account of the election and anointing of David by the Israelite elders, contains the shepherd motif. The elders refer to David’s early career in 1 Sam 18 and say: 2 Sam 5:2 ‫מֹוציא וְ ַה ֵּמ ִבי‬ ִ ‫יתה‬ ָ ִ‫היֹות ָׁש ֥אּול ֶ֙מ ֶל ְ֙ך ָע ֔ ֵלינּו ַא ָּ֗תה ָהי‬ ֙ ְ ‫ם־ׁש ְל ׁ֗שֹום ִּב‬ ִ ַ‫ם־א ְת ֣מֹול ּג‬ ֶ ַ‫ּג‬

‫ּמי ֶאת־יִ ְׂש ָר ֵ֔אל וְ ַא ָ ּ֛תה ִּת ְהֶי֥ה ְלנָ ִג֖יד ַעל־יִ ְׂש ָר ֵ ֽאל׃‬ ֙ ִ ‫ת־ע‬ ַ ‫הוה ְל ָ֗ך ַא ָּ֙תה ִת ְר ֶע֤ה ֶא‬ ֜ ָ ְ‫אמר י‬ ֶ ֹ ‫ֶאת־יִ ְׂש ָר ֵ ֑אל וַ ּ֙י‬

Previously, when Saul was king over us, you were the one who led Israel out and in. And the Lord said to you, “You will shepherd my people Israel, and you will be ruler over Israel.”

“Shepherd” was a metaphorical title of honour for kings and an epithet of gods. It is not by chance that it is used of David.32 However, the quoted passages also reveal a gap in the story of David—a gap that needs to be filled. When did the Lord

32  The motif also occurs in Ps 78:70–71: “He chose David his servant and took him from the sheepfolds; from following the nursing ewes he brought him to shepherd Jacob his people, Israel his inheritance.”

Rewriting David and Goliath?

177

say anything like the words of the elders and how did it happen that the Lord took David from following the sheep? Dealing with these questions, I cannot help thinking of the passage on David’s secret anointing by Samuel in 16:1–13. It is in many ways parallel to the long version of ch. 17. Both passages are about the small shepherd boy, who is disregarded by his family. He is not important enough to be immediately invited to the sacrificial meal arranged by Samuel. Only on Samuel’s demand is he brought back from shepherding the sheep. He is the smallest and useful only as a shepherd. In the long addition of ch. 17, the eldest brother Eliab reproaches David for having left the sheep and come to see the war: “I know your presumption and the evil of your heart” (v. 28). The two passages also have in common the names of the three elder brothers of David and mention the total number of the sons of Jesse as eight. Both passages, in fact, can be read as answers to the problem how the Lord took David from the sheep, first, by sending the prophet to anoint him, and then, by letting his father send him on an errand—just like he had Kish send Saul on an errand—which had far-reaching consequences for the whole nation. David comes directly from the sheep to the anointing scene as well as to the battlefield. That the Lord took him from the sheep was taken very literally! Most exegetes agree that the story about the anointing is one of the latest passages in 1 Samuel. It seems to me that it has an origin similar to the additions in chs. 17–18, although it happened to be added to the book early enough to be present in the Septuagint. These two stories reveal the same kind of strategy, the strategy of “rewriting” or inner-biblical exegesis that uses midrashic elements in order to fill in gaps in the older narrative. Thus, the story of the small shepherd boy David was probably not an independent source-text, but rather, oral exegetical material, midrash-like pieces of narrative that aimed at giving explanations to details of the older traditions and complementing them. The language of this kind of oral material during the late Second Temple period would most probably have been Aramaic. The redactor or scribe who wished to enrich the Hebrew text of the book with these additional narrative items needed to formulate them in Hebrew and in such a way that they would fit into the surrounding text. This was quite an exercise, as knowledge of classical Hebrew was perhaps no longer very active. Taking a closer look at the language of the additional parts of chs. 17–18, we can easily discover formulations that are not standard Hebrew.33 Complementing the older stories in Hebrew, the scribe naturally used the rest of the book as his aid. The v­ ocabulary 33  For instance, the use of the participle instead of the infinitive absolute when expressing continued movement (1 Sam 17:41; two participles, without a finite verb, expressing continued movement at 17:15 is also exceptional). At 1 Sam 17:20, the use of the article with a participle, if it is used predicatively, is incorrect, or otherwise, the following perfect tense is incorrectly preceded by waw copulative.

178

Anneli Aejmelaeus

used also shows that the writer is drawing on other parts of the book.34 This working procedure created many stylistic and linguistic parallels with other parts of the book, which attracted the attention of Auld and Ho.35

Building Bridges to the Torah On the other hand, there are theologically more significant parallels, like 2 Sam 5 and 7 that I just mentioned.36 Moreover, there are also parallels with the Pentateuch, although they have rarely been discussed at all. The Pentateuchal parallels show how very familiar the scribes were with the Torah, even to the exact wording of its various passages. The Torah was Scripture par excellence and the centre of scribal education and interpretative activity. Creating bridges between the Torah and the other books was obviously part of the interpretative assignment and an important factor in its motivation.37 At a closer look, we do find bridges to the Torah also in our story. The most significant of them is the motif of a father sending his son on an errand. Auld and Ho saw here a parallel with the story of Saul, but there is another parallel, which is even more striking, namely the story of Jacob sending Joseph—who is also a shepherd! (cf. Gen 37:2)38—to his brothers to see how they are and to bring back the message to the father (Gen 37:14). This brings Joseph to the beginning of those events at the end of which—after much struggle—he is to save his people. A striking parallel to David! Furthermore, the Joseph story gives a model for the enmity of the elder brothers against a younger one who has big thoughts about himself. The brothers were jealous of Joseph, because the father loved him more than his brothers. Joseph also had dreams about a great future. In the case of David, it is God who loves David and has chosen him before his elder brothers, as related in the story of the anointing (1 Sam 16:6–12). When Samuel is mustering the tall and handsome elder sons of Jesse, he is directly forbidden by the Lord to look at the outward 34  For instance, ‫ מערכה‬1 Sam 17:20, 21, 22, 23, 26, 48 (< 17:8, 10, 36, 45); ‫ נתש‬17:20, 22, 28 (< 10:2, 12:22); ‫ קשר‬nif. 18:1 (< 22:8, 13); ‫ נבא‬hitp. 18:10 (< 10:5, 6, 10, 13; 19:21, 23, 24). In each case, there is a slight shift in the usage. 35  See above note 22. 36  Ps 78:70–71 may also have played a role in the process; these verses are like a summary of 2 Sam 5:2 and 7:8. Also to be considered is Ps 151, which shows especially the detail that appears in the addition of 1 Sam 17:51 that David drew the sword of Goliath and beheaded him with it. 37  When readings out of the non-Pentateuchal books were introduced in the synagogue, they were supposed to be connected to the reading from the Torah which they were chosen to accompany. In view of this development, it is no wonder that the Pentateuch played a role in the “midrashic process”. 38  Curiously enough, Gen 37:2 and 1 Sam 17:34 use the same formulation ‫רעה היה בצאן‬. Speaking of shepherding, ‫ בצאן‬is otherwise only found in 1 Sam 16:11, 19.

Rewriting David and Goliath?

179

a­ ppearance of the boys. “The Lord sees into the heart” (v. 7). In David’s heart Eliab however only sees evil: David thinks too highly of himself, and Eliab suspects—correctly—that David seeks for an opportunity to become a hero. In view of the parallelism with the story of Joseph and his brothers, I think that the behavior of Eliab towards David coming to the battlefield is a reference to ch. 16, to David’s anointing witnessed by the brothers. Consequently, if the long expansion 17:12–31 presupposes the anointing story, the introduction of the father and the family in the beginning does not prove that this was the beginning of an independent source. The scribe probably just imitated the language of the introduction of Saul’s father, while introducing a digression to the story. For the content of the digression, I think, the Pentateuchal model was more essential.

Conclusion There are also other parallels with Pentateuchal passages, but I think I have already made my point. The longer version of the story of David and Goliath had its origin in scribal interpretation, which aimed at filling gaps in the older narrative and highlighting certain features of the story. Rather than being a combination of two independent accounts, the longer story represents the phenomenon of “rewriting” and was developed by complementation of the shorter story. The midrashic features in the narrative were created on the basis of other parts of the story of David as well as passages in the Torah. Creating bridges to the Torah was an important part of the interpretative activity; scriptural parallels, reminiscences, or allusions added scriptural flavour to the text. All this is part of the “midrashic process”, but we could also see that this process had begun already before the creation of the longer version of the story of David and Goliath.

180

Anneli Aejmelaeus

Bibliography Aejmelaeus, A., “Übersetzung als Schlüssel zum Original”, in On the Trail of the Septuagint Translators: Collected Essays (Contributions to Biblical Exegesis & Theology 50; Leuven: Peeters, 2007) 56–143. Auld, A.G., I–II Samuel: A Commentary (The Old Testament Library; Louisville, KY: Westminster John Knox, 2011). Auld, A.G. & Ho, C.Y.S., “The Making of David and Goliath”, JSOT 56 (1992) 39–19. Barthélemy, D. & Gooding, D.W. & Lust, J. & Tov, E. The Story of David and Goliath (OBO 73, 1986). Dietrich, W., Samuel (Biblischer Kommentar: Altes Testament; VIII/24 (17,58—16,1), VIII/25 (18,30—17,1), VIII/26 (19,24—18,1); Neukirchener: Neukirchen, 2013–2012). Fishbane, M., Biblical Interpretation in Ancient Israel (Oxford: Clarendon, 1985). Fokkelman, J.P., Narrative Art and Poetry in the Books of Samuel: A Full Interpretation Based on Stylistic and Structural Analysis. Volume II: The Crossing Fates (1 Sam. 13–31 & 2 Sam. 1) (Assen: Van Gorcum, 1986). McCarter, K., I Samuel (Anchor Bible, 1980). Tov, E., “The Composition of 1 Samuel 18–16 in the Light of the Septuagint”, in The Greek and Hebrew Bible: Collected Essays on the Septuagint (VTS 72; Leiden: Brill, 1999) 62–333. Ulrich, E., The Dead Sea Scrolls and the Origins of the Bible (Studies in the Dead Sea Scrolls and Related Literature; Grand Rapids, MI: Eerdmans/Leiden: Brill, 1999). Ulrich, E., “Crossing the Borders from ‘Pre-Scripture’ to Scripture (Rewritten) to ‘Rewritten Scripture’”, in J. Zsengellér (ed.), Rewritten Bible after Fifty Years: Texts, Terms, or Techniques? A Last Dialogue with Geza Vermes (JSJS 166; Leiden: Brill, 2014) 104–83. Vermes, G., Scripture and Tradition in Judaism: Haggadic Studies (Studia Post-Biblica 4; Leiden: Brill, 1961). Vermes, G., “Bible and Midrash: Early Old Testament Exegesis”, in P.R. Ackroyd/C.F. Evans (ed.), The Cambridge History of the Bible (1970) 231–199. Wellhausen, J., Der Text der Bücher Samuelis (Göttingen: Vandenhoeck & Ruprecht, 1871). Wellhausen, J., Die Composition des Hexateuchs und der historischen Bücher des A.T. (Berlin, 21889; 31899).

Drew Longacre Multilinear Genealogical Networks: Expanding the Scope of Textual History Introduction In the study of the Hebrew Bible/Old Testament, description of textual history has always been a complicated and controversial endeavor. Fundamental differences of opinion both plague and enrich the discipline, not infrequently yielding as many descriptions of textual history as textual scholars who actively work on the question. Limitations of materials and methods often preclude definitive conclusions and consensus. But many of our most heated arguments focus primarily on what we call things and how we describe them. Indeed, taxonomy is not unimportant. Labelling and describing texts, categories, and phenomena both reflect and influence how scholars think about textual history. But it is also possible that such debates—though important—obscure points of general agreement. In particular, a greater appreciation in contemporary textual scholarship for the originality of editorial contributions and resulting texts has led to a proliferation of perspectives on describing the origins of the written pentateuchal traditions. Some scholars continue to stress the essential similarities between our pluriform texts, arguing for the existence of an Urtext, original text, initial text, final form, archetype, or some similar conception. Others stress the uniqueness of each editorial stage in the development of textual traditions, supposing a multiplicity of equally valuable editions, determinative texts, or even multiple “originals” or “urtexts”. But a close reading of scholars with diverse approaches to the textual history of the Pentateuch suggests to me that there may actually be a generally unstated common ground shared between many of them. I suggest that a simple methodological principle has the potential to shed much light on current controversies and significantly clarify descriptions of textual history, as illustrated with the book of Exodus.

Implicit Common Ground To find this common ground, we must often read between the lines. When visualizing or mapping the relationships between texts envisaged by various scholars, similarities may emerge that are not immediately evident on the surface level of their presentation. Let us illustrate with examples from the book of Exodus. To

182

Drew Longacre

start the discussion, Eugene Ulrich presents the history of Exodus in twelve “editions”, each of which is worthy of consideration in its own right as reflecting literary originality.1 Though the whole point of Ulrich’s discussion is to demonstrate the pluriformity of the text of Exodus in multiple irreducible editions, his presentation of the history of the text assumes the linear development of each of these editions from a previous edition, and all documented editions can ultimately be traced back to a single documented edition, which is itself the result of previous editorial work.2 In other words, Ulrich’s editions are only irreducible in the sense that he defines them as distinct literary entities based on significant editorial intervention. But they are indeed reducible in the sense that the interrelationships between these editions can be traced backwards to a point of singular historical textual convergence. As he notes, All the text traditions of a given book are genetically related; that is, all surviving manuscripts can be envisioned simply as dots on a chart, but each is derived from some other earlier text by a direct line, and all texts as they are traced back are eventually shown to be interconnected. Thus, for each book the full chart looks like a tree, with the earliest form of the book as the trunk, which then diverges into a series of branches.3

1  E. Ulrich, “The Evolutionary Production and Transmission of the Scriptural Books”, in H. Von Weissenberg/J. Pakkala/M. Marttila (ed.), Changes in Scripture: Rewriting and Interpreting Authoritative Traditions in the Second Temple Period (Berlin: De Gruyter, 2011) 47–64, on pp. 53–5. On this analysis, the documented history of the text of Exodus begins with a Hebrew text with the tabernacle account of chapters 35–40 arranged as in the Old Greek (OG). The Masoretic text (MT) reflects a subsequent edition resulting from the rearrangement and supplementation of chapters 35–40. As far as I can tell from his publications, Ulrich envisions a linear derivation of the edition reflected in MT from the edition reflected in the OG, such that the former was created on the basis of the latter. The same is true of his next edition, wherein the so-called “pre-Samaritan” harmonizations were interpolated into the edition reflected in the MT. Ulrich proposes two additional editions (the Samaritan Pentateuch [SP] and 4Q[Reworked] Pentateuch), both of which are based on the previous edition, but presumably develop independently of each other and in parallel. 2  In his work on the book of Daniel, Ulrich argues that the editions of the Masoretic text and Old Greek developed independently from a now-undocumented edition of the text, but that edition is still singular and written; see E. Ulrich, “The Parallel Editions of the Old Greek and Masoretic Text of Daniel 5”, in E.F. Mason et al. (ed.), A Teacher for all Generations: Essays in Honor of James C. VanderKam (JSJSup 153; Leiden: Brill, 2012) 201–17. For an alternative approach without supposing a common source text, see I. Young, “The Original Problem: The Old Greek and the Masoretic Text of Daniel 5”, in R.F. Person/R. Rezetko (ed.), Empirical Models Challenging Biblical Criticism (Ancient Israel and Its Literature 25; Atlanta: SBL Press, 2016) 271–301. 3  E. Ulrich, The Dead Sea Scrolls and the Developmental Composition of the Bible (VTSup 169; Leiden: Brill, 2015), 44–5.

Multilinear Genealogical Networks:Expanding the Scope of Textual History

183

If we map this presentation out visually, it bears a marked similarity in general shape—if not in all particulars—to what might be expected from more traditional stemmata of manuscripts or diagrams of the development of traditions, such as those illustrating the perspectives of Frank Moore Cross and others.4 Emanuel Tov’s preference for a series of “determinative” texts can be analyzed in a very similar way.5 On the one hand, Tov states that “now more than ever it seems to me that there never was an ‘archetype’ or ‘original text’ of most scripture books”,6 recognizing that there are often a number of editorial stages in the development of a given text, which might be labelled equally “determinative”. Nevertheless, he consistently presents these determinative texts as being themselves interrelated, such that he associates the view of a series of determinative (or original) texts with a theory of a single original text collectively as one of two possible models for the history of the text.7 Ulrich and Tov, then, despite their many differences, both seem to operate with models of textual history that envisage literarily distinct but genealogically interrelated states of the text that can all be traced back to a single point of common ancestry. Along the same lines, Ronald Hendel writes, “Any edition may differ considerably from previous editions … Yet there is a textual continuity and a genealogical relationship among a book’s editions. The same kind of genealogical relationship arguably exists for the multiple editions of the biblical books.”8 Similarly, The genealogical method of ascertaining textual genealogy assumes that all the manuscripts of a given book belong to a single multibranched community of descent. Each manuscript of a book—for example, the book of Genesis—belongs to the genealogy of that book and not to independent genealogies that generated 4  F.M. Cross, “The Contribution of the Qumrân Discoveries to the Study of the Biblical Text”, in F.M. Cross/S. Talmon (ed.), Qumran and the History of the Biblical Text (Cambridge, MA: Harvard University Press, 1975) 278–92; id., “The Evolution of a Theory of Local Texts”, in F.M. Cross/S. Talmon (ed.), Qumran and the History of the Biblical Text (Cambridge, MA: Harvard University Press, 1975) 306–20; id., “The History of the Biblical Text in the Light of Discoveries in the Judaean Desert”, in F.M. Cross/S. Talmon (ed.), Qumran and the History of the Biblical Text (Cambridge, MA: Harvard University Press, 1975) 177–95. 5  E. Tov, Textual Criticism of the Hebrew Bible (3rd revised and expanded edn; Minneapolis: Fortress Press 2012), 167–9. 6  E. Tov, “Modern Editions of the Hebrew Bible”, in J.C. Paget/J. Schaper (ed.), The New Cambridge History of the Bible, Volume 1: From the Beginnings to 600 (Cambridge: Cambridge University Press, 2013) 365–85, on p. 380. 7 Tov, Textual Criticism, 161. 8  R.S. Hendel, “The Oxford Hebrew Bible: Its Aims and a Response to Criticisms”, Hebrew Bible and Ancient Israel 2, no. 1 (2013) 63–99, on p. 84. For an illustration of Hendel’s application of this perspective to the history of the text of Exodus, see id., “Assessing the Text-Critical Theories of the Hebrew Bible”, in T.H. Lim/J.J. Collins (ed.), The Oxford Handbook of the Dead Sea Scrolls (Oxford: Oxford University Press, 2010) 281–302, on p. 299.

184

Drew Longacre

more-or-less identical books of Genesis. This assumption seems self-evident, but it has been contested by some scholars … However, as Emanuel Tov and others have pointed out, there is no evidence for such a phenomenon in biblical books. The manuscripts are simply too similar …, and their variants are accountable by the normal mechanisms of scribal change. The variations among the manuscripts do not correspond to the types and degree of variance characteristic of oral transmission or manuscripts derived from oral formulaic composition.9

Bruno Chiesa likewise argues that the texts of the Hebrew Bible can be analyzed with traditional philological methods to “arrive at one Text, despite any plurality of ‘texts’”.10 Similar perspectives can be documented in many detailed treatments of the witnesses to the text of Exodus.11 While each of these scholars have important differences in the ways they conceptualize and describe the histories of the texts, there appears to be an important similarity between them. In particular, all of them seem to me to be operating with a model which supposes a common lineage of the preserved texts that can be traced back to a single point of historical convergence. Even though many scholars question the meaningfulness of the concept of a single “original” text, most do not deny the existence of a single text from which the documented textual tradition of a given book of the Hebrew Bible/Old Testament derives. This is very different from the alternative model of Shemaryahu Talmon and others who suggest independent, irreducibly pluriform genealogical origins for our traditions from the very earliest stages.12 As such, it seems to me that there is a greater 9  R.S. Hendel, Steps to a New Edition of the Hebrew Bible (Text-Critical Studies 10; Atlanta: SBL Press, 2016), 230–1. 10  B. Chiesa, “Textual History and Textual Criticism of the Hebrew Old Testament”, in J.C. Trebolle Barrera/L. Vegas Montaner (ed.), The Madrid Qumran Congress: Proceedings of the International Congress on the Dead Sea Scrolls, Madrid 18–21 March, 1991 (STDJ 11; Leiden: Brill, 1992) 257–72, on p. 272. 11  J.R. Davila, “Text-Type and Terminology: Genesis and Exodus as Test Cases”, Revue de Qumran 16, no. 61 (1993) 3–37; N. Jastram, “A Comparison of Two ‘Proto-Samaritan’ Texts from Qumran: 4QpaleoExodm and 4QNumb”, Dead Sea Discoveries 5, no. 3 (1998) 264–89; D. Longacre, “A Contextualized Approach to the Hebrew Dead Sea Scrolls Containing Exodus” (Ph.D. diss., University of Birmingham, 2015), 229–36; J.F. Quant, “Rewriting Scripture Inside and Out: A Typology of Rewriting in Variant Editions and Rewritten Scripture” (Ph.D. diss., Emory University, 2014), 61; J.E. Sanderson, An Exodus Scroll from Qumran: 4QpaleoExodm and the Samaritan Tradition (HSS 30; Atlanta: Scholars Press, 1986). M. Segal, “The Text of the Hebrew Bible in Light of the Dead Sea Scrolls”, Materia Giudaica 12 (2007) 5–20, on p. 19, speaks of a “complex web of interrelationships” between the different versions of the biblical books and groups manuscripts only on the basis of secondary readings, so he seems to approach the textual history of the Pentateuch similarly. 12  S. Talmon, Text and Canon of the Hebrew Bible: Collected Studies (Winona Lake, IN: Eisenbrauns, 2010), 220–3, 392–7, 406–8, 413–18. According to Talmon, various independent collections of Pentateuch-like material were gradually standardized into the Pentateuch as it

Multilinear Genealogical Networks:Expanding the Scope of Textual History

185

amount of essential agreement than the diversity of perspectives might suggest, at least from this one important perspective—i.e., the genealogical interrelatedness of different text forms. Scholars may agree on the general shape of a textual tradition, even while differing on how to evaluate the literary relationships and relative values of each distinct stage. The recent stress on the orality of transmission by some scholars, while yielding valuable insights, has hardly undermined the widespread agreement on this point.13

Expanding the Scope of Textual History The natural consequence of the observations made above is that the scope of textual history must be defined broadly enough to incorporate the study of a given text in all of its interrelated versions. Any conception of textual history limited to a single version or edition of a text is unhelpfully restrictive.14 If we only consider particular textual artifacts or versions of texts, we miss important elements of the history of the texts being studying. To appreciate textual history in its fullness requires broadening our horizons to consider the interrelations between all versions of a text. This accords well with the programmatic call of Hermann-Josef Stipp to “einen einheitlichen exegetischen Aspekt der Textentwicklung annehmen, der alle Stadien der Geschichte biblischer Texte im Bereich der Schriftlichkeit umfaßt”.15 is known today, and many of the differences between the surviving witnesses reflect early differences that had not yet been standardized from this diverse set of starting points. Cf. the sympathetic treatment of G.J. Brooke, “The Qumran Scrolls and the Demise of the Distinction between Higher and Lower Criticism”, in J.G. Campbell/W.J. Lyons/L.K. Pietersen (ed.), New Directions in Qumran Studies: Proceedings of the Bristol Colloquium on the Dead Sea Scrolls, 8–10 September 2003 (London: T & T Clark, 2005) 26–42. 13  D.M. Carr, Writing on the Tablet of the Heart: Origins of Scripture and Literature (Oxford: Oxford University Press, 2005), 168, 291–2, goes so far as to suggest that scribes may have retextualized from memory large portions of texts that had been lost during the Babylonian exile, which could theoretically allow for truly textually independent versions of scriptural books. The general uniformity of the documented traditions and severe practical limitations, however, make this exceedingly unlikely; cf. D. Longacre, “Scribal Approaches to Damaged Manuscripts: Not Just a Modern Dilemma”, in P.B. Hartog/A. Schofield/S.I. Thomas (ed.), The Dead Sea Scrolls and the Study of the Humanities. Method, Theory, Meaning: Proceedings of the Eighth Meeting of the International Organization for Qumran Studies (Munich, 4–7 August, 2013) (STDJ 125; Leiden: Brill, 2018) 141–64, on p. 155 n. 41. 14  Admittedly, there may be times when it is pragmatically appropriate to focus on one version of a text against all others, but this must always be conducted with an awareness of the larger text-historical connections. The main argument of this paper concerns primarily the theoretical basis for textual history, rather than the practical methods used to study it. 15  H.-J. Stipp, “Textkritik—Literarkritik—Textentwicklung: Überlegungen zur exegetischen Aspektsystematik”, EThL 66 (1990) 143–59, on pp. 155–6.

186

Drew Longacre

But even this broadened focus itself may be considerably too narrow. Even those who might agree with incorporating all interrelated versions of a text into its textual history may still limit their perspective to the text(s) of a particular abstract literary work, however that may be defined.16 But, in point of fact, the text of such a literary work usually has numerous, diverse interrelationships with other texts not normally considered within the scope of its “textual history”.17 If the study of history includes the contexts and interrelationships of the objects being studied, then these textual precursors, parallels, and derivatives form integral parts of textual history and cannot be justifiably ignored.18 These diverse intertextual relationships—and therefore “textual history”, as I would define it— clearly extend beyond any definition of a literary work. What does it mean, then, to expand the scope of textual history beyond the limits of a single literary work? This redefinition is on one level incredibly simple and intuitive, and yet on another it is a fundamental realignment. Such a textual history must encompass not only unitary texts, but also the history of all of the constituent parts of all of a text’s versions and all other interrelated literary contexts, each of which themselves should be similarly analyzed. Thus, textual history is the historical description of the origins, development, and reception of all the constituent parts of interrelated texts. On this inclusive definition, textual history ceases to be shorthand for the continuous-text textual tradition of a single literary work and becomes a broad historical sub-discipline in its own right. Essentially, it entails continuously striving towards the unachievable goal of a complete description of a nearly universal textual history. 16  Again, I do not deny that there are many pragmatic benefits to such a restricted focus. Indeed, for most editors and readers the work may be the most useful and frequently used textual entity, and in some cases text almost becomes a synonym for work. I am only arguing that, theoretically, full description of textual history cannot stop there. 17  For other attempts to describe the relationships between interrelated texts beyond the traditional limits of literary works, see the usage of such terms as “cluster” by Florentino García Martínez, cited in C. Hempel, The Qumran Rule Texts in Context: Collected Studies (TSAJ 154; Tübingen: Mohr Siebeck, 2013), 141, 149, and “constellation” by H. Najman, “The Idea of Biblical Genre: From Discourse to Constellation”, in J. Penner/K.M. Penner/C. Wassen (ed.), Prayer and Poetry in the Dead Sea Scrolls and Related Literature: Essays in Honor of Eileen Schuller on the Occasion of Her 65th Birthday (Leiden: Brill, 2012) 307–21, in the study of the diverse body of literature from the Second Temple period. These approaches tend to focus on looser thematic parallels between texts, rather than the close verbal agreement in view in this paper, but they share a similar objective to that envisaged here. Other creative approaches highlighting the flexibility of texts beyond the limits of the literary work include B.W. Breed, Nomadic Text: A Theory of Biblical Reception History (Bloomington, Ind.: Indiana University Press, 2014), and E. Mroczek, The Literary Imagination in Jewish Antiquity (Oxford: Oxford University Press, 2016), though they tend to minimize the real genealogical relationships emphasized in this paper. 18  In a similar vein, some Septuagint scholars might consider quotations in Josephus, Philo, or patristic writers to be part of the textual history of the respective books, though the scope envisioned in this paper is even broader than this relatively inclusive conception.

Multilinear Genealogical Networks:Expanding the Scope of Textual History

187

Distinction between Genealogical and Literary Analysis In order to advance this grander, more comprehensive view of textual history, I suggest that it is helpful to consciously maintain a methodological distinction between classifying texts as literary works and studying the flow of text between interrelated texts. Much of the discussion about textual history focuses on the level of the significant literary distinctions between documented texts. For example, how much and/or in what ways must a text change before it should be classified as a new literary work, composition, or edition of the same work as its base text? Or is the concept of a distinct literary “work” or “edition” even valid for ancient literary cultures where gradual development is common, perhaps even the norm? Discussions of Urtexts, original texts, initial texts, final texts, determinative texts, literary editions, and the like today typically reflect differences of opinion on the literary nature, identity, and value of various texts. Scholars evaluate the relative merits and originality of each text, sometimes resulting in a preference for a single form of the text and other times resulting in a relatively egalitarian appreciation for multiple texts. These debates are exceedingly important for how we understand the history of these texts, and in no way would I suggest that these crucial questions should be ignored. But I would suggest that we should also pay close attention to the genealogical relationships between texts without respect to the literary significances of their differences. Copies of a work, new editions of a work, and new works based on prior works all have one thing in common: a genetic dependence upon earlier texts, whether in small or large degree. This fact of genealogical relationships can easily be obscured or ignored when focusing on the question of whether two or more texts reflect the same literary work or are distinct compositions. Behind every full-bodied description of textual history lies an explicit or implicit genealogical skeleton of intertextual connections that holds it together. By refocusing attention on this aspect of textual history, we can temporarily distance ourselves from the complex controversies over the literary natures of the texts and shed light on the relatively simple question of where they came from. Let us propose then, a methodological distinction between the genealogical and literary aspects of the analysis of preserved texts (see Tables 1 and 2). Literary analysis determines the definitions and limits of textual traditions as literary works and whether two texts are of the same or different works (or potentially whether some other form of literary classification is more appropriate). Genealogical analysis examines the genetic relationships between all interrelated texts regardless of whether or not they belong to the same work. The literary analysis of texts concerns their intended purposes and received significances in relation to prior texts, normally evident primarily in the types and extents of changes introduced into the received texts. The genealogical relationships of texts concern their interconnections via series of source-recipient relationships as simply texts

188

Drew Longacre

without regard for the literary significances of the differences. Literary analysis seeks to isolate groups or clusters of like texts as distinct, meaningful literary entities, while genealogical analysis is the glue that binds together sometimes vastly different interrelated texts in complex networks of connections. Literary analysis charts discontinuities in literary traditions, while genealogical analysis traces the continuity of the flow of text between interrelated texts. The relationships between texts and their literary natures are two interrelated aspects of textual history, and description of textual history must explicitly address both perspectives in order to provide an adequate description for the observable phenomena. Table 1 Genealogical Analysis

Literary Analysis

Flow of text between interrelated texts Genetic relationships between texts without respect to the literary significances of their differences Genetic relationships between all interrelated texts regardless of whether or not they belong to the same work Binds together different interrelated texts Interconnections via series of sourcerecipient relationships Complex network of connections Continuity of the text flow

Classifications of texts as literary works Literary nature, identity, and value of various texts Definitions and limits of textual traditions as literary works and whether two texts are of the same or different works Separates texts into distinct literary entities Intended purposes and received significances in relation to prior texts Distinct clusters of like texts Discontinuity of literary traditions

Table 2 Literary Analysis (same work?)

Genealogical Analysis (source-recipient relationship?)

Yes

No

Yes

No

The recipient is a copy or version of the same work as the source One text is an independent (re)textualization of the same work as the other text

The recipient is a new work in some sense derived from the source One text is a new work textually independent of the other text

Multilinear Genealogical Networks:Expanding the Scope of Textual History

189

Note well that I am not using “genealogical” and “literary” as synonyms for the frequently discussed “textual” and “literary” criticisms (German Textkritik and Literarkritik) respectively. When using the term “literary” in this paper, I speak of the nature or identity of a text as a work of literature. This usage is, of course, related to questions of source and redaction criticism, but is not identical. Neither should my “genealogical” be hastily identified with traditional definitions of textual criticism, since the former incorporates all types of changes at all stages of development, regardless of how many witnesses attest to a given change or whether the changes are intentional or accidental. The categories used in this paper answer different research questions from the traditional disciplinary distinctions between textual and literary criticism and are simply not comparable. If we recognize this methodological distinction between literary and genealogical aspects of textual history, there will be several important ramifications. First, we will be in a much better position to understand when disagreements relate to the general shape and structure of the history of the text and when they are differences in terminology, conception, or literary valuation. For instance, as demonstrated above, many prominent textual scholars seem to operate with similar ideas about the general structure of the history of some texts, even when they disagree significantly on what terms to use to describe different states of the text and how each stage should be appreciated. The proposed distinction identifies common ground more clearly and focuses on what exactly are the real differences. Second, since the transmission of texts and their component parts often crosses the boundaries of literary works, it is essential that textual interrelationships be examined beyond these traditional limits. It is not only continuous-text copies of the entire Hebrew text of Exodus that partake in the textual history of the book, but so also all of its sources, editions, translations, quotations, and otherwise derivative works. The end of Jubilees may adopt only a small proportion of the text of Exodus, but it participates in the history of the book. It witnesses indirectly to a text of Exodus and is at the same time dependent on a text of Exodus for its own composition. This dynamic does not necessarily undermine the concept of a literary work, but it does mean that there is textual continuity that pays no heed to one’s definition and limits of a work. Even if, in the end, we retain the concept of a literary work, we must be willing to think of textual history in terms bigger than any one such work. The proposed distinction provides a methodological justification for expanding our sights to see the big picture of textual history in all its complexity, without necessarily losing the ability to order it appropriately. And third, the proposed distinction also has something to contribute to the vexed question of the relationship between textual criticism and the so-called “literary” criticism (or the equivalent German Literarkritik) alluded to above. Most definitions that make or refuse to make such a distinction focus either on: 1) the supposed differences between compositional and transmissional stages

190

Drew Longacre

in the development of texts, or 2) the evidence available and methods used by scholars to evaluate the textual evidence. The consideration of genealogical relationships as a separate category, however, may provide a third way. It allows us to describe textual history holistically including all of its parts, without automatically excluding any forms of related text, even while leaving open questions about how significant textual differences are to the identities of the texts and how many witnesses might survive to a given text form. My proposed distinction between genealogical and literary analysis (i.e., in the sense defined above for this paper) would apply equally well, whether trying to detect sources behind a single early text, observing ongoing editing in the late Second Temple period, or sorting out a mass of manuscript evidence and derivative literature. Each of these stages could be presented together as a unified whole, without necessarily rejecting literary differences within that textual history.

Source-Recipient Relationships I would suggest that the lines of genealogical relationship can ideally be traced (or at least conceptualized) along series of source-recipient relationships. In any act of written transmission, there is at least one source and at least one recipient. Any text which is taken over—in part or in whole, altered or unchanged—in the creation of a new text can be considered a source. Any new text which reproduces and incorporates source text can be considered a recipient. A single recipient may inherit text from multiple sources, and a single source may pass on text to multiple recipients. A single text may function as both source and recipient in relation to different texts.19 The source-recipient relationship remains valid, regardless of whether the recipient text is considered to reflect the same literary entity as the source text. By framing the discussion in such inclusive terms, we will be in a good position to describe comprehensively the genealogical relationships between a wide variety of interrelated texts.

19  Given our focus on constituent parts of texts, it may even be possible in complex situations for a text to function both as source and recipient in relation to a single other text at different times and in different ways.

Multilinear Genealogical Networks:Expanding the Scope of Textual History

191

Figure 1

As just described, this model provides a conceptual framework for analysis of the relationships within sets of two specific texts in direct source-recipient relationships. In traditions such as those of the books of the Hebrew Bible/Old Testament, however, one or both texts of many of these sets will be lost to modern scholars, yielding many source-recipient sets with only one known member and multiple completely undocumented source-recipient sets—forever lost and unrecoverable—intervening between preserved texts.20 In such cases, the distance between distant source and recipient texts separated by an indeterminable number of lost intermediate stages can be collapsed into indirect source-recipient relationships (see Figure 1), albeit with the loss of granularity inherent in such simplifications. Material limitations (e.g., lost texts and fragmentary texts) may lower the resolution at which textual scholars can retrace textual history, but they do not fundamentally invalidate the conceptual framework proposed here, since even lost and unrecoverable direct source-recipient relationships were indeed real relationships, which impacted the history of the texts.

Complicating Factors in Source-Recipient Relationships A few complicating factors should be accounted for in the proposed model, which can be illustrated using Exodus as a test case. First, we must account for the possibility that a new text could be the recipient of text from multiple source texts. Such would be the case, for instance, when a new literary work draws on multiple different sources. Furthermore, it would account for the possibility of 20  Thanks to Emanuel Tov for stressing this limitation to me.

192

Drew Longacre

mixture between texts within the same tradition. It will also be helpful to note the proportion of the recipient text that is derived from each of its multiple sources to see if any one source text predominates. A good example of this phenomenon from the Exodus tradition would be the harmonistic influx of readings from a Deuteronomy text in the development of the so-called “pre-Samaritan” tradition (e.g., Exod 18:25; 20:19a, 21b; 32:10), in which the Exodus tradition retains a controlling influence on the resulting mixed text. On the other hand, 4Q175 (4QTestimonia) simply juxtaposes a quotation from Exodus with quotations of other books. Second, in many cases the adoption of source text by recipients is selective, such that only part of the source text is inherited in the recipient text. In cases where the source text is simply excerpted (e.g., 4QExodd, 4QExode, tefillin, and mezuzot) the source text remains dominant in the recipient, even if incomplete. But in cases where source text is selectively brought into a recipient text which uses additional sources or incorporates newly composed or textualized material (e.g., so-called “rewritten scripture” like Jubilees), it is helpful to know both the proportion of the source text that is adopted in the recipient text and the proportion of the recipient text that is received from the source text. Third, when a recipient text incorporates newly composed text or textualizes non-written source material, this material cannot be mapped via source-recipient relationships between written texts, so I suggest simply noting the proportion of newly composed or textualized material to received written source text. In many cases it is simply impossible to know whether text without a known written precursor is newly composed or textualized or whether it is inherited. For instance, interpolations such as the expanded Song of Miriam in Exodus 15 according to 4Q365 (4Q[Reworked] Pentateuchc) may fall into this category. Nearly every copying act involves the influence of prior content knowledge and oral traditions and intermediating processes of memory (as scribes recall and write what they have read in their exemplars), the effect of which on the flow of text is inherently difficult to recognize, analyze, and quantify. Fourth, when a recipient changes the source text adopted from its source(s), we should seek to measure the extent of the changes, whether accidental or intentional. In these cases, it would be helpful for scholars to include the proportion of the source text that is adopted without change to that which is altered. For establishing the genealogical relationships, however, the quantity and nature of the changes is less important than the observation of their occurrence.21 Thus, the harmonistic tradition of 4QpaleoExodm and the book of Jubilees are both dependent on texts of Exodus and make changes to their base texts, even though 21  Of course, the nature of the changes may affect scholars’ practical ability to reliably ascertain genetic relationships and the directions of changes (e.g., polygenetic changes or variants which cannot be confidently adjudicated), but this does not undermine the principle.

Multilinear Genealogical Networks:Expanding the Scope of Textual History

193

they do so in radically different ways and degrees. In this way, we can trace the historical development of the texts without prematurely prioritizing certain types of changes (e.g., intentional vs. accidental, large vs. small) in our descriptions, which seems to me inappropriate in a period where the text is accumulating both copyist errors and editorial interventions simultaneously. Since a new text can have multiple source texts, use source texts selectively, incorporate newly composed or textualized material, and introduce changes into received source texts, even the genealogical delineation of source-recipient relationships is not a simple process. In fact, focused attention on the genealogical aspect of source-recipient relationships will yield a fuller, more intricate description than has often been the case. Indubitably, it will yield a complex network of intertextual connections that will ignore traditional literary boundaries and extend almost without limits. Despite its increased complexity and the inevitable disagreements and limitations even at this level, however, this genealogical skeleton will provide a much more explicit reference point for textual scholars to flesh out full-bodied descriptions of textual history than is often the case.

Textual History and the Pentateuch In this short paper, I have not advanced the discussion on any particular textual issues, nor is there enough space to give even a comprehensive survey of the results of previous analyses. Nevertheless, a practical illustration may be helpful at this point to concretize the abstract methodological discussion up to this point. Let us take from the pentateuchal and related literature a few selected examples of specific textual developments that are widely recognized in the scholarly literature to show in small part how such a view of textual history would be realized. The point of this illustration is not dependent upon the accuracy of these sometimes-contested examples, but merely to provide practical clarity to a new paradigm that I suggest would apply equally well, regardless of how the traditions actually developed. An inherent part of the proposed view of textual history is that it will inevitably yield extremely complex and multilinear directional networks of interrelationships that cannot be easily summarized in a simple, mono-linear way. Nevertheless, to oversimplify for a moment, Figure 2 below provides a helpful selective illustration of one way several interrelated texts might be connected as nodes within a unified text-historical description. Most scholars suppose that the book of Exodus as we know it was composed from the conflation and editing of several different sources and/or supplements. At some point during this process, at least some of this text was reused and adapted within the book of Deuteronomy alongside other sources and newly composed material. The tradition of Deuteronomy fragmented into numerous texts that differed in relatively small degrees.

194

Drew Longacre

At some point, the Exodus tradition split into the tradition reflected in the Old Greek (OG-Ex) and the proto-Masoretic (Proto-MT-Ex) text.22 4QExode excerpts a small portion of the latter, and the medieval Masoretic text of Exodus (MT-Ex) preserves proto-MT-Ex well. A pre-Samaritan (Pre-SP-Ex) tradition also developed from Proto-MT-Ex with many internal harmonizations and interpolations from a text of Deuteronomy. 4QpaleoExodm reflects an early form of this text, and the medieval Samaritan Pentateuch (SP-Ex) develops Pre-SP-Ex even further. Other derivative texts like the 4Q(Reworked) Pentateuch (4Q[R]P) manuscripts, Jubilees, and the Temple Scroll apparently inherited text from the Pre-SP-Ex forms of multiple books of the Pentateuch, and perhaps also from each other (e.g., if the Temple Scroll is dependent on the 4Q[R]P manuscript 4Q365 + 4Q365a as a source). Type of SouceRecipient Relationship J Source

JE

Source

Proto-MT-Ex

Proto-MT-DT

Source

Pre-SP-Dt

4Q (Reworked) Pentateuch Temple Scroll

Predominant

Exodus-Archetype?

Deuteronomy-Archetype?

Source

P

Combined Book of Exodus

Combined Book of Deuteronomy

OG-Dt

Minor Substantial

E

OG-Ex

Pre-SP-Ex

Jubilees

4QExod-e

4QpaleoExod-m

Ezekiel the Tragedian Source Source Source 7QpapLXXExod Source

Eusebius of Caesarea

Codex Vaticanus SP-Ex

Josephus Source

MT-Ex Copy Copy

Citation of Eusebius

Figure 2: Partial Illustration of Textual History and the Pentateuch

22  For an overview of the Greek tradition of Exodus and its relation to known Hebrew texts, see D. Longacre, “Exodus in the Second Temple Period”, in S. Ehorn/S. Whittle (ed.), Exodus in the New Testament (London: T & T Clark, forthcoming).

Multilinear Genealogical Networks:Expanding the Scope of Textual History

195

Within the Greek tradition, the texts of some early manuscripts appear to have been partially assimilated to the Proto-MT-Ex form of their respective books (e.g., 7QpapLXXExod), and some of these manuscripts contained multiple books of such mixed character (e.g., Codex Vaticanus). A form of the Greek text was also selectively adopted and reused by early authors such as Ezekiel the Tragedian and Flavius Josephus in their works. The works of these authors themselves were used by later authors, such as Eusebius of Caesarea as some of many sources. Each of these authors in turn have their own processes of textual transmission. Thus far, we have primarily been discussing the genealogical relationships between a wide variety of interrelated texts. In making each of these connections, we can determine that there is a (in)direct genetic dependence or continuity in the flow of text from each source to its corresponding recipient(s). This is naturally true for those which are merely copies or revised editions of the same work, but so also for those which are sources, excerpts, citations, or secondary adaptations of a work. From the perspective of genealogical analysis, each of these interrelated texts have something to say about the other, and their relationships should be explicitly explored. Literary analysis, on the other hand, looks at this genealogical skeleton and tries to make sense of it in light of literary conventions concerning the nature of texts. In other words, it asks why the texts developed in the way they did, and how each text was intended and received within its ancient and modern contexts. For instance, I have argued elsewhere that—despite the complex interrelationships evident between these texts—the high degree of continuity between all continuous-text manuscripts containing the book of Exodus (in all its versions) and relatively low levels of verbatim agreement with other related texts creates a supercluster of Exodus texts and suggests that the book of Exodus was treated as a distinct literary work already in the Hellenistic and early Roman periods.23 In other words, of this broad schema indicated by the genealogical analysis, a subset of those texts is best explained as belonging to a recognizable literary work of definite (if often unclear) boundaries. Sometimes, as in the case of the 4Q(Reworked) Pentateuch manuscripts, it remains unclear exactly how they should be analyzed, but in many cases (e.g., Ezekiel the Tragedian, Josephus, Eusebius, and I would also suggest Jubilees and the Temple Scroll) new texts should clearly be recognized as new works literarily distinct from the book of Exodus to which they are genetically related. Exploring the genealogical relationships between all these texts without respect for supposed literary boundaries helps to see the big picture of textual history more fully and in all its beautiful complexity. Analyzing the literary nature of these interrelated texts allows us to bring order out of this vast array of sometimes chaotic intertextual connections. Neither of these aspects of text-historical 23 Longacre, Contextualized Approach, 195–204.

196

Drew Longacre

description can realistically stand alone, because they are mutually illuminating. Neither can take temporal or logical precedence over the other, because they necessarily impact each other. But by consciously distinguishing genealogical and literary aspects of the study of textual history, we put ourselves in the best possible position to optimize historical descriptions. Such a distinction forces us to consciously consider both continuity and discontinuity in textual history.

Conclusion As should be obvious from the above illustration, the material and theoretical limitations of the work will never permit arrival at a comprehensive, exhaustive, and perfectly accurate description of textual history, but I am convinced it will help us speak more clearly about the work perpetually in progress that is the scholarly study of ancient texts. Limitations of material and methods will forever diminish the detail and accuracy with which textual scholars will be able to retrace textual history, and it must always be remembered that “our knowledge of the textual landscape in the period around the turn of the era is far from perfect”.24 Nevertheless, within these limitations, I suggest that the model proposed above has the potential to optimize what we can know about textual history and provide a clearer conceptual framework for discussions of textual history in the future. An intentional focus on multilinear directional networks of genealogical relationships will both clarify and expand our descriptions of textual history.

Bibliography Breed, B.W., Nomadic Text: A Theory of Biblical Reception History (Bloomington, IN: Indiana University Press, 2014). Brooke, G.J., “The Qumran Scrolls and the Demise of the Distinction between Higher and Lower Criticism”, in J.G. Campbell/W.J. Lyons/L.K. Pietersen (ed.), New Directions in Qumran Studies: Proceedings of the Bristol Colloquium on the Dead Sea Scrolls, 8–10 September 2003 (London: T & T Clark, 2005) 26–42. Carr, D.M., Writing on the Tablet of the Heart: Origins of Scripture and Literature (Oxford: Oxford University Press, 2005). Chiesa, B., “Textual History and Textual Criticism of the Hebrew Old Testament”, in J.C. Trebolle Barrera/L. Vegas Montaner (ed.), The Madrid Qumran Congress: Proceedings of the 24  Andrés Piquer Otero, “Hebrew Bible(s) and Greek Witnesses? A First Look at the Makeup of 2 Kings for the Oxford Hebrew Bible”, in M.K.H. Peters (ed.), XIV Congress of the IOSCS, Helsinki, 2010 (Septuagint and Cognate Studies 59; Atlanta: Society of Biblical Literature, 2013) 691–704, on p. 704.

Multilinear Genealogical Networks:Expanding the Scope of Textual History

197

International Congress on the Dead Sea Scrolls, Madrid 18–21 March, 1991 (STDJ 11; Leiden: Brill, 1992) 257–72. Cross, F.M., “The Contribution of the Qumrân Discoveries to the Study of the Biblical Text”, in F.M. Cross/S. Talmon (ed.), Qumran and the History of the Biblical Text (Cambridge, MA: Harvard University Press, 1975) 278–92. Cross, F.M., “The Evolution of a Theory of Local Texts”, in F.M. Cross/S. Talmon (ed.), Qumran and the History of the Biblical Text (Cambridge, MA: Harvard University Press, 1975) 306–20. Cross, F.M., “The History of the Biblical Text in the Light of Discoveries in the Judaean Desert”, in F.M. Cross/S. Talmon (ed.), Qumran and the History of the Biblical Text (Cambridge, MA: Harvard University Press, 1975) 177–95. Davila, J.R., “Text-Type and Terminology: Genesis and Exodus as Test Cases”, Revue de Qumran 16, no. 61 (1993) 3–37. Hempel, C., The Qumran Rule Texts in Context: Collected Studies (TSAJ 154; Tübingen: Mohr Siebeck, 2013). Hendel, R.S., “Assessing the Text-Critical Theories of the Hebrew Bible”, in T.H. Lim/J.J. Collins (ed.), The Oxford Handbook of the Dead Sea Scrolls (Oxford: Oxford University Press, 2010) 281–302. Hendel, R.S., “The Oxford Hebrew Bible: Its Aims and a Response to Criticisms”, Hebrew Bible and Ancient Israel 2, no. 1 (2013) 63–99. Hendel, R.S., Steps to a New Edition of the Hebrew Bible (Text-Critical Studies 10; Atlanta: SBL Press, 2016) 230–1. Jastram, N., “A Comparison of Two ‘Proto-Samaritan’ Texts from Qumran: 4QpaleoExodm and 4QNumb”, Dead Sea Discoveries 5, no. 3 (1998) 264–89. Longacre, D., “A Contextualized Approach to the Hebrew Dead Sea Scrolls Containing Exodus” (Ph.D. diss., University of Birmingham, 2015). Longacre, D., “Exodus in the Second Temple Period”, in S. Ehorn/S. Whittle (ed.), Exodus in the New Testament (London: T & T Clark, forthcoming). Longacre, D., “Scribal Approaches to Damaged Manuscripts: Not Just a Modern Dilemma”, in P.B. Hartog/A. Schofield/S.I. Thomas (ed.), The Dead Sea Scrolls and the Study of the Humanities. Method, Theory, Meaning: Proceedings of the Eighth Meeting of the International Organization for Qumran Studies (Munich, 4–7 August, 2013) (STDJ 125; Leiden: Brill, 2018). Mroczek, E., The Literary Imagination in Jewish Antiquity (Oxford: Oxford University Press, 2016). Najman, H., “The Idea of Biblical Genre: From Discourse to Constellation”, in J. Penner/K.M. Penner/C. Wassen (ed.), Prayer and Poetry in the Dead Sea Scrolls and Related Literature: Essays in Honor of Eileen Schuller on the Occasion of Her 65th Birthday (Leiden: Brill, 2012) 307–21. Piquer Otero, A., “Hebrew Bible(s) and Greek Witnesses? A First Look at the Makeup of 2 Kings for the Oxford Hebrew Bible”, in M.K.H. Peters (ed.), XIV Congress of the IOSCS, Helsinki, 2010 (Septuagint and Cognate Studies 59; Atlanta: Society of Biblical Literature, 2013) 691–704.

198

Drew Longacre

Quant, J.F., “Rewriting Scripture Inside and Out: A Typology of Rewriting in Variant Editions and Rewritten Scripture” (Ph.D. diss., Emory University, 2014). Sanderson, J.E., An Exodus Scroll from Qumran: 4QpaleoExodm and the Samaritan Tradition (HSS 30; Atlanta: Scholars Press, 1986). Segal, M., “The Text of the Hebrew Bible in Light of the Dead Sea Scrolls”, Materia Giudaica 12 (2007) 5–20. Stipp, H.-J., “Textkritik—Literarkritik—Textentwicklung: Überlegungen zur exegetischen Aspektsystematik”, EThL 66 (1990) 143–59. Talmon, S., Text and Canon of the Hebrew Bible: Collected Studies (Winona Lake, IN: Eisenbrauns, 2010). Tov, E., “Modern Editions of the Hebrew Bible”, in J.C. Paget/J. Schaper (ed.), The New Cambridge History of the Bible, Volume 1: From the Beginnings to 600 (Cambridge: Cambridge University Press, 2013) 365–85. Tov, E., Textual Criticism of the Hebrew Bible (3rd revised and expanded edn; Minneapolis: Fortress Press 2012). Ulrich, E., The Dead Sea Scrolls and the Developmental Composition of the Bible (VTSup 169; Leiden: Brill, 2015). Ulrich, E., “The Evolutionary Production and Transmission of the Scriptural Books”, in H. Von Weissenberg/J. Pakkala/M. Marttila (ed.), Changes in Scripture: Rewriting and Interpreting Authoritative Traditions in the Second Temple Period (Berlin: De Gruyter, 2011) 47–64. Ulrich, E., “The Parallel Editions of the Old Greek and Masoretic Text of Daniel 5”, in E.F. Mason et al. (ed.), A Teacher for all Generations: Essays in Honor of James C. VanderKam (JSJSup 153; Leiden: Brill, 2012) 201–17. Young, I., “The Original Problem: The Old Greek and the Masoretic Text of Daniel 5”, in R.F. Person/R. Rezetko (ed.), Empirical Models Challenging Biblical Criticism (Ancient Israel and Its Literature 25; Atlanta; SBL Press, 2016) 271–301.

Part III Writing and Rewriting in Translation

Katja Kujanpää Adjusted to the Argument: Tracing Paul’s Motives for Modifying the Wording of Scriptural Quotations Introduction With his numerous direct quotations,1 Paul is an early witness to the text of the Septuagint. His testimony, however, always needs to be carefully scrutinised before it can be of any text-critical use, for he frequently alters the wording of his scriptural quotations.2 That the majority of Paul’s quotations derive from the Septuagint, not from a Hebrew or an Aramaic text, for example, is a matter of scholarly consensus.3 However, the textual plurality of the first century is reflected in some quotations, and it appears that in certain cases Paul reproduced a Greek text that had already been revised in order to bring it closer to a Hebrew text.4 Therefore, studying Paul’s quotation practice not only results in a deeper understanding of his argumentation, but it can also shed light on the transmission of the Septuagint in the first century c.e. The significance of his quotat ions is found not so much in tracing the Urtext of the Septuagint, but in reconstructing second­ ary readings that were circulating in Paul’s time. Contrary to scribes, translators, and editors, Paul does not transmit complete literary works. Instead, he is an author who uses short extracts of sacred writings in new compositions, interweaving direct quotations with his own discourse. Consequently, Paul’s reasons for modifying the wording of scriptural texts tend to 1  This article concentrates on “direct” or “explicit” quotations. A scriptural reference is defined as a quotation if it has 1) an introduction formula, or 2) an established formula used for textual interpretation (cf. Rom 10:6: “that is”), or 3) a clear syntactical or stylistic tension with the surrounding text (such as an abrupt change of personal pronouns), or 4) significant verbal correspondence with a certain scriptural passage. The last criterion is disputed among scholars and open to various interpretations. More important than having a certain amount of words quoted in a row is the frequency of words and forms. However, all of the examples used in this article fulfil the first criterion. 2  Estimations of the frequency of the alterations depend on the definition of a quotation and on the criteria that scholars use to establish the wording that served as the basis for the changes. For well-informed calculations, see D.-A. Koch, Die Schrift als Zeuge des Evangeliums: Untersuchungen zur Verwendung und zum Verständnis der Schrift bei Paulus (BHT 69; Tübingen: J.C.B. Mohr 1986), 186; C.D. Stanley, Paul and the Language of Scripture: Citation Technique in the Pauline Epistles and Contemporary Literature (SNTSMS 74; Cambridge: Cambridge University Press, 1992), 348–9. 3  A seminal study for establishing this was A.F. Kautzsch, De Veteris Testamenti Locis a Paulo Apostolo Allegatis (Whitefish, MT: Kessinger Publishing, 2010 [reprint, original 1869]). For more recent investigations, see Koch, Schrift, 48–88; Stanley, Paul and the Language, 37–51. 4  See below n. 52.

202

Katja Kujanpää

be somewhat different from those of scribes, for in Paul’s letters the context of the scriptural passage has radically changed. Since this change of context creates tensions, in numerous instances Paul modifies the wording of the quotation to make it accord better with the surrounding argumentation. However, if it were possible to “peel off ” all Pauline alterations and reconstruct the reading that served as the basis for these alterations, one would gain access to Greek readings that were circulating in the middle of the first century c.e. Whil e such an endeavour may at first sound like a wild goose chase because of the textual plurality at that time, the situation is perhaps not that hopeless. Yet in this article, I will not undertake the task of reconstructing the Vorlage of Paul’s quotations, but concentrate instead on a step that should precede all such reconstructions. The aim of distinguishing Paul’s alterations from pre-­ Pauline variant readings requires an analysis of Paul’s authorial intention when he quotes scriptures.5 I will demonstrate how it is often possible to connect deviations from the wording of the Septuagint with certain authorial concerns. It is crucial to examine both Paul’s techniques of altering the wording of quotations and the motives behind the alterations, for this enables one to sketch a profile of Paul as an author quoting scriptures. Familiarity with his quotation practice may then help to also trace the origin of his readings in cases that have major textual problems. For example, when the manuscripts of the Septuagint are divided and the minority reading agrees with Paul, the question arises whether the reading is a pre-Pauline variant or whether the manuscripts of the Septuagint have been harmonised with Paul’s quotation, the wording of which Paul himself has altered. To decide between these alternatives, one should analyse what Paul’s motives for both quoting and modifying the wording could have been and whether there are parallel cases of similar motives elsewhere in his letters. In other words, it is helpful to know Paul’s “style”. Tracing Paul’s authorial intentions when he quotes scriptures does not mean that one should attempt to get inside his head. I do not propose a psychological approach, but an argumentative and rhetorical one: it is necessary to examine what functions the quotations perform in the discourse and how they are related to their immediate new context on one hand and the argumentation as a whole on the other.6 When Paul’s reading deviates from the reading of the Septuagint 5  Cf. Wilk’s emphasis on authorial intention: F. Wilk, “Letters of Paul as Witnesses to and for the Septuagint Text”, in W. Kraus/R.G. Wooden (ed.), Septuagint Research: Issues and Challenges in the Study of the Greek Jewish Scriptures (SCS 53; Atlanta: SBL, 2006) 253–71, on pp. 259, 261. 6  In my opinion, tracing Paul’s authorial intentions is less problematic than tracing the intentions of an anonymous author of a writing representing a different genre. This is because of the nature of Paul’s writings: the letters are argumentative and intended to persuade the audience, not multi-layered narratives to be interpreted. Moreover, they are letters by a historical author to a certain audience and can be dated with reasonable accuracy. This makes the

Adjusted to the Argument: Tracing Paul’s Motives

203

(as reconstructed in critical editions), one should approach these deviations with the following questions: Do they make sense in the new context of quotation? Does Paul’s version of the quotation create thematic or verbal links with its surroundings? Does it solve grammatical, lexical or theological problems that may have arisen with the reading of the Septuagint? Does it increase the coherence or the rhetorical effectiveness of the argumentation? In the following, I examine three quotations from Rom 9–11 and demonstrate how deviations from the wording of the Septuagint are related to the new context of the quotations in Romans. The quotations to be analysed exemplify the different techniques that Paul employs to modify the wording of the quotations and how the modifications render the quotations more compatible with the surrounding argumentation. Instead of producing a comprehensive list of all techniques and motives that occur in Paul’s letters, the case studies only function as examples. Studying Paul’s alterations to the wording of quotations necessitates careful text-critical analysis that encompasses the textual variance in both the Septuagint and the New Testament.7 In most cases, the textual tradition of New Testament does not pose major problems (either because of the lack of significant variants or because there are convincing arguments for their secondary origin), but there are a handful of quotations on which the witnesses are strongly divided and deciding between the alternatives becomes more difficult.8 As for the source text of the quotation, while the critically reconstructed text of the Septuagint and the s­ ituation substantially different from trying to reconstruct the intentions of, for example, the author of Esther or its Greek translator. 7  It is not possible to offer an extensive discussion of methodology here. For that, see Koch, Schrift, 11–24; Stanley, Paul and the Language, 31–61; Wilk, “Letters of Paul”, 253–263; R.J. Wagner, Heralds of the Good News: Isaiah and Paul “In Concert” in the Letter to the Romans (NTS 101; Leiden: Brill, 2002), 5–8, 16–28; J.D.H. Norton, Contours in the Text: Textual Variation in the Writings of Paul, Josephus and the Yahad (LNTS 430; New York: T & T Clark, 2011), 1–56. 8  Rom 9:26 is a good example: the reading of NA28 (‫ א‬A D K L P Ψ 33. 81. 104. 365. 630. 1175. 1241. 1505. 1506.) follows the Septuagint by reading οὗ ἐρρέθη αὐτοῖς, whereas the witnesses P46 F G ar b d* syp read οὗ ἐὰν κληθήσονται. The testimony of P46 is significant, for according to Günther Zuntz, the papyrus contains less adaptations to the Septuagint than numerous other manuscripts: G. Zuntz, The Text of the Epistles: A Disquisition upon the Corpus Paulinum: The Schweich Lectures of the British Academy 1946 (London: Oxford University Press, 1953). The syntax of the minority reading does not follow the grammar of classical Greek, for ἐάν is followed by the future instead of the subjunctive aorist (see the detailed analysis of Wagner, Heralds, 84–5 n. 127). In consequence, the origin of the variant is difficult to explain if assumed to be secondary: it is far more probable that Paul’s wording has later been adapted towards the wording of the Septuagint that represents better Greek than that the change has occurred vice versa. The substitution of ἐρρέθη with κληθήσονται accords well with Paul’s overall argument where the verb καλέω is in a key position, occuring in three other places in verses 24–6 alone. As will be argued below, Paul also replaces the verb λέγω with καλέω at the beginning of verse 25. Therefore, there are good arguments to conclude that the minority reading is probably the original one (Zuntz, The Text of the Epistles, 174; Wagner, Heralds, 84).

204

Katja Kujanpää

preserved textual variants of its manuscript tradition offer a good starting point for comparison, the other Greek textual traditions (e.g. Aquila, Symmachus, and Theodotion, patristic quotations), the Hebrew variants of Qumran, the Masoretic text, and early translations (before all Old Latin, the Targums, and the Peshitta) should also be included. The possibility that the textual tradition that Paul used suffered extinction, leaving no other witnesses, also has to be considered. When all of the preserved text-critical material on one hand and the function and the context of the quotation on the other are taken into account, it is possible to attribute with reasonable probability certain deviations from the Septuagint to Paul’s own editorial activity.9 This is certainly the case with the textual examples in this article; in order to avoid circular reasoning, I have deliberately chosen cases in which already the textual evidence strongly suggests Pauline origin. Analysing Paul’s authorial intention in cases like these is of considerable importance, for the observations may then be extended to quotations the textual profile of which is more obscure. My treatment of the examples is selective in this article, for one quotation may contain several text-critical problems on different levels: although the Pauline origin of certain alterations may be almost beyond doubt, the same quotation may also contain a pre-Pauline variant reading (an approximation towards the Hebrew, for example). In the following, I discuss only changes that are directly related to the theme of this article. Before turning to the case studies, it is useful to briefly discuss the process of recontextualisation in light of modern quotation theory, for it illuminates how modifications can be necessitated by the new context of the quotation.

From One “Network of Relations” to Another When Paul presents a quotation from scriptures, the passage he quotes is detached from its immediate literary context and integrated into a new textual entity. In the field of literary criticism, Meir Sternberg has examined what happens to a quotation when it is taken from its original context and inserted into a new one.10 Sternberg argues that a shift in the meaning of the quotation is inevitable in this recontextualisation process, for a quotation always belongs to “a network of relations”: the quoted passage has a frame that encloses and regulates it. When the 9  Cf. Wilk, “Letters of Paul”, 263–4. Attention to Paul’s authorial intention is also central for the question of “memory lapses”. Koch’s and Stanley’s comprehensive studies on all quotations in undisputed Pauline letters (see n. 2 above) indicate that deviations from the source text should not primarily be explained by Paul quoting from memory and not getting the wording quite right, for often it is obvious that the deviant wording is connected to the way in which Paul uses the quotation; see Koch, Schrift, 186–90; Wagner, Heralds, 14. 10  M. Sternberg, “Proteus in Quotation-Land: Mimesis and the Forms of Reported Discourse”, Poetics Today 3 (1982) 107–56.

Adjusted to the Argument: Tracing Paul’s Motives

205

passage is extracted from the framing elements that influence its interpretation and inserted into a new frame with different regulating elements, there is always a change in the meaning of the quotation.11 Sternberg concludes: “However accurate the wording of the quotation and however pure the quoter’s motives, tearing a piece of discourse from its original habitat and reconstructing it within a new network of relations cannot but interfere with its effect.”12 The concepts “frame” and “network of relations” are helpful for describing the differences between the old and new literary contexts of a scriptural passage quot­ ed by Paul. For example, a verse in Isaiah is situated in a network of relations that specifies who is speaking to whom, whom the personal pronouns refer to, what the topic is, and how it is related to the broader narrative. Moreover, verbal links may connect the passage with a certain thematic framework. When Paul inserts the verse into his letter, a different network of relations surrounds the quotation, and what made sense in the original context may appear obscure or irrelevant in the new one. In general, Paul takes great care to frame quotations with elements that tie them to the rest of the argumentation. The most important framing element is the introduction formula. Occasionally Paul uses established, formulaic expressions (“as it is written”),13 but more often he creates a new introduction according to his argumentative needs. Introduction formulae offer the audience additional information by specifying the content, speaker, addressee, or location of the quotation. Most of them feature conjunctions (γάρ, ὡς, καθώς, δέ) that indicate how the quotation is related to Paul’s own words or other quotations, offering confirmation or indicating a change of topic or speaker, for example. After presenting a quotation, Paul occasionally makes his own summarisations or conclusions about its message (cf. Rom 11:5). The frame of a quotation may also contain so-called catchwords, expressions that have a pivotal role in the argumentation and connect passages with each other.14 These verbal links that tie the quotation to nearby sentences strengthen the cohesion of the argumentation. However, in numerous cases Paul does not limit his activity to creating an appropriate frame for the quotation, but he also changes something in its wording 11  Ibid., 108, 131, 152. 12  Ibid., 145. 13  Paul uses the expression “as it is written” (καθὼς γέγραπται) 18 times in his undisputed letters. In non-Jewish literature the expression is rare, whereas in Jewish it is in common use (cf. 2 Kings 14:6). In Qumran, the equivalent phrase ‫ כאשר כתוב‬is used (Koch, Schrift, 25, 29). 14  Using catchword connections is sometimes presented as a typical rabbinic method: see, for example, E.E. Ellis, Prophecy and Hermeneutic in Early Christianity: New Testament Essays (WUNT 18; Tübingen: Mohr, 1978), 214–16; W.R. Stegner, “Romans 9:6–29—a Midrash”, JSNT 22 (1984) 37–52, on p. 40; R. Jewett, Romans: A Commentary (Hermeneia; Minneapolis: Fortress, 2006), 571. However, the phenomenon is also well-attested in non-Jewish literature: see J.S. Kloppenberg, The Formation of Q: Trajectories in Ancient Wisdom Collections (Philadelphia: Fortress 1987), 48, 268, 282. For catchwords in general, see S.J. Patterson, The Gospel of Thomas and Jesus (Sonoma: Polebridge 1993), 99–102.

206

Katja Kujanpää

to make it more compatible with his argumentation. This may be due to an obvious discrepancy between the quotation and its new context. For example, as will be illustrated below, the personal pronouns in the quotation are sometimes incompatible with its surroundings. In other cases it appears that a change in the wording would not have been absolutely necessary, but it enhances the rhetorical effect of the quotation, making it appear more relevant. In the following, I include examples of both situations.

Adjusted to the Argument: Textual Examples Coherence by Catchwords It is characteristic of Paul to enhance the consistency of his argumentation by using catchwords. In Rom 9:6–29, the verb “to call” (καλέω) functions as a catchword that ties different parts of the argumentation together; it is found in Rom 9:7, 12, 24, 25–6. This last occurrence is a combined quotation from Hos 2:23 and 1:10b. At the beginning of the book of Hosea, God orders the prophet to give symbolic names to his children: the daughter is to be called “No Mercy” (Οὐκ-ἠλεημένη) and the younger son “Not My People” (Οὐ-λαόσ-μου). Although these names refer to Israel’s rejection, that is only temporary, for Hos 2:23 is a promise of the future reversal of Israel’s fate: “And I will have mercy on ‘No Mercy’ and I will say to ‘Not My People’: ‘You are my people.’” Paul quotes this verse and combines it with Hos 1:10, but his wording deviates from the Septuagint: Rom 9:24: Οὓς καὶ ἐκάλεσεν ἡμᾶς οὐ μόνον ἐξ Ἰουδαίων ἀλλὰ καὶ ἐξ ἐθνῶν, Rom 9:25 (NA28)

Hos 2:23 (LXX Göttingen)15

καλέσω τὸν οὐ λαόν μου λαόν μου καὶ τὴν οὐκ ἠγαπημένην ἠγαπημένην·

καὶ ἐλεήσω τὴν Οὐκ-ἠλεημένην16 καὶ ἐρῶ τῷ Οὐ-λαῷ-μου Λαός μου εἶ σύ

15  The footnotes do not reproduce the apparatus of the Göttingen edition, but contain only variants that will be discussed in the main text or that are otherwise relevant for the matters at hand. 16  B V 407 Co Cyr.p Hil stand for the reading ἀγαπήσω τὴν Οὐκ-ἠγαπημένην, and 239 and Aeth combine the two readings. This is a fine example of a textual problem that raises the

Adjusted to the Argument: Tracing Paul’s Motives

207

Rom 9:26 (NA28): καὶ ἔσται ἐν τῷ τόπῳ οὗ ἐρρέθη αὐτοῖς· οὐ λαός μου ὑμεῖς, ἐκεῖ κληθήσονται υἱοὶ θεοῦ ζῶντος (Hos 1:10b).17 Figure 1: Rom 9:25/Hos 2:23

The Septuagint has the parallel clauses “I will have mercy on ‘No Mercy’” and “I will say to ‘Not My People’: ‘You are my people’”, whereas in Romans “No Mercy” and “Not My People” are both subordinate to the verb καλέσω. In consequence, the direct address (“You are my people”) is changed to indirect (“I will call ‘Not My People’ ‘My People’”). There is no support for this syntactical pattern in the textual tradition of the Septuagint or in other versions. The probability of the alteration being Paul’s is increased by the observation that he has a clear motivation for preferring καλέω and substituting the original verbs with it. First, in the preceding verse, Paul has stated that God “has called us” from among the Jews and Gentiles. The repetition of the same verb increases the relevance of the quotation, as it seems to confirm exactly what he has just argued. Second, the verb joins— together with the phrase “my people”—the two halves of the combined quotation.18 On the contrary, it is highly improbable that the reading of Romans represents a pre-Pauline variant reading, for in the immediate literary context of the verse in Hosea there appears to be no incentive that could have inspired a pre-Pauline question whether the reading is a pre-Pauline textual variant or whether it results from a harmonisation of the Septuagint manuscripts with Romans. The case demonstrates how important it is to analyse the relationship between the quoted passage and its context, both original and subsequent. Many commentators have suggested that the verb ἐλεέω would have fitted Paul’s argument better than ἀγαπάω, for ἐλεέω is repeated five times in Rom 9:15–23, whereas ἀγαπάω occurs only in the quotation from Malachi in Rom. 9:13: “Jacob I loved, but Esau I hated.” However, it is perfectly conceivable that Paul wished to create a verbal connection with Rom 9:13, for the connection implies that a reversal of divine exclusion is possible: God’s call can make the once “Not Beloved” or hated into “Beloved”. Paul’s authorial intention can thus be reconstructed in different ways to support either alternative. As for the context of the verse in the Septuagint, Hos 1–2 refers to the symbolic naming of Hosea’s children several times. However, only some of the witnesses reading ἠγαπημένην in Hos 2:23 (V 407 and certain Coptic versions) consistently also read ἠγαπημένην when the name of the daughter occurs in Hos 1:6, 8 and 2:1. Yet in verses Hos 1:7 and 2:4 these witnesses use ἐλεέω to render the finite verb (Wagner, Heralds, 81–2 n. 120). By doing so, they break the verbal correspondence between divine action (to have mercy) and the symbolic name of Hosea’s daughter (No Mercy), thus ruining the word play. It is difficult to imagine how and why a reading such as this would have originated if it were pre-Pauline. Therefore, the variation of the name in B was probably caused by unsystematic harmonisation with the New Testament. The scribes corrected the name in 2:23, but left it intact in other places, whereas the scribes behind the text of V and 407 were more consistent. 17  For the minority reading of the verse, see n. 8. 18 Koch, Schrift, 105, 167; Stanley, Paul and the Language, 110; B. Fuß, “Dies ist die Zeit, von der geschrieben ist...”: Die expliziten Zitate aus dem Buch Hosea in den Handschriften von Qumran und im Neuen Testament (NTAbh 37; Münster: Aschendorff, 2000), 182.

208

Katja Kujanpää

scribe to alter the syntax and replace the verbs. It is reasonable to conclude that Paul rewrites the syntax and substitutes the original verbs with καλέω in order to produce an additional catchword connection that makes his argumentation more consistent and creates the impression that the combined quotation seamlessly supports his statement in 9:24. However, despite this extensive rewriting of the syntax of the verse, its contents and message about the reversal of status remain essentially the same.19 Nothing in the contents of the quotation necessitates modification, but it is the way of expression and place of emphasis Paul adjusts.

Prioritising Relevant Elements In addition to the syntactical and lexical changes in Rom 9:25, the order of the names of Hosea’s children is also reversed in Romans so that “Not My People” comes first. As was the case with the modifications already discussed, there is no textual support for such a reversal in the Septuagint or in other versions. Again the immediate literary context of the quotation in Romans offers a clue why Paul would have changed the order of the names: in verse 9:24 he affirms that the called ones also come from amongst Gentiles, and the reversal of the status of “Not My People” into “My People” in the quotation appears to correspond to this calling of Gentiles. Since “Not My People” is connected with the statement that immediately precedes the quotation, it is a more significant item in the quotation than the other name.20 Accordingly, Paul switches the order to advance the more relevant expression. The probability that the modification is intentional is increased by the obser­ vation that Paul makes similar moves elsewhere; he also switches the place of parallel elements in a quotation in Rom 10:20 and 11:3.21 In Rom 11:3 he quotes 3 Kgdms 19:10, where Elijah is speaking to the Lord just before the theophany on Horeb. The passage does not belong to the so-called kaige sections of Kingdoms that were revised to achieve closer correspondence with the Hebrew text.22

19  Cf. Stanley’s remark that “the extent to which the wording of a passage has undergone modification is no sure sign of how far Paul has deviated from the ‘original sense’ in his application/interpretation of a given biblical verse” (Stanley, Paul and the Language, 111). 20 Koch, Schrift, 105; Stanley, Paul and the Language, 110. 21  On Rom 10:20, see F. Wilk, Die Bedeutung des Jesajabuches für Paulus (FRLANT 179; Göttingen: Vandenhoeck & Ruprecht, 1998), 35–6; Wagner, Heralds, 211. 22 The kaige sections defined by Thackeray are 2 Kgdms 11:2–3 Kgdms 2:11 and 3 Kdms 22:1–4 Kdms 25:30 (H.St.J. Thackeray, “The Greek Translators of the Four Books of Kings”, JTS 8 [1907] 262–78, on p. 263).

Adjusted to the Argument: Tracing Paul’s Motives

Rom 11:3

209

3 Kgdms 19:1023

κύριε, τοὺς προφήτας σου ἀπέκτειναν, τὰ θυσιαστήριά σου κατέσκαψαν, κἀγὼ ὑπελείφθην μόνος καὶ ζητοῦσιν τὴν ψυχήν μου.

τὰ θυσιαστήριά σου κατέσκαψαν καὶ τοὺς προφήτας σου ἀπέκτειναν24 ἐν ῥομφαίᾳ,25 καὶ ὑπολέλειμμαι26 ἐγὼ μονώτατος, καὶ ζητοῦσι τὴν ψυχήν μου λαβεῖν αὐτήν27.

Figure 2: Rom 11:3/3 Kgdms 19:1028

The sins of the people, destroying the altars and killing the prophets, are found in Romans in reversed order. Since the only witnesses supporting this word order are Justin and Origen,29 it can in all probability be ascribed to Paul. The killing of prophets seems to be an idea that can be linked to the experience of some early Christ-followers and has relevance to their identity (cf. 1 Thess 2:15; Matt 10:17–21; 23:29–39; Luke 11:47–51; 13:34–5).30 Paul further enables this connection with his own time by eliminating the expression “by the sword”;31 this specification of the manner of killing is unnecessarily concrete and inappropriate in the contemporary situation. In contrast to “killing the prophets”, demolishing the altars of God would probably not have offered similar connections to the situation

23 The text is that of Rahlfs with noteworthy variant readings cited in the footnotes. The readings are taken from the critical apparatus of the forthcoming Göttingen edition of 3–4 ­Kgdms by Julio Trebolle and Pablo Torijano, reproduced with their permission. 24  Justin and Origen add the vocative κύριε and follow Paul’s order of the parallel clauses, but they are in all probability dependent on Paul, cf. Stanley, Paul and the Language, 148 n. 217. On the phenomenon of patristic authors harmonising readings of the Old and the New Testament, see M. Meiser, “Relevanz der Kirchenväterzitate für die Textgeschichte der neutestamentlichen Zitate aus der Septuaginta”, in M. Karrer et al. (ed.), Von der Septuaginta zum Neuen Testament: Textgeschichtliche Erörterungen (ANTF 43 Berlin: De Gruyter, 2010) 283–316. 25  The words ἐν ῥομφαίᾳ are omitted only in Priscilian’s quotation, probably due to the influence of Romans. 26  L (the Lucianic manuscripts 19-82-93-108-127) reads here ὑπελείφθην. 27  The words λαβεῖν αὐτήν are omitted in the Sahidic and Ethiopic versions and in a quotation from Justin but in no Greek manuscript. All three are probably dependent on Paul. 28  Note the variation between 19:10 and the almost identical verse 19:14: σε/τὴν διαθήκην σου and κατέσκαψαν/καθεῖλαν. The verb reveals that Paul is quoting 19:10. 29  See above n. 24. 30 Stanley, Paul and the Language, 149. 31 Koch, Schrift, 75; Stanley, Paul and the Language, 150.

210

Katja Kujanpää

of Paul and his audience, which is why he reverses the order of the clauses and begins with the more interesting accusation.32 Prioritising certain elements in a quotation is not specifically a Pauline technique. There are examples of modifications in the Dead Sea Scrolls and in Greek Jewish literature where the position of an important element is advanced in a direct quotation.33 It is thus a practice that accords with the literary conventions of Paul’s time. Therefore, the parallel cases of switched elements in his letters cannot be used as an argument for the Pauline origin of an alteration. Instead, what they demonstrate is, first, that the modification could be Pauline, for the technique belongs to his toolkit, and second (if other reasons already suggest Pauline origin), that the modification is probably intentional, for it is less plausible that Paul repeatedly remembers clauses in reversed order.

Adding Clarifications Romans 11:3 also demonstrates how Paul can compensate for the loss of information caused by the change in the network of relations. In the quotation, an unidentified person is complaining to someone about violent persons who have attacked the addressee’s altars and prophets. Paul crafts a lengthy introduction formula that contains exceptionally many elements that guide the interpretation of the quotation: “Or do you not know what the Scripture says in Elijah [narratives], how it appeals to God against Israel” (ἢ οὐκ οἴδατε ἐν Ἠλίᾳ τί λέγει ἡ γραφή, ὡς ἐντυγχάνει τῷ θεῷ κατὰ τοῦ Ἰσραήλ). The core is a typical formulaic introduction, λέγει ἡ γραφή,34 but Paul has tailored its other parts. The introduction begins with a rhetorical question (“Or do you not know”) that anticipates proof or reasoning of some kind. Paul also specifies the location of the quotation: it can be found in the Elijah narratives.35 The rest of the introduction identifies whom the words are directed at and what they concern: they are addressed to God and contain an accusation “against Israel”. By offering these specifications, Paul selectively invokes certain key aspects of the original context that are central for following the heavily abridged dialogue in Rom 11:3–4.36 Yet although the introduction formula supplies all of this information about location, subject matter, 32  E. Käsemann, An die Römer (HNT 8a; Tübingen: J. C. B. Mohr [Paul Siebeck], 1974), 290; Koch, Schrift, 74 n. 83, 104. 33  For examples, see Stanley, Paul and the Language, 301, 318, 327, 331. 34  Paul also uses the expression λέγει ἡ γραφή in Rom 4:3; 9:17; 10:11; 11:2, and Gal. 4:30. 35 Koch, Schrift, 27 n. 17. Usually Paul mentions the name of the alleged author, not the location of the quotation, but this introduction formula has parallels in Rom 9:25 and 1 Cor 9:9. Cf. also Mark 12:26 and Luke 20:37. 36  In contrast, certain aspects of the narrative in 3 Kgdms 19:10–18 are either irrelevant or even counterproductive for Paul’s purposes: there is no reference in Romans to the anointing of the kings of Syria and Israel or to the killing of the renegades (3 Kgdms 19:15–17).

Adjusted to the Argument: Tracing Paul’s Motives

211

and the addressee of the words of the quotation, for the sake of clarity Paul also makes an alteration to the actual wording of the quotation. In Romans, the quotation begins with the vocative κύριε, which has no equivalent in the Septuagint.37 The vocative is in all probability a Pauline clarification intended to help readers follow the dialogue between Elijah and the Lord, for the vocative makes it unambiguous that second-person singular forms (“your prophets”, “your altars”) refer to the Lord. It is much more characteristic of Paul to omit words in a quotation than to add any, but there are several parallel cases of additions like this.38

Concise Conflated Quotations Paul not only combines texts from different sources and presents them one after another as one quotation (as is the case in Rom 9:25–26), but he also conflates passages, inserting elements from another scriptural text into a quotation. The source texts become tightly interwoven, so that they usually form one single sentence. Instead of presenting two separate quotations or a single combination, Paul creates a completely new textual entity. In Rom 11:7–8, Paul is wrestling with the question of God’s faithfulness in a situation where the majority of Israel is mysteriously insensitive and unreceptive towards the gospel. In 11:7 Paul explains that this unreceptiveness is caused by God: “Israel failed to obtain what it was seeking. The elect obtained it, but the rest were made insensible (ἐπωρώθησαν).” That the passive form is to be interpreted as passivum divinum is made unambiguous in the following catena of two quotations (Rom 11:8–10).39 Together the quotations not only offer scriptural support for Paul’s claim that God himself has made insensible part of his people, but they also describe the nature and consequences of this insensibility. The latter quotation is from Ps 68:23–24 (69:23–24 MT), but the first one is a conflation. While the bulk of the quotation derives from Deut 29:4 (29:3 MT), the expression καρδίαν εἰδέναι is replaced with πνεῦμα κατανύξεως (“spirit of stupefaction”),40 which instead comes from Isa 29:10. 37  On Justin’s and Origen’s quotations (that contain the vocative), see n. 24 above. 38  See Koch, Schrift, 132–9. 39  C.K. Barrett, A Commentary on the Epistle to the Romans (BNTC; London: Adam & Charles Black, 1962), 210; H. Hübner, Gottes Ich und Israel: Zum Schriftgebrauch des Paulus in Römer 9–11 (FRLANT 136; Göttingen: Vandehoeck & Ruprecht, 1984), 106; O. Hofius, “Das Evangelium und Israel: Erwägungen zu Römer 9–11”, ZTK 83 (1986) 297–324, on p. 303 n. 20. 40 The Hebrew word ‫ ַּת ְר ֵּד ָמה‬used in Isa 29:10 can be rendered as ‘deep sleep’; see C.E.B. Cranfield, A Critical and Exegetical Commentary on the Epistle to the Romans (ICC; vol. 2; Edinburgh: T & T Clark 1983), 550. However, κατάνυξις should, in light of normal Greek usage, probably be translated as ‘bewilderment’ or ‘stupefaction’ (LSJ) rather than ‘deep sleep’. The Greek translator of Isaiah may have noticed that in Isa 29:9–10 the spirit actually results in people being “faint and confused” (ἐκλύθητε καὶ ἔκστητε), not asleep.

212

Katja Kujanpää

Rom 11:8 NA28

Deut 29:4 (29:3 MT) (LXX Göttingen)

Isa 29:10 (LXX Göttingen)

ἔδωκεν αὐτοῖς ὁ θεὸς πνεῦμα κατανύξεως, ὀφθαλμοὺς τοῦ μὴ βλέπειν καὶ ὦτα τοῦ μὴ ἀκούειν, ἕως τῆς σήμερον ἡμέρας.

καὶ οὐκ ἔδωκεν κύριος ὁ θεὸς ὑμῖν καρδίαν εἰδέναι καὶ ὀφθαλμοὺς βλέπειν41 καὶ ὦτα ἀκούειν ἕως τῆς ἡμέρας ταύτης

ὅτι πεπότικεν ὑμᾶς κύριος πνεύματι42 κατανύξεως καὶ καμμύσει τοὺς ὀφθαλμοὺς αὐτῶν καὶ τῶν προφητῶν αὐτῶν καὶ τῶν ἀρχόντων αὐτῶν οἱ ὁρῶντες τὰ κρυπτά

single underline = agreement between Rom 11:8 and Deut 29:4 double underline = agreement between Rom 11:8 and Isa 29:10 bold = agreement between all three Figure 3: Rom 11:8/Deut 29:4 + Isa 29:10

In addition to this conflation, the wording in Romans also deviates from Deut 29:4 in several other ways. Here I will concentrate on the most consequential modification, the relocation of the negation in the sentence and the effect it causes together with the conflation. While in Deut 29:4 the negation precedes the finite verb (οὐκ ἔδωκεν), in Rom 11:8 it is the infinites that are negated (τοῦ μὴ βλέπειν, τοῦ μὴ ἀκούειν). This greatly affects the way in which the divine action is described. In Deut 29:4, Moses is addressing the people of Israel and recounting their experience in the desert. Although Israel had seen the great deeds of the Lord in Egypt, the Lord had not given them understanding (Deut 29:2–3). The situation has lasted “until this day”, that is, until the day Moses is addressing the people. On that day Moses places the two options, life or death, clearly before the people as they stand before God, entering into the covenant (Deut 29:10–12; 30:15). When Paul omits the word οὐκ and negates the infinitive forms instead, the nature of the divine action changes. In Romans, God actively gives a spirit of stupefaction that causes lack of understanding, whereas in Deuteronomy it is simply stated that God “has not given” understanding. The Pauline origin of these modifications is relatively secure, for there is no evidence at all for a pre-Pauline reading that would omit the negation οὐκ and negate the infinitives instead. Although the idea that God actively deprives Jews of understanding is present in 41  The article τοῦ before βλέπειν occurs only in A (which, inconsistently, does not have an article before εἰδέναι or ἀκούειν). 42  The following witnesses read πνεῦμα: S 93 309 301 538 Or. X 51 Wirc. Spec. According to Eusebius, Symmachus also uses the accusative form, but with a different main verb.

Adjusted to the Argument: Tracing Paul’s Motives

213

the gospels,43 there is no evidence whatsoever that the actual wording of Rom 11:8 would be dependent on an early Christian reformulation of Deut 29:4.44 The removal of the negation from the beginning of the sentence strengthens the parallel with Isa 29:10, in which God’s action against Israel is expressed without negations. Therefore, it is probable that Isa 29:10 has offered Paul the stimulus to reformulate the sentence. As a result of the relocation of the negation, the three parallel items that God has not given his people become three items that God has given. The “heart of understanding” of Deut 29:4 would have accorded well with the other items in his list, as long as it was negated like the other infinitives; God has given a heart that would not understand.45 Why has Paul replaced the expression? The reason for the conflation lies probably more in the connotations of πνεῦμα κατανύξεως than Paul’s dissatisfaction with καρδίαν εἰδέναι. Later in the argument Paul presents Israel’s unbelief and disobedience as a temporary condition that will change (11:11–12, 15, 25–7). Isaiah’s concept of “spirit of stupefaction” may have appealed to him, because it sounds like resulting in a temporary and passing condition that comes from outside: when the time comes, Israel will awaken from its stupor.46 Thus, the concept offers Paul a medium to explain the source of Israel’s current disbelief and to present it as provisional. When he conflates the two passages, he gives priority to a phrase that he considers more enlightening. Why does Paul not present Deut 29:4 and Isa 29:10 one after another instead of conflating them? It seems that he had an idea about the interplay of Deut 29:4 and the following Psalm quotation. However, the unmodified wording of Deut 29:4 would perhaps have had too harsh of a ring in its new context in Rom 11. In their original literary setting in Deuteronomy, the words are not as severe, for they explain Israel’s former lack of understanding, whereas in Romans the focus is on Israel’s current unbelief. These different frames of the quoted words influence their message. If Paul’s explanation for the reasons behind Israel’s unbelief was that “God did not give them a heart to understand and eyes to see and ears to hear”, it would sound more like a categorical statement without much hope of 43  See the use of Isa 6:9–10 in Mark 4:12/Luke 8:10/Matt 13:13–15; John 12:40; Acts 28:26 (Stanley, Paul and the Language, 159 n. 252). 44  Ibid., 159. 45  Jewett suggests that the phrase would have caused confusion because of Paul’s earlier references to “senseless heart” (ἡ ἀσύνετος αὐτῶν καρδία) in 1:21 and “impenitent heart” (ἀμετανόητον καρδίαν) in 2:5 (Jewett, Romans, 662). However, these references do not contain anything that contradicts Paul’s message in Rom 11, and, moreover, they are far away in the letter. 46  Cf. Stanley, Paul and the Language, 161; Wilk, Bedeutung, 54; S.-L. Shum, Paul’s Use of Isaiah in Romans: A Comparative Study of Paul’s Letter to the Romans and the Sibylline and Qumran Sectarian Texts (WUNT II 156; Tübingen: Mohr Siebeck, 2002), 234; M.A. Seifrid, “Romans”, in G.K. Beale/D.A. Carson (ed.), Commentary on the New Testament Use of the Old Testament (Grand Rapids: Baker Academic, 2007) 607–94, on p. 670.

214

Katja Kujanpää

change. Instead, when he interweaves the “spirit of stupefaction” with the quotation and modifies the negations, he prepares the way for the development of his argumentation: although God now actively gives the spirit of stupefaction that causes the unreceptive condition, there is hope that he will allow the stupefaction to pass in the future. In summary, Paul conflates the two texts to include in the quotation an idea that is consistent with the following argumentation. What was observed with regard to Rom 11:8 also applies to other conflated quotations (e.g. Rom 9:33; 11:26–27): neither quotation would have exactly the effect that Paul wishes to create, but when conflated, they form a concise entity that is more suitable for his argumentation than either of the texts would be separately.

Consistency in Grammatical Forms The quotation in Rom 11:8 also exemplifies how the change of the network of relations may cause grammatical inconsistencies. If a quotation lacks an introduction formula, these inconsistencies can be useful, for they signal to the audience that Paul is quoting from an external source.47 However, in order to avoid unnecessary confusion caused by variation in tenses and persons, he frequently harmonises grammatical forms in a quotation with those of surrounding statements. In Rom 11:8, Paul changes the personal pronoun from the second-person plural to the third-person plural, from ὑμῖν to αὐτοῖς. This substitution can be attributed to Paul with confidence, for there is absolutely no support for it in the textual transmission of the Septuagint and it is obvious why he needs to modify the pronoun. In Deuteronomy, Moses is giving his speech to the people of Israel, whereas in Rom 11 Paul speaks of the unbelieving Israel in the third person (cf. Rom 11:7, 9–14). For Paul it is crucial that the audience of Romans does not feel itself addressed by a quotation that is meant to account for the unbelief of the majority of Israel.48 By changing the pronoun, he makes it unequivocal that αὐτοῖς refers to the “others” (οἱ λοιποί) of the previous verse, who have not acquired what they sought (11:7). Moreover, in the next verses (9–10), Paul quotes Ps 68:23–24 (69:23–24 MT) in such a form that the third-person plural pronoun occurs four times. For Paul’s purposes it is important that both quotations seem to refer to the same group, “the others”. After his adjustment, the personal pronoun of Deut 29:4 is consistent with what stands both before and after it. Furthermore, the accusative (instead of dative) form of πνεῦμα κατανύξεως in Rom 11:8 is in all probability related to the change of the network of relations, although in this case the textual tradition of the Septuagint is divided. In contrast 47  Cf. the definition of quotation in n. 1; see also Koch, Schrift, 13; Stanley, Paul and the Language, 37. 48  Cf. Koch, Schrift, 111; Stanley, Paul and the Language, 159–60.

Adjusted to the Argument: Tracing Paul’s Motives

215

to the personal pronoun discussed above, here the expression from Isa 29:10 is not harmonised with Paul’s statements surrounding the quotation but with the other scriptural text in the conflated quotation, Deut 29:4. Since Paul conflates the two texts and begins the quotation with ἔδωκεν (from Deut 29:4)—and not with πεπότικεν as in Isa 29:10—he has to replace the dativus instrumentalis with an accusative.49 As this alteration is obviously necessitated by the conflation, it is improbable that Paul was influenced by the minority reading of the Septuagint with πνεῦμα in the accusative.50

Conclusions This article has presented examples of different types of deliberate changes that Paul makes to the wording of scriptural quotations: how he substitutes original verbs in order to create a catchword connection, switches the place of parallel items in order to present the more relevant one first, adds a word for the sake of clarity, conflates two passages to create a quotation that better suits his argument, and changes grammatical forms to improve the consistency between the quotation and the rest of the argumentation. These alterations only serve as examples, for the list of Paul’s techniques of modifying wording could be extended to include different kinds of omissions and changes in word order, for example. Similarly, the list of Paul’s motives for committing the changes is far from exhaustive. All of the modifications examined here are connected to the change in the network of relations (to use Sternberg’s terminology) that surrounds the quoted passage. The modifications of grammatical forms are directly necessitated by the new frame of the quotation, but there are numerous cases in which Paul could have managed with the original wording as well. However, the modifications improve the aptness of the quotations and the consistency and flow of the argumentation.

49  Cf. Wilk, Bedeutung, 18 n. 6. 50  See above n. 42. Even if the accusative form of the minority reading were pre-Pauline, there is still no need to assume that Paul’s Vorlage contained it (cf. Wagner, Heralds, 243 n. 73). The variant is not especially widespread, and, according to Koch, in general Paul’s quotations of Isaiah are closest to the text type represented by uncials A and Q (Koch, Schrift, 48–51, 170 n. 48), which here contain the dative form. The variant may be a linguistic improvement (ibid., 170 n. 48; Wagner, Heralds, 243 n. 73) that corresponds to the normal usage of the verb, for ποτιζω is typically used with a double accusative; see BDR 155:7 n. 7; R. Helbing, Die Kasussyntax der Verba bei den Septuaginta: Ein Beitrag zur Hebraismenfrage und zur Syntax der Koinē (Göttingen: Vandenhoeck & Ruprecht, 1928), 49. Cf. Gen 19:32; Judg 4:19; 1 Kgdms 30:11; and Sir 15:3. The construction where the drink is in the dative is rarer: see 3 Macc 5:2 (Bauer). Therefore, it appears that Paul ended up with the same form as some revisers of the Septuagint text, but for different reasons; Paul had to adapt the case to match the verb, whereas the revisers were motivated by improving the language.

216

Katja Kujanpää

In most of the cases addressed above, there are no serious textual problems, for the manuscript evidence suggests that the changes do not represent pre-Pauline variants, but should be attributed to Paul. This, however, is not always the case. In some cases it is probable that Paul’s wording represents a Greek text that had been revised in order to bring it closer to a Hebrew text.51 Moreover, the transmission of both the text of Romans and of the Septuagint shows traces of harmonisations. Where Paul’s reading of a quotation differed from that of the Septuagint, scribes occasionally “corrected” the wording. The reverse phenomenon took place when scribes gave priority to Paul’s wording and harmonised the wording of a Septuagint manuscript with it.52 This two-way influence in the textual transmission of Paul’s letters and the Septuagint has produced several intriguing cases in which one has to ask which alternative is more probable: did Paul quote a pre-Pauline variant reading of the Septuagint, or did the influence of the New Testament cre51  A well-known example is Paul’s conflated quotation in 9:33, the middle part of which derives from Isa 8:14. Paul’s wording follows the Hebrew closely and greatly resembles the trans­ lations of Aquila, Symmachus, and Theodotion, but it clearly diverges from the Septuagint, the translation of which completely rewrites the passage. Rom 9:33 (Isa 28:16 + 8:14)

Isa 8:14 LXX

Isa 8:14 MT

ἰδοὺ τίθημι ἐν Σιὼν λίθον προσκόμματος καὶ πέτραν σκανδάλου, καὶ ὁ πιστεύων ἐπ᾿ αὐτῷ οὐ καταισχυνθήσεται.

καὶ ἐὰν ἐπ᾿ αὐτῷ πεποιθὼς ᾖς, ἔσται σοι ‫וְ ָהיָ ה‬ εἰς ἁγίασμα, καὶ οὐχ ὡς λίθου ‫ּול ֶא ֶבן נֶ גֶ ף‬ ְ ‫ְל ִמ ְק ָּדׁש‬ προσκόμματι συναντήσεσθε αὐτῷ οὐδὲ ‫ּולצּור ִמ ְכׁשֹול‬ ְ ὡς πέτρας πτώματι

The Hebrew wording contains a striking tension: “And he [=God] will become a sanctuary and a stone of offense and a rock of stumbling.” That God becomes at the same time a sanctuary and a stone of offense causes an interpretative problem, which the Greek translator solves by inserting a conditional clause, a new main verb, and negations before the stone and the rock: “And if you trust/believe in him, he will become a sanctuary for you, and you will not encounter him as a stone of stumbling, nor as a rock of falling…” Paul’s wording has neither such an ifclause nor negations before the stone and the rock, but instead preserves the harshness of the Hebrew. In addition, with its genitive constructions (λίθον προσκόμματος, πέτραν σκανδάλου) Paul’s syntax diverges from that of the Septuagint (λίθου προσκόμματι, πέτρας πτώματι) and represents an approximation towards the Hebrew construct chains (Koch, Schrift, 60; Stanley, Paul and the Language, 123). The same syntax can also be found in Aquila, Symmachus and Theodotion, which makes it probable that Paul renders here unaltered the wording of his Vorlage. Another deviation from the Septuagint is Paul’s use of σκάνδαλον instead of πρόσκομμα. As σκάνδαλον seems to be a word favoured by Paul (Rom 11:9; 14:13; 16:17; 1 Cor 1:23; Gal 5:11), the possibility cannot be excluded that he substituted the word himself. However, since σκάνδαλον is also used by Aquila (and, according to Eusebius, by Symmachus as well), it is more probable that the word derives from the pre-Pauline Hebraizing revision (Wilk, Bedeutung, 23 n. 14). Traces of Hebraising revision can be found especially in quotations from Isaiah, but the phenomenon is not limited to that book. On the phenomenon in general, see Koch, Schrift, 57–9; Stanley, Paul and the Language, 14–15; Wilk, Bedeutung, 19–20. 52  For an example of the former phenomenon, see n. 8, and, for the latter, n. 16.

Adjusted to the Argument: Tracing Paul’s Motives

217

ate the textual variant in the manuscript tradition? Or with regard to the text of Romans, which one is the original reading, the one agreeing with the Septuagint or the one deviating from it?53 It is crucial to examine the less problematic cases, as has been done in this paper, to form a general idea of Paul’s “style”. Which techniques of modification are characteristic of him, and what are his typical reasons for changing something in the wording? In text-critically ambiguous cases, it is helpful to have parallels with both similar modifications in other quotations and similar motives for committing the change. If there are several probable cases in which Paul changes personal pronouns, these parallels can serve in an ambiguous case as an additional argument in favour of the Pauline origin of the alteration. Similarly, if it can be demonstrated that catchword connections are important for Paul and that he alters the wording of quotations in order to create them, this observation may help to determine whether a reading is more likely to originate from Paul or a scribe transmitting the text. Such arguments cannot be used as proof of Pauline origin, for Paul was not the only author in antiquity to alter personal pronouns in a quotation or create catchword connections. However, conformity with Paul’s well-attested tendencies and practices increases the probability that Paul is behind the modifications. In conclusion, analysis of Paul’s intentions in all cases in which he modifies the wording of quotations is integral to solving the textually obscure quotations. Finally, the fact that Paul alters the wording of quotations does not imply that he considered the exact wording to be insignificant. On the contrary, the frequency of alterations rather indicates that the wording of the quotations is important, and sometimes even crucial, for his argument.54 For the effectiveness of argumentation, it is advantageous that there are verbal links with his own statements, that interesting matters are presented first in the quotations, that the quotations bring into discussion new concepts and themes, that it is clear who is speaking to whom, that there are no syntactical problems in the quotations, and that personal pronouns refer to the right groups. Therefore, Paul’s attitude towards the wording is not indifferent—he shows great care in modifying wording so that the quota­ tion expresses exactly what he intends it to convey. Consequently, although the scriptures of Israel are without doubt authoritative for Paul, it appears that their authority does not lie in the immutability of their wording. If Paul has to decide 53  Martin Karrer and Ulrich Schmid rightly emphasise that in general in early Christian codices the Old Testament and the New Testament were copied independently of each other, that is, without harmonising tendencies: M. Karrer/U. Schmid, “Old Testament Quotations in the New Testament and the Textual History of the Bible—the Wuppertal Research Project”, in M. Karrer, et al. (ed.), Von der Septuaginta zum Neuen Testament: Textgeschichtliche Erörterungen (ANTF 43 Berlin: De Gruyter, 2010) 165–96, on pp. 167, 185. However, this overall picture cannot be used as a rule of thumb to solve each particular case. 54  Cf. Koch, Schrift, 347: “Gerade weil der Wortlaut der Zitate selbst für Paulus eine so große Bedeutung hat, verändert er ihn z. T. massiv.”

218

Katja Kujanpää

between preserving the original wording or highlighting the relevance and suitability of the quotation for the matter at hand, Paul tends to choose relevance. This readiness to actualise the quotation by adjusting its wording enables him to use quotations in important argumentative functions.

Bibliography Barrett, C.K., A Commentary on the Epistle to the Romans (BNTC; London: Adam & Charles Black, 1962). Bauer, W., Griechisch-deutsches Wörterbuch zu den Schriften des Neuen Testaments und der frühchristlichen Literatur (6th edn, Berlin: de Gruyter, 1988). Cranfield, C.E.B., A Critical and Exegetical Commentary on the Epistle to the Romans (Vol. 2, ICC Edinburgh: T & T Clark 1983). Ellis, E.E., Prophecy and Hermeneutic in Early Christianity: New Testament Essays (WUNT 18; Tübingen: Mohr, 1978). Fuß, B., “Dies ist die Zeit, von der geschrieben ist...”: Die expliziten Zitate aus dem Buch Hosea in den Handschriften von Qumran und im Neuen Testament (NTAbh Neue Folge 37; Münster: Aschendorff, 2000). Helbing, R., Die Kasussyntax der Verba bei den Septuaginta: Ein Beitrag zur Hebraismenfrage und zur Syntax der Koinē (Göttingen: Vandenhoeck & Ruprecht, 1928). Hofius, O., “Das Evangelium und Israel: Erwägungen zu Römer 9–11”, ZTK 83 (1986) 297–324. Hübner, H., Gottes Ich und Israel: Zum Schriftgebrauch des Paulus in Römer 9–11 (FRLANT 136; Göttingen: Vandehoeck & Ruprecht, 1984). Jewett, R., Romans: A Commentary (Hermeneia; Minneapolis: Fortress, 2006). Karrer, M. & U. Schmid, “Old Testament Quotations in the New Testament and the Textual History of the Bible—the Wuppertal Research Project”, in M. Karrer/S. Kreuzer/M. Sigismund (ed.), Von der Septuaginta zum Neuen Testament: Textgeschichtliche Erörterungen (ANTF 43 Berlin: De Gruyter, 2010) 165–96. Kautzsch, A.F., De Veteris Testamenti Locis a Paulo Apostolo Allegatis (Whitefish, MT: Kessinger Publishing, 2010 [reprint, original 1869]). Kloppenberg, J.S., The Formation of Q: Trajectories in Ancient Wisdom Collections (Studies in Antiquity and Christianity; Philadelphia: Fortress 1987). Koch, D.-A., Die Schrift als Zeuge des Evangeliums: Untersuchungen zur Verwendung und zum Verständnis der Schrift bei Paulus (BHT 69; Tübingen: Mohr 1986). Käsemann, E., An die Römer (2nd edn, HNT 8a; Tübingen: Mohr Siebeck, 1974). Meiser, M., “Relevanz der Kirchenväterzitate für die Textgeschichte der neutestamentlichen Zitate aus der Septuaginta”, in M. Karrer/S. Kreuzer/M. Sigismund (ed.), Von der Septuaginta zum Neuen Testament: Textgeschichtliche Erörterungen (ANTF 43; Berlin: De Gruyter, 2010) 283–316. Norton, J.D.H., Contours in the Text: Textual Variation in the Writings of Paul, Josephus and the Yahad (LNTS 430; New York: T & T Clark, 2011).

Adjusted to the Argument: Tracing Paul’s Motives

219

Patterson, S.J., The Gospel of Thomas and Jesus (Sonoma: Polebridge 1993). Seifrid, M.A., “Romans”, in G.K. Beale/D.A. Carson (ed.), Commentary on the New Testament Use of the Old Testament (Grand Rapids: Baker Academic, 2007) 607–94. Shum, S.-L., Paul’s Use of Isaiah in Romans: A Comparative Study of Paul’s Letter to the Romans and the Sibylline and Qumran Sectarian Texts (WUNT II 156; Tübingen: Mohr Siebeck, 2002). Stanley, C.D., Paul and the Language of Scripture: Citation Technique in the Pauline Epistles and Contemporary Literature (SNTSMS 74; Cambridge: Cambridge University Press, 1992). Stegner, W.R., “Romans 9:6–29—a Midrash”, JSNT 22 (1984) 37–52. Sternberg, M., “Proteus in Quotation-Land: Mimesis and the Forms of Reported Discourse”, Poetics Today 3 (1982) 107–56. Thackeray, H.St.J., “The Greek Translators of the Four Books of Kings”, JTS 8 (1907) 262–78. Wagner, R.J., Heralds of the Good News: Isaiah and Paul “In Concert” in the Letter to the Romans (NTS 101; Leiden: Brill, 2002). Wilk, F., Die Bedeutung des Jesajabuches für Paulus (FRLANT 179; Göttingen: Vandenhoeck & Ruprecht, 1998). Wilk, F., “Letters of Paul as Witnesses to and for the Septuagint Text”, in W. Kraus/R.G. Wooden (ed.), Septuagint Research: Issues and Challenges in the Study of the Greek Jewish Scriptures (SCS 53 Atlanta: SBL, 2006) 253–71. Zuntz, G., The Text of the Epistles: A Disquisition upon the Corpus Paulinum: The Schweich Lectures of the British Academy 1946 (London: Oxford University Press, 1953).

Andrés Piquer Otero Creative Philology and Glosses: Secondary Versions of Kingdoms and Lexical Accumulation or Mutation1 მელნად ვიხმარე გიშრის ტბა და კალმად მე ნა რხეული, ვინცა ისმინოს, დაესვას ლახვარი გულსა ხეული. For ink I have taken a pool of jet and for stylus a crystal waving; whoever hears, a jagged sword his heart will pierce.

Introduction Being in Georgia for the first time, I wanted to begin this paper with a couple of lines from Šota Rustaveli’s The Knight in the Panther Skin, as it is somehow fitting with some of the ideas underlying the research I am presenting here. Although by no means knowledgeable in medieval Georgian literature, I think that Rustaveli’s introduction of the epic—in discussing the craft and aim of writing from remarkable Platonic perceptions, which seem to share quite a bit with the art of the troubadours one Mediterranean Sea away—can be a nice example of the issue at hand when dealing with the sequence of textual traditions, from the early to the late: We always tend to believe that we know better and therefore that we can make our texts endure. I mean, even though we scholars, as scribes and translators in earlier eras, are painfully aware of how much lore has been lost to time, we still think that we are aptly prepared to do a great job. And, in the world of books, as in a few others, a great job is a job which will endure. Be it the latest critical edition, a Renaissance Polyglot, the Hexapla, or a manuscript which is just a number and scant catalogue data in a Verzeichnis, the crafter of a text tries to create the clearest, most correct or complete version of a book or whole Bible. In this, now as then, indicators of quality may change, of course, but a shared mark of distinction seems to be the access in the editorial process to multiple earlier editions, be they versions, individual codices or even commentaries. Capacity of choice—agency in choosing the best reading according to the editor’s interests— is therefore an ideal situation, which, nevertheless, leads to different scenarios. In some cases, it involves a full-fledged textual revision. A typical case would be the kaige recension of the Septuagint. A better text (from the point of view of 1  The research leading to this paper has been carried out within the Research Project Edición electrónica políglota sinóptica de 1–2 Reyes (FFI2013-40920-P), funded by the Secretary of Research of the Spanish Government.

222

Andrés Piquer Otero

the kaige editors) was thus achieved by modifications of readings.2 This time I am not looking at that kind of activity, crucial as it is for unraveling the history of the Greek text(s) and their Vorlagen, but to later phenomena, which involve both access to a growing number of texts as centuries passed by and to different cultural and scholarly mindsets, which have led to the production of extended or composite readings, a form of longer text which, clearly not belonging to an earlier stage of the book’s history, may witness: 1st  important textual evidence somehow hidden in what appears as a mere gloss or alternate rendering created by a scribe/translator; 2nd exegetical reading traditions reflected in the translation of particular terms or, again, the creation of detailed readings or glosses; 3rd approaches which could be considered “philological” in a creative way, as they imply an intervention into the text in order to clarify terms or somehow produce an output more in agreement with style, meaning, or what the scholars of the time might have conceived as a “proper” biblical text.

This paper, therefore, aims at presenting my ongoing research on a proposed typology of long or composite readings oriented towards a textual-historical analysis of the secondary versions of Kings. I have selected some remarkable phenomena which involve, this time, the Coptic versions of the Septuagint and the Arabic text of Walton’s Polyglot. The cases presented may shed some light on how the creation of long readings has an impact both in textual criticism and in the study of translators’ and scribes’ cultural background.

Double Readings from Different Text-Types in the Sahidic Version3 First, I will examine some cases where the Coptic translator has created a doublet which is relevant for the textual history of the Septuagint, as it may combine a 2  See D. Barthélemy, Les Devanciers d’Aquila (Leiden: Brill, 1963). For a present-day consideration, also including the problem of secondary versions, see A. Piquer Otero/P. Torijano Morales/J. Trebolle Barrera, “Septuagint Versions, Greek Recensions and Hebrew Editions: The Text-Critical Evaluation of the Old Latin, Armenian and Georgian Versions in III–IV Regnorum”, in H. Ausloos et al. (ed.), Translating a Translation: The LXX and its Modern Translations in the Context of Early Judaism (BETL 213; Leuven: Peeters, 2008) 251–81. 3  The main problem when facing the Coptic traditions of 3–4 Kgdms is the fragmentary state of the evidence. There is no single complete manuscript, and most fragments are short, 3–4 chapters in extent in the most fortunate circumstances. These fragments span several centuries and come from many different codices and other materials, such as lectionaries or katameros. Lacking a published full, detailed inventory, codicological data have to be gleaned from different inventory journal papers since the early 20th century. For references and a detailed overview, see P. Nagel, “Old Testament, Coptic Translations of ”, in Aziz Z. Atiya (ed.),

Creative Philology and Glosses: Secondary Versions of Kingdoms

223

kaige or Hexaplaric reading with Old Greek materials. Some cases are straightforward and hard to deny, but even those seem to imply a certain amount of creativity or stylistic desire in the translator’s mind. A good example of this is 3 Kgdms 1:34 and 39. In both cases, the Sahidic text has ϩⲉⲛⲥⲁⲗⲡⲓⲅⲝ ⲛⲁϣⲧⲁⲡ, “trumpets of bone”. This seems a combination between the B text and majority Greek reading κερατίνῃ with the Lucianic and Old Latin ἐν σάλπιγγι (de tuba.) This double reading is not attested in the rest of the Greek textual tradition and therefore seems to indicate a creative action by the Coptic translator, who drew from manuscript sources in order to produce a “fuller and clearer” text. A similar case indicates that this is not just an isolated example. In a different fragment, the text of 3 Kgdms 22:17 shows an analogous procedure: the majority Greek reads ποίμνιον, “flock”, whilst the L group has πρόβατα, “sheep”. The Coptic renders ⲟⲩⲟϩⲉ ⲛ̅ⲧ̅ⲃ̅ⲛⲉ, “a flock of sheep”. As in the earlier case, two different textual traditions are merged, not as a mere gloss or clumsy marginalia, but by creating a distinct nominal phrase. These cases are important both as further textual evidence for the different text-types (in the form of doublets, as usual in late materials4) and also for the interests and mindsets of the crafters of the Sahidic translation of the biblical text. There are other instances where a similar procedure has operated. In 3 Kgdms 21:11, the vast majority of the Greek tradition renders the king of Israel’s boast as μὴ καυχάσθω ὁ κυρτὸς ὡς ὁ ὀρθός, “let not the hunchback boast as the straight one”. Only in the Syro-Hexapla an alternate version in the Greek tradition (attributed to Aquila) has been preserved: ‫ܠܐ ܢܫܬܒܗܪ ܗܘ ܕܡܬܐܣܪ ܐܝܟ ܗܘ ܕܡܬܫܪܐ‬, easily retroverted into Greek as μὴ καυχάσθω ὁ ζωννύμενος ὡς ὁ λυόμενος, “let not the girt one boast as the loose one.” The Coptic fragment has included both readings as ⲙ̅ⲡⲣ̅ⲧⲣⲉⲡⲉⲧⲟ ⲛ̅ϫⲟ ϣⲟⲩϣⲟⲩ ⲙ̅ⲙⲟⲥ ⲛ̅ⲑⲉ ⲙ̅ⲡⲉⲧⲥⲟⲩⲧⲱⲛ ⲉϥⲁϩⲉⲣⲁⲧ̅ϥ ⲁⲩⲱ ⲡⲉⲧⲥⲏϭ ⲙ̅ⲡ̅ⲣ̅ⲧⲣⲉϥⲣ̅ ⲑⲉ ⲙ̅ⲡⲉⲧⲙⲏⲣ, “let not the hunchback boast as the straight one who stands, nor the palsied one do as the one who is girt.” The Coptic translation, in agreement with the Aquila reading, is definitely wrong and likely influenced by the majority version in the previous sentence, losing somehow the military reference inherent to the action of girding and adding another depiction of physical incapacity, perhaps understanding the simple λυόμενος as παραλυόμενος. Nevertheless, this double reading presents additional evidence for the Aquila tradition and also shows the mindset and practice of Coptic translators, with an adaptation

The Coptic Encyclopedia Vol. 6 (New York: Macmillan) cols. 1836a–1940a; A. Piquer Otero, “Coptic, Arabic, Ethiopic”, in A. Salvesen (ed.), The Oxford Handbook of the Septuagint (New York: Oxford University [forthcoming]). 4  On the textual relevance of double readings in Kings, see e.g. J. Trebolle Barrera, Centena in libros Samuelis et Regum: Variantes textuales y composición literaria en los libros de Samuel y Reyes (Madrid: CSIC, 1989); Piquer Otero/Torijano Morales/Trebolle Barrera, “Septuagint Versions, Greek Recensions, and Hebrew Editions”.

224

Andrés Piquer Otero

which, though erroneous, clearly attempts to create a linked parallel literary unit by combining both textual traditions.5 In other cases, no tampering of this kind has operated; in 4 Kgdms 6:5, the translator seems to be baffled by two readings in the Greek tradition which make full sense in the context of the woodcutter losing his ax into the river. The reading attested by the L group, along with Hexaplaric codices (N 158 247) and materials assigned to the Three in the Syro-Hexapla, καὶ αὐτὸ κεχρημένον, “and it was borrowed”, is combined with the reading of the B text and the rest of the Greek codices, καὶ αὐτὸ κεκρυμμένον, “and it became hidden (lost)”, which, in all likelihood, seems to be an intra-Greek corruption due to itacism and the confusion between kappa and chi. Again, the Coptic translator is creating a double reading with available textual data: ⲛ̅ⲧⲁⲓ̈ϫⲓ̅ ⲡⲁⲓ̈ ⲉⲡⲉⲩϣⲁⲡ ⲁⲩⲱ ⲉⲓⲥϩⲏⲏⲧⲉ ⲁϥⲱⲙⲉⲥ, “and I got it on loan and lo, it has sunk!” He does display some level of literary acumen in connecting the concept of “becoming hidden” to the context and in creating a form of pathos by the addition of “lo” with the particle ⲉⲓⲥϩⲏⲏⲧⲉ; one could even assume that the Egyptian cultural context played a role in the preservation and interpretation of this reading. Sinking under water and the capacity to recover lost items or people from a river or lake is a recurring topic both in Egyptian fiction since the Middle Kingdom6 (with remarkable examples from the Demotic era7) and in ritual practice in the Graeco-Roman period,8 and that could explain the interest of the Coptic translator or scribe in introducing this double reading. Elisha’s role as a thaumaturge was thus underscored within the cultural coordinates of the translator.

Onomastics, Toponymy, and Glosses This leads in turn to other more complex cases, where the Coptic translation could be incorporating the remains of variant Greek traditions, but one cannot be certain of it. I mean, the Sahidic text preserves readings which can be found 5  I have treated this case in A. Piquer Otero, “The Secondary Versions of Kings: Variants and Renderings Between Vorlagen and Ideology”, in S. Kreuzer/M. Meiser/M. Sigismund (ed.), Die Septuaginta: Orte und Intentionen. 5. Internationalen Septuaginta-Tagung Wuppertal 2014 (Tübingen: Mohr Siebeck, 2018), 244–55. 6  One of the most remarkable examples is to be found in Papyrus Westcar, see the edition and comments in M. Lichtheim, Ancient Egyptian Literature Vol. 1. The Old and Middle Kingdoms (Berkeley: UC Press, 1973), 214–21. 7  Raising a dead corpse and spirit from the depths of the river is a recurring element of the composition in the Setne I tale of the Demotic era. See M. Lichtheim, Ancient Egyptian Literature Vol. 3. The Late Period (Berkeley: UC Press, 1973), 125–49. 8  See R. Ritner, “Necromancy in Ancient Egypt”, in L. Ciraolo/J. Seidel (ed.), Magic and Divination in the Ancient World (Ancient Magic and Divination 2; Leiden: Brill-Styx, 2002) 89–96.

Creative Philology and Glosses: Secondary Versions of Kingdoms

225

in different passages of the Septuagint and fall in the realm of the practices of Septuagint translators or editors, but are not attested in Greek in a particular verse. The situation is frequent with divine onomastics and with some toponymy. In 3 Kgdms 19:10, ⲡ̅ϫⲟⲉⲓⲥ ⲡ̅ⲛⲟⲩⲧⲉ ⲛⲛ̅ϭⲟⲙ ⲡⲡⲁⲛⲧⲟⲕⲣⲁⲧⲱⲣ, “the Lord, God of Hosts, All-Powerful”, LXX τῷ κυρίῳ παντοκράτορι, is clearly an expansion, as is the similar case in 19:14, ⲡ̅ϫⲟⲉⲓⲥ ⲡⲡⲁⲛⲧⲟⲕⲣⲁⲧⲱⲣ ⲡ̅ϫⲟⲉⲓⲥ ⲛⲛ̅ϭⲟⲙ ⲡ̅ⲛⲟⲩⲧⲉ ⲙ̅ⲡ̅ⲓ̅ⲏ̅ⲗ̅, “the All-Powerful Lord, Lord of Hosts, God of Israel”, which also renders LXX τῷ κυρίῳ παντοκράτορι. Although some Greek witnesses show partial expansions (in v. 14 L adds θεῷ, with no reference to Israel; in v. 10 Latin Origen and Theodoretus add θεῷ Ἰσραηλ), no other source includes the doublet for παντοκράτορι, combining a literal with a non-literal translation, both of them attested throughout the Septuagint.9 Here, the Sahidic text seems to be intentionally accumulating divine names, perhaps for completion’s sake, perhaps—though such research would go beyond the scope of this paper—in relation to the relevance of the divine name, and the epithet “Lord of Hosts” in particular, within the wider context of Coptic magical formulae in Late Antiquity.10 A similar situation may be found with some place names, which, in the vast majority of Greek witnesses, have come across as transcriptions of the Hebrew. In 4 Kgdms 14:7, the Hebrew toponym ‫יא־ה ֶמּ ַלח‬ ַ ֵ‫ ג‬is transcribed as γαιμελα or other intra-Greek variants thereof. The Sahidic text, nevertheless, reads ⲡⲓ̈ⲁ ⲙ̅ⲡⲉϩⲙⲟⲩ, “valley of salt”, a literal translation only attested in Hexaplaric sources (Armenian and Syro-Hexaplaric, together with Eusebius’ Onomasticon) as an Aquila and/or Symmachus reading. Thus, one has to decide whether the Coptic was inserting here a Hexaplaric reading of the Three as its main text or whether a tendency to translate Hebrew toponyms was prevalent at some point in Coptic scribal practice, regardless of the presence of a Greek source along those lines or not. This could be confirmed by the case of 4 Kgdms 15:29, where the Sahidic ⲁⲃⲉⲗ ⲡⲏⲓ̈ ⲙ̅ⲙⲁⲁⲭⲁ, “Abel, House of Maacha”, is a literal rendering of the Hebrew underlying the Greek transcription Αβελβαιθαμααχα. No Greek source has produced a translation of βαιθ as “house”. Additionally, the Sahidic text is longer, as ⲡⲏⲓ̈ ⲙ̅ⲙⲁⲁⲭⲁ is followed by ⲁⲩⲱ ⲃⲉⲑⲁⲙ ⲙ̅ⲛ ⲇⲉⲃⲗⲁⲑⲁ, “and Betham and Deblatha”. It is hard to ascertain the provenance of these two new toponyms inserted into the list, but it is possible that the origin of the longer reading was a conflation of the attempt to translate (Αβελ)βαιθαμααχα with a return to a transcription of the Greek, as ⲃⲉⲑⲁⲙ would reproduce the βαιθαμ element clearly. The doublet then 9  See Piquer, “The Secondary Versions”, for a more detailed treatment; see also, though not always in agreement with the proposed conclusions, the study on the typology of renderings of the Tetragrammaton in Z. Talshir, “The Representation of the Divine Epithet ‫ צבאות‬in the Septuagint and the Accepted Division of the Books of Kingdoms”, JQR 1–2 (1987) 57–75. 10  See the global treatment in M.W. Meyer/R. Smith, Ancient Christian Magic. Coptic Texts of Ritual Power (Princeton: Princeton University, 1999).

226

Andrés Piquer Otero

became corrupted and the second half understood as additional towns. The surprising appearance of ⲇⲉⲃⲗⲁⲑⲁ, a transcription of the Greek form of the Hebrew ‫ר ְב ָלה‬,ִ Δεβλαθα, could be due to a scribe’s attempt to make sense of the text in a later phase, perhaps by choosing a toponym more familiar due to the prominence of Riblah in the text of Jeremiah regarding the end of Judah’s monarchy.11 In any case, these two examples show that at times it is hard to ascertain whether the translators into Coptic were reflecting some Greek sources (i.e., Hexaplaric readings, as in the case of 14:7) lost to us or if some of the innovations introduced into the text (mistaken as they may be) were actually attempts by the Coptic scribes to do philological/historical analysis in order to produce a better text.12 A similar approach is useful for placing in their right (and complex) contexts some cases of glosses, which seem to be typical of the Coptic mindset in translating and transmitting the LXX. Glosses are typically used to explain difficult Greek words or unclear passages or ideas, but some of the cases are not straightforward, or, at least, indicate a modicum of sophistication. In 3 Kgdms 19:4, the plant under which Elijah lies, ραθμ, is presented as such with an explanation: ⲟⲩⲥϣⲏⲛ ⲉϣⲁⲩⲙⲟⲩⲧⲉ ⲉⲣⲟϥ ϫⲉ ϩⲣⲁⲧⲙⲉⲛ, “a tree which is called Hratmen”. The gloss would be a typical case of explanation of a transcribed, untranslated, word, but in this case, again, one has to decide between independent activity or a shared tradition, as the Old Latin sub virgultis Raphem and the Ethiopic ታሕተ፡ዕፀ፡ተርሜን፡ (“under a Tarmen tree”) not identical readings in this case, but attempts to explain the transcribed word. In this case, the Sahidic and other versions would agree in spirit, if not in the literal wording, with the procedure attested in some Greek manuscripts (121 247 ἀρκεύθου ῥαθμεν; 158* ἀρκείθου ῥαθαμειν.) At times, it requires some thought and research to determine, if possible, the perception of the original word leading to a gloss. For instance, in 4 Kgdms 14:13 the term γωνίας, “corner”, is transcribed as ⲕⲟⲛⲓⲁ, and then supplemented by the clause ⲉⲧⲉ ⲧⲉⲧⲟ ⲛⲕⲟⲟϩ, “which was a corner”. It is likely that the gloss came about as the text was reinterpreted, taking the Greek transcribed term ⲕⲟⲛⲓⲁ as the name of the gate, thus creating a symmetry between the first gate in the measurement, the gate of Ephraim, and the gate of the corner, here the “gate of Konia”. In other cases, the text follows the literary structure of a gloss with a clearly grammatical aim of intervention: to smoothen out a rough or complicated text. In 4 Kgdms 1:6, ἐν τῇ Βααλ μυῖαν θεὸν Ακκαρων, certainly a complex Greek syntax, taking Baal as a feminine dative with an instrumental function, is rendered as ϩⲙ̅ ⲡⲃⲁϩⲁⲗ ⲡⲉⲧⲟⲩⲙⲟⲩⲧⲉ ⲉⲣⲟϥ ϫⲉ ⲡⲁϥ ⲛⲛⲟⲩⲧⲉ ⲁⲕⲕⲁⲣⲱⲛ, “by Bahal, who is 11  See Jer 39:6; 52:9, 10, 26, 27. 12  The problem of Hebraistic tendencies in (later) Coptic manuscript traditions have been treated in R. Vollandt, ‫הבראיסטים קופטים בימי הביניים? על מסירת התפסיר של רס"ג‬, Tarbiz 83 (2015) 71–86.

Creative Philology and Glosses: Secondary Versions of Kingdoms

227

called the divine fly of Akkaron”. Besides cleaning up the grammar, the gloss has ideological import, as the Coptic version does away with Baalzebub as a specific deity, and thus understands the reference to the fly as an epithet of the generic Baal, maybe placing the episode fully within the Baal-Yahweh dispute of the Elijah narratives.13 A close look at glosses in a wider context allows us also to detect that the translation process resulting in the witnesses which have come to us in the Sahidic corpus is long and the result of several different hands, even in books in such a fragmentary state as 3–4 Kgdms. In 4 Kgdms 2:12, the Coptic translates the Greek ἅρμα as ⲡⲉⲛⲓⲟⲭⲟⲥ, “charioteer”. This explanatory reading, regardless of whether it was purely Coptic in nature or part of a wider reading tradition (as could be indicated by OL agitator), clashes with the text of 3 Kgdms 22:34, where one scribe felt the need to explain the Greek ἡνιόχος with the clause ⲉⲧϫⲱⲣ̅ⲙ̅ ⲉⲡⲉϥϩⲁⲣⲙⲁ, “who rode his chariot”. The need to explain the Greek word indicates that the explanation must have entered the Sahidic tradition at a different stage or phase than the free explanatory rendering of ἅρμα using that very same word. Thus, the process of glossing seems to be richer and less straightforward than mere explanation into Coptic of obscure Greek words; it may reflect literary and ideological interests, as well as being a potential means for approaching the history and variety of Coptic translations.

Textual Adaptation and Expansion This attitude extends also to cases which cannot be interpreted, structurally speaking, as glosses, but as other forms of textual alteration. They include issues of translation technique, more or less mechanical, which I will not detail here (e.g. the tendency to suppress the particle καὶ or changes in word-order due to Coptic syntax),14 but also some general tendencies which fall within the realm of style. One of the most salient ones is the recurring production of translations with two verbs when the Greek original only has one. The number of cases attested, despite the fragmentary state of the corpus of 3–4 Kgdms, is remarkable: 13  See the treatment of this theophoric in A. Piquer Otero/P. Torijano Morales, “Between the Search in the Word and the Asking to God: Two Mantic Verbs in the Textual History of Samuel-Kings”, in K. De Troyer/T.M. Law/M. Liljeström (ed.), In the Footsteps of Sherlock Holmes: Studies in the Biblical Text in Honour of Anneli Aejmelaeus (Leuven: Peeters, 2014) 299–330. 14  On Coptic (Sahidic) translational issues, see E. Perttilä, “How to Read the Greek Text behind the Sahidic Coptic”, in A. Voitila/J. Jokiranta (ed.), Essays on Septuagint, Hebrew Bible and Dead Sea Scrolls in Honour of Raija Sollamo (JSJSup 126; Leiden: Brill, 2008) 367–78; E. Perttilä, Sahidic 1 Samuel: A Daughter Version of the Septuagint 1 Reigns (DSI 8; Göttingen: Vanderhoeck und Ruprecht, 2017).

228

Andrés Piquer Otero

3 Kgdms 19:9; 21:16; 22:15; 4 Kgdms 2:20; 4:12, 19, 31; 5:6; 6:4; 9:18, 19; 12:10, 17, 21; 13:14; 14:5 (2x), 10, 11, 19 (2x). A more detailed examination of this list is relevant: a) some cases focus on creating a more logical sequence of actions, as in 4 Kgdms 2:20, ⲁⲩϫⲓⲧⲥ ⲁⲩⲉⲓⲛⲉ, “and they took it and brought (it)” for καὶ ἔλαβον πρὸς αὐτόν, or 9:18 ⲁⲙⲟⲩ ⲛ̅ⲧⲟⲕ ⲛⲅ̅ⲟⲩⲁϩⲕ ⲛ̅ⲥⲱⲓ̈, “come and defect to me” for ἐπίστρεφε εἰς τὰ ὀπίσω μου. b) other passages seem to introduce an intensive element, as seems to be the case of 4 Kgdms 14:10, ⲕⲛⲁϩⲉ ϭⲉ ⲡⲉ ⲛ̅ⲅ̅ⲑ̅ⲃ̅ⲃⲓⲟ, “and thou shalt fall and be humiliated” for καὶ πεσῇ σὺ. c) finally, there is a peculiar usage which seems to show something which could be considered a philological awareness of style: the recurring tendency to produce two-verb clauses, which could be trying to emulate or mimic the Semitic solecisms of wayyaqom wa… or wayyešeb wa… by adding forms of ϩⲙⲟⲟⲥ (sit) or ⲧⲱⲟⲩⲛ (rise) to a clause which has a single verb in the rest of the LXX tradition, e.g. 4 Kgdms 12:11, ⲁⲩϩⲙⲟⲟⲥ ⲉⲩϫⲓ ⲏⲏⲡⲉ, “and they sat down to count the silver”; 3 Kgdms 22:15 ⲧⲱⲟⲩⲛ ⲛⲅ̅ⲃⲱⲕ, “rise and go”, 4 Kgdms 13:14, ⲁϥⲧⲱⲟⲩⲛ ⲁϥⲉⲓ ϣⲁⲣⲟϥ, “and he rose and came to him.” The Sahidic even creates an idiomatic-stylistic translation of the royal assassination formula in 4 Kgdms 14: καὶ ἐπάταξεν in v. 5 is translated as ⲁϥⲧⲱⲟⲩⲛ ⲁϥⲙⲟⲩⲟⲩⲧ, “he arose and slew”, and the following clause with the participle of the same verb, a remembrance of the rebellion, is rendered as a relative clause with the same pair, ⲛ̅ⲧⲁⲩⲧⲱⲟⲩⲛ ⲁⲩⲙⲟⲩⲟⲩⲧ, “those who had arisen and slain…”. As in the following verse the verb πατάσσω is translated simply with ⲙⲟⲩⲟⲩⲧ, so it seems that in v. 5 the author tried to achieve a strict parallelism, by producing a 2-verb translation in both cases, even when the first clause does not refer to royal assassination. Also, he must have taken into account that, besides the stylistic usage in producing the 2-verb idiom, the verb ⲧⲱⲟⲩⲛ would be connected to the notion of arising and rebelling against somebody, as witnesses in the expansion to v. 19: ⲁⲩⲉⲓⲣⲉ ⲉⲣⲟϥ ⲛ̅ⲟⲩϣⲟϫⲛⲉ ⲁⲩⲧⲱⲟⲩⲛ ⲉϫⲱϥ, “and they made a conspiracy and arose against him.” Besides this stylistic procedure, other interventions in the text also seem to indicate a philological/literary interest, at times paired with a need to explain the text. 4 Kgdms 16:3 seems remarkable, as the Coptic translator explains the practice of “passing a child through the fire” with elements of Egyptian context: ⲡⲉϥϣⲏⲣⲉ ⲇⲉ ⲁϥⲉⲓⲛⲉ ⲙ̅ⲙⲟϥ ϩⲛ ⲟⲩⲥⲁⲧⲉ ⲉⲩϫⲓ ⲉⲓⲉⲣⲃⲟⲟⲛⲉ ⲙ̅ⲙⲟϥ, “and he put his child in the fire in order to take the evil eye from him.” In the context of the idolatry of the kings condemned in this paragraph, the translator added materials which fit within the contemporary Coptic system of beliefs, as the popular folk-conception of the evil eye is well-attested and ⲉⲓⲉⲣⲃⲟⲟⲛⲉ is the technical term used to refer to it.15 Further research in this direction could shed some light on the reasons which 15  See W.E. Crum, A Coptic Dictionary (Oxford: Clarendon, 1939), 39.

Creative Philology and Glosses: Secondary Versions of Kingdoms

229

led the scribe to link fire with exorcism or repulsion of the evil eye in the context of Coptic magic in Late Antiquity. In other cases, reasons are connected to awareness of textual problems and a desire to smooth them out. In 4 Kgdms 2:14, Elisha’s prayer ends with a transcription of the Hebrew, αφφω (from ‫)אף הוא‬. Preciously few Greek manuscripts and Hexaplaric readings translate the transcription as καίπερ αὔτος or καὶ νῦν. The Sahidic text does translate, ⲧⲉⲛⲟⲩ, “now”, but then it proceeds to create a long reading with no parallel in extant sources: ⲧⲉⲛⲟⲩ ⲇⲉ ⲛ̅ⲧⲟϥ ⲡⲉ ⲡⲛⲟⲩⲧⲉ ⲙⲙⲟⲩⲛⲉⲓⲟⲟⲩⲉ, “and now he is also the God of waters.” Similarly, in 9:16, the Coptic translator adapted his translation in order to produce a more clear and precise word, substituting the national reference to Syrians οἱ Αραμιν with ⲛⲉⲇⲟⲝⲟⲧⲏⲥ, “the archers”. This desire to produce details could come from an intention to increase variation from the parallel materials in verse 15. The last two examples show an extreme in the adaptation or textual intervention of the translator-copyist in the biblical text. On the other hand, cases which could well follow in this line of interpretative/stylistic tampering are more puzzling when they show an element of agreement between versions. This seems to be the case of 4 Kgdms 6:12, where the Sahidic reads ⲙⲛ̅ ⲗⲁⲁⲩ ⲛ̅ϩⲏⲧⲛ̅ ⲡⲁϫⲟⲉⲓⲥ ⲡ̅ⲣⲣⲟ ϥⲓ ⲙ̅ⲡⲉⲕϣⲟϫⲛⲉ ⲉⲃⲟⲗ, “None of us, my lord the king, gave away your plan”, instead of the majority Greek οὐχί κύριέ μου βασιλεῦ. At first sight, this could be analyzed typologically as one of the previous instances of stylistic-literary retouching of the narrative, until one examines the wide scope of versions of this passage. The Old Latin (Vind.) reads here nemo te prodet rex; whereas the Latin Chrysostom presents nullus tua consilia denudauit. Coincidence of three different sources in producing closely related renderings of the same beginning of the verse is remarkable, and it is necessary to think carefully in order to envisage a scenario where these developments could take place. Either there was a reading tradition of the Greek which was shared or a variant Greek textual form which has been lost in the rest of Greek witnesses.

Eclecticism in the Arabic Text(s)? Usually, little attention is given, from the point of view of textual criticism, to the Arabic versions of the Bible. Certainly, diffusion and exposition of the Arabic texts in the West has been complicated by the eclectic character of most of the codices and materials used as the basis for the Renaissance Polyglots (Paris and London), or by the lack of publication of many fragmentary texts (namely the abundant biblical passages, either fragments of biblical codices or in commentaries, from the Cairo Genizah.) Most of the attention, even today, has been given to the books of the Pentateuch, due to the cultural importance in the Middle

230

Andrés Piquer Otero

Ages of Sa‘adya Ga’ōn’s Tafsīr and also to the abundance of Pentateuch codices and materials.16 Recent research in this area, leading to proposals about quite a fluid scenario of exchange between texts at the avail of different communities of Bible-readers both in Egypt and Syria in the medieval period, could be promising if extended to other biblical books, even if evidence is more scarce or harder to come up with. Here, I have just chosen to give a small look at the Arabic text of Kings in Walton’s Biblia Polyglotta. The text was analyzed in the early 19th Century By E. Roediger17 and his views are summarized in the still-fundamental Geschichte der Christlichen arabischen Literatur by G. Graf.18 His description reads: Die eine Vorlage ist Peš für I und II, III Kg 1–11 (13/14. Jh.?), IV Kg 12,17–25,30 (von einem anderen Übersetzer, 14. Jh.), die andere unmittelbar die hebräische Masorah, jedoch unter Benutzung auch eines chaldäischen (aramäischen) Targums, für III Kg 12,1–IV Kg. 12,16 übersetz von einem Juden aus Damaskus oder Umgebung.

The few cases I am treating here deal with the second section detailed in Graf ’s paragraph. Although his proposal seems to be supported by the data, my aim is to offer some readings which indicate a far more complex textual scenario in medieval Arabic versions. Although these translations are, in the vast majority of cases, too late for impacting the text-critical work of an eclectic edition, their importance is considerable when approaching the cultural context of Middle Ages biblical transmission and the vision of relationships between textual traditions in the Arabic-speaking world. First, it has to be noted that, regardless of the clear relationship of the Vorlage of the Arabic text with a philo-Masoretic tradition, the incipit of 2 Kings follows the naming conventions of the Greek-Christian tradition, ُ‫ٱلرا ِب ُع ِمن اَ ْس َف ِار ٱلْ ُملُوك‬ َّ ‫اَ ِلس ْف ُر‬, “Fourth Book of the Books of the Kings.” Obviously, the addition of incipits and colophons may be later and does not impact on the textual Vorlage itself, but it is definitely evidence of the fluidity of textual materials of the Bible in the East Mediterranean, where a Jewish version (speaking in terms of Vorlage) could be incorporated into a Christian context, as in the more studied and better-documented cases of usages of Sa‘adya’s Tafsīr by Coptic communities in Egypt.19

16  See R. Vollandt, Arabic Versions of the Pentateuch: A Comparative Study of Jewish, Christian, and Muslim Sources (Leiden: Brill, 2015). 17  E. Roediger, De origine et indole arabicae Librorum V.T. historicorum interpretationis (Halle, 1829). 18  G. Graf, Geschichte der christlichen arabischen Literatur. Erster Band. Die Übersetzungen (Vatican City: Biblioteca Apostolica Vaticana, 1944), 111. 19  See S.H. Griffith, The Bible in Arabic: The Scriptures of the ‘People of the Book’ in the Language of Islam (Princeton: Princeton University, 2013), 146–48; Vollandt, “‫הבראיסטים‬.”

Creative Philology and Glosses: Secondary Versions of Kingdoms

231

On a closer look, the text does indeed present, as indicated by Graf, a large amount of Targumic elements, saliently readings which involve an exegetical or euphemistic approach. Other cases do indicate a deeper relationship with an Aramaic text closely similar to Targum Jonathan, so it is necessary to keep the issue open on whether the translator was using a Hebrew text with the aid of the Targum, the Targum with occasional usage of MT, or was bound on creating a mixed translation by comparing both textual sources. Exegetical readings and euphemisms could be easily inserted into a more barebones MT-like text, and just a survey of 2 Kgs 1–2 gives a good list thereof, e.g.: 1:2 ‫ َطاغُ ْو َت‬in agreement with TJ ‫ָטעוּת‬ 1:3

‫ِسائِيل‬ َ ْ ‫ َك َّن لَي َْس إ َ ٌِل قَد اَقَا َم َس ْكيَنَتَ ُه َحال َّ َت ِف َءآلِ إ‬, a literal translation of TJ ‫ְה ָלא ְא ָלה ַקיָ ים‬

‫ ִד ְשׁ ִכינְ ֵתיה ָשׁריָ א ְביִ ְשׂ ָר ֵאל‬with a clear anti-anthropomorphic bent. 2:10 ‫ َك ُروح ِ ن ُ ُب َّو ِت َك‬, with TJ ‫בוּא ָתך‬ ְ ְ‫רוּח נ‬ ַ ‫ְב‬ 2:12 ‫ِسائِي َل ب َِص َل ِت ِه ِم ْن َم َرا ِكبِ ِ م َوفُ ْر َسانِ ِم‬ َ ْ‫إ‬

َ ‫ َي َس ِّي ِدي َي َس ِّي ِدي‬, in agreement with TJ ‫ַר ִבי ַר ִבי‬ ِ‫آأل ْج َر ُد ِل َءل‬

‫וּפ ָר ִשׁין‬ ָ ‫צלֹותיה ֵמ ְר ִת ִכין‬ ֵ ‫ְד ָטב ֵליה ְליִ ְשׂ ָר ֵאל ִב‬

These samples should suffice to indicate the interest of the producer of this Arabic text of 2 Kings in the Targumic materials and in his basically literal approach to the translation of the Targum paraphrases. On the other hand, as I have anticipated, some readings are harder to classify in this sense and seem to indicate that an Aramaic text was favored when translating more standard material. For instance, in 2 Kgs 1:2, the Arabic reads ‫ َم َر ِض هَا َذا‬, in agreement with TJ ‫רעי ֵדין‬ ִ ‫ִמ ַמ‬ against MT (no possessive suffix). This could point either to a direct usage of the Targum, to an interest in correcting MT via the Targumic reading, or to a variant in the Masoretic Text when compared to the textus receptus as attested by several medieval manuscripts, but in any case it requires placing the Arabic version in a wider context and analyzing it accordingly. Similarly, the variant in 2:24, ‫ ِذئ َبان‬, “two wolves” instead of “two bears”, seems to be more likely to happen when reading an Aramaic text, where the similarity between debba, “bear” and deba, “wolf ”, is more marked. The relationship between the Arabic text and Aramaic versions, nevertheless, seems to go beyond the likely Targum Vorlage, as there are some readings where the text departs from agreement with either MT or TJ in order to follow, against Graf ’s proposal, the Peshitta text. This is, for instance, the case in 2 Kgs 2:14, a problematic passage which I have already commented on in the previous section on Coptic variant readings. The problematic element ‫ ַאף הּוא‬in MT is treated in the Arabic as a temporal marker for the following sentence: ... َ‫“ فَ ِع ْند‬And then he struck the waters…” This reading agrees with the Peshitta text, but departs from

232

Andrés Piquer Otero

MT and from TJ, and it is also found in the Coptic text before the free Sahidic addition discussed above begins. A reading with no affiliation to MT or TJ is remarkable, because it may point in the direction of a more complex history for the translation and/or transmission of the Arabic text of 2 Kings20 treated here. Another couple of examples will underscore this situation. َ , “the widow, lady of the In 1 Kgs 17:17, the Arabic text reads ‫آأل ْر َم َ ِل َصا ِحبَ ِة آل َبيْت‬ house”, a reading which only agrees with the Coptic Bohairic text, ⲛ̅ϯⲭⲏⲣⲁ ⲧϭⲟⲓⲥ ⲙ̅ⲡⲓⲏⲓ. The reading is interesting because it opens up the scenario of connections between versions of the text of Kings. On one hand, it is possible that a Bohairic text displays influence from Arabic materials, as penetration of Arabic Bibles into Egypt was quite marked since around the 10th century and Arabic started displacing Coptic in the Egyptian Church around the 11th century.21 On the other hand, this particular reading is remarkable for affecting a cultural-ideological element, the idea of a widowed woman being mistress of her household, something that is somehow diluted in other versions by using the generic “woman” instead of “widow”. The Vulgate, which translates (or, rather, adapts) into mulieris matris familiae is significant evidence in that direction. In any case, despite the possibility of a parallel development, this coincidence should give us pause in order to review other possible connections between Coptic and Arabic texts in the Middle Ages. Finally, there is one striking coincidence between the Arabic of the Polyglots and the Septuagint text, in particular the Lucianic text-type. in 2 Kgs 2:10, the Arabic reading ‫ون‬ ْ َ‫ َوإ ِْن ل َ ْم ت‬is quite reminiscent of the L text καὶ εἀν δὲ μὴ ἴδης οὐ َ ‫ري فَ َ ْل يَ ُك‬ μὴ γένηται against the rest of the manuscript tradition. Even though the explication of a verb here could, again, be a case of parallel development, accumulation of textual coincidences should entice us to thoroughly study the possibility of cross-influences between different versions, whether they are in particular readings or in general stylistic tendencies, such as, for instance, the recurring usage ً ِ‫ قَائ‬as an introduction to direct speech in passages where the of the masdar form ‫ال‬ Greek text has λέγων against other lines of the textual tradition (e.g. 2 Kgs 1:3). Influence of the Septuagint text on the Arabic text via the Syro-Hexapla is a likely scenario, which should be further explored in the context of the Historical Books. All in all, this small list of samples indicates that a straightforward identification of an Arabic version with a given text-type should always be weighed against the possibility of cross-influence from other lines of textual tradition and translations of the Old Testament in the complex situation of the Middle Ages. Arabic 20 See R. Vollandt, “Some Historiographical Remarks on Medieval and Early-Modern Scholarship of Biblical Versions in Arabic: A Status Quo”, Intellectual History of the Islamicate World 1 (2013) 25–42. 21  See R.S. Richter, “Greek, Coptic and the Language of the Hijra: The Rise and Decline of the Coptic Language in Medieval Egypt”, in H.M. Cotton/R.G. Hoyland (ed.), From Hellenism to Islam: Cultural and Linguistic Change in the Roman Near East (Cambridge: Cambridge University, 2012) 401–46.

Creative Philology and Glosses: Secondary Versions of Kingdoms

233

can aid in understanding, for instance, certain Coptic readings (especially in the Bohairic dialect) and, in turn, can be better analyzed if all textual traditions, both in the spheres of MT and of LXX, are taken into account and examined together.

Conclusions The materials which I have presented are a selection of meaningful cases which should make us reflect when we study and evaluate versions of the Bible of a considerably late date, comparing both translations in different moments in Late Antiquity and important points of revision and dissemination with other textual traditions during the Middle Ages (Coptic and Arabic, though this scenario should be extended to Ge’ez22 and, in all likelihood, to Georgian and Armenian, languages which I have not covered in this paper).23 The introduction of variant readings at a late date is a complex phenomenon, and it has at times been overlooked by text critics, as traditionally our aim is to reconstruct or at least point in the direction of the earlier forms of a given reading or, hopefully, of a whole book. Nevertheless, it is important to take into account these late readings and, to say it mildly, their status of a “mixed bag” of elements, which should be studied with proper care, as the variegated interests and perspectives of Late Antique and medieval scribes hold important information, which I have tried to outline above: 1) variant readings from the ancient tradition embedded and adapted into their text with a purpose of textual richness and “perfection”; 22) explanations which are meaningful for tracing reading traditions of Vorlagen behind particular renderings or important information on the contexts of the translators and scribes themselves; 3) something which could be described as a “philological inclination”, which leads scribes to attempt stylistic and literary adaptation in order to produce a clearer text, at times from the point of view of exegesis, but at times merely from the angle of literary coherence or from the attempt to achieve linguistic accuracy (as in translations from the Hebrew or imitation of Semitic style seen above); 22  For the convoluted textual history (through multiple recensions) of the Ge’ez text, see A. Piquer Otero, “The Septuagint in Translation: Coptic, Arabic, Ethiopic”, in A. Salvesen/ T.M. Law (ed.), Oxford Handbook of the Septuagint (Oxford: Oxford University [forthcoming]); M.A. Knibb, Translating the Bible: The Ethiopic Version of the Old Testament (Oxford: Oxford University, 1999). 23  I have presented some cases connected to Georgian in Piquer, “Secondary Versions”; other Georgian and Armenian samples appear in A. Piquer Otero, “Who Names the Namers? The Interpretation of Necromantic Terms in Jewish Translations of the Bible”, in A. Piquer Otero/P. Torijano Morales (ed.), Textual Criticism and Dead Sea Scrolls Studies in Honour of Julio Trebolle Barrera. Florilegium Complutense (SJSJ 157; Leiden: Brill, 2012) 241–76.

234

Andrés Piquer Otero

4) active usage of other textual traditions in a scenario of exchange and fluidity of texts between different communities (Christian Arabic, Judaeo-Arabic, Syriac, Coptic24), which, even if it does not have a radical impact on the history of transmission (though such was the case in the Ethiopic version and the role of Syro-Arabic recensions), may be important for understanding the versions of the Eastern Mediterranean in the medieval period and, as a mirror or shadow, for further assessing the cultural dynamics of the age, against certain traditional ideas of textual immobility and exclusiveness.

As I commented in the introduction, we all want our words to endure, be they poetry and courtly love, or an edition or other scholarly work. We are part of the history of the text and should understand our forerunners from one millennium ago or more as part of this same spirit. Words and works endure according to fate or change, but, as researchers of text and history, what we always should keep in mind and heart is that the attempt to endure in the text and the mechanics thereof are, if we do not forget to read them and practice them critically, the things which, age by age, will remain and persevere the most, beyond the words themselves, in jet and supple glass.

Bibliography Barthélemy, D., Les Devanciers d’Aquila (Leiden: Brill, 1963). Crum, W.E., A Coptic Dictionary (Oxford: Clarendon, 1939). Graf, G., Geschichte der christlichen arabischen Literatur. Erster Band. Die Übersetzungen (Vatican City: Biblioteca Apostolica Vaticana, 1944). Griffith, S.H., The Bible in Arabic: The Scriptures of the ‘People of the Book’ in the Language of Islam (Princeton: Princeton University, 2013). Knibb, M.A., Translating the Bible: The Ethiopic Version of the Old Testament (Oxford: Oxford University, 1999). Lichtheim, M., Ancient Egyptian Literature. Vol. 1. The Old and Middle Kingdoms (Berkeley: UC Press, 1973). Lichtheim, M., Ancient Egyptian Literature. Vol. 3. The Late Period (Berkeley: UC Press, 1973). Meyer M.W. & Smith, R., Ancient Christian Magic: Coptic Texts of Ritual Power (Princeton: Princeton University, 1999). Nagel, P., “Old Testament, Coptic Translations of ”, in Aziz Z. Atiya (ed.), The Coptic Encyclopedia Vol. 6 (New York: Macmillan) cols. 1836a–1940a. Perttilä E., “How to Read the Greek Text behind the Sahidic Coptic”, in A. Voitila/J. Jokiranta (ed.), Essays on Septuagint, Hebrew Bible and Dead Sea Scrolls in Honour of Raija Sollamo (JSJSup 126; Leiden: Brill, 2008) 367–78. 24  Although with a marked emphasis on the Pentateuch, this approach is experiencing remarkable developments in present-day biblical scholarship. See Vollandt, Arabic Versions.

Creative Philology and Glosses: Secondary Versions of Kingdoms

235

Perttilä, E., Sahidic 1 Samuel: A Daughter Version of the Septuagint 1 Reigns (DSI 8; Göttingen: Vanderhoeck und Ruprecht, 2017). Piquer Otero, A., “Coptic, Arabic, Ethiopic”, in A. Salvesen (ed.), The Oxford Handbook of the Septuagint (New York: Oxford University [forthcoming]). Piquer Otero, A., “The Secondary Versions of Kings. Variants and Renderings Between Vorlagen and Ideology”, in S. Kreuzer/M. Meiser/M. Sigismund (ed.), Die Septuaginta: Orte und Intentionen. 5. Internationalen Septuaginta-Tagung Wuppertal 2014 (Tübingen: Mohr Siebeck, 2018), 244–55. Piquer Otero, A., “The Septuagint in Translation: Coptic, Arabic, Ethiopic”, in A. Salvesen/T.M. Law (ed.), Oxford Handbook of the Septuagint (Oxford: Oxford University, [forthcoming]). Piquer Otero, A., “Who Names the Namers? The Interpretation of Necromantic Terms in Jewish Translations of the Bible”, in A. Piquer Otero/P. Torijano Morales (ed.), Textual Criticism and Dead Sea Scrolls Studies in Honour of Julio Trebolle Barrera. Florilegium Complutense (SJSJ 157; Leiden: Brill, 2012) 241–76. Piquer Otero, A. & Torijano Morales, P., “Between the Search in the Word and the Asking to God. Two Mantic Verbs in the Textual History of Samuel-Kings”, in K. De Troyer/ T.M. Law/M. Liljeström (ed.), In the Footsteps of Sherlock Holmes: Studies in the Biblical Text in Honour of Anneli Aejmelaeus (Leuven: Peeters, 2014) 299–330. Piquer Otero, A., Torijano Morales, P. & Trebolle Barrera, J., “Septuagint Versions, Greek Recensions, and Hebrew Editions: The Text-Critical Evaluation of the Old Latin, Armenian, and Georgian Versions of III-IV Regnorum”, in H. Ausloos/J. Cook/F. García Martínez/ B. Lemmelijn/M. Vervenne (ed.), Translating a Translation: The LXX and its Modern Translations in the Context of Early Judaism (BETL 213; Leuven: Peeters, 2008) 251–81. Richter, R.S., “Greek, Coptic and the Language of the Hijra: The Rise and Decline of the Coptic Language in Medieval Egypt”, in H.M. Cotton/R.G. Hoyland (ed.), From Hellenism to Islam: Cultural and Linguistic Change in the Roman Near East. (Cambridge: Cambridge University, 2012) 401–46. Ritner, R., “Necromancy in Ancient Egypt”, in L. Ciraolo/J. Seidel (ed.), Magic and Divination in the Ancient World (Ancient Magic and Divination 2; Leiden: Brill, 2002) 89–96. Roediger, E., De origine et indole arabicae Librorum V.T. historicorum interpretationis (Halle, 1829). Talshir, Z., “The Representation of the Divine Epithet ‫ צבאות‬in the Septuagint and the Accepted Division of the Books of Kingdoms”, JQR 1–2 (1987) 57–75. Trebolle Barrera, J., Centena in libros Samuelis et Regum. Variantes textuales y composición literaria en los libros de Samuel y Reyes (Madrid: CSIC, 1989). Vollandt, R., Arabic Versions of the Pentateuch: A Comparative Study of Jewish, Christian, and Muslim Sources (Leiden: Brill, 2015). Vollandt, R., “Some Historiographical Remarks on Medieval and Early-Modern Scholarship of Biblical Versions in Arabic: A Status Quo”, Intellectual History of the Islamicate World 1 (2013) 25–42. Vollandt, R., ‫הבראיסטים קופטים בימי הביניים? על מסירת התפסיר של רס"ג‬, Tarbiz 83 (2015) 71–86.

S. Peter Cowe Scribe, Translator, Redactor: Writing and Rewriting Scripture in the Armenian Versions of Esther, Judith, and Tobit Introduction Modernism witnessed the unification of the economy, military, educational and judicial systems, and culture within the stereotypical nation-state.1 As a result, equality of opportunity and uniform standards of quality control are indissociable elements of our discourse in an era where globalizing trends extend such perspectives, undermining the sovereignty of the national government through the web of multinational conglomerates, international organizations (United Nations, World Bank, International Court of Justice, etc.), and ecological issues that impel us to grapple with the concept of world citizenship.2 Clearly this is a far cry from the tenor of life in the Ancient Near East or the eastern Mediterranean in the Late Antique period, that concerns us in this panel. On the contrary, theirs was a life experience characterized by variation and diversity in every dimension, dwelling in states with dynastic rule, devolved administration, core identities regionally based, with a hierarchy of dialects and languages, and an artisan model of fabrication foregrounding individual skill and imagination rather than the consistency and uniformity of automated machine production.3 Within such a context it is understandable that the literary sphere was also categorized primarily by diversity and the lack of uniformity in its paradigms of literary creation and transmission, both sacred and secular.4 As a result, it is vital to hold these norms in mind as we investigate texts emanating from this

1  For an overview of these developments, see E. Gellner, Nations and Nationalism (Ithaca, NY: Cornell University Press, 1983), 53–62, and A.S. Smith, The Ethnic Origins of Nations (Oxford: Oxford University Press, 1986), 129–52. 2  For the impact of international organizations and multinational companies, see R.H. Robbins, Global Problems and the Culture of Capitalism (Boston: Pearson/Allyn & Bacon, 1999), 131–44. 3  For a case study, see G.E. Areshian, “Sasanian Imperialism and the Shaping of Armenian Identity (Interdisciplinary Verification and Ambivalence of empire-nation Relationship)”, in id. (ed.), Empires and Diversity: on the Crossroads of Archaeology, Anthropology, and History (Los Angeles: The Cotsen Institute of Archaeology Press, 2013) 185–203. 4  Diversity of genre; variation in transmission of sacred and secular texts. Lack of close supervision of the process. Large majority of the population was illiterate as ubiquitous until the modern period.

238

S. Peter Cowe

matrix as a salient corrective to modern presuppositions.5 On the other hand, it is equally important to gain a firmer sense of the parameters of variation tolerated in such societies and hence to appreciate the criteria for determining the mainstream from the marginal, the canonical from the uncanonical, the degree of authorial and textual integrity maintained and acceptable scribal intervention, what should be recopied from what should be consigned to physically disintegrate and disappear from the tradition.6 At the same time, the scholars who are largely involved in studying this corpus of ancient texts and their transmission tend to be institutionally conditioned to valorize one perspective to the comparative detriment of the other, being inured to the methodology applied in producing a critical edition. As such, they focus single-mindedly on reconstituting as far as possible what is regarded as the original text of a given work, and hence advance from the plethora of readings preserved in different manuscripts in a diversity of languages back through the successive layers of tradition to the form regarded as primary. This procedure frequently leads to consciously or unconsciously viewing later textual strata as encrustations that must be scraped away by various tools in order to reveal the early structures underlying these. Consequently, the transmission process tends to be considered rather one-sidedly as one of distortion, aberration, and the perversion of pristine purities. Hence the importance of this conference which pursues exactly the opposite path of seeking to comprehend the phenomenon, tracing patterns of transmission, analyzing and classifying the relevant data, exploring the forms of agency active in the endeavor, the milieus in which the tradents functioned, and the outlook of the communities they served. Thus within the time and place under discussion change emerges not as the indubitable disruption of a final literary and textual form but potentially the vehicle for maintaining the resonance of a work within a continually evolving community of readers and hearers with different needs to address.7 Within the context of religious literature the organizers’ focus on the books Esther, Judith, and Tobit provides rich material for discussion from this viewpoint. The Hebrew Torah was recognized as sacred scripture by the 4th century  b.c.e.

5  For the growing significance of uniformity of print culture, see E.L. Eisenstein, The Printing Revolution (2nd edn; Cambridge: Cambridge University Press, 2005), 50–63. 6  A good example of this phenomenon is the disappearance of manuscripts of the Hebrew and Aramaic texts of the ‘Apocrypha’ after failing to achieve canonical status. Falling out of mainstream Judaism, they were preserved in the Christian ‘periphery’ not in the original but in a series of translations. 7  On the diffusion of literacy, see W.V. Harris, Ancient Literacy (Cambridge: Harvard University Press, 1989), and R.L. Fox, “Literacy and Power in Early Christianity”, in A.K. Bowman/ G. Woolf (ed.), Literacy and Power in the Ancient World (Cambridge: Cambridge University Press, 1996) 126–48.

Writing and Rewriting Scripture in the Armenian Versions

239

and the prophetic corpus by the following century.8 Consequently, the protracted evolution those books had undergone, often from oral matrices through various literate stages, was largely behind them, the intermediate strata generally no longer accessible in documentary form and thus only partly amenable to reconstitution by diverse scholarly disciplines.9 Thereafter, the degree of transformation the proto-Massoretic text experienced in most books was comparatively limited. This contrasts with the situation of the third scriptural portion, the Writings, the group to which the three books under investigation belong. Many of the texts falling into that category were still being composed in the Hellenistic period and did not achieve authoritative status until the 2nd century c.e. Hence they feature a later transmission history, from which a number of documentary variants survive, providing us with more objective evidence, with which to attempt to reconstruct the trajectory of their literary and textual development.

The Writings within the Multifaceted Jewish Interface with Hellenism Granted the sociopolitical and cultural changes concomitant with Alexander’s empire and its successor states, the scriptural evolution referred to above was variously impacted by its encounter with Hellenism. Knowledge of Greek became widespread in the large diasporan Jewish communities of Antioch and Alexandria, as indicated by the project of translating the Torah into that idiom in the 3rd century b.c.e. This was followed by the other scriptural texts thereafter. The translation technique of books rendered later in the cycle tends to reflect the morphological and syntactic structures of the proto-Massoretic text in Greek guise. In fact, most of the books in the Writings corpus fall into this category (1–2 Chronicles, Ezra-Nehemiah [1–2 Esdras], Ruth, Lamentations, Song of Songs, and Ecclesiastes, the last of which is so literal that it has been argued to be the product of Aquila). In parallel, a series of revisions was undertaken to approximate the older Greek versions to the form of Hebrew text in current circulation (e.g. Daniel). The only exceptions to this tendency are the two wisdom books Job, rendered in a fine literary register not dependent on Hebrew idiom, and Proverbs, which, though oscillating widely in its translation technique, embraces

8  E. Ulrich, The Dead Sea Scrolls and the Developmental Composition of the Bible (SVT 169; Leiden: Brill, 2015), 201–2. 9  One of the most important exceptions to this trend is Jeremiah, of which the Greek version represents an earlier form of the Hebrew than the Massoretic Text.

240

S. Peter Cowe

Hellenism to such a degree as to incorporate into its text proverbs of extraneous Greek origin.10 To further contextualize Judith and Tobit we have to consider that this same timeframe witnessed the composition of a library of additional works that met with a divergent reception as Judaic communal identity diversified in the early centuries c.e. Though mainly of Semitic provenance, for various reasons those books did not establish themselves within the Rabbinic Jewish tradition, yet entered into emergent Christian usage and ultimately gained acceptance into the latter canon. Like the majority of books we have discussed so far, most of this second category experienced a rather literal isomorphic rendering into Greek (e.g. Ben Sira, 1 and 3 Maccabees, Judith). In contrast, Baruch is arguably of hybrid provenance, the first section (1:1–3:8) probably representing the translation of a Hebrew text, while the remainder appears an original composition in Greek. The final grouping comprises two works written purely in Greek, 2 Maccabees and Wisdom, the latter probably one of the last books in the corpus to be composed.11 The impact of this movement is that the structure of the Septuagint, as the motley collection of those Greek texts has come to be known, is typical of the unsystematic diversity associated with literary composition and textual circulation at that time and place. As alluded to above, in certain cases the translation process encompassed several phases, following different trajectories in translation technique. Thus, revision of the more dynamic rendering of the Old Greek of Daniel engendered the more literal Thedotionic version of the 1st or 2nd century  c.e. Meanwhile, the G1 and GIII versions of Tobit reformulate the Semitic hue of the GII narrative in a style more appealing to a sophisticated readership fluent in Greek.12 A further dimension that must be treated is the element of editing or rewriting that reveals itself in certain books. Both Greek versions of Daniel contain a set of traditions concerning the sage that, though mainly betraying a Semitic origin, were not accepted as scripture in that form (Susanna, Song of Azarias, Song of the Three, Bel and Dragon). Similarly, Esther, the third of the books of particular concern in this paper, is not only extant in two Greek versions (Old Greek and Alpha Text) that share certain affinities, but features six additional sections lacking in the Hebrew, the integrating of which involved significant redaction of both Greek forms. The complexity of levels and discontinuities among the three versions of Esther (MT, OG, AT) is such that conjectural reconstructions have all but exhausted the mathematical possibilities for determining the sequence of 10  J. Cook, The Septuagint of Proverbs-Jewish and/or Hellenistic Proverbs? Concerning the Hellenistic Colouring of LXX Proverbs (VTSup 69; Leiden: Brill, 1997), 402. 11  Scholarly opinions vary, but Wisdom probably dates to the 1st century b.c.e. 12  On the three versions see R. Hanhart, Tobit (SVTG VIII, 5; Göttingen: Vandenhoeck & Ruprecht, 1983), 29–36.

Writing and Rewriting Scripture in the Armenian Versions

241

strata.13 It is interesting that the existence of yet another Greek version closer to the Massoretic Text has recently been postulated as the parent text for the oldest Church Slavonic version of the book.14 As the books of Esther, Judith, and Tobit, have been classified by modern scholarship as belonging to the genre of historical novel, fictionally developing the narrative and characterization against a historical backdrop,15 one wonders whether this literary typology exposed the works to greater intervention and reformulation by later tradents.16

Background to the Armenian Version To sum up the argument so far, we observe that around the turn of the eras, in addition to copying and revising earlier Greek translations of the Writings, Jewish literati were also active in rewriting and composing new works in Greek which were also appropriated by Christian tradents and transmitted by them in parallel at least into the 2nd century c.e. alongside the writings associated with their new revelation, which would become the New Testament.17 To contextualize the Armenian version and sketch its intellectual genealogy it is important to extend the narrative of scriptural transmission over the next two centuries of momentous significance for the Christian community as it gradually spread across the Mediterranean world. Extending beyond the sphere where Greek was lingua franca necessitated the creation of a succession of Latin translations in North Africa and Europe in the 2nd century and Coptic and Syriac in the following in Upper Egypt and Mesopotamia respectively.18 In view of the lack of organization, coordination, and an overarching structure to administer and regulate what was still an illicit religious group, local autonomy prevailed, with the result that when in the later 5th–6th centuries distinct Coptic and Syrophone ecclesiastical jurisdictions were formed, they adopted as authoritative those books in circulation in their

13  These will be explored in a little more detail in the final section of the paper. 14  A terminus ad quem for the translation is provided by the two earliest manuscripts dated c. 1400. For details, see H.G. Lunt/ M. Taube, The Slavonic Book of Esther: Text, Lexicon, Linguistic Analysis, Problems of Translation (Harvard Series in Ukrainian Studies; Cambridge, MA: Harvard Ukrainian Research Institute, 1998). 15  See A. Berlin, “The Book of Esther and Ancient Storytelling”, JBL 120/1 (2001) 3–14. 16  This raises the issue of the transmission of the Writings at this period in contrast to Torah and the Prophets. 17  For a discussion of this transition, see M. Hengel, The Septuagint as Christian Scripture: Its Prehistory and the Problem of its Canon (Edinburgh: T & T Clark, 2002), 57–128. 18  The Peshitta Pentateuch was begun on Jewish initiative, and the rest of the Old Testament was largely rendered from Hebrew, so that only the books of the Apocrypha derive from Greek.

242

S. Peter Cowe

communions at that time, thus manifesting inner-Christian diversity not only in language and idiom but also canon.19 Although now distinguished in liturgy and ethos, the 2nd–3rd centuries witnessed a continuing interaction between Judaism and Christianity, which led to a widening awareness of the incongruities between the current Hebrew and Greek texts of Scripture. Concurrently, the Church began its own interchange with Hellenic paideia and largely embraced it,20 resulting in Origen’s monumental enterprise of comparing the contemporary Massoretic Text with the four major Greek versions, employing Aristarchian critical signs to characterize Greek pluses and minuses over against the Hebrew and create a composite recensional text of the Septuagint. Origen’s Hexapla, in turn, became the foundation for a further series of textual initiatives in the other eastern metropolis of Antioch. Granted the city’s commercial importance at the hub of a number of trade routes, its local Greek text was transposed to North Africa by sea where it impacted the Old Latin, while land routes carried it along the coast to Cilicia and inland to greater Syria and Armenia. A generation after Origen, Lucian of Antioch conceived the project of redacting the local Septuagint text. Granted the impact of the Second Sophistic movement on education and culture in the Roman empire in the 2nd and 3rd centuries favored Atticism, a return to the classical register of Athens in the 5th century b.c.e., rather than the koinē of the Hellenistic period in which the Septuagint was couched,21 Lucian preferred to adopt readings from Symmachus to revise the Old Greek text in contrast to Aquila and Theodotion because of his elegant Greek style and idiomatic expression.22 These considerations highlight Christianity’s penetration not only horizontally, but also vertically to touch all social strata of the empire and, naturally, in so doing to undergo the respective influence of various political and cultural factors. In addition to linguistic and stylistic concerns, Lucian’s recension is also typified by a solicitude for reviewing parallel passages in different books (e.g. SamuelKings and Chronicles) to piece together the disparate elements to gain an 19  See W.-P. Funk, “The Translation of the Bible into Coptic”, in J.C. Paget/J. Schaper (ed.), New Cambridge History of the Bible 600–1450, vol. 1 (Cambridge: Cambridge University Press, 2013) 536–46, and Williams, “The Syriac Versions of the Bible”, in J.C. Paget/J. Schaper (ed.), New Cambridge History of the Bible 600–1450, vol. 1 (Cambridge: Cambridge University Press, 2013) 527–35. 20  From this perspective Philo in the 1st century foreshadowed figures like Justin Martyr in the following in presenting his religious ideas in the thought structure of pagan philosophy. 21  On this movement, see T. Whitmarsh, The Second Sophistic (Oxford: Oxford University Press, 2005), and S. Swain, Hellenism and Empire: Language, Classicism and Power in the Greek World, AD 50–250 (Oxford: Oxford University Press, 1996). 22  See B.M. Metzger, “Theories of the Translation Process”, Bibliotheca Sacra 150 (1993) 140–50, on p. 142.

Writing and Rewriting Scripture in the Armenian Versions

243

understanding of the fuller dimensions of the biblical narrative. This was coupled with a desire to clarify metaphorical expressions in the text by careful comparison of the Greek and Hebrew versions. In view of Antioch’s fame for rhetoric, especially in the era of Libanius, Lucian’s successors in the second half of the 4th century who are associated with the ‘school’ of Antioch primed their exegetical acuity with a thorough training in oratorical skills.23 The result, as elaborated in the commentaries of Chrysostom, Theodore, and Theodoret among others, was a more rigorous investigation of what in effect constituted text by scrutinizing different Greek translations and utilizing the Syriac to gain insights into the Hebrew before synthesizing these components by way of paraphrase at the beginning of their comments on each verse. Intertextuality and explication of figures of speech formed an integral part of this process, which was then informed by theological perspectives (e.g. Christological typology that underscored the unity between the Old and New Testaments) and hermeneutical imperatives such as the refutation of pagans and heretics. It is out of this matrix that the Armenian Version emerged at the beginning of the 5th century, largely based on a Greek text of early Antiochian (pre-Lucianic) and Lucianic complexion, its approach guided by the principles of Antiochene interpretation.24 Thus, although we usually conceive of commentary as unilaterally dependent on text, in this case we observe the cyclical pattern by which its insights are ploughed back and reinscribed in text. As such, we are afforded a rare insight into the application of a known and developed form of scriptural exegesis embedded within the texture of an actual biblical translation. As we shall see, the translators could therefore be quite interventionist in their approach, not only reformulating the expression, but continuing the process we observed earlier in Hebrew and Greek of rewriting passages or whole books on occasion.25 Bearing 23  Granted the loose character of the ties binding the main exponents of the Antiochene methodology, see the application of network theory to better account for their activities in A.M. Schor, Theodoret’s People: Social Networks and Religious Conflict in Late Roman Syria (Berkeley: University of California Press, 2011). For the argument that a similar informal setting characterized its counterpart, the catechetical ‘school’ of Alexandria, see R. van den Broeck, “The Christian ‘School’ of Alexandria in the Second and Third Centuries” in J.W. Drijvers/ A.A. MacDonald (ed.), Centres of Learning: Learning and Location in Pre-Modern Europe and the Near East (Leiden: Brill, 1995) 39–47. 24  S.P. Cowe, “The Bible in Armenian”, in A.E. Matter/R. Marsden (ed.), New Cambridge History of the Bible 600–1450, vol. 2 (Cambridge: Cambridge University Press, 2012) 143–61, on pp. 147–54. 25  In this connection, it is interesting that the possibility has been mooted of a final Christian stage to the redaction of the Alpha Text of Esther. See K.H. Jobes, The Alpha-Text of Esther: Its Character and Relationship to the Massoretic Text (JBL Dissertation Series 153; Atlanta, GA: Scholars Press, 1996), 226–7. An example of the latter procedure is the Armenian Version of Lamentations, which involves a large amount of rewriting, particularly in chapter 2, on which see S.P. Cowe, “Rhetoric, Theology, and Antiochene Exegesis in the Armenian Version

244

S. Peter Cowe

in mind the 4th century advances in ecclesiastical organization and hierarchical structure,26 it is important to note that the translation was an official enterprise under the sanction of the king and chief-bishop of Greater Armenia and featuring a well-trained group of scholars under a single leader that was responsible for the undertaking.27 In this sense it bears some affinity with Jerome’s contemporaneous activities, revising the Old Latin of the Gospels at the request of Pope Damasus I and producing a new translation of the Hebrew scriptures supported by the Hexapla and Greek exegetical materials. The early Armenian version (Arm1) underwent a revision (Arm2) in the aftermath of the Council of Ephesus (431) conducted on the basis of manuscripts with a different textual complexion and according to an isomorphic translation technique.28 However, the variant readings in the Zohrabean edition of 1805 from which the Armenian evidence has been collated against the Göttingen editions of the Septuagint suggest that in our three books the revision was rather perfunctory, leaving most of the early version relatively intact.29

Armenian Indigenization of Data From the viewpoint of the Armenian version, it is noteworthy that the subject matter of these books directly relates to its geopolitical setting in that the plot of Esther unfolds in Persia, while that of Tobit is oscillates between the poles of Mesopotamia and Media, still within a broad Persian ambience, and, though the action of Judith takes place in Israel, it is set against the broader movement of Hellenism being disseminated throughout the wider Seleucid realm.30 Bearing in mind Armenia’s traditional status in this timeframe as a borderland between the Roman and Persian empires, much of the overall context of the narratives was intimately familiar to the translators, leading them to exploit this in various

of Lamentations”, in P.F. Fumagalli (ed.), Dies Academicus del “Classe di Studi sul Vicino Oriente dell’Accademia Ambrosiana” (Analecta Orientalia Ambrosiana IV; Roma: Bulzoni Editore, 2015) 143–65. 26  See C. Rapp, Holy Bishops in Late Antiquity: The Nature of Christian Leadership in an Age of Transition (Berkeley: University of California Press, 2005). 27  For an English translation of the appropriate passages from the contemporary source Koriwn’s “Life of Maštoc‘”, see Ē. Pivazyan, Vark‘Maštoc‘i [The Life of Maštoc‘] (Erevan: Erevan State University Publication, 1980), 277–83. 28  Cowe, “The Bible in Armenian”, 154–8. 29  On the variation in the degree of revision of the Arm1 stratum, see Cowe “The Bible in Armenian”, 155–60. 30  On Seleucid ideology, see P.J. Kosmin, The Land of the Elephant Kings: Space, Territory, and Ideology in the Seleucid Empire (Cambridge, MA: Harvard University Press, 2014).

Writing and Rewriting Scripture in the Armenian Versions

245

ways.31 Several of the Armenian noble houses were of Persian descent as attested by the onomasticon.32 Thus Ahasuerus, the name of the monarch in the Massoretic Text of Esther, which the Old Greek renders as Artaxerxes, as for example at A(11):1, is found in its Armenian reflex of Artašēs rather than a transliteration of the Greek: ἔτους δευτέρου βασιλεύοντος Ἀρταξέρξου յամին երկրորդի ի թագաւորութեանն Արտաշէսի (in the second year in the reign of Artašēs)

The form is quite widespread in Armenian,33 being associated with the Artaxiad dynasty that ruled Armenia from 118 b.c.e. to 6 c.e. Similarly, focus on their target readership prompted the translators to communicate the date, which appears in the same verse, according to the Hebrew calendar by allusion to its Armenian counterpart.34 Esther A(11):1 τῇ μιᾷ τοῦ Νισα յառաջնում աւուր ամսոյն ադարայ, որ է հայերէն արեգ (on the first day of the month of Adar, which is Areg in Armenian)

Significantly the Armenian version reads the twelfth Hebrew month Adar in place of Nisan, the first, and renders it by the eighth Armenian month Areg, associated with the Sun, which falls approximately in March. The same concern for reader comprehension impelled the intervention at Jdt 6:11(8) clarifying the general location of the heroine’s village:

31  R.H. Hewsen, Armenia: A Historical Atlas (Chicago: University of Chicago Press, 2001), 29–95. 32  N.G. Garsoïan, “The Emergence of Armenia”, 46–50 and “The Aršakuni Dynasty”, 64–81 in R.G. Hovannisian (ed.), The Armenian People from Ancient to Modern Times, vol. 1 (New York: St. Martin’s Press, 1997). To contextualize this within the broader scope of Iranian loanwords in Armenian, see R. Schmitt/H.W. Bailey, “Armenia and Iran. iv. Iranian Influences in Armenian”, in Encyclopaedia Iranica (www.iranicaonline.org/articles/armenia-iv). 33  H. Ačar˙ean, Anjnanunneri bar˙aran [Prosopographical Dictionary], 5 vols. (Beirut: Sevan Publishing, 1972), vol. 1, 305–309. On the significance of the dynasty, see H. Manandyan, Tigrane II & Rome: nouveaux eclairissements à la lumière des sources originales (Lisbon: Imprensa Nacional, 1963). 34  For the background to the term, see H. Martirosyan, Etymological Dictionary of the Armenian Inherited Lexicon (Leiden: Brill, 2010), 135. Note that the Alpha Text glosses the Hebrew months with their Macedonian equivalents.

246

S. Peter Cowe

ὑποκάτω Βαιτυλουά ի ներքոյ Բետիուղայ որ է Շամրին (below Betiułay, which is Šamrin, i.e. Samaria)35

Unintended Scribal Errors in the Zohrapean Base Manuscript The origins of our modern perception of the Bible in its sociocultural and religious role are predicated upon the example of the Luther Bible of the 16th century. Its mass production via the press, wide dissemination through booksellers, relative uniformity of text, and linguistic and financial accessibility commended it to the public, on whose increased levels of literacy its success was founded, levelling socioeconomic and educational distinctions to such a degree that a doctrine of Sola Scriptura and the autonomy of the reader could be countenanced.36 Conditions in the ancient and medieval world, however, were very different. Bearing in mind what was said earlier about the ubiquity of variation and diversity, when considering the domain of textual transmission more narrowly, we can define two contrasting phenomena, the unintended errors of copyists whose goal is the maximal representation of the work they are copying and the intended interventions of redactors who seek to modify the form and content of a writing and thereby transform its impact on the reading public.37 We will review several examples of both categories. While the redactor usually functions at the macro level, leaving his or her footprint throughout the work, the former operates at the micro level frequently creating minor omissions occasioned by lack of coordination in the movement of the eye from the exemplar to the copy or encountering unfamiliar names and foreign toponyms, etc.38 Manuscript variants cited in the apparatus of the Zohrapean edition facilitate the correction of such minor scribal errors.39 Thus before his death Tobit asks his son to gather his sons for him to impart to them his final counsel, however the 35  The Armenian form of toponym is a transliteration of the Syriac Šmryn. 36  For the socioeconomic background to these developments, see B. Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism (London: Verso, 1991), 38– 46 and E.L. Eisenstein, The Printing Revolution in Early Modern Europe (2nd edn; Cambridge: Cambridge University Press, 2005), 145–84. 37  For perspectives on copying, see K. Haines-Eitzen, “The Social History of Early Christian Scribes”, in B.D. Ehrman/ M.W. Holmes, (ed.), The Text of the New Testament in Contemporary Research (2nd edn; NTTSD 42; Leiden: Brill, 2013) 479–95. 38  See B.M. Metzger, The Text of the New Testament (2nd edn; Oxford: Clarendon Press, 1968), 190. 39  For appropriate caveats on use of Zohrapean’s apparatus, see C.E. Cox, “Introduction”, in Y. Zohrapean, Astuatsashunch‘ matean hin ew nor ktakarants‘ (a facsimile reproduction of the 1805 Venetian edition with an introduction by Claude Cox) (Delmar, NY: Caravan Books, 1984) v–xxvi, on pp. xii–xix.

Writing and Rewriting Scripture in the Armenian Versions

247

transmission history behind the base text, presumably under the influence of the reference to his one son, retains the singular with regard to Tobit’s grandchildren. Tob 14:3 (5) τούς . . . υἱούς

զորդի Ztxt (son) զորդիս Zap (sons)

The original reading can be reinstated from other witnesses. The opposite situation is reflected at Jdt 9:13(18) where the heroine’s prayer turns to God’s indwelling of Mt. Sion in Jerusalem. κατά . . . κορυφῆς Σιών ի վերայ . . . բարձունս Սիոնի Ztxt (on the heights of Sion) բարձուն Zap (height)

There a dittography of the sibilant in the Armenian text creates an ungrammatical plural form, which can be corrected by the variant in the apparatus. On other occasions the putative reading may be restored through conjectural emendation. As the editor’s access to Old Testament witnesses was restricted to nine manuscripts, it is hoped that wider collations will corroborate these conjectural forms.40 Jdt 9:9(12) ὁ διενήθην κράτος

զօրութիւն` զոր խորհեցայս առնել (the power that I have considered activating)

Armenian usage demands the prefacing of the definite direct object with the proclitic զ-,41 but the copyist has omitted this through haplography at Jdt 9:12. Jdt 9:9(12) δὸς ἐν χειρίχειρί

տուր ձեռն (give in the hand)

In the same verse where the protagonist is appealing for divine strength to enact her plan the preposition consisting of a single character has been inadvertently omitted. Jdt 9:13(18) κατὰ . . . οἴκου ἡγιασμένου σου ի վերայ . . . տան քոյ (your holy house)

40  The fact that of the nine manuscripts at Zohrapean’s disposal only one was not typologically a full Bible, suggests that most of his witnesses were congers also textually, thus reducing the edition’s potential to represent the diversity of the version’s transmission history. 41 Meillet, Altarmenisches Elementarbuch (Heidelberg: Carl Winters Universitäts-Buchhandlung, 1913), 73, 79–80.

248

S. Peter Cowe

Likewise, in the next verse where Judith invokes the Temple the epithet ‘holy’ has fallen out of the text by parablepsis through homoeoteleuton.42 Granted that in the ideal world all witnesses would reproduce the autograph in all respects, shared deviations from this norm imply a genealogical relation between the codices under comparison and hence are constitutive of textual families. Moreover, these are transferred from one level of tradition to another, for example, by means of translation, and hence systematic affiliations of this sort help determine the version’s parent texts.

Shared Secondary Readings in constituting Textual Families One of the most fecund generators of variants is foreign names or terms, of which scripture is replete, as at Tob 14:10(12): Ἁμάν] αδαμ B d 46-55-318-319 393 Syo Arm

Ադամ (Adam)

There the Armenian adduces the variant Adam in concert with Codex Vaticanus, the d group, and a series of codices mixti, with which the version evinces a particular affinity.43 In view of the significant phonetic developments Greek was undergoing at this period, it is natural that that process should also affect the textual transmission. The next two examples target the phenomenon of itacism by which the phonetic values of a series of monophthongs and diphthongs coalesced with the original vowel ‘i’.44 In the first instance from Tob 14:7 (9) the original distinction between the vowels η and υ was lost by the early Byzantine period, resulting in the confusion between the first and second person plural pronouns. ἡμῶν] ὑμων 46c Arm

ձերոց (your)

The latter form may have entered Armenian from a textual stream such as that represented by the corrector of miniscule 46 It is, however, impossible to discern whether the reading represents the objective lexeme in the translator’s exemplar or simply his mental image. 42  For this regular scribal error, see Metzger, The Text of the New Testament, 189 and J.R. Royse, “Tendencies in the Transmission of the Text of the New Testament”, in B.D. Ehrman/ M.W. Holmes (ed.), The Text of the New Testament in Contemporary Research: Essays on the Status Quaestionis Second Edition (NTTSD 42; Leiden: Brill, 2013) 461–78, on p. 462. 43  These affinities are easily observed in Hanhart’s apparatus. Note especially Armenian affiliation with Greek MSS 46, 311, and 126. 44 Pring, The Oxford Dictionary of Modern Greek, vii, xiii, and Metzger, The Text of the New Testament, 190–2.

Writing and Rewriting Scripture in the Armenian Versions

249

A similar confusion between the vowels η and ι at Jdt 16:14 (17) has led to the transformation of the reading κτίσις (creation) in the critical text into the variant κτῆσις (possession): πᾶσα ἡ κτίσις σου (κτησις MS 381) b

ամենայն ստացուածս քո (all your possessions)

That lexeme is patently secondary in Judith’s final paean devoted to God’s creation and sustenance of the natural and human realm. The variant is witnessed by miniscule 381 of the b subgroup in concert with the Armenian. The third example illustrates a parallel phonetic development, the gradual identification of the value of the diphthong αι with that of the monophthong ε, which occurred by the second century c.e.45 Jdt 2:27 (17) πεδία] παιδια B A 314 98c 670 Aeth Arm զմանկունս (children)

The presence of the term νεανίσκους (young men) in the next clause as a further object of Holophernes’ destruction probably played a role in generating the variant.

Translation and the Exigencies of Armenian Grammar Clearly the translator’s task is much more complex than that of the scribe in that the process of transmitting a literary work from one language to another raises questions of compatibility between the media on at least three levels: the exigencies of grammar relating to the basic building blocks of expression, issues of idiom and the norms of usage, and the accepted canons of style. Since by definition a one-to-one match between languages is excluded, translation is obviously the art of the possible, often requiring the sacrifice of one facet of the text in preference for another more valorized by the exponent’s explicit or implicit philosophy of translation.46 In addition to the standard grammatical incongruencies between Greek and Armenian that have often been commented on (participles, compound verbs, prepositions) I would like to discuss some others which are reflected in our three books.47 45 Pring, The Oxford Dictionary of Modern Greek, vii. 46  See J.D. Waard/E.A. Nida, From One Language to Another: Functional Equivalence in Bible Translating (Nashville, TN: Nelson, 1986) 36–44. 47  For a more detailed discussion of the limitations of the earlier purely grammatically orient­ ed approach, see S.P. Cowe, “The Armenian Version of the New Testament”, in B.D. Ehrman/ M.W. Holmes (ed.), The Text of the New Testament in Contemporary Research (2nd edn; NTTSD 42; Leiden: Brill, 2013) 253–92, on pp. 277–9.

250

S. Peter Cowe

Although probably a feature of Proto-Armenian, grammatical gender was lost by the classical form represented in writing.48 Thus at Esther 2:22 the protagonist has just learned from Mordecai of the regicidal plot and must alert the king. καὶ αὐτὴ ἐνεφάνισεν

և Եսթեր եցոյց (and Esther demonstrated)

As the third person pronoun is undifferentiated,49 the Greek feminine form cannot be replicated, necessitating her designation by name. Tob 3:10 ἀκούσασα

իբրև լուաւ աղջիկն (when the girl heard)

Similarly, in the above case, as Armenian lacks the means of denoting Sarra by the feminine form of a participle, she is referred to more directly by a noun. Armenian kinship terms make a fascinating field of study, as illustrated by Tob 14:13(15): ἔθαψεν τοὺς πενθεροὺς αὐτοῦ թաղեաց նա զաներն իւր և զզոքանչն (he buried his father-in-law and mother-in-law)

One of their fundamental characteristics is their individual distinctiveness, which precludes the formation of collectives as that of πενθεροὺς (parents-in-law) in the Greek, necessitating the specific mention of each. Another facet of Armenian discontinuity with Greek grammar relates to a category of verbs featuring a medio-passive form with active meaning like Latin deponents. The aorist tense of the Armenian verb “to listen” falls within that classification.50 Consequently, at Tob 3:16 (24) the passive is rendered by a periphrasis, which therefore does not constitute a variant reading. εἰσηκούσθη ἡ προσευχὴ ἀμφοτέρων լսելի եղեն աղօթք երկոցունն (lit. the prayer of both became audible)

48  For a discussion of the loss of grammatical gender, see R. Godel, An Introduction to the Study of Classical Armenian (Wiesbaden: Ludwig Reichert Verlag, 1975), 92–9. 49  On the undifferentiated third person pronouns, see A. Meillet, Altarmenisches Elementarbuch (Heidelberg: Carl Winters Universitäts-Buchhandlung, 1913), 60. On the artificial attempt to create a distinctive feminine form նէ by Armenian grammarians on the basis of Greek categories, see G. Awetik‘ean, Nor bar˙girk‘ haykazean lezui [New Dictionary of the Armenian Language] 2 vols. (Venice: St. Lazar’s Press, 1836–1837), vol.1, 424. 50 Meillet, Altarmenisches Elementarbuch, 105.

Writing and Rewriting Scripture in the Armenian Versions

251

Translation and Armenian Idiom Even in specific cases where morphological compatibility exists between the base and target languages, translators must also consider the dictates of established idiom in selecting the most appropriate rendering, which may not necessarily be amenable to literal isomorphic alignment. It is striking that early Armenian did not possess a denominative verb of praying but, like Syriac and Persian,51 expressed the action by means of the periphrasis of ‘standing at prayer’. Having been blinded, as narrated in chapter 2, it is likely that Tobit thereafter led a more sedentary life. Therefore, when, after an altercation with his wife that underlined the degree of his current hardships and suffering, he is moved to pray about his situation, the segue to this would inevitably be to rise to his feet, as indicated in the Armenian version at 3:1: καὶ προσηυξάμην և յարեայ կացի յաղօթս (and I rose [and] stood at prayer)

Consequently, the presence of the form յարեայ (I rose) in situ should not be interpreted as a representing a plus in its parent text, but rather as providing a smooth transition from the previous scene to the current. Tob 3:15(17) ἵνα συντηρήσω ἐμαυτὴν αὐτῷ γυναῖκα զի պահեցից ես զանձն իմ նմա կնութեան (for me to keep myself for him for wifehood)

In a similar prayer seeking release from her inability to lead a normal family life because of the untimely death of a succession of husbands, Sarra despairs of attaining her goal as her father has no more kinsmen for her to marry. Since she has no further prospects, she might as well be dead. In Greek she articulates the possibility of marriage as ‘keeping herself for him as wife’. However, although grammatically the sentiment could have found direct expression in Armenian, there the accepted expression for marriage is one of ‘wifehood’. The currency of the idiom is corroborated by an embassy recorded in the mid-5th century historian P‘awstos to the effect that Շապուհ թագաւորն Պարսից կամի տալ զդուստր իւր կնութեան Արշակայ թագաւորին Հայոց (the Persian king Shapur wishes to give his daughter in marriage (lit. for wifehood) to the Armenian king Aršak).52

51  D. Kölligan, “Armenisch-deutsches Text”, in O. Zwierlein, Die Urfassungen der Martyria Polycarpi et Pionii und das Corpus Polycarpianum, vol. 1 (Untersuchungen zur antiken Literatur und Geschichte 116; Berlin: De Gruyter, 2014) 46–66, n. v, 55. 52  Ps. P‘awstos, Buzandaran Patmut‘iwnk‘ (The Epic Histories) also known as Patmut‘iwn Hayoc‘ (History of Armenia) Attributed to Pawstos Buzandac‘i (Delmar, NY: Caravan Books, 1984), 113.

252

S. Peter Cowe

As we have already noted, most of the Septuagint and related Greek texts preserved witness a penchant to represent as closely as possible the morphology and syntax of the Massoretic Text with little or no regard for the idiom of the target language. This frequently results in the creation of Hebraisms in the texture of the translation (e.g. parataxis, literal parallelism of the waw conservative, construct infinitive, repetition of prepositions with two or more referents, expressions not compatible with Greek usage, etc.). At the same time, changes in readership and ambiance in Christianity as it became first licit (Edict of Milan, 313), then the official religion of the empire (380), led to moves to articulate its message in terms more accessible to the educated classes. Whereas the new faith had initially circulated in Armenia in Greek and Syriac, the officially sponsored translation of the Bible into Armenian was clearly intended to make Scripture speak to the population in the vernacular, as stated by the hagiographer Koriwn, himself a participant in the task.53 The Old Greek version of Judith belongs to the mainstream and hence features many instances of interlinear translation technique as at Jdt 16:13 (16): θαυμαστὸς ἐν ἰσχύι

սքանչելի ես զօրութեամբ (you are wondrous with might)

Here the point at issue concerns the function of the Hebrew preposition ‫ ב‬to denote manner, relation, or perspective, while Armenian renders this directly through the instrumental case alone. Another role of the same Hebrew preposition is to convey means or instrumentality, a feature which Indo-European inflected languages tend to represent directly by a case associated with that semantic sphere.54 Thus at the opening of Judith’s song of praise in v. 1 where the protagonist invokes the musical instruments accompanying festivities, typical of his general procedure, the Greek translator again maintains quantative parity with the underlying Hebrew text first by assigning the Hebrew preposition the Greek equivalent ἐν and then by repeating it with each instrument cited. ἐν τυμπάνοις . . . ἐν κυμβάλοις

թմբկօք . . . ծնծղայիւք (with drums . . . with cymbals)

In contrast, his Armenian counterpart preserves the idiom of the target language in conveying the sense simply via the equivalent terms in the instrumental case.

53  Ē. Pivazyan, Vark‘Maštoc‘i [The Life of Maštoc‘] (Erevan: Erevan State University Publication, 1980), 282. 54  For the situation in Armenian, see Meillet, Altarmenisches Elementarbuch, 44–58.

Writing and Rewriting Scripture in the Armenian Versions

253

Rhetorical Explication of Figures of Speech The translators of the early stratum of the Armenian version (Arm1), which is largely represented in our books as a result of a rather superficial revision mentioned above, were significantly influenced by the exegetical practice of the School of Antioch, as already described. A frequent facet of this enterprise, which, we have noted, was already of concern to the redactor Lucian, is the elucidation of metaphor in more straightforward, literal terms.55 Figures of speech are explicated, not replicated. Let us review the phenomenon in action in a set of examples also drawn from Sarra’s predicament in Tobit chapter 3. Tob 3:9 τί ἡμᾶς μαστιγοῖς; արդ զմեզ զի՞ տանջես (so why are you afflicting us?)

While her maids complain at Sarra’s continual harping on her loss of seven husbands on her wedding night in the evocative metaphor of beating or scourging, all the more unpleasant for them since they are not responsible, the connotation is of course a verbal drubbing, and hence it is rendered into Armenian in more neutral language. Tob 3:12 (14) τοὺς ὀφθαλμούς μου καὶ τὸ πρόσωπόν μου εἰς σὲ δέδωκα բոլորով սրտիւ իմով (with my whole heart)

Similarly, at the opening of Sarra’s prayer in v. 12 the young woman refers to her physical posture to convey the sense of her directing her whole self to God in prayer, a psychological reality the translator captures in referring to the heart. Meanwhile, in the following verse the young woman is so overwhelmed by her plight that she yearns to be released from it in death, a procedure she portrays figuratively as being ‘set free from the earth’. Tob 3:13 (15) εἶπον ἀπολῦσαί με ἀπὸ τῆς γῆς ասացի հրաժարել հոգւոյս իմարմնոյս իմմէ (I said for my soul to separate from my body)

The translator, who incidentally interpreted the initial imperative in terms of its first person aorist homograph, renders the idea more literally in terms of the separation of the soul from the body. In the final case, in concluding his prayer of rejoicing in chapter 13, Tobit waxes lyrical about the streets of Jerusalem spontaneously bursting into a song of praise to the Creator.

55  Cowe, “The Bible in Armenian”, 151. See also id., “Rhetoric, Theology, and Antiochene Exegesis”, 152–4.

254

S. Peter Cowe

Tob 13:18(23) ἐροῦσιν πᾶσαι αἱ ῥῦμαι αὐτῆς Ἁλληλοιά ասասցեն յամենայն փողոցս Երուսաղէմի ալէլուիա (they will say in all the streets of Jerusalem Halleluiah)

Here, too, the Armenian reformulates the expression in a more literal mode, transferring agency to the inhabitants of the city.

Rhetorical Harmony and Balance of Expression Another facet of the rhetorical proclivities of the Armenian translator is his attention to balance and harmony of expression, particularly in poetic passages or elevated prose. Attuned to the basic structure of Hebrew poetry in parallelismus membrorum,56 he employs this as a constructional principle in rendering the Greek into Armenian, enhancing this characteristic in his version where he finds it lacking in his Greek exemplar. A typical example of his treatment at the micro level is afforded by the maid’s comment at Tob 3:9: εἰ ἀπέθαναν, βάδιζε μετ᾽αὐτῶν

ընդէ՞ր և ոչ դու ընդ նոսա կորեար (why didn’t you perish along with them?)

Where the Greek formulation is rather abrupt and uncoordinated, the Armenian reframes the remark as one coherent, integrated whole. Our next example explores the translator’s handling at the macro level of what might be termed a ‘purple’ passage literally and figuratively. The description of the courtyard of Ahasuerus’ palace at Esther 1:6, as it was elaborately decked out for a reception for the population of Susa, is a highly instructive example of a growing text.57 κεκοσμηνένῃ βυσσίνοις καὶ καρπασίνοις τεταγμένοις1 և էր զարդարեալ բեհեզովք և կերպասովք` զորս էր կարգեալ և կազմեալ2 and it was adorned with muslins and silks that he had arranged and fitted 1 τεταμένοις] τεταγμένοις S* 93 Arm(vid) 2 կազմեալ (emendation)] կազմեալս Z

56 O’Connor, Hebrew Verse Structure (Winona Lake, IN: Eisenbrauns, 1980), 87–137. See also Cowe, “Rhetoric, Theology, and Antiochene Exegesis”,151–2. 57  The narrative below features the parallel Greek and Armenian texts followed by the latter’s English translation in a three-line sequence. The layout adopted renders the translator’s expansionist approach more visually perceptible. As the Armenian represents a more critically established textual form correcting various secondary errors in Zohrapean’s continuous text, a fourth line has been added as an apparatus to document the variant readings.

Writing and Rewriting Scripture in the Armenian Versions

255

ἐπὶ σχοινίοις βυσσίνοις καὶ πορφυροῖς ἐπὶ κύβοις χρυσοῖς զլարեօքն ծիրանեօք և բեհեզովք. և զորս էր կապեալ3 ի խոյակսն ոսկիս, about cords of purple and muslin and that he had attached to the golden 3 կարգեալ Ztxt] կապեալ Ζap καὶ ἀργυροῖς, ἐπὶ στύλοις παρίνοις καὶ λιθίνοις և ի ծիրանիս որ4 կային ի վերայ սեանցն պատուականաց. և ի խարսխացն and purple capitals which stood atop the exquisite columns and on the golden 4 ուր Ztxt] որ Zap κλῖναι χρυσαῖ καὶ ἀργυραῖ ἐπὶ λιθοστρώτου ոսկեղինաց ուր5 կային գահոյք բազումք ոսկեղէնք և արծաթեղէնք ընդելուզեալք bases where stood many couches gold and silver inlaid with 5 որ Z] ուր (emendation) σμαραγδίτου λίθου καὶ πινίνου καὶ παρίνου λίθου, καὶ στρωμναὶ διαφανεῖς ποικίλως մարգարտովք և ակամբք պատուականօք` լցեալս պաստառւօք. և գոյնագոյն pearls and precious stones replete with fine linens and variegated διηνθισμέναι, κύκλῳ ῥόδα πεπασμένα նկարօք, և պատուական հանդերձիւք ի վերայ վարդայատակ գաւթին: designs and costly raiments on the rose-floored courtyard

Already ornate in the Hebrew version, the translator of the LXX text rendered the scene even more elaborate by adding a number of details, such as that the rings or rods (a point we will return to) by which the exquisite fabrics were hung were of both silver and gold (ἐπὶ κύβοις χρυσοῖς καὶ ἀργυροῖς). Likewise the pillars were not merely of marble but also of other stones (ἐπὶ στύλοις παρίνοις καὶ λιθίνοις), and while the Massoretic Text is silent concerning the cloths that adorned the couches where the guests were to recline, the Greek indicates they were elaborately embroidered (στρωμναὶ διαφανεῖς ποικίλως διηνθισμέναι). Naturally, the Armenian translator’s approach is informed by Antiochene exegesis. Let us start with the crux of the passage, the reference to rings or rods by which the curtains were affixed to the columns. The Hebrew term for that (‫ )לילג‬is found in this sense only here in scripture, and hence was apparently unfamiliar to the Greek translator who rendered it by κύβος, which in turn appears only here and at Job 38:38 in the Septuagint in an unrelated context. The notion of a cube on a column to which various hangings were attached clearly

256

S. Peter Cowe

suggested to his Armenian counterpart that the reference was to the capital (ի խոյակսն), which would often be picked out in attractive colors, and from there fullness of expression demanded a reference to the base as well (ի խարսխացն), which was also of gold (ոսկեղինաց). Similarly, maintaining a logical sequence required the Greek portrayal of the scene to be revised since, after following the Hebrew order moving from the columns to the couches and floor, it then adds a note about the embroidered textiles, which were festooned on the couches, necessitating a reversal of movement. Consequently, the Armenian interprets the term λιθόστρωτος not in the nominal sense of mosaic pavement but as a verbal adjective meaning ‘strewn with stones’, which he then applies to the rich inlay work embellishing the couches (ընդելուզեալք մարգարտովք և ակամբք պատուականօք) before turning to the drapes spread on them and finally treating the floor. Here, too, the Greek is rather abrupt, loosely appending a note on roses scattered around (κύκλῳ ῥόδα πεπασμένα), whose connection with the preceding is uncertain. Some have taken it to refer to designs on the drapes,58 while it is equally possible to construe the phrase as relating to actual roses strewn on the floor to heighten the opulence of the surroundings.59 Conscious of the ambiguity from his perspective as to whether the reference is to actual roses or to a motif on the tiling and of the need to provide the scene with a fitting conclusion, the translator determined to coin a new term (վարդայատակ) ‘rose-floored’ constructed so as to allow both interpretations, which has remained a hapax legomenon in Armenian literature.60 Moreover, throughout the verse the translator has been exercised to sustain rhetorical harmony and balance of expression in the various clauses, making connections explicit where these were lacking, as in the verb (կապեալ) for attaching the fabrics to the columns, and creating an overall unity of luxury and splendor by the leitmotif of պատուական (precious, costly) that recurs three times in the narrative.

Intertextuality A regular feature of Antiochene exegesis, as has been alluded to, is to define context by intertextuality within the broader scriptural landscape, setting the articu58  K.H. Jobes, “Esther”, in A. Pietersma/ B.G. Wright (ed.), A New English Translation of the Septuagint (Oxford: Oxford University Press, 2007) 424–40, on p. 427. 59  It was a custom among various peoples to strew rose petals on the floor before dignitaries so as to obviate their defiling their feet on the ground. For the practice, see M. Touw, “Roses in the Middle Ages”, Economic Botany 36/1 (1982) 71–83, on p. 71. 60  It is one of the characteristics of early Armenian literature already apparent in the Bible translation for writers to create neologisms, especially in the form of compound adjectives, many of which did not establish themselves in subsequent literary usage.

Writing and Rewriting Scripture in the Armenian Versions

257

lation of a particular thought or action in one verse in dialogue with wider parallels.61 At the micro level it may be witnessed in the prelude to Sarra’s prayer at Tob 3:11, where her stance near the window triggered associations with Daniel’s similar pose in an analogous situation of deep existential angst. ἐδεήθη πρὸς τῇ θυρίδι կայր նա յաղօթս ընդ պատուհանն որ էր ընդդէմ Երուսաղէմի (she stood at prayer by the window which was toward Jerusalem) Dan 6:10 և պատուհանք բաց էին նմա . . . ընդդէմ Երուսաղէմի (and his windows were open toward Jerusalem)

Characteristic of the application of the principles of Antiochene exegesis at the macro level by the exponents of Arm1 is their approach to Judith’s prayer in 9:2–14. The form of intertextualization employed there is primarily grounded in the book of Esther, particularly the heroine’s own prayer in C:12–30 (Armenian 14:1–19), by which the translators recreate the sequence of thought and enhance the rhetorical balance of the piece. It is interesting that the Greek formulation of this passage in turn is dependent on Moses’ prayer at Deut 9:26–29, thus highlighting some of the continuities of intertextual rewriting at work within the early centuries of scriptural transmission in different geographical and linguistic environments.62 Let us begin with the prelude at Jdt 9:1 where the Armenian observes the more logical order of events: καὶ ἐπέθετο σποδὸν ἐπὶ τἠν κεφαλὴν αὐτῆς καὶ ἐγύμνωσεν ὅν ἐνεδεδύκει σάκκον և ի բաց առ զքուրձն որ զգեցեալ էր ինքեան, և ցանեաց մոխիր ի վերայ գլխոյ իւրոյ: (and she removed the sackcloth, which she had been wearing, and strewed dust over her head) Esther 4:1 on Mordecai: պատառեաց զպատմուճան իւր, և զգեցաւ քուրձ, և ցանեաց մոխիր զգլխով իւրով (he rent his cloak, and put on sackcloth, and strewed dust all over his head)

There Judith, like Mordecai, removes her garment before scattering ash on her head, in contrast to the Greek text in situ. Moreover, the term ցանեաց (‘strewing’)

61  Cowe, “The Bible in Armenian”, 152, and id., “Rhetoric, Theology, and Antiochene Exegesis”, 154–8. 62  K.H. Jobes, The Alpha-Text of Esther: Its Character and Relationship to the Massoretic Text (JBL Dissertation Series 153; Atlanta, GA: Scholars Press, 1996), 230.

258

S. Peter Cowe

to connote the action repeats the Armenian lexical choice in Esther over against the Greek verb ἐπέθετο (‘placing upon’ sc. her head). The heroine’s subsequent appeal at Jdt 9:4 (3) juxtaposes her self-reference as a widow in parallel to the Greek with the additional appellation as a maidservant, a term recurring in the Armenian text at v. 12. εἰσάκουσον ἐμοῦ τῆς χήρας լուր ձայնի աղախնոյ քոյ և նայեաց ի խնդրուածս այրւոյս (hear the voice of your maidservant and look to the petitions of me a widow) Jdt 9:12(18) εἰσάκουσον τῆς δεήσεώς μου լուր աղօթից աղախնոյ քոյ (hear the prayer of your maidservant)

The term աղախին (maidservant) occurs twice in the Armenian version of Esther’s prayer as an apposite equivalent for its counterpart δούλη in the Greek text at C:28 (14:17) cited below and in the following verse: οὐκ ἔφαγεν ἡ δούλη σου ոչ կերաւ աղախին քո (your maidservant did not eat)

The effect of the additions in Judith chapter 9 is therefore to draw the composition more firmly within the scriptural language of prayer and petition. A further aspect of the process is the subsidiary addition of the term ձայն (voice) at Jdt 9:4 (3) referenced above which finds its analogue in Mordecai’s prayer at Esther C:1–10 (Armenian 13:8–18). There too the Armenian translator has added it to the official’s appeal in v.10. to provide a more fitting conclusion: εἰσάκουσον τῆς δεήσεώς μου 13:17 լուր Տէր ձայնի աղօթից իմոց (hear, Lord, the voice of my prayer)

In turn, the latter clearly evokes the lexicon of the Psalter: cf. Ps 27(LXX 26):7 εἰσάκουσον, κύριε, τῆς φωνῆς μου լուր Տէր ձայնի իմում (Lord, hear my voice) 28(LXX 27):2 εἰσάκουσον τῆς φωνῆς τῆς δεήσεώς μου լուր Տէր ձայնի խնդրուածոց իմոց (Lord, hear the voice of my petitions)

One final aspect of the reformulation I should like to review relates to the activity elicited of God and the object to which that is directed. The Greek text at Jdt 9:14(19) presents the heroine’s general intercession for God to make himself known in a way that would reveal his divinity:

Writing and Rewriting Scripture in the Armenian Versions

259

ποίησον ἐπὶ πᾶν τὸ ἔθνος . . . ἐπίγνωσιν τοῦ εἰδῆσαι ծանո զբարձր բազուկ քո ժառանգութեան քոյ (make known your high upheld arm to your inheritance)

Meanwhile, the Armenian translator interprets this in light of the conclusion of Esther’s prayer at C:25. Although the Greek version there refers to God’s hand in undifferentiated fashion, its Armenian counterpart gives the expression more point by a powerful evocation of the Exodus tradition of salvation where the arm substitutes for the hand as a symbol of more active intervention: ἡμᾶς δὲ ῥῦσαι ἐν χειρί σου 14:14 զմեզ փրկեա բարձր բազկաւ քով (save us with your high upheld arm)

This becomes obvious when we compare the common reading of both Greek and Armenian texts at Exod 6:6: λυτρώσομαι ὑμας ἐν βραχίονι ὑψηλῷ ապրեցուցից զձեզ բարձր բազկաւ (I will deliver you with a high upheld arm)

In this way the Armenian translator of Judith transforms the protagonist’s prayer through the mediation of the parallel composition in Esther, thereby setting it in continuity with the archetypal situation of God’s self-revelation to his people in Exod 6:2–6.63

Theological Concerns Inspired by Antiochene perspectives, the Armenian translators exhibit distinct theological concerns, which they creatively weave into their rendering by their selection of phrasing, additions, and targeted omissions.64 An instructive example of this treatment is Mordecai’s prayer in Esther C:2–10. This the translator achieves in part by scriptural intertextuality, integrating the thought structures of the prayer into that of the Psalmist and other well-known passages. 63  Several scholars have commented on the centrality of traditions in the construction of Esther. See G. Gerleman, “Studien zu Esther: Stoff—Struktur—Stil—Sinn”, BS 48 (1966) 1–48, D.J.A. Clines, The Esther Scroll: The Story of the Story (JSOTSS 30; Sheffield: University of Sheffield, 1984), 155, and Jobes, The Alpha-Text of Esther, 121, 177–80, 230. 64  For the significance of divine providence, see Cowe, “Rhetoric, Theology, and Antiochene Exegesis”, 160–4, and, for an anti-idolatry diatribe, id. “The Armenian Version of the Epistle of Jeremiah: Parent Text and Translation Technique”, in C.E. Cox (ed.), VII Congress of the International Organization for Septuagintal and Cognate Studies, Leuven 1989 (SCSS 31; Atlanta, GA: Scholars Press, 1991) 373–91.

260

S. Peter Cowe

One of the tenets he underscores is divine omnipotence, which he protects against possible diminution, such as by association with other gods, as at v. 2 (13:9): βασιλεῦ πάντων κρατῶν . . . ἐν ἐξουσίᾳ σου τὸ πᾶν ἐστιν թագաւոր ամենակալ . . . քում հրամանիդ սպասէ ամենայն ոք (king almighty . . . upon your command does everyone wait)

There the ‘syncretic’ phrase βασιλεῦ πάντων κρατῶν (king of all powers) is substituted by the exclusivist technical term թագաւոր ամենակալ (king almighty), a distinction sustained in the continuation where the phrase ἐν ἐξουσίᾳ σου τὸ πᾶν ἐστιν (everything is in your authority) is transformed into քում հրամանիդ սպասէ ամենայն ոք (upon your command does everyone wait), which is modeled on Acts 23:21:65 սպասեն քում հրամանի (they wait upon your command)

Note in both cases the pointed initial placing of the possessive for maximum effect. Similarly, in v. 4 (13:11) the translator significantly subverts potential opposition to the deity: ἀντιτάξεταί σοι τῷ κυρίῳ հակառակ կայ քում տէրութեանդ (stands opposed to your rule)

Instead of implying adversaries of such equal magnitude as to contend against the Lord himself, the translator refers to the deity indirectly in terms of his lordship or rule, implying that the challenge is presented by an entity already subsumed under that dominion. The translator then develops his argument at v. 22 (14:11) by neutralizing the envisioned relinquishing of divine rule to other divinities: μὴ παραδῷς, κύριε, τὸ σκῆπτρόν σου τοῖς μὴ οὖσιν մի մատներ զվիճակ ժառանգութեան քոյ ի ձեռս այնոցիկ որ ոչն իսկ են բնաւ (do not transfer the lot of your inheritance into the hands of those who indeed inno-way exist)

Concurrently, he refocuses attention from power struggle to care for those under God’s governance, i.e. Israel, emphasizing the non-existence of those entities by the nuanced use of the particles in bold. 65  Cf. Ps 119 (LXX 118):82.

Writing and Rewriting Scripture in the Armenian Versions

261

Turning to the human sphere, the potential for reverencing it above the divine is left open by Mordecai in the Greek text, while denying his own inclination to pursue that course of action (13:14): οὐ προσκυνήσω οὐδένα πλὴν σοῦ τοῦ κυρίου μου զի մի տաց փառս մարդոյ, այլ քեզ միայն ճշմարտիդ Աստուծոյ (so I should not give glory to man, but to you alone, the true God)

However, the possibility of such veneration is expressly rejected in the translator’s reformulation, which explicitly contrasts the categories of human and divine, reserving worship exclusively for the latter, grounding his monotheistic message in other scriptural passages. Isa 65: 16 εὐλογήσουσιν γὰρ τὸν θεὸν τὸν ἀληθινόν օրհնեսցեն զաստուած ճշմարիտ (they shall bless the true God) John 17:3 ἵνα γινώσκωσιν σὲ τὸν μόνον ἀληθινὸν θεόν զի ծանիցեն զքեզ միայն ճշմարիտ աստուած (that they may know you the only true God)

In this connection, it is important to recall that religious pluralism was a feature of the landscape which the Armenian translators sought to engage with in their scriptural hermeneutic.66

Macro level Redaction of the Old Greek of Esther The multiplicity of layers in the narrative development of Esther and the imprecision of the methodologies employed to decisively distinguish priority within the sequence of strata has led to the proliferation of schemas, each affording valuable insights, but marred by areas of less plausible argumentation.67 Several scholars maintain that the Alpha Text represents a form of Hebrew text predating the

66  On the background, see Cowe, “The Bible in Armenian”, 143–4. As a result, distinguishing licit and illicit cult, locus, ministers, and paraphernalia became a major preoccupation of the Armenian translators, on which, see id., “Tendentious Translation and the Evangelical Imperative: Religious Polemic in the Early Armenian Church”, Revue des études arméniennes 22 (1990–91) 97–114. 67  For a succinct presentation of the various viewpoints, see R. Kossmann, Die Esthernovelle vom Erzählen zur Erzählung: Studien zur Traditions- und Redaktionsgeschichte des Estherbuches (SVT LXXIX; Leiden: Brill, 2000), 313–21.

262

S. Peter Cowe

Massoretic, from which the Old Greek version derives68 while others hold the opposite permutation that the Old Greek is the earlier of the two translations and that the Alpha Text marks a redaction of that and therefore lacks independent contact with the Hebrew.69 Meanwhile, Milik, beginning with the Qumran fragments 4Q550a–f, contends that these form the nucleus of the Esther narrative, which, after various early developments, emerges as the Alpha Text, of which the Old Greek is a later redaction, while the Massoretic Text represents the latest textual form.70 In contrast to the earlier strata of Esther, the position of the Armenian Version in the book’s transmission history is clear, in that it rests largely on the Old Greek with some secondary affinities with the Alpha Text. However, we have already indicated that the translator’s approach can be interventionist and have just reviewed several of his rhetorical and theological predilections. In concluding, I should like to succinctly present examples of a much more thoroughgoing multifaceted redactional focus that informs all aspects of his narrative in such a way as to significantly revisualize the tale. Though the Hebrew and Greek texts were certainly composed by Jews and represent a perspective sympathetic to them and their welfare in the Persian realm, it is striking how the Armenian Version underscores their defense by heightening the esteem and preferment in which Esther and Mordecai are held by the authorities, undermining the opposition of Haman and the two eunuchs, and offering a ringing endorsement of the king’s reversal of the latter’s edict for their destruction.71 Thus, while scholars have observed and variously interpreted the absence of any explicit divine intervention in the Massoretic Text72 and have 68 Clines, The Esther Scroll, 140–151; C.V. Dorothy, The Books of Esther: Structure, Genre and Textual Integrity (JSOT SS 187; Sheffield: Sheffield Academic Press, 1997); K.H. Wynn, “The Sociohistorical Contexts of the Recensions of Esther” (Ph.D. dissertation; Southern Baptist Theological Seminary, 1990); M.V. Fox, The Redaction of the Books of Esther (Atlanta: Scholars Press, 1991); Jobes, The Alpha-Text of Esther, 223–33. 69  K. De Troyer, The End of the Alpha Text of Esther: Translation and Narrative Technique in MT 8:1–17, LXX 8:1–17, and AT 7:14–41 (SBL SCSS 48; Atlanta, GA: Society of Biblical Literature, 2000), 397–403. 70  J.T. Milik, “Les modèles araméens du livre d’Esther dans la Grotte 4 de Qumrân”, Revue de Qumrân 59/15 (1992) 321–99. 71  It is significant that here the Armenian Version maintains continuity with similar initiatives on the part of earlier tradents. An instructive example is the heightening of the contrast between moral and reprehensible conduct in the Old Greek of Proverbs, on which see J. Cook, “Proverbs”, in A. Pietersma/ B.G. Wright (ed.), A New English Translation of the Septuagint (Oxford: Oxford University Press, 2007) 621–47, on pp. 621–2. For the variety in characterization of the protagonists and antagonist in different versions this book, see A.K. Fountain, Literary and Empirical Readers of the Books of Esther (SBL 43; New York: Peter Lang, 2002), 75–129. 72 For the perspective of synergy between human agency and divine intervention, see Clines, The Esther Scroll, 155–157, and, for the veiled aspect of divine presence in the Massoretic Text, H.M. Wahl, “‘Glaube ohne Gott?’ Zur Rede vom Gott Israels im hebräischen Buch

Writing and Rewriting Scripture in the Armenian Versions

263

tended to view the two Greek texts’ more tangible references as a corrective from a more conventional religious viewpoint, it is significant that the Armenian text marks a further step in that direction to reinforce divine providence. Meanwhile, the Jews are perceived not as a vulnerable ethno-religious minority in the Persian realm, but as the people of God in a spatiotemporal sphere, which he controls. Let us now consider the means by which the translator/redactor achieves this goal.

More Strongly Anti-Jewish Opposition 3:8 οἱ δὲ νόμοι αὐτῶν ἔξαλλοι παρὰ πάντα τὰ ἔθνη որք են ապիրատք և խորամանգք քան զամենայն ազգս (who are more iniquitous and deceitful than all nations)

In treating Haman’s opposition at 3:8, rather than focusing on the distinctiveness of Jewish law, the translator undermines the credibility of his charges by reporting Haman’s subjective slander of the Jews for their immoral conduct.73 2:21 ἐλυπήθησαν οἱ δύο εῦνοῦχοι . . . ὅτι προήχθη Μαρδοχαῖος տրտմեցան երկու ներքինիք . . . վասն զի նախանձէին ընդ Մուրթքէի` առաւել քան զնոսա պատուելոյ ի թագաւորէն (the two eunuchs were sorrowful . . . because they were envious of Mordecai at his being honored more than them by the king)

Similarly, the translation at 2:21 explores the eunuchs’ antagonism to Mordecai at a more profound level, predicating it on malicious envy at his promotion by the king. 7:9 ὕλον ἡτοίμασεν Αμαν Μαρδοχαίῳ τῷ λαλήσαντι περὶ τοῦ βασιλέως փայտ մի կազմեաց Մուրդքի սիրելւոյ արքայի (he has set up a pole for Mordecai the king’s friend)

Moreover, instead of having the eunuch underscore Mordecai’s status at 7:9 by emphasizing the service he had done Ahasuerus by alerting him to the plot on his life, the translator chooses rather to focus on the monarch’s newfound affection for him, leading to his designation as friend. Esther”, BZ 45 (2001) 37–54, and B. Schmitz, “‘… am Ende ihres Weges Den zu schauen, an dem man stirbt, wenn man ihm naht’ (Rainer Maria Rilke): Die Rede von Gott in den EsterErzählungen”, in R. Egger-Wenzel/K. Schöpflin/J.F. Diehl (ed.), Weisheit als Lebensgrundlage (DCLS 15; Berlin: Reiterer, 2013) 275–96. 73  One aspect of the change is to remove emphasis on the law as the prime institution of Jewish life as being less significant for a non-Jewish readership.

264

S. Peter Cowe

Esther’s Favor 2:9 ἐχρήσατο αὐτῇ καλῶς καὶ ταῖς ἅβραις αὐτῆς դարմանէր զԵսթեր առաւելագոյն քան զայլ ընկերս նորա, և զնաժիշտս նորա նոյնպէս (he cared for Esther more than her other counterparts, and her maids likewise [sc. more than their counterparts])

A parallel favor is also extended to Esther, whose distinction from the other candidates for the king’s hand is already expressed at 2:9 where the chief of the harem affords preferential treatment not only to her but extends it to her maids.

Haman’s Reversal of Fortune 6:12 Αμαν δὲ ὑπέστρεψεν . . . λυπούμενος և Համան գնաց . . . լի տրտմութեամբ (and Haman went . . . full of sorrow) 6:13 ἤρξαι ταπεινοῦσθαι ἐνώπιον αὐτοῦ . . . οὐ μὴ δύνῃ αὐτὸν ἀμύνασθαι, ὅτι θεὸς ζῶν μετ᾽αὐτοῦ և եղև իշխան, հնազանդեա նմա. զի ոչ կարես դու նմա չար հատուցանել, քանզի Տէր Աստուած ընդ նմա է, և նա մեծացոյց զնա: (and he [sc. Mordecai] has become a prince, subordinate yourself to him; for you are unable to render him any harm, as the Lord God is with him and he has magnified him.)

As the reversal of fortune begins in chapter 6, Haman’s situation in the Armenian version becomes more untenable than in the Greek. First of all he is not only sorrowful but ‘full’ of sorrow at v.12 as he returns home after escorting Mordecai on his parade through the city. Then, if the Old Greek develops the Hebrew text to present his wife as a spokesperson for Yahwism in stating that the Living God is with him so that he should pragmatically acknowledge his ascendance, since he cannot ignore him or sideline him, the Armenian version develops the parameters of the discourse. Far from ignoring Mordecai, Haman has actively schemed to kill him, however he now learns his agency is ineffective against the power of divine providence which has exalted his adversary and raised him to be a leading figure in the state.

Writing and Rewriting Scripture in the Armenian Versions

265

Revision of Haman’s Edict 8:11 χρῆσθαι τοῖς νόμοις αὐτῶν . . . βοηθῆσαί τε αὑτοῖς καὶ χρῆσθαι τοῖς ἀντιδίκοις αὐτῶν καὶ τοῖς ἀντικειμένοις αὐτῶν 8:10 որպէս զի կացցեն հրեայք ողջ և անարատք 8:11 իւրեանց օրինօք . . . օգնականք և թիկունք լինել առ ի հատուցանել նոցա զփոխարէն թշնամեաց իւրեանց. (for the Jews to live safely and irreproachably by their laws . . . to assist and support in repaying to them the recompense of their enemies)

The same process encompasses the subversion of Haman’s edict against the Jews. While the Hebrew text highlights the rights the new royal ordinance affords them to kill their assailants and confiscate their property and the Old Greek emphasizes permission to live in conformity to their laws in addition to self-defense, the Armenian version actively instructs state officials to support the community (clearly reading αὑτοῖς as αὐτοῖς) in attaining a secure lifestyle and facilitating the settling of grievances against their enemies, thus placing the onus on the civil service.74 9:2 ἀπώλοντο οἱ ἀντικείμενοι τοῖς Ἰουδαίοις. οὐδεὶς γὰρ ἀντέστη 9:1 կորնչէին որ միանգամ ընդդիմադարձք լինէին հրէիցն: 9:2 և ոչոք էր որ ընդդէմ դառնայր նոցա ամենևին (whoever had opposed the Jews were perishing, and there was no one who stood against them in any way)

The impact of the new edict at 9:2 is also more emphatic in Armenian where reference to both groups, their previous adversaries and the majority of the current population, is absolutized.

Euphemistic Presentation of Jewish Retaliation 9:14–15 ἐξέθηκεν τοῖς Ἰουδαίοις . . . τὰ σώματα τῶν υἱῶν Αμαν κρέμασαι . . . καἰ ἀπέκτειναν ἄνδρας τριακοσίους հան . . . առ հրէայսն . . . զմարմինս որդւոցն Համանայ հանել ի փայտ . . . և կուտեցին արս երեքհարիւր (he removed . . . to the Jews . . . the bodies of Haman’s sons to string up . . . and they piled up three hundred men)

74  For the issue of Gentile authorities protecting the Jewish community, see Clines, The Esther Scroll, 174.

266

S. Peter Cowe

Moreover, while the Hebrew and Greek texts plainly state that the Jewish community of the capital hanged Haman’s sons and killed three hundred others by different means, the Armenian translator is exercised to minimize their involvement in retaliation, indirectly alluding to such activities by the euphemisms of ‘stringing up’ and ‘piling up’ to mitigate possible adverse reactions to the blood-letting.75

Conclusions Having reviewed a series of examples illustrating textual change at different levels within the context of the Armenian version of Esther, Judith, and Tobit, what can be said by way of summatory remarks? Bearing in mind what was said about the ubiquity of variation and diversity in the ancient world, when considering the domain of textual transmission and its parameters more narrowly, it is clear that the scribe and redactor occupy the extremities of the spectrum, the former aiming at maximal representation of the work being copied and the latter intent on modifying the form and content of a writing and thereby transforming its impact on the reading public. Within that range the translator occupies the wide middle ground, desiring on the one hand to remain faithful to certain central aspects of the text selected for treatment, but on the other conscious of his or her responsibility to make the latter communicate to a readership in terms they will understand and appreciate. In this way the translator is rarely an antiquarian, but follows an exegesis informed to an important degree by hermeneutics. As a result, we are not to imagine the translation process as occurring in a vacuum, nor as the interface purely between text and translator, but as one inevitably mediated by the readings of intervening generations that shape the act of producing meaning in a dynamic process that we can only imperfectly reconstruct. Our survey of the Armenian version of Esther, Judith, and Tobit has tried to demonstrate the impact of Antiochene rhetoric and theological engagement on the translation both at the micro level of individual verses and passages as well as in a more significantly redactional mode in the reenvisioning of the action, both polarizing the characterization of the positive and negative characters, idealizing the former and denigrating the latter in order possibly to approximate to hagiographical and romance models, while underscoring God’s omnipotent direction of affairs that never leaves the outcome in doubt. While the aspect of rendering the figurative in terms of the literal relates to translation’s function as explication, the application of intertextual interpretation undoubtedly speaks to the 75  Once more it is important to consider that in the present context, as the people of God, they have become ideals for emulation by the largely non-Jewish readers.

Writing and Rewriting Scripture in the Armenian Versions

267

three books’ increasing ‘canonization’ or reception into the scriptural ambience, which leads to the conformation of some of the individual aspects of prayers, for example, as well as more extensive aspects of the Esther narrative to discursive norms pervading that literary context in continuity with earlier strata of the book’s transmission history.76 I hope that in these ways the Armenian version will assist us in exploring the phenomenon of change in the transmission of the biblical text and in determining its parameters and modalities.

Bibliography Ačar˙ean, H., Anjnannuneri bar˙aran [Prosopographical Dictionary], 5 vols. (Beirut: Sevan Publishing, 1972). Anderson, B., Imagined Communities: Reflections on the Origin and Spread of Nationalism (London: Verso, 1991). Areshian, G.E., “Sasanian Imperialism and the Shaping of Armenian Identity (Interdisciplinary Verification and Ambivalence of empire-nation Relationship)”, in id. (ed.), Empires and Diversity: on the Crossroads of Archaeology, Anthropology, and History (Los Angeles: The Cotsen Institute of Archaeology Press, 2013) 185–203. Awetik‘ean, G., Nor bar˙girk‘ haykazean lezui [New Dictionary of the Armenian Language] 2 vols. (Venice: St. Lazar’s Press, 1836–1837). Berlin, A., “The Book of Esther and Ancient Storytelling”, JBL 120/1 (2001) 3–14. Clines, D.J.A., The Esther Scroll: The Story of the Story (JSOTSS 30; Sheffield: University of Sheffield, 1984). Coogan, M., A Brief Introduction to the Old Testament: The Hebrew Bible in its Context (Oxford: Oxford University Press, 2009). Cook, J., The Septuagint of Proverbs-Jewish and/or Hellenistic Proverbs? Concerning the Hellenistic Colouring of LXX Proverbs (VTSup 69; Leiden: Brill, 1997). Cook, J., “Proverbs”, in A. Pietersma/ B.G. Wright (ed.), A New English Translation of the Septuagint (Oxford: Oxford University Press, 2007) 621–47. Cowe, S.P., “Tendentious Translation and the Evangelical Imperative: Religious Polemic in the Early Armenian Church”, Revue des études arméniennes 22 (1990–91) 97–114. Cowe, S.P., “The Armenian Version of the Epistle of Jeremiah: Parent Text and Translation Technique”, in C.E. Cox (ed.), VII Congress of the International Organization for Septuagintal and Cognate Studies, Leuven 1989 (SCSS 31; Atlanta, GA: Scholars Press, 1991) 373–91. Cowe, S.P., “The Bible in Armenian”, in A.E. Matter/R. Marsden (ed.), New Cambridge History of the Bible 600–1450, vol. 2 (Cambridge: Cambridge University Press, 2012) 143–61.

76  On the issue of adapting narratives to a broader scriptural norm, see Clines, The Esther Scroll, 169–171.

268

S. Peter Cowe

Cowe, S.P., “The Armenian Version of the New Testament”, in B.D. Ehrman/ M.W. Holmes (ed.), The Text of the New Testament in Contemporary Research (2nd edn; NTTSD 42; Leiden: Brill, 2013) 253–92. Cowe, S.P., “Rhetoric, Theology, and Antiochene Exegesis in the Armenian Version of Lamentations”, in P.F. Fumagalli (ed.), Dies Academicus del “Classe di Studi sul Vicino Oriente dell’Accademia Ambrosiana” (Analecta Orientalia Ambrosiana IV; Roma: Bulzoni Editore, 2015) 143–65. Cox, C.E., “Introduction”, in Y. Zohrapean, Astuatsashunch‘ matean hin ew nor ktakarants‘ (a facsimile reproduction of the 1805 Venetian edition with an introduction by Claude Cox) (Delmar, NY: Caravan Books, 1984) v–xxvi. De Troyer, K., The End of the Alpha Text of Esther: Translation and Narrative Technique in MT 8:1–17, LXX 8:1–17, and AT 7:14–41 (SBL SCSS 48; Atlanta, GA: Society of Biblical Literature, 2000). De Waard, J. & Nida, E.A., Functional Equivalence in Bible Translating from one Language to Another (Nashville, TE: Thomas Nelson, 1986). Dorothy, C.V., The Books of Esther: Structure, Genre and Textual Integrity (JSOT SS 187; Sheffield: Sheffield Academic Press, 1997). Eisenstein, E.L., The Printing Revolution in Early Modern Europe (2nd edn; Cambridge: Cambridge University Press, 2005). Fountain, A.K., Literary and Empirical Readings of the Books of Esther (SBL 43; New York: Peter Lang, 2002). Fox, M.V., The Redaction of the Books of Esther (Atlanta: Scholars Press, 1991). Fox, R.L., “Literacy and Power in Early Christianity”, in A.K. Bowman/G. Woolf (ed.), Literacy and Power in the Ancient World (Cambridge: Cambridge University Press, 1996) 126–148. Funk, W.-P., “The Translation of the Bible into Coptic”, in J.C. Paget/J. Schaper (ed.), New Cambridge History of the Bible 600–1450, vol. 1 (Cambridge: Cambridge University Press, 2013) 536–46. Garsoïan, N.G., “The Emergence of Armenia” and “The Aršakuni Dynasty”, in R.G. Hovannisian (ed.), The Armenian People from Ancient to Modern Times, vol. 1 (New York: St. Martin’s Press, 1997) 37–62 and 63–94. Gellner, E., Nations and Nationalism (Ithaca, NY: Cornell University Press, 1983). Gerleman, G., “Studien zu Esther: Stoff—Struktur—Stil—Sinn”, BS 48 (1966) 1–48. Godel, R., An Introduction to the Study of Classical Armenian (Wiesbaden: Ludwig Reichert Verlag, 1975). Haines-Eitzen, K., “The Social History of Early Christian Scribes”, in B.D. Ehrman/ M.W. Holmes, (ed.), The Text of the New Testament in Contemporary Research (2nd edn; NTTSD 42; Leiden: Brill, 2013) 479–495. Hanhart, R., Iudith (SVTG VIII, 4; Göttingen: Vandenhoeck & Ruprecht, 1979). Hanhart, R., Esther (2nd edn; SVTG XVI, 3; Göttingen: Vandenhoeck & Ruprecht, 1983). Hanhart, R., Tobit (SVTG VIII, 5; Göttingen: Vandenhoeck & Ruprecht, 1983). Harris, W.V., Ancient Literacy (Cambridge: Harvard University Press, 1989).

Writing and Rewriting Scripture in the Armenian Versions

269

Hengel, M., The Septuagint as Christian Scripture: Its Prehistory and the Problem of its Canon (Edinburgh: T & T Clark, 2002). Hewsen, R.H., Armenia: A Historical Atlas (Chicago: University of Chicago Press, 2001). Jobes, K.H., The Alpha-Text of Esther: Its Character and Relationship to the Massoretic Text (JBL Dissertation Series 153; Atlanta, GA: Scholars Press, 1996). Jobes, K.H., “Esther”, in A. Pietersma/ B.G. Wright (ed.), A New English Translation of the Septuagint (Oxford: Oxford University Press, 2007) 424–40. Kölligan, D., “Armenisch-deutsches Text”, in O. Zwierlein, Die Urfassungen der Martyria Polycarpi et Pionii und das Corpus Polycarpianum, vol. 1 (Untersuchungen zur antiken Literatur und Geschichte 116; Berlin: De Gruyter, 2014) 46–66. Kosmin, P.J., The Land of the Elephant Kings: Space, Territory, and Ideology in the Seleucid Empire (Cambridge, MA: Harvard University Press, 2014). Kossmann, R., Die Esthernovelle vom Erzählen zur Erzählung: Studien zur Traditions- und Redaktionsgeschichte des Estherbuches (SVT LXXIX; Leiden: Brill, 2000). Lunt, H.G. &Taube, M., The Slavonic Book of Esther: Text, Lexicon, Linguistic Analysis, Problems of Translation (Harvard Series in Ukrainian Studies; Cambridge, MA: Harvard Ukrainian Research Institute, 1998). Manandyan, H., Tigrane II & Rome: nouveaux eclairissements à la lumière des sources originales (Lisbon: Imprensa Nacional, 1963). Martirosyan, H., Etymological Dictionary of the Armenian Inherited Lexicon (Leiden: Brill, 2010). Meillet, A., Altarmenisches Elementarbuch (Heidelberg: Carl Winters Universitäts-Buchhandlung, 1913). Metzger, B.M., The Text of the New Testament (2nd edn; Oxford: Clarendon Press, 1968). Metzger, B.M., The Early Versions of the New Testament (Oxford: Clarendon Press, 1977). Metzger, B.M., “Theories of the Translation Process”, Bibliotheca Sacra 150 (1993) 140–150. Milik, J.T., “Les modèles araméens du livre d’Esther dans la Grotte 4 de Qumrân”, Revue de Qumrân 59/15 (1992) 321–399. O’Connor, M.P., Hebrew Verse Structure (Winona Lake, IN: Eisenbrauns, 1980). Pivazyan, Ē., Vark‘Maštoc‘i [The Life of Maštoc‘] (Erevan: Erevan State University Publication, 1980). Ps. P‘awstos, Buzandaran Patmut‘iwnk‘ (The Epic Histories) also known as Patmut‘iwn Hayoc‘ (History of Armenia) Attributed to Pawstos Buzandac‘i (Delmar, NY: Caravan Books, 1984). Rapp, C., Holy Bishops in Late Antiquity: The Nature of Christian Leadership in an Age of Transition (Berkeley: University of California Press, 2005). Robbins, R.H., Global Problems and the Culture of Capitalism (Boston: Pearson/Allyn & Bacon, 1999). Royse, J.R., “Tendencies in the Transmission of the Text of the New Testament”, in B.D. Ehrman/ M.W. Holmes (ed.), The Text of the New Testament in Contemporary Research: Essays on the Status Quaestionis Second Edition (NTTSD 42; Leiden: Brill, 2013) 461–478. Schmitt, R. & Bailey, H.W., “Armenia and Iran. iv. Iranian Influences in Armenian”, in Encyclopaedia Iranica (www.iranicaonline.org/articles/armenia-iv).

270

S. Peter Cowe

Schmitz, B., “‘… am Ende ihres Weges Den zu schauen, an dem man stirbt, wenn man ihm naht’ (Rainer Maria Rilke): Die Rede von Gott in den Ester-Erzählungen”, in R. Egger-Wenzel/ K. Schöpflin/J.F. Diehl (ed.), Weisheit als Lebensgrundlage (DCLS 15; Berlin: Reiterer, 2013) 275–96. Schor, A.M., Theodoret’s People: Social Networks and Religious Conflict in Late Roman Syria (Berkeley: University of California Press, 2011). Smith, A.S., The Ethnic Origin of Nations (Oxford: Oxford University Press, 1986). Swain, S., Hellenism and Empire: Language, Classicism and Power in the Greek World, AD 50– 250 (Oxford: Oxford University Press, 1996). Touw, M., “Roses in the Middle Ages”, Economic Botany 36/1 (1982) 71–83. Ulrich, E., The Dead Sea Scrolls and the Developmental Composition of the Bible (SVT 169; Leiden: Brill, 2015). van den Broeck, R., “The Christian ‘School’ of Alexandria in the Second and Third Centuries” in J.W. Drijvers/A.A. MacDonald (ed.), Centres of Learning: Learning and Location in PreModern Europe and the Near East (Leiden: Brill, 1995) 39–47. Waard, J.D. & Nida, E.A., From One Language to Another: Functional Equivalence in Bible Translating (Nashville, TN: Nelson, 1986). Wahl, H.M., “‘Glaube ohne Gott?’ Zur Rede vom Gott Israels im hebräischen Buch Esther”, BZ 45 (2001) 37–54. Whitmarsh, T., The Second Sophistic (Oxford: Oxford University Press, 2005). Williams, P.J., “The Syriac Versions of the Bible”, in J.C. Paget/J. Schaper (ed.), New Cambridge History of the Bible 600–1450, vol. 1 (Cambridge: Cambridge University Press, 2013) 527–35. Wynn, K.H., “The Sociohistorical Contexts of the Recensions of Esther” (Ph.D. dissertation; Southern Baptist Theological Seminary, 1990). Zohrapean, Y., Astuatsashunch‘ matean hin ew nor ktakarants‘ (a facsimile reproduction of the 1805 Venetian edition with an introduction by Claude Cox) (Delmar, NY: Caravan Books, 1984).

Jean-Marie Auwers The Intermediate Version of the Book of Tobit in its Greek Dress The Book of Tobit combines a number of stories within one narrative: the story of Tobit who, despite his piety, goes blind; that of Sarra who successively loses seven husbands, even before being united with them; and that of Tobias, Tobit’s son, who undertakes a long journey in the company of the angel Raphael. In the course of that journey, he finds love (in the person of Sarra) and discovers the remedy which will allow him to put the evil demon who has killed Sarra’s first seven husbands to flight and restore his father’s sight. The book is attested in various textual forms, agreeing on the major story lines, but diverging on detail. In fact, the transmission of the Book of Tobit is unusually complicated.1

Semitic Original The Book was composed in a Semitic language (Hebrew or Aramaic). Most specialists today opt for an Aramaic original, but the question remains open.2 The Semitic original is lost: the four (or probably five) fragmentary Aramaic manuscripts of the Book of Tobit found at Qumran (4Q196–199 and 4Q196a) and the Hebrew manuscript (4Q200), do not attest the whole book (far from it).3 The medieval Aramaic and Hebrew forms of the Book of Tobit are translations from Greek or from the Latin Vulgate and differ considerably from the Qumran fragments.4 St. Jerome claimed to have translated into Latin an Aramaic model that 1  The most recent status quaestionis is given by L. Stuckenbruck/S. Weeks, “Tobit”, in J.K. Aitken (ed.), The T & T Clark Companion to the Septuagint (London: Bloomsbury T & T Clark, 2015) 237–60. See also J.A. Fitzmyer, Tobit (Berlin: Walter de Gruyter, 2003), 3–17 and M. Hallermayer, Text und Überlieferung des Buches Tobit (Berlin: Walter de Gruyter, 2008), 6–32. For one verse in particular, the complexity of the problem is clearly illustrated by C.M. Tuckett, “Tobit 12,8 and 2 Clement 16,4”, ETL 88 (2012) 129–44. 2  Stuckenbruck/Weeks, “Tobit”, 240 and G. Tonoli, L’originale del libro di Tobia. Studio filologico-linguistico (Madrid: CSIC, Instituto de filología, 2004). 3  J. Fitzmyer (ed.), “Tobit”, in M. Broshi et al. (ed.), Qumran Cave 4: XIV: Parabiblical texts, Part 2 (DJD 19; Oxford: Clarendon, 1995) 1–76; M. Hallermayer/T. Elgvin, “Schøyen Ms. 5234: Ein neues Tobit-Fragment vom Toten Meer”, RevQ 22 (2006) 451–60. This manuscript has been published as a fragment of 4Q196, but upon closer inspection, many letters differ and it may represent a fifth Aramaic manuscript, which perhaps does not come from Qumran and for which another, mysterious fragment exists (in private hands). There is also an additional Genizah fragment identified as Tobit; it is to be published by Moshe Lavee. I thank Eibert Tigchelaar for that piece of information (e-mail dated 15/06/2015). 4 Fitzmyer, Tobit, 11–14.

272

Jean-Marie Auwers

had been translated for him into Hebrew by a man “conversant in both languages” (Aramaic and Hebrew).5 In truth, it has been shown that Jerome did nothing more than rewrite an ancient Latin version according to his tastes and ideas. Jerome’s dependence with respect to a form of the Old Latin version is established for the following reasons: the style is not Jerome’s; the translation’s vocabulary is not what he habitually uses; for certain verses, Jerome’s translation coincides almost exactly with the Vetus Latina, even to the point of reproducing its corruptions. Here is an example.6 When Tobias, accompanied by the angel Raphael, arrives at his distant cousin Ragouel’s, they are warmly greeted. Ragouel kills a ram from his flock to welcome them and has it prepared. Before sitting down to eat, the guests perform their ablutions: “they bathe and wash”, says Sinaiticus (7:9: ἐλούσαντο καὶ ἐνίψαντο), “they wash themselves” (loti sunt), say certain Old Latin witnesses (VL 7, cf. 109: cum loti fuissent)—but loti sunt is rapidly corrupted into locuti sunt (“they talked”) in another part of the Old Latin tradition (VL 62).7 And it is that corruption (within the Latin tradition) which is to be found in Jerome’s translation: Postquam autem locuti sunt, praecepit Raguhel occidi arietem et parari convivium (“So after they talked, Ragouel ordered a ram to be slaughtered and a banquet to be prepared”). The Aramaic text on whose basis Jerome claims to have worked, if it existed, escapes every effort at reconstitution.

Greek I and Greek II In Greek, the Book of Tobit is attested in full (or almost in full) in two principal forms: a short recension (also called Greek I) and a longer one (or Greek II). The two recensions are linked. Either the short recension abridges the long one or the long recension develops the short one.8 Scholars who favour(ed) the short version’s priority argue(d) in this way: Greek I is more stable and less fluctuating among its related versions; its brevity is skillful, whereas the long recension is guilty of mere expansiveness; and certain geographical errors in Greek I, which 5 Jerome, Prologus Tobiae: “Quia vicina est Chaldeorum lingua sermoni hebraico, u ­ triusque linguae peritissimum loquacem repperiens, unius diei laborem arripui et quicquid ille mihi hebraicis verbis expressit, haec ego accito notario, sermonibus latinis exposui”, in R. Weber (ed.), Biblia sacra iuxta Vulgatam versionem (5th edn by R. Gryson; Stuttgart: Deutsche Bibelgesellschaft, 2007), 676. The main reference work on that topic is the book of V.T.M. Skemp, The Vulgate of Tobit Compared with Other Ancient Witnesses (SBLDS 180; Atlanta: SBL, 2000). 6  Cf. L. Rosso, “Un antica variante del libro di Tobit (Tob., VII, 9)”, RSO 50 (1976) 73–89; S. Gathercole, “Tobit in Spain: Some Preliminary Comments on the Relations Between the Old Latin Witnesses”, in M. Bredin (ed.), Studies in the Book of Tobit: A Multidisciplinary Approach (London: T & T Clark, 2006) 5–11. 7  Most manuscripts (VL 123 133 134 135 144 145 148 149 150) have laverunt. See the same corruption (lotus > locutus) in 2:5, 9 (VL 62 130). 8  Cf. J.D. Thomas, “The Greek Text of Tobit”, JBL 91 (1972) 463–71.

The Intermediate Version of the Book of Tobit in its Greek Dress

273

differ from some in Greek II, indicate that the short recension is earlier. But, particularly since J.D. Simpson’s article in 1912–13,9 most scholars share the view that the short Greek recension is a reworked form of the earlier Greek long recension, produced in an effort to improve the Greek phraseology and literary character of the Tobit story. The differences between Greek II and Greek I are most easily accounted for by a process of curtailment and the elimination of Semitisms that are found in Greek II. Greek I is seen to be a modification of the long recension because it reflects the general presuppositions and ideas, the historical conditions, the religious characteristics and theological developments of an age long subsequent to that in which the long recension was written.10 Discovery of the Qumran manuscripts from 1952 has provided a decisive argument in favour of the long recension’s priority: the Aramaic and Hebrew fragments preserved in the Qumran Scrolls normally agree with the long form of the book found in Greek II and in the Old Latin version,11 although in a few instances the Aramaic or the Hebrew agree with Greek I or give a text non-attested in Greek.12 Greek I is preserved by the vast majority of manuscripts (Vaticanus, Alexandrinus, Venetus and a host of minuscule manuscripts). On the contrary, Greek II is preserved in only two manuscripts: Sinaiticus contains the whole book, except for two lacunae, one in chapter 4 (v. 7–19), the other in chapter 13 (v. 6–10). The other manuscript is an eleventh-century one (Rahlfs 319) which contains the long recension for just three chapters (3:6–6:16). The first lacuna of Sinaiticus is thus compensated for by ms 319, but that is not the case for the second lacuna.

The Old Latin Version In the indirect tradition, the long recension is also attested by the ancient Latin version, no longer accessible to us in its original form today but in various recensions preserved in fifteen Old Latin manuscripts that have come down to us.13 The interest in the ancient Latin tradition for establishing the long text of Tobit is that much greater given that the direct tradition is poor and that Sinaiticus 9  J.D. Simpson, “The Chief Recensions of the Book of Tobit”, JTS 14 (1912–13) 516–30. 10 Fitzmyer, Tobit, 5–6. Cf. Stuckenbruck/Weeks, “Tobit”, 252–3. 11 Fitzmyer, Tobit, 9–10. 12  Hence the thesis of T. Niklas/C. Wagner, “Thesen zur textlichen Vielfalt im Tobitbuch”, JSJ 34 (2003) 141–59, on p. 151: “Die Tobit-Fragmente aus Qumran müssen nicht als Zeugen für die Priorität von S interpretiert werden. Vielmehr lassen sich Indizien aufzeigen, die auf eine freie und vielfältige Überlieferung des Tobit-Buches bereits in der semitischen Ursprache hindeuten”. 13  For a status quaestionis about the Vetus Latina of Tobit, see J.-M. Auwers, “La tradition vieille latine du livre de Tobie: Un état de la question”, in G.G. Xeravitz/J. Zsengellér (ed.), The Book of Tobit: Text, Tradition, Theology (JSJSup 98; Leiden: Brill, 2005) 1–21.

274

Jean-Marie Auwers

presents quite a number of defects which the Old Latin tradition permits to correct (e. g. in 11:4, κ[ύριο]ς is clearly a mistake we may correct in κύων after the Vetus Latina reading canis), and omissions, which it allows to re-establish.14 As already pointed out, Sinaiticus omits five verses of chapter 13 by haplography (v. 6–10), a lacuna which is not compensated for by the other Greek witness to the long text. Based on 4Q196, Joseph Fitzmyer has been able to reconstitute the Aramaic text of v. 6 entirely, and partially that of v. 7. As for vv. 8–10, there are only snippets in Aramaic.15 Those verses of the long recension are only completely conserved by the Vetus Latina. One cannot purely and simply identify Sinaiticus with the original long recension, the latter being reconstructed on the basis of two Greek witnesses (S and 319), but also other witnesses, principally the ancient Latin tradition, as well as the Aramaic and Hebrew fragments (although all these witnesses present irreconcilable differences),16 as well as Greek III.

Greek III A third version, called the “intermediate recension”, is indeed preserved by two Greek manuscripts from the fourteenth century (Rahlfs 106–107 = Greek III) for a little less than half the book (6:8b—12:22 or 13:2, cf. infra). In these two Ferrara manuscripts, as in the secondary manuscripts which contribute nothing towards establishing the text,17 the Book of Tobit is complete, but those manuscripts only attest a particular recension for a part of the Book; the rest of the text conforms to Greek I. This intermediate version is also partially (7:11—12:22) attested by the majority Syriac version.18 It was supposed that Greek III must originally have been complete and that, if Greek manuscripts and the Syriac version only conserve about one half, it was the result of an accident in the manuscript tradition. This has been confirmed by Anna Kharanauli and Natia Dundua, who

14  J.R. Busto Saiz, “Algunas aportaciones of the Vetus Latina para a a nueva edición crítica del libro of Tobit”, Sef 38 (1978)53–69; Auwers, “La tradition vieille latine“, 13–16. Errors and omissions in Sinaiticus are not necessarily attribuable to the copist of this manuscript, but may go back to the copied model; cf. S. Weeks, “Restoring the Greek Tobit”, JSJ 44 (2013) 1–15, on p. 2, n. 3. 15  Fitzmyer (ed.), “Tobit”, 26–8; Hallermayer, Text und Überlieferung des Buches Tobit, 73–9. 16  Cf. Niklas/Wagner, “Thesen zur textlichen Vielfalt im Tobitbuch”, 144–55. 17  R. Hanhart (ed.), Tobit (Septuaginta VIII,5; Göttingen: Vandenhoeck & Ruprecht, 1983), 10. 18  The only exception is ms 27 of Deir as-Suryan in Wadi Natrun, which preserves the Syro-Hexaplaric text (8f1 in Lebram’s edition; SyO in Hanhart’s edition). Cf. H. Lebram, “Die Peshitta zu Tobit 7,11–14,15”, ZAW 69 (1957) 185–211; H. Lebram (ed.), Tobit (Vetus Testamentum syriace 4/6; Leiden: Brill, 1972).

The Intermediate Version of the Book of Tobit in its Greek Dress

275

have recently shown that this recension is integrally conserved in Georgian in the Oshki Bible, edited by Ts. Kurtsikidze in 1970.19

P. Oxy. viii 1076 In 1911, A.S. Hunt published a parchment fragment in uncial writing from the sixth century (P. Oxy. viii 1076 = Rhalfs 910), which contains snippets of Tob 2:2–4, 8.20 As this little fragment sometimes corresponds to Greek I, sometimes to Greek II, and sometimes to neither of them, the editor had proposed seeing a witness to Greek III in it and concluded that this recension dated to the sixth century at the latest.21 Would the Oshki Bible confirm Hunt’s hypothesis or not? Let us take v. 8, for instance. Greek I and Greek II are rather close to one another: Greek I

Greek II

καὶ οἱ πλησίον ἐπεγέλων λέγοντες Οὐκέτι φοβεῖται

καὶ οἱ πλησίον μου κατεγέλων λέγοντες Οὐ φοβεῖται οὐκέτι; ἤδη γὰρ ἐπεζητήθη τοῦ φονευθῆναι περὶ τοῦ πράγματος τούτου· φονευθῆναι περὶ τοῦ πράγματος τούτου· καὶ ἀπέδρα, καὶ ἰδοὺ πάλιν θάπτει τοὺς καὶ ἀπέδρα, καὶ πάλιν ἰδοὺ θάπτει τοὺς νεκρούς. νεκρούς.

And the neighbors were laughing, saying: “He is no longer afraid to be murdered for this thing. And he ran away and behold he is burying the dead again”.

And my neigbhbors were laughing in scorn, saying: “Is he still unafraid? For he was already sought out to be murdered for this thing. And he ran away and again behold he is burying the dead”.

19  Printed edition by Tsiala Kurtsikidze, ӡveli aγtkmis ap’ok’ripebis kartuli versiebi (X–XVIII ss. xelnac’erta mixedvit) (2 vol.; Tbilisi: Mecniereba, 1970), 1. Electronic edition: http://titus. uni-frankfurt.de/texte/etcs/cauc/ageo/at/oskijer/oskij.htm accessed 10 March 2015. Cf. A. Kharanauli, “The Georgian Book of Tobit”, in E. Tov (ed.), Textual History of the Bible, vol. 1 (Leiden: Brill, forthcoming) and the article of N. Dundua in this volume. 20  A.S. Hunt, “A New Recension of Tobit ii”, in A.S. Hunt (ed.), The Oxyrhynchus Papyri, vol. VIII (London: Egypt Exploration Fund, 1911) 6–9. 21  See the discussion in R. Hanhart, Text und Textgeschichte des Buches Tobit (Abhandlungen der Akademie der Wissenschaften in Göttingen, Phil.-Hist. Klasse III. 130 = Mitteilungen des Septuaginta-Unternehmens 17; Göttingen: Vandenhoeck & Ruprecht, 1984), 44–5; C.J. Wagner, Polyglotte Tobit-Synopse. Griechisch—Lateinisch—Syrisch—Hebräisch—Aramäisch (Abhandlungen der Akademie der Wissenschaften in Göttingen, Phil.-Hist. Klasse III. 258 = Mitteilungen des Septuaginta-Unternehmens 28; Göttingen: Vandenhoeck & Ruprecht, 2003), xiv–xv; Niklas/Wagner, “Thesen zur textlichen Vielfalt im Tobitbuch”, 144–50; S. Weeks/S. Gathercole/ L. Stuckenbruck (ed.), The Book of Tobit: Texts from the Principal Ancient and Medieval Traditions (Berlin: Walter de Gruyter, 2004), 15–17; Weeks, “Restoring the Greek Tobit”, 7–8.

276

Jean-Marie Auwers

V. 8 is not complete in the parchment. We read only: ἐκινδύνευσεν ἀποθανεῖν καὶ ἀπέδρα, καὶ άπώλεσεν πάντα τὰ ὑπάρχο[ν]τα αὐτοῦ καὶ ἰδοὺ He was in mortal danger and (he) ran away and lost all his property and behold…

The words καὶ ἀπώλεσεν πάντα τὰ ὑπάρχοντα αὐτοῦ are taken from chapter 1 where we learn that in fact Tobit has braved Sennacherib’s interdict, that he has risked his life to bury his Jewish brothers, that he has had to flee and that all of his property has been seized (v. 19–20). The papyrus shows a different text than Greek I and Greek II. But the Georgian text of the Oshki Bible is also different: და მოძმენი მეკიცხევდეს მე და იტყოდეს: არღარა ეშინის ამას მოკლვისაგან, ამის საქმისათჳს ივლტოდა და კუალად მკუდართავე დაჰფლავს. And the neigbhbors reproached me, saying: “He is no longer afraid of being murdered; he had run away for this very thing and again he’s burying the dead”.22

We see that the Georgian does not present the same characteristics as the Greek fragment. It does not witness to the same Greek recension. Consequently, if the Oshki Bible attests the intermediate recension, we can no longer consider the date of the document (6th century) as the terminus ante quem of this recension.23

Where Does Greek III End? Robert Hanhart considers Greek III (and the Syriac version) ending with the conclusion of chapter 12 (v. 22).24 For Stuart Weeks, the Ferrara manuscripts stop attesting the intermediate recension in 13:2.25 The first two verses of chapter 13 in these two manuscripts coincide with neither Greek I, nor Greek II:

22  I would like to thank particularly Dr. Tamar Pataridze and Natia Dundua for the help they granted me in translating the Georgian version. 23  Tucket has recently tried to show that, for Tob 12:8, readings we find only in Greek III (where almsgiving is rated above prayer and fasting) are supported by the evidence of 2 Clement 16,4, which “may have implications for the date of the Greek III version of Tobit” (“Tobit 12,8 and 2 Clement 16,4”, quotation on p. 144). Weeks does not agree: “At best, we are dealing with an indirect reference to, or paraphrase of this much cited passage, which should not be taken as conclusive evidence for the early existence of the ‘Third’ recension” (“Some Neglected Texts of Tobit: The Third Greek Version”, in Bredin (ed.), Studies in the Book of Tobit, 12–42, on p. 14, n. 11). 24 Hanhart, Text und Textgeschichte des Buches Tobit, 14. 25  Weeks, “Some Neglected Texts of Tobit”, 15. See the discussion in Wagner, Polyglotte Tobit-Synopse, xxii and Weeks/Gathercole/Stuckenbruck, The Book of Tobit, 14.

The Intermediate Version of the Book of Tobit in its Greek Dress

Greek I 1 Καὶ Τωβὶτ ἔγραψε προσευχὴν εἰς ἀγαλλίασιν καὶ εἶπεν Εὐλογητὸς ὁ Θεὸς ὁ ζῶν εἰς τοὺς αἰῶνας καὶ ἡ βασιλεία αὐτοῦ,

Greek II

277

Rahlfs 106

1 Καὶ εἶπεν

1 Καὶ ἔγραψε Τωβὶτ τὴν προσευχὴν ταύτην εἰς ἀγαλλίασιν καὶ εἶπεν· Εὐλογητὸς ὁ θεὸς ὁ εὐλογητὸς ὁ θεὸς ὁ ζῶν ζῶν εἰς τὸν αἰῶνα εἰς τοὺς αἰῶνας, ὅτι εἰς καὶ ἡ βασιλεία αὐτοῦ, πάντας τοὺς αἰῶνας ἡ βασιλεία αὐτοῦ, 2 ὅτι αὐτὸς μαστιγοῖ 2 ὅτι αὐτὸς μαστιγοῖ 2 ὅτι αὐτὸς μαστιγοῖ καὶ ἐλεεῖ, κατάγει εἰς καὶ ἐλεᾷ, κατάγει ἕως καὶ ἐλεᾷ, κατάγει εἰς ᾅδην ᾅδην καὶ ἀνάγει, ᾅδου κατωτάτω τῆς γῆς, κάτω καὶ αὐτὸς ἀνάγει καὶ αὐτὸς ἀνάγει ἐκ τῆς ἐκ τῆς ἀπωλείας ἐν τῇ ἀπωλείας τῆς μεγάλης, μεγαλωσύνῃ αὐτοῦ, καὶ οὐκ ἔστιν ὃς ἐκφεύξεται καὶ οὐκ ἔστιν οὐδέν, ὃ καὶ οὐκ ἐστιν ὃς ἐκφεύξεται τὴν χεῖρα αὐτοῦ. ἐκφεύξεται τὴν χεῖρα αὐτοῦ. τὴν χεῖρα αὐτοῦ. 1. And Tobit wrote a prayer 1. And he said: 1. And Tobit wrote this of rejoicing and said: prayer of rejoicing and said: Blessed be God who lives Blessed be God who lives Blessed be God who lives forever—and his kingdom. forever—and his kingdom. forever for his kingdom is for all ages. 2. For he afflicts and he 2. For he afflicts and he 2. For he afflicts and he shows mercy, He brings shows mercy, He brings shows mercy, He brings down to Hades down to Hades in the low- down to Hades in the low; est part of the earth and brings back and brings back from the he brings back from the great destruction destruction in his greatness and there is no one who and there is nothing that and there is no one who will will escape his hand. will escape his hand. escape his hand.

Comparison with the Georgian may indicate another solution. Ms 106 1 Καὶ ἔγραψε Τωβὶτ τὴν προσευχὴν ταύτην εἰς ἀγαλλίασιν καὶ εἶπεν· εὐλογητὸς ὁ θεὸς ὁ ζῶν εἰς τοὺς αἰῶνας, ὅτι εἰς πάντας τοὺς αἰῶνας ἡ βασιλεία αὐτοῦ, And Tobit wrote this prayer of rejoicing and said:

Georgian და ტობი დაწერა ლოცვაჲ მათა მიმართ სიხარულად და თქუა: კურთხეულ არს ღმერთი, რომელი ცხოველ არს საუკუნეთა, და სუფევაჲ მისი ჰგიეს უკუნისამდე. And Tobit wrote the prayer of rejoicing for them and said:

278

Jean-Marie Auwers

Ms 106 Blessed be God who lives forever for his kingdom is for all ages. 2. ὅτι αὐτὸς μαστιγοῖ καὶ ἐλεᾷ, κατάγει εἰς ᾅδην κάτω καὶ αὐτὸς ἀνάγει ἐκ τῆς ἀπωλείας ἐν τῇ μεγαλωσύνῃ αὐτοῦ, καὶ οὐκ ἐστιν ὃς ἐκφεύξεται τὴν χεῖρα αὐτοῦ. For he afflicts and he shows mercy, He brings down to Hades in the low; he brings back from the destruction in his greatness and there is no one who will escape his hand.

Georgian Blessed be God who lives forever – and his kingdom is forever. რამეთუ მან ტანჯის და შეიწყალის, შთაიყვანის ჯოჯოხეთა და აღმოიყვანის, და არავინ არს, რომელი განერეს ჴელსა მისსა. For he afflicts and he shows mercy, He brings down to Hades and brings back and there is no one who will escape his hand.

We see that the most characteristic element of ms 106 for v. 1 (ὅτι εἰς πάντας τοὺς αἰῶνας) is attested in the Georgian.26 By contrast, the most characteristic elements of v. 2 (κάτω and ἐν τῇ μεγαλωσύνῃ αὐτοῦ) are absent from the Georgian. The conclusion could be that the Ferrara manuscripts stop attesting the intermediate recension in 13:1.

Mixed or Intermediate Recension? It would be preferable to refer to an intermediate (or medium-length) recension rather than to a mixed (or genetically intermediary) recension. Greek III is intermediate between the other two recensions as to overall length (but there are passages where Greek III is briefer than Greek I), but it would be wrong to imagine that it mixes the other two textual forms or that it is a sort of eclectic edition, “a mediating redaction”27 or “a compromise between the other two Greek versions”.28 From this point of view, the option chosen by A.E. Brooke and N. McLeen to provide the readings of Greek III in Greek I’s apparatus (as if the Ferrara manuscripts were Greek witnesses to Greek I), or by R. Hanhart to provide them in Greek II’s apparatus (as if those manuscripts were Greek witnesses to Greek II),29 is misleading. Greek III deserves to be edited for its own sake, as C. Wagner and 26  Greek III probably attests the oldest reading here (cf. infra). 27  D.C. Simpson, “The Book of Tobit”, in R.H. Charles (ed.), Apocrypha and Pseudepigrapha of the Old Testament (2 vol.; Oxford: Clarendon, 1913) 1.174–241, quotation p. 176. R. Hanhart considers Greek III “als sekundäre Bearbeitung der Textformen GI und GII” (Text und Textgeschichte des Buches Tobit, 48). 28 Fitzmyer, Tobit, 5. 29 Hanhart, Text und Textgeschichte des Buches Tobit, 14; cf. Hanhart (ed.), Tobit, 33.

The Intermediate Version of the Book of Tobit in its Greek Dress

279

S. Weeks and his colleagues have done,30 following ms 10631 (as before them O.F. Fritzsche and F. Vigouroux).32

Greek III, Greek II, and Vetus Latina Admittedly, Greek III is closely linked to Greek II. 11,8 provides a good example. The angel Raphael explains to Tobias how he can heal his blind father with fish gall: Greek I

Greek II

Greek III

σὺ οὖν ἔγχρισον τὴν χολὴν εἰς τοὺς ὀφθαλμοὺς αὐτοῦ, καὶ δηχθεὶς διατρίψει καὶ ἀποβαλεῖται τὰ λευκώματα καὶ ὄψεταί σε.

ἔμπλασον τὴν χολὴν τοῦ ἰχθύος εἰς τοὺς ὀφθαλμοὺς αὐτοῦ, καὶ ἀποστύψει τὸ φάρμακον καὶ ἀπολεπίσει τὰ λευκώματα ἀπὸ τῶν ὀφθαλμῶν αὐτοῦ, καὶ ἀναβλέψει ὁ πατήρ σου καὶ ὄψεται τὸ φῶς. Apply the gall of the fish to his eyes, and the medicine will contract the white films and peel them off from his eyes, and your father will look up and see the light.

ἔμπασον τὴν χολὴν εἰς τοὺς ὀφθαλμους αὐτοῦ, καὶ ἀποστύψει τὸ φάρμακον καὶ ἀπολεπίσει τὰ λευκώματα ἀπὸ τῶν ὀφθαλμῶν αὐτοῦ, καὶ ἀναβλέψει ὁ πατήρ σου καὶ ὄψεται τὸ φῶς.

You thus anoint his eyes with the gall, and when it smarts, he will rub them and will shed the white films and he will see you.

Apply the gall to his eyes, and the medicine will contract the white films and peel them off from his eyes, and your father will look up and see the light.

Greek III is identical to Greek II, apart from the absence of τοῦ ἰχθύος (of the fish). That’s why, from the point of view of textual criticism, Greek III is important when it supports the Vetus Latina against Sinaiticus. There are good reasons for thinking that the agreement of Greek III and the Vetus Latina provides access to the long text, corrupted in the Sinaiticus. The examples are rather numerous, so I will provide only some examples.

30 Wagner, Polyglotte Tobit-Synopse; Weeks/Gathercole/Stuckenbruck, The Book of Tobit. 31  On the respective merit of manuscripts 106 and 107, see Hanhart (ed.), Tobit, 34. 32 O.F. Fritzsche, Libri apocryphi Veteris Testamenti graece (Leipzig: Brockhaus, 1871) 128–165; F. Vigouroux, La Sainte Bible polyglotte (8 vol.; Paris: Roger et Chernivicz, 1900–9), 3.488–516 (according to Paris. Suppl. Gr. 609 and the readings of ms 106 mentioned in the apparatus criticus of Holmes-Parsons, cf. p. 465).

280

Jean-Marie Auwers

At 7:13, whereas Sinaiticus does not specify that the act of marriage between Tobit and Sarra was sealed, the Vetus Latina does so (et signavit).33 It is also found in Greek III (καὶ ἐσφραγίσαντο). That confirms that the detail is original in the long recension. At 8:7, Tobias says a prayer before consuming his union with Sarra. More explicitly than Greek I and Sinaiticus, Greek III expresses a wish for fertility (καὶ δὸς ἡμῖν, κύριε, τέκνα καὶ εὐλογίαν) which corresponds to the Vetus Latina (et da nobis filios in benedictione).34 At 8:20, the words καὶ ὤμοσεν αὐτῷ, absent from Greek I and Greek II, have a correspondent in the ancient Latin version (et iuravit illi). At 6:11, the reading καλή, absent from Sinaiticus and Greek I, is confirmed by Greek III, ms 319, the Old Latin version (speciosam)35 and 4Q197 (‫)שפירה‬. At 6:12, the reading ἀγαπᾷ, absent from Sinaiticus and Greek I, is confirmed by Greek III, ms 319, the Old Latin version (diligit)36 and 4Q197 (‫)רחם‬. At 13:1, the words ὅτι εἰς πάντας τοὺς αἰῶνας, proper to Greek III within the Greek tradition as has been said above, have an equivalent in the Vetus Latina and in a fragment of Qumran.37 The Greek III probably attests the oldest reading here.38 In 9:3–4, the situation is different. Tobias explains to Raphael why, despite the insistence of his father-in-law Ragouel (who wants Tobias to stay with him), he is in a hurry to return to his father Tobit, who will be worried that his son has not returned. Greek I

Greek II

διότι ὠμόμοκε Ραγουὴλ μὴ σὺ γὰρ γινώσκεις ὅτι ἔσται ἐξελθεῖν με, ἀριθμῶν ὁ πατὴρ τὰς καὶ ὁ πατήρ μου ἀριθμεῖ τὰς ἡμέρας, καὶ ἐὰν χρονίσω ἡμέρας, καὶ ἐὰν χρονίσω ἡμέραν μίαν, λυπήσω μέγα, ὀδυνηθήσεται λίαν αὐτὸν λίαν· καὶ θεωρεῖς τί ὤμοσεν Ραγουηλ, καὶ οὐ δύναμαι παραβῆναι τὸν ὅρκον αὐτοῦ.

Greek III σὺ γὰρ γινώσκεις ὅτι ἔσται ὁ πατήρ μου ἀριθμῶν, καὶ ἐὰν χρονίσω μίαν ἡμέραν, λυπήσω αὐτόν. καì Ῥαγουὴλ ὤμοσε μεῖναί με δεκατέσσαρας ἡμέρας παρ´ αὐτῷ, καὶ οὐ δυνήσομαι ἀθετῆσαι τὸν ὅρκον αὐτοῦ.

33  et signavit VL 7 62 109 123 130 133 134 135 144 145 148 149 150. 34  et da nobis filios in benedictione VL 123 133 134 144 145 148 149 150; et da nobis in benedictione filios 135; om. 7 62; lac. 130; offerentes tibi hostias laudis et salvationis nostrae 109. 35  speciosam VL 7 62 109 123 133 134 135 144 145 148 149 150; speciosa forma 143 (= καλὴ τῷ εἴδει, G III) ; bonam 130. 36  diligit 7 62 123 130 133 134 135 144 145 148 149 150; om. 143. 37  Vetus Latina: quoniam (om. 143) in (om. 7 134) omnia (om. 135) saecula VL 7 109 23 130 134 135 143 144 145 148 149 150; om. 62 per homoeotel.; 4Q200 (‫)אשר לכול העולמים‬. 38  Cf. Wagner, Polyglotte Tobit-Synopse, xxiii.

The Intermediate Version of the Book of Tobit in its Greek Dress

Greek I … because Ragouel has sworn that I should not leave. But my father is counting the days, and if I delay long, he will be greatly distressed.

Greek II

281

Greek III

For you know that my father is counting the days, and if I delay one day, I will grieve him very much. And you see how Ragouel has sworn,

For you know that my father is counting, and if I delay one day, I will grieve him. And Ragouel has sworn that I should remain fourteen days with him and and I cannot go against his I cannot violate his oath. oath.

It is quite clear that Greek III follows Greek II rather than Greek I. The omissions of τὰς ἡμέρας, λίαν and θεωρεῖς τί are insignificant. At the end of the verse, Greek III employs another verb, ἀθετῆσαι (“violate”), instead of παραβῆναι (“go against”). The most remarkable element is the addition, in Greek III, of the words μεῖναί με δεκατέσσαρας ἡμέρας παρ´ αὐτῷ, without corresponding in Greek I, Greek II and the Old Latin version. Greek III recalls the contents of the oath made in 8:20 and so facilitates the reader’s understanding.

Greek III and Greek I Greek III sometimes agrees with Greek I against Greek II, but this involves minor details, and these agreements may have occurred purely by chance or as a result of the desire—shared by the redactors of Greek I and Greek III—to rewrite the book in a more readable, coherent and succinct way.39 For example, in 6:13 (Raphael speaks to Tobias about Ragouel): Greek I οὐ μὴ δῷ αὐτὴν ἀνδρὶ ἑτέρῳ κατὰ τὸν νόμον Μωυσῆ

Greek II οὐ μὴ δυνηθῇ Ραγουὴλ κωλῦσαι αὐτὴν ἀπὸ σοῦ ἢ ἐγγυᾶσθαι ἑτέρῷ, ὀφειλήσειν θάνατον κατὰ τὴν κρίσιν τῆς βίβλου Μωυσέως

Greek III οὐ δυνάται Ραγουὴλ ἀντειπεῖν σοι, ὅτι σὺ ἄρχεις αὐτὴν παρὰ πάντα τὰ ἔθνη, ὅτι γινώσκει ὅτι ἐὰν δώσει αὐτὴν ἀνδρὶ ἑτέρῳ, ὀφειλήσει θάνατον κατὰ τὸν νόμον Μωυσῆ

39  Weeks is right in writing: “The relatively few remaining agreements [of Greek III] with the Short Greek are mostly very minor, and some may be coincidental. They provide little or no basis, therefore, for the old idea that the Third Greek was written as a bridge between the Short and Long” (“Some Neglected Texts of Tobit”, 18). See also Stuckenbruck/Weeks, “Tobit”, 257.

282

Jean-Marie Auwers

Greek I He cannot give her to another man, according to the law of Moses.

Greek II

Greek III

Ragouel cannot keep her from you or promise her to another; he would be liable to death according to the decree of the book of Moses.

Ragouel cannot contradict you for you have authority over her more than anybody, for he knows that if he give her to another man, he would be liable to death according to the law of Moses.

Of course the elements in italics (“give her to another man”, and “the law of Moses”) are common to Greek I and Greek III, but are hardly noticed in the midst of words proper to Greek III or common to Greek II and Greek III.

The Vocabulary of Greek III Greek III favours a certain more literary vocabulary. It contains 44 words absent from the other two textual forms.40 In 7:1, παρὰ τὴν θύραν τῆς αὐλῆς (Greek II) becomes παρὰ τὴν αὐλαίαν θύραν. Greek III sometimes employs a refined vocabulary. In 7:7, Greek III does not say, as does Greek I, that Tobit has lost his eyes (ἀπώλεσεν τοὺς ὀφθαλμούς) or, as does Sinaiticus, that he has gone blind (ἐτυφλώθη); Greek III says that Tobit has become disabled (ἐπηρώθη). In employing the term δικαίωμα (6:12 bis; 6:18; 8:7), Greek III insists on Tobias’s right to marry Sarra and inherit her father’s possessions. Greek III favours the term κοιτών in talking about the nuptial chamber (7:16; 8:1, 12, 13), whereas the other two versions use the term ταμιεῖον (Greek II: 7:15, 16; 8:1, 4; Greek I: 7:15) which only appears once in Greek III (7,15).41 The demon who has killed Sarra’s seven previous husbands and whom Tobias must disarm is very present in the other two versions, where he is usually designated by the term δαιμόνιον (3:8, 17; 6:8, 14, 16, 18; 8:3). He is less often mentioned in Greek III, which refers to an “impure spirit” (πνεῦμα ἀκάθαρτον: 6:14, 15; 8:3) and only once to an “impure demon” (ἀκάθαρτον δαιμόνιον: 8:7).

Variations in the Scenario The characters are not presented in the same way. In Greek III the angel Raphael does not remind Tobias of the obligation Tobit had imposed on him to take a wife 40 J.-M. Auwers/B. Kindt, “Tobie et Esther dans le Thesaurus Duplex Septuaginta”, in W. Kraus/S. Kreuzer/M. Meiser/M. Sigismund (ed.), Die Septuaginta—Text, Wirkung, Rezeption (WUNT 325; Tübingen, Mohr Siebeck, 2014) 242–54. 41  We also find νυμφών in the three recensions in 6:14, 17.

The Intermediate Version of the Book of Tobit in its Greek Dress

283

from his house or at least from the Jewish people (6:16). The endogamic marriage obligation plays less of a role in Greek III: Greek I Οὐ μέμνησαι τῶν λόγων, ὧν ἐνετείλατό σοι ὁ πατήρ σου, ὑπὲρ τοῦ λαβεῖν σε γυναῖκα ἐκ τοῦ γένους σου; Do you not remember the words with which your father commanded you to take a wife from among your people?

Greek II Οὐ μέμνησαι τὰς ἐντολὰς τοῦ πατρός σου, ὅτι ἐνετείλατό σοι λαβεῖν γυναῖκα ἐκ τοῦ οἴκου τοῦ πατρός σου; Do you not remember your father’s commandments, that he commanded you to take a wife from your father’s house?

Greek III Οὐ μέμνησαι πάσας τὰς ἐντολὰς τοῦ πατρός σου;

Do you not remember all your father’s commandments?

In 7:6, the three recensions agree in saying that Ragouel, upon learning that Tobias is the son of a cousin that he has not seen in years, “leaps to embrace him, weeping” with emotion. The attitude of the two women of the family, Edna and her daughter Sarra, is described in v. 8. Their reaction differs depending on the recension. According to Greek I and Greek II, they weep over the hardships of the unfortunate Tobit, gone blind. According to Greek III, they embrace Tobias, which creates a better transition to what follows immediately (“and they welcomed them warmly”)42 and helps us to understand Tobias’s haste in encouraging his companion Azarias to begin the marriage negotiations. In 8:17, the Greek III (ms 106) shows the full magnitude of divine benevolence in adding that it also extends to the parents: Greek I εὐλογητὸς εἶ ὅτι ἠλέησας δύο μονογενεῖς.

Greek II

Greek III (Rahlfs 106)

καὶ εὐλογητὸς εἶ ὅτι ἠλέησας δύο μονογενεῖς.

εὐλογητὸς εἶ, κύριε, ὅτι ἠλέησας δύο μονογενεῖς καὶ τοὺς γονεῖς αὐτῶν. Praise to you, for you have And praise to you, for you Praise to you, Lord, for you taken pity on two only have taken pity on two have taken pity on two only children. only children. children and their parents.

42  In the other two recensions, the transition is abrupt: “Edna his wife and Sarra his daughter also wept and they welcomed them warmly” (Greek I, cf. Greek II).

284

Jean-Marie Auwers

In 11:11, Tobias approaches his father to heal him in following the angel’s instructions. He says to him, according to Greek I and Greek II: “Take courage, father” (Θάρσει, πάτερ). He doesn’t tell him what he’s going to do. Greek III corresponds more to what the reader expects: “and (Tobit) said: what are you doing, my son? And Tobias says: It’s a remedy to heal you, father” (Καὶ εἶπε Τί τοῦτο ἐποίησας τέκνον; καὶ εἶπε Τωβίας Φάρμακον ἰάσεώς ἐστι, πάτερ). In the same chapter 11, the Greek III displaces v. 5 and integrates it into v. 9. For that reason, the account’s chronology is a bit different. In Greek I and Greek II, on approaching Nineveh, where Tobias’s parents live, Raphael proposes to Tobias that they should run before the rest of the caravan carrying the fish gall, which they do, followed by a dog (v. 2–4). The narrative then transports the reader to Hanna, who sits looking intently down the road for her boy. When she sees her son from afar, she tells her husband that the boy is arriving with his companion (v. 5–6). The narrative returns to Raphael who explains how to use the fish gall to Tobias (v. 7–8). It once again assumes the viewpoint of Hanna who runs towards her son (v. 9), then assuming Tobit’s viewpoint who, seeking to come out, bangs into the door (v. 10). In Greek III, the account is less sophisticated. The camera is more static. It is first focused on Raphael, the dog and Tobias, to whom the angel explains how he should use the fish gall to restore his father’s sight (v. 2–8). The camera then moves to Hanna who, upon seeing the dog run ahead, runs to tell her husband that their son is back and, immediately afterwards, rushes towards Tobias to fall on his neck (v. 9). Then it is the father, Tobit, who comes out of the house on hearing his son’s voice (v. 10). Clearly the writer to whom we owe Greek III sought a more simple presentation of the facts. The demon is much less present in Greek III. This is particularly clear in the scene where Tobias, having entered the nuptial chamber, renders the demon harmless (8:1–3). We need only parallel Greek III with Greek II. Greek II Καὶ ὅτε συνετέλεσαν τὸ φαγεῖν καὶ πιεῖν, ἠθέλησαν κοιμηθῆναι. καὶ ἀπήγαγον τὸν νεανίσκον καὶ εἰσήγαγον αὐτὸν εἰς τὸ ταμίειον. καὶ ἐμνήσθη Τωβιας τῶν λόγων Ραφαηλ καὶ ἔλαβεν τὸ ἧπαρ τοῦ ἰχθύος καὶ τὴν καρδίαν ἐκ τοῦ βαλλαντίου, οὗ εἶχεν, καὶ ἐπέθηκεν ἐπὶ τὴν τέφραν τοῦ θυμιάματος. καὶ ἡ ὀσμὴ τοῦ ἰχθύος ἐκώλυσεν, καὶ ἀπέδραμεν τὸ δαιμόνιον ἄνω εἰς τὰ μέρη Αἰγύπτου, καὶ βαδίσας Ραφαηλ συνεπόδισεν αὐτὸν ἐκεῖ καὶ ἐπέδησεν παραχρῆμα.

Greek III Καὶ ἐγένετο ὅτε συνετέλεσαν τοῦ εὐφρανθῆναι, εἰσήγαγον Τωβίαν εἰς τὸν κοιτῶνα. καὶ ἐμνημόνευσε Τωβίας τῶν λόγων Ῥαφαὴλ, καὶ ἔλαβε τὴν καρδίαν τοῦ ἰχθύος καὶ ἐξέβαλε τὸ ἀκάθαρτον πνεῦμα καὶ ἕφυγεν εἰς τὰ ἄνω μέρη τῆς Αἰγύπτου, καὶ Ῥαφαὴλ ἔδησεν αὐτὸ ἐκεῖ.

The Intermediate Version of the Book of Tobit in its Greek Dress

285

Greek II

Greek III

When they had finished eating and drinking, they wanted to go to sleep; so they led the young man out and led him into the room. Then Tobias remembered the words of Raphael, and he took the fish’s liver and heart out of the bag where he had them and put them on the embers of the incense. And the odor of the fish became a hindrance, and the demon fled into the upper parts of Egypt. Raphael went and bound him there hand and foot and tied him up at once.

And it came, when they had finished being delighted, they led Tobias into the room.

Then Tobias remembered the words of Raphael, and he took the fish’s heart.

And the impure spirit cast himself outside and fled into the upper parts of Egypt. And Raphael bound him there.

The demon is less present and less courageous too, for Tobias need only bring out the fish heart to drive the demon off. The quasi-magical ritual of putting to flight disappears in Greek III. At 8:10–18, according to Greek II, once Tobias and Sarra have entered nuptial chamber, Ragouel, fearing the demon’s intervention, calls upon his servants to dig a tomb with him for he who may soon be dead and orders his wife Edna to send a maidservant into the chamber to see if Tobias is still alive. The maidservant announces that he is alive. Whereupon, Ragouel pronounces a benediction formula and orders his servants to go and fill in the tomb. A difference exists in the Greek I narrative. Ragouel digs the tomb himself (no doubt for added discretion), but once the danger is past, he orders his servants to go and fill in the tomb (as in Greek II). In Greek III, Ragouel orders his servants to dig a tomb in order that, should Tobias die, he might bury the hapless young man before sunrise (“I’ll bury him during the night, and nobody will know it, or I would be an object of shame and mockery”). In reading Greek III, we almost get the impression that Ragouel is more worried about his own reputation than the life of his son-in-law. We see here that the redactor of Greek III sometimes introduces material on his own. As in the other two versions, the maidservant makes her report to Ragouel and Edna (for that matter, in Sinaiticus, the couple have opened the chamber door for the maidservant together and handed her a torch), in Greek III the maidservant makes her report to Edna alone, who then tells her husband. Finally, the servants come on their own and Ragouel asks them to fill in the tomb before daybreak. This is not really the same scenario.43 43  Cf. Weeks, “Some Neglected Texts of Tobit”, 20.

286

Jean-Marie Auwers

At the end of this study, I must say that I share the conclusion of Stuckenbruck and Weeks: “Greek III tells a better story than either Greek I or Greek II, and it is hard to believe that that was not its creator’s intention.”44 As we have seen, he has rewritten the book in a more readable, coherent and succinct way. The personality of the main characters (including the demon) is quite different. It is also why the intermediate recension of Tobias deserves to be studied in its own right. The same may be said about Jerome’s translation; he too rewrote the book according to his own criteria—and Jerome’s criteria were different from those of the scribe to whom we owe Greek III. But for both of them, the Book of Tobias was a story to be told, rather than a text to be transmitted.

Bibliography Auwers, J.-M., “La tradition vieille latine du livre de Tobie. Un état de la question”, in G.G. Xeravitz/J. Zsengellér (ed.), The Book of Tobit: Text, Tradition, Theology. Papers of the First International Conference on the Deuterocanonical Books, Pápa, Hungary, 20–21 May 2004 (JSJSup 98; Leiden: Brill, 2005) 1–21. Auwers, J.-M. & Kindt, B., “Tobie et Esther dans le Thesaurus Duplex Septuaginta”, in W. Kraus/S. Kreuzer/M. Meiser/M. Sigismund (ed.), Die Septuaginta—Text, Wirkung, Rezeption. 4. Internationale Fachtagung veranstaltet von Septuaginta Deutsch (LXX.D), Wuppertal 19.–22. Juli 2012 (WUNT 325; Tübingen: Mohr Siebeck, 2014) 242–54. Busto Saiz, J.R., “Algunas aportaciones of the Vetus Latina para a a nueva edición crítica del libro of Tobit”, Sef 38 (1978) 53–69. Fitzmyer, J.A., Tobit (Berlin: Walter de Gruyter, 2003). Fitzmyer, J. (ed.), “Tobit”, in M. Broshi et al. (ed.), Qumran Cave 4: XIV: Parabiblical texts, Part 2 (DJD 19; Oxford: Clarendon, 1995) 1–79. Fritzsche, O.F. (ed.), Libri apocryphi Veteris Testamenti graece (Leipzig: Brockhaus, 1871). Gathercole, S., “Tobit in Spain: Some Preliminary Comments on the Relations Between the Old Latin Witnesses”, in M. Bredin (ed.), Studies in the Book of Tobit: A Multidisciplinary Approach (Library of Second Temple Studies 55; London: T & T Clark, 2006) 5–11. Hallermayer, M. & Elgvin, T., “Schøyen Ms. 5234: Ein neues Tobit-Fragment vom Toten Meer”, RevQ 22 (2006) 451–60. Hallermayer, M., Text und Überlieferung des Buches Tobit (Deuterocanonical and Cognate Literature Studies 3; Berlin: Walter de Gruyter, 2008). Hanhart, R. (ed.), Tobit (Septuaginta VIII,5; Göttingen: Vandenhoeck & Ruprecht, 1983). Hanhart, R., Text und Textgeschichte des Buches Tobit (Abhandlungen der Akademie der Wissenschaften in Göttingen, Phil.-Hist. Klasse III. 130 = Mitteilungen des Septuaginta-Unternehmens 17; Göttingen: Vandenhoeck & Ruprecht, 1984).

44  Stuckenbruck/Weeks, “Tobit”, 257.

The Intermediate Version of the Book of Tobit in its Greek Dress

287

Hunt, A.S. , “A New Recension of Tobit ii”, in A.S. Hunt (ed.), The Oxyrhynchus Papyri, vol. VIII (London: Egypt Exploration Fund, 1911) 6–9. Kharanauli, A., “The Georgian Book of Tobit”, in E. Tov (ed.), Textual History of the Bible, vol. 1 (Leiden: Brill, forthcoming). Kurtsikidze, T. (ed.), ӡveli aγtkmis ap’ok’ripebis kartuli versiebi (X–XVIII ss. xelnac’erta mixedvit) (2 vol.; Tbilisi: Mecniereba, 1970). Lebram, H., “Die Peshitta zu Tobit 7,11–14,15”, ZAW 69 (1957) 185–211. Lebram, H. (ed.), Tobit (Vetus Testamentum syriace 4/6; Leiden: Brill, 1972). Niklas, T. & Wagner, C., “Thesen zur textlichen Vielfalt im Tobitbuch”, JSJ 34 (2003) 141–59. Rosso, L., “Un antica variante del libro di Tobit (Tob., VII, 9)”, RSO 50 (1976) 73–89. Simpson, J.D., “The Chief Recensions of the Book of Tobit”, JTS 14 (1912–13) 516–30. Simpson, D.C., “The Book of Tobit”, in R.H. Charles (ed.), Apocrypha and Pseudepigrapha of the Old Testament (2 vol.; Oxford: Clarendon, 1913) 174–241. Skemp, V.T.M., The Vulgate of Tobit Compared with Other Ancient Witnesses (SBLDS 180; Atlanta: SBL, 2000). Stuckenbruck, L. & Weeks, S., “Tobit”, in J.K. Aitken (ed.), The T & T Clark Companion to the Septuagint (London: Bloomsbury T & T Clark, 2015) 237–60. Thomas, J.D., “The Greek Text of Tobit”, JBL 91 (1972) 463–71. Tonoli, G., L’originale del libro di Tobia: Studio filologico-linguistico (Madrid: CSIC, Instituto de filología, 2004). Tuckett, C.M., “Tobit 12,8 and 2 Clement 16,4”, ETL 88 (2012) 129–44. Vigouroux, F. (ed.), La Sainte Bible polyglotte (8 vol.; Paris: Roger et Chernivicz, 1900–9). Wagner, C.J. (ed.), Polyglotte Tobit-Synopse. Griechisch—Lateinisch—Syrisch—Hebräisch—­ Aramäisch (Abhandlungen der Akademie der Wissenschaften in Göttingen, Phil.-Hist. Klasse III. 258 = Mitteilungen des Septuaginta-Unternehmens 28; Göttingen: Vandenhoeck & Ruprecht, 2003). Weeks, S. & Gathercole, S. & Stuckenbruck, L. (ed.), The Book of Tobit: Texts from the Principal Ancient and Medieval Traditions (Fontes et Subsidia ad Bibliam pertinentes 3; Berlin: Walter de Gruyter, 2004). Weeks, S., “Some Neglected Texts of Tobit: The Third Greek Version”, in Bredin (ed.), Studies in the Book of Tobit: A Multidisciplinary Approach (Library of Second Temple Studies 55; London: T & T Clark, 2006) 12–42. Weeks, S., “Restoring the Greek Tobit”, JSJ 44 (2013) 1–15.

Natia Dundua What Can the Georgian Translation of the Book of Tobit Tell about GIII? Introduction The books of the Apocrypha, as known, often are presented with diverse textual forms. Consequently, their translations mostly reflect the colorful nature of the Vorlage as well. The Book of Tobit is a good example of this. The paper deals with the problems of the textual history of Tobit and attempts to find a place for the Georgian translation of the book in this complicated textual history.

Greek Sources of the Book of Tobit Three textual forms of Tobit in Greek have survived, two of which are preserved completely (GI and GII), and one partially (GIII [6:9–12:22]). The first textual form is attested in most of the Greek manuscripts, and on it are based the Ethiopic, Coptic-Sahidic, Syro-Hexaplaric and Armenian translations. The second form— in S “Codex Sinaiticus”, 910 (22 πολ]λα–8 και ιδου: cf. Text und Textgeschichte des Buches Tobit =TGT), 319 (36–616 τουτου) and the Old Latin (VL). The third is found in group d (69–12fin) and the Syriac (Sy) (711 καὶ νῦν–12fin).1 Fragments of four Aramaic manuscripts (4Q196–199) and one Hebrew manuscript (4Q200) of Tobit were among the findings of Qumran Cave 4 in 1952, which mostly support GII and VL.2 In his edition R. Hanhart presents the following textual groups of the Greek manuscripts of Tobit:3 GI : B A V (990: 1214–19 with omissions) a b c d (11–68 131–1415) 46 55 108 126 311ˊ 318 319 (11–35 616 ὅτι–1415) 392 488 535 583 Sy (11–711 νύκτα Sy, 711 ἀλλά–1222 SyO, 131–18 SyO Sy, 141–15 SyO) Sa Aeth Arm Recensions: a = 71-74-76-130-236-314-370-402-542-762 74ˊ = 74-76 b = 64-98-243-248-381-728-731 1  R. Hanhart, Septuaginta, Tobit (Göttingen: Vandenhoeck & Ruprecht, 1983). 2  J. Fitzmyer, Tobit (Berlin: Walter de Gruyter, 2003). 3  R. Hanhart, Septuaginta, 32.

290

Natia Dundua

c = 58-249-670 249ˊ = 249-670 d = 106-107 Codices mixti: 46 55 108 126 311 318 319 392 488 535 583 746 311ˊ = 311-746. GII: S (910: 22 πολ]λα– 8 και ιδου: cf. TGT) 319 (36–616 τουτου) La. GIII : d (69–12fin) Sy (711 καὶ νῦν–1222).

GI represents the so-called short version of Tobit, GII the long version, while GIII has an intermediate length between the two textual forms. These versions are genetically linked but they do not represent just re-writings of the story. The relationship among the three textual forms has been established by R. Hanhart as follows: ...deren teilweise Übereinstimmungen zwar dermaßen eindeutig sind, daß gegenseitige Abhängigkeit mit Sicherheit angenommen werden muß, deren Unterschiede aber so tiefgreifend sind, daß ihr Verhältnis zueinander nicht als Rezension sondern als selbständige Textform bestimmt werden muß.4

The issues of the priority among these textual forms have long been discussed by various scholars and on the question of which is the earlier one, critics are still divided.5 For my study, as will be shown below, most interesting is GIII. Overall, there are two main suggestions among scholars concerning GIII: 1. that it is a mediating form between GI and GII, and 2. that it is derived from GII and has secondary readings from GI. Robert Hanhart writes about GIII: ...der Text dieser Zeugen seinem Charakter nach als ein Mittelglied zwischen Textform und Rezension bestimmt werden muss. Ursprünglicher Text könnte bei diesen Zeugen darum nicht in dem Sinn vorliegen, daß ihre Textform gegenüber den beiden anderen die primäre wäre, sondern höchstens in dem Sinn, daß sie als 4  R. Hanhart, Text und Textgeschichte des Buches Tobit (Göttingen: Vandenhoeck & ­Ruprecht, 1984), 11. 5  On these problems one might consult the following literature: J.R. Busto Saiz, El doble texto griego del libro de Tobit y su insercion en la historia de G (Madrid: Universidad Complutense, 1975); J. Fitzmyer, Tobit: Commentaries on Early Jewish Literature (Berlin: Walter de Gruyter, 2003); R. Hanhart, Septuaginta, Tobit (Vetus Testamentum Graecum Auctoritate Academiae Scientiarum Gottingensis editum vol. VIII, 5 Tobit; Göttingen: Vandenhoeck & Ruprecht, 1983); idem., Text und Textgeschichte; N. Fernández Marcos, “The Origins of the Septuagint: The Double Text of Tobit”, in: The Septuagint in Context, Introduction to the Greek Versions of the Bible (Leiden: Brill, 2000), etc.

What Can the Georgian Translation of the Book of Tobit Tell about GIII

291

Bearbeitung der beiden anderen Textformen zuweilen das Ursprüngliche bewahrt oder wiederhergestellt hätte.6

Concerning GIII Stuart Weeks in his article “Some Neglected Texts of Tobit: the Third Greek Version” notes that it should be “an attempt to improve the Long Greek version stylistically”.7 Editors of Tobit publish the text of GIII in different ways. For example, BrookeMcLean-Thackeray8 give it in the apparatus criticus of GI and Hanhart in the apparatus criticus of GII,9 while Wagner and Weeks present the text separately.10

Georgian Sources of the Book of Tobit There are two translations of Tobit in Georgian: I. The oldest one (=OGeo) is preserved in the following manuscripts: 1st O, the Old Testament Codex, called the Oshki Bible (Ath-1, Athos, Iviron, 978 c.e.; 529r–531v);11 2nd the fragment m (no. 7125, Yerevan, The Mesrop Mashtots Institute of Ancient Manuscripts, twelfth–thirteenth centuries); 3rd  I (A-570, Tbilisi, NCM, 1460 c.e.; 95v–103r); 4th  F (A-646, Tbilisi, NCM, fifteenth–sixteenth centuries; 404v–426v); 5th  D (H-885, Tbilisi, NCM, seventeenth century; 39v–47r);

6 Hanhart Text und Textgeschichte, 14. 7  S. Weeks, “Some Neglected Texts of Tobit: the Third Greek Version”, in M. Bredin (ed.), Studies in the Book of Tobit, A Multidisciplinary Approach (London: T & T Clark, 2006) 12–42, on pp.19, 23. 8  A.E. Brooke/N. McLean/H.St.J. Thackeray, The Old Testament in Greek: VIII-I Esther, Judith, Tobit (Cambridge: Cambridge University Press, 1940). 9  R. Hanhart, Septuaginta, Tobit (Göttingen: Vandenhoeck & Ruprecht, 1983). 10  C.J. Wagner, Polyglotte Tobit-Synopse (MSU 28; Göttingen: Abhandlungen der Akademie der Wissenschaften in Göttingen, 2003); S. Weeks/L. Stuckenbruck/S. Gathercole (ed.), The Book of Tobit: Texts from the Principle Ancient and Medieval Traditions, with Synopsis, Concordances, and Annotated texts in Aramaic, Hebrew, Greek, Latin and Syriac (Berlin: Walter de Gruyter, 2004). 11  In the Oshki Bible the Book of Tobit is written by a different hand from the rest of the text (the hand of the copyist seems to be not later than eleventh century, cf. Kurtsikidze, 1973) and is placed at the end of the first volume. The text of Tobit here is not complete, containing only text from 5:1 to 12:1. It is noteworthy that the book of Judith, which precedes Tob in the Oshki Bible, also lacks the last part (16:13–25). It should be investigated whether 1) there was a full text of Tob in the Oshki manuscript, which was lost later (together with Jdt), or 2) it has never been there, and this incomplete text was added later in order to fill the collection.

292

Natia Dundua

6th  S, the Bible Codex, called the Mcxeta Bible (A-51, Tbilisi, NCM, the turn of the seventeenth-eighteenth centuries; 304v–309r), which represents a revision of the OGeo according to the Latin Vulgate.12 II. A later translation of Tobit is supposed to have been made from the Old Church Slavonic, and it is included in the Bakar Bible (=B, 1743 c.e.; 417–24), publ ished in Moscow.13

The Old Georgian translation of the Book of Tobit shows varied relations to the three textual types of the Greek Tobit in the different parts of the text. Where scholars separate GIII (6:9–12:22), OGeo follows this textual type unambiguously. In those parts where the features of GIII are not so obvious and, therefore, the editors of the text present it variously, in OGeo the readings of the two textual types are intertwined in such a complex manner that it is impossible to say which of them the Georgian is following. Meanwhile, there are Georgian variants, which have no parallels in the extant sources. However, it should be noted that by character they do not seem to be changes due to the translation technique and/or the nature of the Georgian language, but more likely reflect the Greek original. Present research, which is based on the method of word for word collation of the Georgian Tobit to all sources presented in the Göttingen Editio Major of the Septuagint, has shown that the Georgian version follows the GIII, which varies between GI and GII and has its own variants as well. The GIII textual form of Tobit, as already mentioned, has reached us just by several sources in Greek and Syriac, but in both languages the texts are not complete (d [6:9–12:fin], Sy [7:11 καὶ νῦν—12:fin]), whereas the old Georgian translation is preserved completely and is typologically very similar to the GIII along the entire text. This supports the assumption that a complete GIII textual type should have existed from which Tobit was translated into Georgian. What kind of text could the complete GIII have been? How could such a text be referred to? To what extent can the Georgian translation clarify the textual variety? And how can OGeo be helpful in this confusing situation? In order to answer these questions, I will analyze the old Georgian translation of Tobit piecemeal (1. 6:9–12:22; 2. 1:1–6:8; 3. 13:1–14:15) in accordance with the other sources.

12  Ts. Kurtsikidze, ʒveli aγtkmis ap’ok’ripuli c’ignebis kartuli versiebi [Georgian Versions of the Old Testament Apocryphal Books] Book II (Tbilisi: Mecniereba, 1973). 13 Kurtsikidze, ʒveli aγtkmis ap’ok’ripuli, 166–70.

What Can the Georgian Translation of the Book of Tobit Tell about GIII

293

6:9–12:22 GIII First of all, let us see the part where all three textual forms are preserved in Greek and other daughter versions (i.e. 6:9–12:22). The Georgian translation obviously follows the GIII textual form in this part. For clarity, I will present several instances, where the character and differences among the three textual types and agreement of OGeo with GIII are most remarkable, for example:14 11:115 GI

καὶ

ἐπορεύετο

×

×

μέχρις

οὗ

×

GII

»

×

×

×

×

ὡς

×

GIII

»

ἐπορεύθησαν

τὴν ὀδὸν

αὐτῶν

×

×

καὶ

OGeo

და

წარვიდეს = GIII

გზასა

მათსა

×

×

და

14  Here and below only several instances are chosen and presented from many typologically similar examples. 15  Abbreviations and Sigla: × do not have » the same as in the upper row > omission + addition ⌒ omitted text in cases of homoioteleuton ] marker of lemma cf. confer fin end om omission p page pr addition in front of lemma GI first Greek textual form GII second Greek textual form GIII third Greek textual form d 106–107 La Old Latin translation Sy Syriac translation VL Vetus Latina Arm Armenian translation OGeo Old Georgian translation Jdt Book of Judith Tobit Book of Tobit

294

Natia Dundua

GI

ἐγγίσαι

αὐτοὺς ×

GII

ἤγγισαν

×

GIII

ῆλθον

OGeo

εἰς Νινευη

×

×

×

εἰς Κασεριν

ἥ

ἐστιν

κατέναντι Νινευη

×

εἰς Καισάρειαν

×

×

×

×

მოვიდეს = GIII

×

კესარიადმდე ×

×

×

×

GI

ὅτι

×

ἐμαστίγωσας

καὶ

ἠλέησάς

με

GII

»

αὐτὸς

ἐμαστίγωσέν

×

×

»

GIII

»

×

×

×

×

×

OGeo

რამეთუ

×

×

×

×

×

GI

×

×

ἰδοὺ

βλέπω

Τωβιαν

τὸν υἱόν μου

GII

καὶ

×

»

»

»

»

»

GIII

×

×

»

»

×

»

»

OGeo

×

ვიხილე ძე ჩემი [(I) saw my son]16

აჰა ესერა

ვხედავ

×

ძესა

ჩემსა

GI

×

×

×

×

×

GII

×

×

×

×

×

GIII

εὐλογητὸν

τὸ ὄνομα

κυριου

τοῦ θαυμαστὰ

ποιοῦντος

OGeo

კურთხეულ არს

სახელი

უფლისაჲ

საკჳრველთ

მოქმედისაჲ

11:14

16  English translations of the Georgian variants are my own.

What Can the Georgian Translation of the Book of Tobit Tell about GIII

295

11:15 GI

καὶ

εἰσῆλθεν

×

×

×

×

ὁ υἱὸς

αὐτοῦ

×

GII

»

»

Τωβιας

×

×

×

×

×

×

GIII

»

»

Τωβὶτ

καὶ

Ἄννα

ἠ γυνὴ ×

αὐτοῦ

εἰς τὸν οῖκον

OGeo

და

შევიდეს

ტობი =GIII

და

ანა

ცოლი ×

მისი

სახიდ

GI

×

×

χαίρων

×

×

GII

×

×

»

καὶ

εὐλογῶν

GIII

καὶ

αὐτοὶ

ἔχαιρον

»

ηὐλόγουν

OGeo

და

217 მათ

1 უხაროდა

და

აკურთხევდეს

GI

×

×

×

×

×

×

GII

τὸν θεὸν

×

×

×

ἐν ὅλῳ

τῷ αὐτοῦ στόματι

GIII

»

περὶ πάντων

τῶν αὐτοῖς γεγενημένων

×

×

×

OGeo

ღმერთსა

ყოვლისა მისთჳს

რავდენი ქმნა

მათთჳს ×

×

×

×

The difference between these three textual forms is so obvious that it is not necessary to present further arguments that the Georgian follows GIII.

17  The numbers here and elsewhere mark the transposition.

296

Natia Dundua

d Group and Sy Now let us clear up concretely which group of manuscripts is followed by the Georgian translation. The word for word collation has shown that OGeo is very close to the d group, for example: 6:1318 GI

καὶ

×

×

×

×

×

GII

»

εἶπεν

δεδικαίωταί

σοι

λαβεῖν

αὐτήν

d

×

×

×

×

×

×

OGeo

×

×

×

×

×

×

GI

×

νῦν

ἄκουσόν

μου

×

καὶ

GII

καὶ

×

»

»

ἄδελφε

»

d

»

νῦν

»

»

×

»

OGeo

და

აწ

ისმინე

ჩემი

×

და

GI

λαλήσω

τῷ πατρὶ

αὐτῆς

×

×

×

×

GII

»

»

×

περὶ τοῦ κορασίου

τὴν νύκτα

ταύτην

ἵνα

d

»

×

περὶ αὐτῆς

×

×

×

καὶ

OGeo

ვიტყოდი

×

მისთჳს =d

×

×

×

და =d

18  In this part Sy is not preserved.

What Can the Georgian Translation of the Book of Tobit Tell about GIII

297

GI

×

×

×

×

καὶ

ὅταν

ὑποστρέψωμεν

GII

λημψόμεθά

σοι

αὐτὴν

νύμφην

»

»

ἐπιστρέψωμεν

d

ἁρμωσόμεθά

»

»

σήμερον

»

»

»

OGeo

მოგითხოო =d

×

2იგი

1დღეს =d

და

ოდეს

მოვიქცეთ

GI

ἐκ Ῥάγων

ποιήσομεν

τὸν γάμον

×

×

διότι

GII

»

»

»

αὐτῆς

×

καὶ

d

»

»

»

»

οἶδα

ὅτι

OGeo

ჰრაგავით + მაშინ [then]

ვყოთ

ქორწილი

მისი

ვიცი

რამეთუ =d

GI

ἐπίσταμαι

Ραγουηλ

×

×

×

×

GII

»

×

×

×

×

×

d

οὐ δύναται

Ραγουηλ

ἀντειπεῖν

σοι

ὅτι

σὺ

OGeo

ვერ ჴელეწიფოს =d

ჰრაგუელს

სიტყჳსგებად შენდა რამეთუ შენდა + არს [is]

GI

×

×

×

×

×

×

ὅτι

GII

×

×

×

×

×

×

»

d

ἄρχεις

αὐτῆς

παρὰ πάντα

τὰ ἔθνη

ὅτι

γινώσκει

»

OGeo

თჳსობაჲ მისი

უფროჲს ყოველთა

კაცთა

რამეთუ იცის

×

298

Natia Dundua

GI

οὐ μὴ

δῷ

×

×

αὐτὴν

ἀνδρὶ

GII

»

δυνηθῇ

Ραγουηλ

κωλῦσαι

»

×

d

ἐάν

δώσει

×

×

»

ἀνδρὶ

OGeo

უკუეთუ =d

მისცეს =d

×

×

იგი

2ქმარსა

GI

×

×

ἑτέρῳ

GII

ἀπὸ σοῦ

ἢ ἐγγυᾶσθαι

»

d

×

×

»

OGeo

×

×

1სხუასა

GI

×

×

×

κατὰ τὸν νόμον

×

GII

ὀφειλήσειν

×

θάνατον

κατὰ τὴν κρίσιν

τῆς βίβλου

d

ὀφειλεί

σοι

»

κατὰ τὸν νόμον

×

OGeo

თანამდებ + არს [is]

×

სიკუდილისა სამართლისაებრ

GI

Μωυσῆ

×

ἢ ὀφειλέσει

θάνατον

ὅτι

τὴν κληρονομίαν

GII

Μωυσέως

διὰ τὸ γινώσκειν

×

×

»

2 κληρονομία

d

»

×

×

×

×

×

OGeo

მოსესისა

×

×

×

×

×

წიგნისა

What Can the Georgian Translation of the Book of Tobit Tell about GIII

299

GI

σοὶ

καθήκει

λαβεῖν

×

ἢ πάντα

GII

1»

»

»

τὴν θυγατέρα αὐτοῦ

παρὰ πάντα

d

×

×

×

×

×

×

OGeo

×

×

×

×

×

×

GI

ἄνθρωπον

×

×

×

×

×

GII

»

καὶ

νῦν

ἄκουσόν

μου

ἄδελφε

d

×

×

×

×

×

×

OGeo

×

×

×

×

×

×

GI

×

×

×

×

×

×

×

GII

καὶ

λαλήσομεν

περὶ τοῦ κορασίου

τὴν νύκτα

ταύτην

καὶ

μνηστευσόμεθά

d

×

×

×

×

×

×

×

OGeo

×

×

×

×

×

×

×

GI

×

×

×

×

×

GII

σοι

αὐτήν

καὶ

ὅταν

ἐπιστρέψωμεν

d

×

×

×

×

×

OGeo

×

×

×

×

×

×

300

Natia Dundua

GI

×

GII

×

×

×

×

×

ἐκ Ῥάγων λημψόμεθα

αὐτὴν

καὶ

×

ἀπάξομεν

d

×

×

×

»

λαβόντες

2ἐπάξωμεν

OGeo

×

×

×

და

მოვიყვანოთ

2 მივიყვანოთ

GI

×

×

×

×

GII

αὐτὴν

μεθ᾽ ἡμῶν

εἰς τὸν οἶκόν

σου

d

1»

μεθ᾽ ἑαυτῶν

πρὸς τὸν πατερα

»

OGeo

1იგი + და [and] ×

მამისა =d

შენისა

A similar picture is found in 7:11–12:22, where GIII together with d is also offered by Sy. Here again Georgian mainly follows d, for example: 8:1 GII, Sy

Καὶ

×

ὃτε

συνετέλεσαν

×

d

»

ἐγένετο

ὅταν

»

τοῦ εὐφρανθῆναι

OGeo

და

იყო,

ვითარცა

აღასრულეს

განცხრომაჲ იგი სიხარულისაჲ19

GII, Sy

τὸ φαγεῖν

καὶ

πιεῖν,

ἠθέλησαν

κοιμηθῆναι

d

×

×

×

×

×

OGeo

×

×

×

×

×

19  Eng.: merriment of the joy.

What Can the Georgian Translation of the Book of Tobit Tell about GIII

301

8:19 GII, Sy

εῖπεν συντελεῖν αὐτους, καὶ

ἤρξαντο

παρασκευάζειν

×

d

×

×

×

×

ἔταξεν

ἑτοιμάζειν

αὐτούς

OGeo

×

×

×

×

ბრძანა =d მომზადებაჲ =d მათი

However, in some cases OGeo deviates from d and agrees with La, Sy and even GI, for example: 7:15 αὐτήν d] virginem nostram LaX ასული ჩუენი OGeo; cf. filiam tuam Sy│ 8:2 ἐκ τοῦ βαλλαντίου, οῦ εῖχεν d] > Sy OGeo = GI│ 8:18 γενέσθαι d] + ne quis sciat Sy რაჲთა არავინ აგრძნეს OGeo│ 9:2 τέσσαρας οἰκέτας καὶ δύο καμήλους d] duos camelos et quattuor viros Sy cf.:520 ორნი აქლემნი და ოთხნი მონანი OGeo│ 10:12 ἐν παραθήκῃ d] > LaG Sy OGeo│ 11:9 ἀποθανοῦμαι d] pr non dolet mihi si et add quoniam ecce vidi te Sy არღარა ვწუხდე მე რამეთუ ვიხილე პირი შენი OGeo│ 11:18 παρεγένοντο d] + illuc Sy მისა OGeo.

106 The d group includes manuscripts 106 and 107, and OGeo follows 106 more frequently than 107 (the fact that together with the readings of 106 Sy is also often presented speaks to the popularity of these variants), for example: 7:2 καί] + εγενετο οταν εισηλθοσαν 106 ვითარცა შევიდეს სახიდ OGeo│ 8:6 τὸν ἀδάμ] pr ανθρωπον 106 კაცი OGeo│ 8:13 ἀπέστειλαν τὴν παιδίσκην] απεστειλεν εδνα τ. π. 106 Sy მიავლინა ედნა მჴევალი OGeo, cf. απηλθεν η παιδισκη (una ex ancillis pro η π. LaR) 107 LaR│ 8:17 μονογενεῖς GII] + και τους γονεις αυτων 106 Sy და მშობელნი მათნი OGeo│ 9:6 καὶ εἰσῆλθονGII] > 106 Sy OGeo│ 10:6 μή GII] μη θης επι την καρδιαν σου κακον 106 Sy ნურას დაიდებ გულსა შენსა ბოროტსა OGeo│ 10:10 πάντων τῶν ὑπαρχόντων αὐτῷ GII] παντα των υπ. αυτω το ημισυ 106 Sy ყოველთაგან მონაგებთა თჳსთა ზოგი OGeo│

20  This variant in Sy is due to the parallel place, whereas the Georgian follows the Greek Vorlage.

302

Natia Dundua

10:12 ὕπαγε πρὸς τὸν πενθερόν σου GII] θυγατερ τιμα τον πενθερον σου και την πενθεραν σου 106 La Sy ასულო, პატივ-ეც მამამთილსა შენსა და დედამთილსა შენსა OGeo│ 10:13 μή GII] pr ουτως 106 რაჲთა OGeo│ 10:14 εἶπεν GII] προσηυξατο λεγων 106 ილოცვიდა და თქუა OGeo, etc.

What is notable, OGeo does not share the secondary readings of d ‒ 125 and 4421 (for example, 7:11 εῖπεν αὐτῶ Ῥαγουήλ] om αὐτῶ d La OGeo = GI; ειπεν αζαριας ταυτα τω ραγουηλ και ειπεν 125; ελαλησε ταυτα αζαριας προς ραγουηλ και απεκριθη ραγουηλ 44), which increases the importance of the Georgian translation.

OGeo contra d On the other hand, OGeo does not contain some variants of d, for example: 6:13 τὴν κρίσιν βίβλου სამართლისაებრ წიგნისა OGeo] τον νομον d cf. GI│ 6:14,15 δαιμόνιον ეშმაკმან OGeo] πνευμα ακαθαρτον d│ 6:16 τὰς ἐντολάς მცნებაჲ OGeo] pr πασας d│ 6:15 αὐτῶν მათსა OGeo] > d.

It should especially be noted that OGeo does not attest the omissions shared by d and 319 La, for instance: 6:11 τῷ παιδαρίῳ ყრმასა მას OGeo] > d 319 LaWX, cf. 𝔙.

Also in 6:16 OGeo preserves the text which is omitted because of homoioteleuton by d, 319, LaGRJX: 6:16 σου, ὃτι ἐνετείλατό σοι λαβεῖν γυναῖκα ἐκ τοῦ οἳκου τοῦ πατρός σου შენთაჲსა რამეთუ გრქუა შენ მოყვანებაჲ ცოლისაჲ სახლისაგან მამისა შენისა OGeo] σου 1˚⌒2˚ d 319 LaGRJX.

At the same time, some Georgian variants, deviated from d, agree with other sources, among them to 31922 and La, which according to R. Hanhart contain GII in this part of the text, for example: 21  For the correspondence of mss 44 and 125 with d, see Hanhart, Septuaginta, Tobit, 31. 22  It should be noted, that OGeo reveals agreements with 319 (and La) in 3:6–6:9, where this manuscript is preserved and which Hanhart gives in the App. of GII, for example: 3:8 καθάπερ ἀποδεδειγμένον ἐστὶν GII] ος εστιν αποδεδειγμενον 319 ვითარცა არს წესი OGeo│ 5:6 πολλάκις ἐγὼ ἐγενόμην ἐκεῖ καὶ ἐμπειρῶ GII] πολλα εγω επισταμαι και εμπειρος 319 La ფრიად ვიცი და მეცნიერ ვარ OGeo│

What Can the Georgian Translation of the Book of Tobit Tell about GIII

303

6:14 λεγόντων αὐτῶν GII იტყოდეს ვინმე OGeo quosdam dicentes La; λεγόντων 319] > d│ 6:15 αὐτήν 319 მისა OGeo] > d, etc.

At the end of the chapter 12 the portion of GI grows in OGeo, and the Georgian shows many agreements with it. In 12:16 the Georgian differs from d and agrees mostly with GI (cf. 12:16 ὅτι ἐφοβήθησαν GI რამეთუ შეეშინა OGeo] καὶ ἐφοβήθησαν GII; > d Sy). In 12:20–22 also the Georgian translation bears the GI text (cf. d and Sy, which reflect GI from 13:1): 12:20 GI

καὶ

νῦν

×

×

×

×

ἐξομολογεῖσθε

τῷ θεῷ

GII

»

»

εὐλογεῖτε

ἐπὶ τῆς γῆς

κύριον

καὶ

»

»

GIII

»

»

ἀναστῆτε

ἐκ τῆς γῆς

×

×

×

×

OGeo

და

აწ

×

×

×

×

აუარებდით

ღმერთსა,

GI

διότι

×

ἀναβαίνω πρὸς τὸν ἀποστείλαντά με

GII

ἰδοὺ

ἐγὼ

»

»

»

GIII

×

×

×

×

×

OGeo

რამეთუ

×

აღვალ

მომავლინებელისა

ჩემისა + უფლისა. [to the Lord]

5:14 οἱ ἀδελφοί σου ἄνθρωποι GII] οι ανθρωποι ουτοι οι αδελφοι ημων 319 კაცნი ესე ძმანი ჩუენნი OGeo│ 5:20 παρὰ τοῦ κυρίου τοῦτο ἱκανὸν ἡμῖν GII] παρ αυτου κυριε 319 domino sic vivamus LaW უფლისა მიერ GeoIDF, ღმრთისა მიერ GeoO, etc.

304

Natia Dundua

GI

καὶ

γράψατε

πάντα

×

τὰ συντελεσθέντα

×

GII

×

»

»

ταῦτα

τὰ συμβάντα

ὑμῖν

GIII

καὶ

»

2»

1»

×

×

OGeo

და

დაწერეთ

2ყოველი

3ესე,

4რაჲცა შეგემთხჳა

5თქუენ. 

GI

εἰς βιβλίον

×

×

×

GII

×

καὶ

×

×

GIII

εἰς βιβλίον

»

τὸν θεον

εὐλογεῖτε

OGeo

1წიგნსა

×

×

×

GI

×

×

×

×

×

×

GII

×

×

ἀνέβη

×

×

×

GIII

ἐγὼ

γὰρ

ἀναβαίνω

πρὸς τὸν ἀποστείλαντά

με

θεον

OGeo

×

×

×

×

×

×

12:21 GI

καὶ

ἀνέστησαν

καὶ

οὐκέτι

×

εἶδον

αὐτόν

×

GII

»

»

»

»

ἠδύναντο

ἰδεῖν

»

×

GIII

»

»

»

οὐκ

×

εἶδοσαν

»

οὐκέτι

OGeo

და

აღ-რაჲდგეს,

×

არღარა

×

იხილეს იგი. 

×

What Can the Georgian Translation of the Book of Tobit Tell about GIII

305

12:22 GI

καὶ

ἐξωμολογοῦντο ×

GII

»

ηὐλόγουν

καὶ ὕμνουν

τὸν καὶ ἐξωμολογοῦντο αὐτῷ θεὸν

GIII

»

»

×

×

»

»

»

×

OGeo

და

აღუარებდეს +იგინი [they]

×

×

×

×

×

×

GI

τὰ ἔργα

×

τὰ μεγάλα

καὶ

θαυμαστὰ

τοῦ θεοῦ

καὶ

GII

ἐπὶ τὰ ἔργα

αὐτοῦ

»

×

×

×

×

GIII

»

»

×

×

×

×

×

OGeo

საქმეთა მათთა დიდდიდთა

და

საკჳრველებათა

მისთა,

და

GI

×

ὡς

ὤφθη

αὐτοῖς

ὁ ἄγγελος

κυρίου

GII

ταῦτα

»

»

»

ἄγγελος

θεοῦ

GIII

×

ἄ

ἐποίησεν

»

×

×

OGeo

×

ვითარ-იგი

ეჩუენა

მათ

ანგელოზი ღმრთისაჲ. =GII

×

×

×

×

×

OGeo Sonderlesarten Most significant are those readings (e.g., transpositions and additions) of Georgian manuscripts, which presuppose changes in the Greek Vorlage, but have not survived in the extant Greek sources, for example, the following categories of readings: 1) 7:1 αὐτούς GI d; αὐτόν GII] + და მანცა მოიკითხნა იგინი [and he also greeted them] OGeo—in OGeo the phrase of greeting is extended;

306

Natia Dundua

2) the sense is defined more precisely in OGeo: 11:10 προσέκοπτεν] + აქა და იქი [here and there] OGeo│ 11:10 Τωβίτ] + პირი თჳსი [his mouth] OGeo; 3) synonyms are presented in OGeo: 8:4 δεηθῶμεν ვევედრნეთ OGeo] pr ვილოცოთ და [(let us) pray and] OGeo│ 9:4 χρονίσω დავაყოვნოთ OGeo] + და დავაკლოთ [and lack] OGeo │ 11:13 εἶδε აღიხილნა OGeo] + და იხილა [and saw] OGeo; 4) in OGeo the biblical style is presented although in Greek there is a simple form: 10:7 με] პირი ჩემი [my mouth] OGeo; 5) in OGeo the narrative is extended to make the context more evident: 10:7 μου] + ღუაწლსა შინა არიან [they are into worry] OGeo, etc.

In order to see the relations between OGeo and GI and GII in the first and the last parts of the text let us analyze them separately via examples.

1:1–6:8 In the first part (1:1–6:8) OGeo varies between the two textual types. For instance, in 1:1–2:12 it contains GI but, on the contrary, in 5:1–10 attests GII. In some cases it compiles GI and GII, for example: 1:5 GI

καὶ

πᾶσαι

αἱ φυλαὶ

ἔθυον

τῇ Βααλ

GII

×

πάντες

οἱ ἀδελφοί μου ×

×

×

OGeo

რამეთუ ყოველნი

ტომნი =GI

×

×

GI

τῇ δαμάλει

καὶ

ὁ οἶκος

Νεφθαλιμ

τοῦ πατρός

μου

GII

×

»

»

»

»

»

OGeo

×

და

სახლი ნეფთალემისი, მამისა

×

×

αἱ συναποστᾶσαι

რომელნი თანა-განდგეს

ჩემისაჲ,

What Can the Georgian Translation of the Book of Tobit Tell about GIII

307

GI

×

×

×

×

×

GII

ἐθυσίαζον

ἐκεῖνοι τῷ μόσχῳ

ὃν

ἐποίησεν

OGeo

უზორვიდეს ×

3რომელი

4ქმნა

GI

×

×

×

×

GII

Ιεροβεαμ

ὁ βασιλεὺς

Ισραηλ

ἐν Δαν

OGeo

5იორობოამ,

6მეფემან

7ისრაელისამან,

1დანს შინა

GI

×

×

×

GII

ἐπὶ πάντων

ὀρέων

τῆς Γαλιλαίας

OGeo

ზედა ყოველთა pr და მაღალთა [and of high]

მთათა 

გალილიაჲსათა.

2ჴბოსა მას + ოქროჲსასა [of gold]

In 2:11–12 the Georgian text partly reflects GI, and partly GII: 2:11 GI

καὶ

×

×

ἡ γυνή

μου

GII

»

ἐν τῷ χρόνῳ

ἐκείνῳ

2»

3»

OGeo

ხოლო

×

×

ცოლი

ჩემი

GI

Αννα

ἠριθεύετο

×

ἐν τοῖς γυναικείοις

GII

1»

»

ἐν τοῖς ἔργοις

τοῖς γυναικείοις

OGeo

ანა

ნადობდა

×

სადედოსა.

308

Natia Dundua

2:12 GI

καὶ

ἀπέστελλε

τοῖς κυρίοις

×

καὶ

ἀπέδωκαν

αὐτῇ

GII

»

»

»

αὐτῶν

»

ἀπεδίδουν

»

OGeo

და

მიუძღუანებნ

უფალსა

მათსა

და

მოიღებნ 

×

GI

×

×

×

×

×

×

GII

τὸν μισθόν

καὶ

ἐν τῇ ἑβδόμῃ

τοῦ Δύστρου

ἐξέτεμε

τὸν ἱστὸν

OGeo

სასყიდელსა. და

მეშჳდესა იგრიკისასა + თთუესა ამის [of the month]

GI

×

×

×

×

καὶ

GII

καὶ

ἀπέστειλεν

αὐτὸν

τοῖς κυρίοις

»

OGeo

და

მიუძღუანნა

იგინი

უფალთა + მათთა [their]

და

GI

×

αὐτοὶ

τὸν μισθὸν

προσδόντες

×

GII

ἔδωκαν

αὐτῇ

»

×

πάντα

OGeo

მოსცეს

მას

სასყიდელი

×

×

GI

καὶ

×

×

×

×

ἔριφον

×

GII

»

ἔδωκαν

αὐτῇ

ἐφ᾽

ἑστίᾳ

»

ἐξ αἰγῶν

OGeo

და

დაურთეს

მას

ზედა ×

თიკანი.

×

მოჰკუეთნა ქსელნი

What Can the Georgian Translation of the Book of Tobit Tell about GIII

309

Οr, as in 2:13–14, the variants of GI and GII occur side by side: 2:13 GI

ὅτε

δὲ

ἦλθεν

πρός με

×

ἤρξατο

GII

2»

1καὶ

εἰσῆλθεν

»

ὁ ἔριφος

»

OGeo

2ვითარცა 1ხოლო

შემოვიდა

ჩემდა,

3თიკანმან მან. 1იწყო

GI

κράζειν

×

×

×

καὶ

εἶπα

αὐτῇ

πόθεν

GII

»

καὶ

ἐκάλεσα

αὐτὴν

»

»

×

»

OGeo

2ბარიდ

×

×

×

და

ვარქუ

მას:

ვინაჲ + მოვა [is coming]

GI

τὸ ἐρίφιον

×

μὴ

κλεψιμαῖόν

ἐστιν

GII

»

τοῦτο

μήποτε

»

»

OGeo

თიკანი

ესე?

ნუუკუე

მპარავი

არს?

GI

ἀπόδος

αὐτὸ

τοῖς κυρίοις ×

GII

»

»

»

OGeo

მიეც

×

უფალთა

GI

×

×

×

φαγεῖν

×

κλεψιμαῖον

GII

ἐξουσίαν

ἔχομεν

ἡμεῖς

»

οὐδὲν

»

OGeo

×

×

×

2ჭამად.  ×

οὐ γὰρ

θεμιτόν ἐστιν

αὐτοῦ

»

×

თჳსთა, + ვისიცა არს, [whose (it) is]

რამეთუ არა

ჯერ-არს

1ნაპარევი

310

Natia Dundua

2:14 GI

ἡ

δὲ

εἶπεν

×

×

δῶρον

δέδοταί

μοι

GII

×

καὶ

λέγει

μοι

αὐτή

δόσει

»

»

OGeo

2მან

1ხოლო

თქუა:

×

×

ნიჭად

მომცეს +ეგე [it]

×

GI

ἐπὶ τῷ μισθῷ

καὶ

οὐκ ἐπίστευον

αὐτῇ

καὶ

ἔλεγον

GII

»

»

»

»

»

»

OGeo

სასყიდელსა ზედა.

და

არა მრწმენა

მისი.

×

ვეტყოდე,

GI

ἀποδιδόναι

αὐτὸ

τοῖς κυρίοις

καὶ

ἠρυθρίων

GII

ἀποδοῦναι

×

»

»

προσηρυθρίων χάριν

τούτου

OGeo

რაჲთამცა მისცა

×

უფალთა + მისთა [her]

და

ვჰრისხევდი

×

×

GI

πρὸς αὐτήν

ἡ

δὲ

ἀποκριθεῖσα

εἶπέν

μοι

×

ποῦ

GII

»

×

εἶτα

»

λέγει

»

καὶ

»

OGeo

მას.

2მან

1ხოლო

მომიგო + და [and]

მრქუა:

×

×

სადა

GI

εἰσιν

αἱ ἐλεημοσύναι

σου

καὶ

×

×

αἱ δικαιοσύναι

GII

»

»

»

×

ποῦ

εἰσιν

»

OGeo

არიან

მოწყალებანი

×

და

×

×

სიმართლენი

»

×

×

What Can the Georgian Translation of the Book of Tobit Tell about GIII

311

GI

σου

ἰδοὺ

×

γνωστὰ

πάντα

μετὰ σοῦ

×

GII

»

ἰδὲ

ταῦτα

2»

×

1»

ἐστιν

OGeo

შენნი?

აჰა ესერა

×

განცხადნა ყოველი

შენ ზედა. ×

Such compilation of the textual types is often manifested in lexical, phraseological or grammatical doublets. Furthermore, there are various kinds of grecisms, semitisms, and expressions that sometimes have parallels also in other books. For example, 1:21 ἔφυγον] წარივლტოდეს იგინი და წარვიდეს OGeo (synonyms) 5:1 ποιήσω] ყოფით ვყო OGeo (figura etymologica) 5:10 ἀδύνατος τοῖς ὀφθαλμοῖς] მდაბალი ვარ და უძლური თუალითა [I am incapable and feeble with eye] OGeo cf. Job 22:29 κύφοντα ὀφθαλμοῖς მდაბალი იგი თუალითა GeoOS, etc.

910 Special attention should be paid to the fragment of 910 (P. Oxyrh. 1076), which, where preserved (2:2–8), is also supposed to be a bearer of the GIII text form.23 According to Hunt this fragment is clearly distinguished from GI on the one hand and from GII on the other. He notes that “both 1076 [i.e. 910, N.D.] and C [i.e. GIII, N.D.] belong to the ‫ א‬type [i.e. GII, N.D.], but are more concise, while at the same time they occasionally add points of their own”24. I will bring the whole survived fragment of 910 in order to see the relationship between 910 and OGeo: 2:2 GI

πολλὰ

καὶ

εἶπα

×

GII

πλείονα

»

»

τῷ Τωβια »

»

παιδίον βάδιζε

910

πολ]λά,

×

»

Τωβια

»

»

×

OGeo

მრავალი და

ძესა

ჩემსა: ×

ვარქუ ×

τῷ υἱῷ μου

×

βάδισον

» წარვედ

23  cf. A.S. Hunt, The Oxyrhynchus Papyri (Part VIII, Oxford: Horace Hart, 1911), 6–9; Hanhart, Septuaginta, Tobit, 44–45. 24 Hunt, The Oxyrhynchus Papyri, 7.

312

Natia Dundua

GI

καὶ

ἄγαγε

ὃν ἐὰν

εὕρῃς

τῶν ἀδελφῶν

ἡμῶν

GII

»

×

ὃν ἂν

»

2»

3»

910

»

ἄγαγε

ὃν ἐὰν

»

ἐκ τῶν ἀδελφῶν

»

OGeo

და

მოიყვანე,

ვინცა

ჰპოო

ძმათაგანი

ჩუენთაჲ

GI

ἐνδεῆ

×

×

ὃς

μέμνηται

GII

1πτωχὸν

ἐκ Νινευητῶν

αἰχμαλώτων

»

»

910

»

×

×

×

×

OGeo

ნაკლულევანი,

×

×

რომელსა

ეჴსენოს

GI

τοῦ κυρίου

×

×

×

GII

×

ἐν ὅλῃ

καρδίᾳ

910

×

×

OGeo

უფალი; ×

GI

×

×

καὶ

ἰδοὺ

μενῶ

GII

κοινῶς

μετ᾽ ἐμοῦ

»

ἰδὲ

προσμενῶ

910

×

×

×

×

παραγενε]σθαι

OGeo

×

×

და

აჰა ესერა,

გელი

×

×

×

×

×

αὐτοῦ καὶ

ἄγαγε

αὐτὸν καὶ

φάγεται

×

×

×

×

×

×

×

×

×

×

×

×

×

×

What Can the Georgian Translation of the Book of Tobit Tell about GIII

313

GI

σε

×

×

×

×

×

GII

»

παιδίον

μέχρι

τοῦ

σε

ἐλθεῖν

910

»

×

×

×

×

×

OGeo

შენ. 

×

×

×

×

×

2:3 GI

×

×

×

GII

καὶ

ἐπορεύθη

910

»

OGeo

×

×

×

×

×

Τωβιας ζητῆσαί

τινα

πτωχὸν

τῶν ἀδελφῶν

ἡμῶν

»

»

×

×

×

×

×

×

×

×

×

×

×

×

×

GI

×

×

GII

καὶ

910

×

×

×

×

×

×

×

×

ἐπιστρέψας λέγει πάτερ καὶ

εἶπα

αὐτῷ ἰδοὺ

ἐγώ παιδίον

×

×

×

×

×

×

×

×

×

×

OGeo

×

×

×

×

×

×

×

×

×

×

GI

καὶ

ἐλθὼν

εἶπεν

×

πάτερ

GII

»

ἀποκριθεὶς

»

×

»

910

»

ἀναστρέψας

»

μοι

×

OGeo

და

მოვიდა +და [and]

მრქუა

მე

მამაო,

314

Natia Dundua

GI

×

εἷς

ἐκ τοῦ γένους

ἡμῶν

ἐστραγγαλωμένος

GII

ἰδοὺ

»

ἐκ τοῦ ἔθνους

»

πεφόνευται

910

»

»

τῶν ἀπὸ τοῦ ἔθνους

»

×

OGeo

×

ერთი + ვინმე [somebody]

ნათესავისა

ჩუენისაჲ

შეგულარძნილი

GI

×

ἔρριπται

ἐν τῇ ἀγορᾷ

×

×

GII

καὶ

»

»

καὶ

αὐτόθι νῦν

ἐστραγγάληται

910

×

×

×

×

×

×

×

OGeo

×

დაგდებულ არს

უბანსა ზედა. 

×

×

×

×

×

×

2:4 GI

κἀγὼ

×

×

×

πρὶν ἢ

γεύσασθαί

GII

καὶ

ἀναπηδήσας

ἀφῆκα

τὸ ἄριστον »

»

910

×

×

×

×

×

×

OGeo

ხოლო

×

×

×

2ვიდრე

3პურის ჭამამდე

GI

με

×

GII

»

910 OGeo

ἀναπηδήσας

ἀνειλόμην

αὐτὸν

αὐτοῦ καὶ

ἀναιροῦμαι

×

»

×

×

»

ἦρα

×

»

1მე

×

2და 1აღვდეგ

აღვიღე

იგი + და შევიღე [and (I) entered]

×

What Can the Georgian Translation of the Book of Tobit Tell about GIII

315

GI

×

×

εἴς τι οἴκημα

×

ἕως οὗ

GII

ἐκ τῆς πλατείας

καὶ

εἰς ἓν τῶν οἰκιδίων

ἔθηκα

μέχρι

910

»

×

εἰς ἓν τῶν οἰκημάτων

×

»

OGeo

×

×

სახლსა,

×

ვიდრემდე

GI

ἔδυ

ὁ ἥλιος

×

×

×

GII

2δύειν

1τοῦ τὸν ἥλιον

καὶ

θάψω

αὐτόν

910

»

τὸν ἥλιον

»

»

»

OGeo

დაჰჴდა

მზე.

×

×

×

As it is seen the Georgian mostly agrees with the readings of 910. However, it should be noted that almost all of such readings correspond to the GI as well. Thus, we cannot assume any direct connections between 910 and OGeo.

13:1–fin Ιn the final part of the book a similar picture is evident as in the beginning. Here, however, OGeo mainly reflects GI, for example: 14:5 GI

καὶ

ὁ οἶκος

τοῦ θεοῦ

ἐν αὐτῇ

οἰκοδομηθήσεται

GII

»

»

»

»

»

OGeo

და

სახლი

ღმრთისაჲ

მას შინა

აღეშენოს

316

Natia Dundua

GI

εἰς πάσας

τὰς γενεὰς

τοῦ αἰῶνος

οἰκοδομῇ

ἐνδόξῳ

GII

×

×

×

×

×

OGeo

ყოველთა ნათესავთა საუკუნეთა შენებულითა დიდებულითა, მიმართ

GI

καθὼς

ἐλάλησαν

περὶ αὐτῆς

οἱ προφῆται

×

GII

»

»

»

»

τοῦ Ισραηλ

OGeo

ვითარცა იტყოდეს

მისთჳს

წინაწარმეტყუელნი

×

In this part also OGeo shares some variants of GII and sometimes mixes the readings of GI and GII, for example: 14:2 GI

ὅτε

×

ἀπώλεσεν

τὰς ὄψεις

GII

»

ἐγένετο

ἀνάπειρος

τοῖς ὀφθαλμοῖς

OGeo

ოდეს

×

წარწყმიდა =GI

თუალნი =GII

GI

καὶ

μετὰ ἔτη ὀκτὼ

ἀνέβλεψεν

×

GII

»

μετὰ τὸ

ἀναβλέψαι

αὐτὸν

OGeo

და

შემდგომად რვისა წლისა =GI

აღიხილნა =GI

×

GI

καὶ

ὅτε

ἀπέθανεν Αννα

×

×

GII

»

»

»

×

ἡ μήτηρ

αὐτοῦ »

»

OGeo

და

ოდეს მოკუდა

ანა,

დედა

მისი,

იგი

14:12 ἔθαψεν

დაჰფლა

αὐτὴν

What Can the Georgian Translation of the Book of Tobit Tell about GIII

317

Unique readings of OGeo, which have no parallels in any extant sources, are evident in the final part of the book too. For example: 14:2 καὶ

GI

×

×

ἦν

ἐτῶν πεντήκοντα

ογδοηκοντα οκτω (vel πηʹ) Α 71-236-402-542-762te 122 126 311 583; om ὀκτώ d

GIvar

GII

»

ἀπέθανεν ἐν εἰρήνῃ ×

»

ἑκατὸν

GIIvar OGeo

ὀκτώ

×

×

×

იყო

×

δώδεκα duo Sy = �te 2

ორმეოცდაათექუსმეტის [56]

d in 1:1–6:8; 13:1–14:15: As we ascertained one more source of the GIII text form and saw that OGeo mostly follows the d recension in 6:9–12:22, it is interesting whether there are correspondences between OGeo and the manuscripts of the d group also in those parts (1:1–6:8; 13:1–14:15), where, according to Hanhart, this recension reflects GI like the a b and c recensions and which “an einigen Stellen nach Textform GII überarbeitet ist”.25 The collation of OGeo with d in 1:1–6:8; 13:1–14:15 showed that when d differs from GI in some instances OGeo shows also similar readings to d, for example: 1:6 τῶν γενημάτων] πρωτογενηματων d: cf. GII პირველთა ნაყოფთა OGeo│ 2:8 om πάλιν d 910 LaJ OGeo│ 3:13 μὴ ἀκοῦσαί με μηκέτι] μηκετι ακουσαι d, cf. 3:1526 μηκετι ακουειν 319 არღარა მესმეს OGeo│

25  “wie weit diese Rezensionselemente mit der Textform GII und innerhalb davon oft der altlateinischen Überlieferung näher stehenden, in diesen Teilen nicht mehr erhaltenen Textform GIII übereinstimmen, läßt sich nicht mehr entscheiden” (Hanhart, Text und Textgeschichte, 57). 26  This variant in d is due to the parallel place, whereas the Georgian follows the Greek Vorlage.

318

Natia Dundua

4:7 σοι] σου A V 370 58-670 d 55 108 318 392 488 583 319 შენთა OGeo│ 4:14 αὐλισθήτω] + εν σοι d SaAC შენ თანა OGeo│ 5:12 σὺ ζητεῖς] επιζητεις d 108 535 ეძიებ OGeo│ 5:16 μισθόν] + σου V a d 126 Aeth Arm 319 შენსა OGeo│ 13:1 καὶ ἡ βασιλεία αὐτοῦ GI GII] οτι εις παντας τους αιωνας d La (quoniam in omnia saecula regnum est illius [cum var] La-R BrevGoth 393, cf. Vulg.) და სუფევაჲ მისი ჰგიეს უკუნისამდე (მარადის F) OGeo│ 13:8 om μου d OGeo│ 14:4 om καὶ 1° a d 55 126 318 392 Arm OGeo│ 14:10 Αμαν] αδαμ B d 46 55 318 319 392 SyO Arm ადამ OGeo.

But differences from d are attested as well, for example: 13:2 ἀνάγει GI GII OGeo] αυτος (>107) αναγει εκ της απωλειας (om εκ τ. απ. 107) εν τη μεγαλωσυνη αυτου d = La (ipse [> Q P R] reducit [revovat X] a perditione maiestate sua [cum var]): cf. S (αὐτὸς ἀνάγει ἐκ τῆς ἀπωλείας τῆς μεγάλης)│ 13:4 om ἐκεῖ მუნ d 583 Sa Arm: cf. 𝔙│ 14:4 ἐν τῇ γῇ OGeo] > 71 d 311 535 583 Sa Arm│ 14:12 τῶν υἱῶν αὐτοῦ GI ძეთა მისთა OGeo] γυναικος αυτου d.

As we assume that OGeo still bears GIII in these parts and shows similar varie­ gated relations to GI and GII as in other parts, it seems that d should also reflect the GIII textual form for the entire text, where different textual types are presented with different intensity. Thus, editing d in a separate form seems to be more expedient in order to show a clearer picture of this recension.

Conclusion To sum up, the combined text of the Georgian translation, as evidenced already by the given examples, often coincides with the texts of various Greek manuscripts or the daughter versions, but it does not reveal a constant parallelism with any of them. Moreover, there are certain passages in the Georgian translation, which have no parallels with any of the extant sources. In spite of this, their character indicates their Greek origin. The aforementioned factors give us sufficient grounds to assume the high probability of the existence of a Greek Vorlage, parent to the OGeo, which represented the GIII textual type even in the first and the last parts of Tobit. GIII for its part should be an incomplete textual form, in different parts of which GI and GII dominated to a different degree, and which has been petrified in the process of development. It is hard to say whether this process was directed to the expansion of GI or, on the contrary, to the abridgement of the GII textual form, or neither the one nor the other, with each of these textual forms developed

What Can the Georgian Translation of the Book of Tobit Tell about GIII

319

independently. One thing is undeniable: the Georgian translation displays the character of the GIII textual form, which cannot be fit neither in the apparatus criticus of GI nor of GII. About the text-critical value of OGeo the following can be said: 1) The GIII textual form presented in Greek minuscules 106–107 (=d) and SyO most consistently in 6:9–12:22 is preserved in the Old Georgian translation more fully than in other extant sources. In the part of the text where d and Sy carry GI, OGeo still attests the compiled text of GI and GII; it sometimes supports Greek witnesses (among them Sc) and, what is more important, also the Vetus Latina (more often). So the Georgian translation represents a more complete picture of GIII in section 6:9–12:22 as well as beyond. 2) The Georgian text attests the older stage of the textual history of GIII than the other witnesses (cf. 6:11, 16, etc.). 3) The variants of OGeo which deviate from d and agree with La and Sy (e.g., 7:15, 10:12, etc.) specify the character of GIII textual form even more ­precisely.

Bibliography Brooke, A.E. & McLean, N. & Thackeray, H.St.J., The Old Testament in Greek: VIII-I Esther, Judith, Tobit (Cambridge: Cambridge University Press, 1940). Busto Saiz, J.R., El doble texto griego del libro de Tobit y su insercion en la historia de G (Madrid: Universidad Complutense, 1975). Dochanashvili, El., Mcxeturi xelnac’eri [Mcxeta manuscript] (Tbilisi: Mecniereba, 1983) 34–53 [Geo.]. Fernández Marcos, N., “The Origins of the Septuagint: The Double Text of Tobit”, in The Septuagint in Context: Introduction to the Greek Versions of the Bible (Leiden: Brill, 2000). Fitzmyer, J., Tobit: Commentaries on Early Jewish Literature (Berlin: Walter de Gruyter, 2003). Hanhart, R., Septuaginta, Tobit (Göttingen: Vandenhoeck & Ruprecht, 1983; 1984). Hanhart, R., Septuaginta, Tobit (Vetus Testamentum Graecum Auctoritate Academiae Scientiarum Gottingensis editum vol. VIII, 5 Tobit; Göttingen: Vandenhoeck & Ruprecht, 1983). Hanhart, R., Text und Textgeschichte des Buches Tobit (Göttingen: Vandenhoeck & Ruprecht, 1984). Hunt, A.S., The Oxyrhynchus Papyri (Part VIII, Oxford: Horace Hart, 1911). Kurtsikidze, Ts., ʒveli aγtkmis ap’ok’ripebis kartuli versiebi [Georgian Versions of the Old Testament Apocrypha (according to the X–XVIII cc. manuscripts)] (Book I, Tbilisi: Mecniereba, 1970), 110–84 [Geo.]. Kurtsikidze, Ts., ʒveli aγtkmis ap’ok’ripuli c’ignebis kartuli versiebi [Georgian Versions of the Old Testament Apocryphal Books] (Book II, Tbilisi: Mecniereba, 1973), 117–74 [Geo.].

320

Natia Dundua

Marcos, N.F., The Origins of the Septuagint: The Double Text of Tobit, in: The Septuagint in Context, Introduction to the Greek Versions of the Bible (transl. Wilfred G.E. Watson; Leiden: Brill, 2000). Wagner, C.J., Polyglotte Tobit-Synopse (MSU 28; Göttingen: Abhandlungen der Akademie der Wissenschaften in Göttingen, 2003). Weeks, S., “Some Neglected Texts of Tobit: the Third Greek Version”, in M. Bredin (ed.), Studies in the Book of Tobit: A Multidisciplinary Approach (London: T & T Clark, 2006). Weeks, S., Stuckenbruck, L. & Gathercole S. (ed.), The Book of Tobit: Texts from the Principle Ancient and Medieval Traditions, with synopsis, Concordances, and Annotated texts in Aramaic, Hebrew, Greek, Latin and Syriac (Berlin: Walter de Gruyter, 2004).

Natia Mirotadze The Old Georgian Version of the Book of Esther—All in One1 The main peculiarity of the Book of Esther is textual plurality. Alongside the MT of Esther, there are three Greek textual forms: o', L and GrLa (preserved only in the old Latin translation), and none of them are identical to the Hebrew. The main difference between the MT and the Greek traditions is six large passages not attested in Hebrew. Therefore, they are considered as apocryphal sections of the book. Moreover, even the Greek tradition is not homogenous, and all three texts (o', L and GrLa) significantly differ from each other. Furthermore, these differences exceed typical recensional/revisional changes.2 The following article attempts to show how the Old Georgian Esther is related to the textual traditions of the book attested in the Greek text forms and Old Latin versions based compositional and structural details of the apocryphal sections. It is obvious that the examples discussed in the article do not cover all the types of peculiarities of GeII throughout the additional chapters.

The Book of Esther in the Old Georgian Tradition There are two independent versions of the book of Esther in Georgian. One of them—GeI—is a translation of the L text and is preserved in only one manuscript, the so-called Oshki Bible3—Ath.-Iver.-1 (978). Until now this is the only known daughter version of this textual form. But the subject of my research is the 1  I would like to thank Prof. Anna Kharanauli for her suggestions and her guidance in writing the paper and Antonia St Demiana for improving the English of the paper. 2  For a detailed discussion on the subject, see the main bibliography: J.-C. Haelewyck (ed.), Hester, Band 7/3 (Freiburg: Herder, 2003–2008); J.-C. Haelewyck, “The Relevance of the Old Latin Version for the Septuagint, with Special Emphasis on the Book of Esther”, JTS (2006) 439–73; R. Hanhart (ed.), Esther, in Septuaginta, Vetus Testamentum Graecum, Auctoritate Academiae Litterarum Gottingensis, editum vol. VIII/3 (Göttingen: Vandenhoeck & Ruprecht, 1966); C.A. Moore, “On the Origin of the LXX Additions to the Book of Esther”, JBL 92 (1973) 382–93; C.A. Moore, Daniel, Esther and Jeremiah: The Additions (AB 44; Garden City, NY: Doubleday, 1977), 180; E. Tov, “The ‘Lucianic’ Text of the Canonical and Apocryphal Sections of Esther: A Rewritten Biblical Book”, in The Greek and Hebrew Bible: Collected Essays on the Septuagint (VTSup 72; Leiden: Brill, 1999) 535–48; K. De Troyer, “The Letter of the King and the Letter of Mordecai: An Analysis of MT & LXX 8.9–13 and AT 7.33–38”, in Textus 21 (2002) 175–207; K. De Troyer, Rewriting the Sacred Text: What the Old Greek Texts Tell Us about the Literary Growth of the Bible (TCSt 4; Leiden: Brill, 2003), 1–28, 59–89. 3  The manuscript was copied in the Oshki Monastery (the southern part of historical Georgia, now in Turkey) especially for the newly founded Georgian Monastery on Mount Athos.

322

Natia Mirotadze

other Old Georgian version of Esther—GeII—which is preserved in five 13th18th century manuscripts:4 1) M—1694.4, Fr.14-1, 7125 d(4)r—13th cent.5; 2) I—A-570—1460; 3) F—A-646—15th-16th cent.; 4) D—H-885—17th cent.; 5) S—A-516—17th-18th cent.7 Despite the late date of these sources, the version is an old translation. A few factors point to this: 1) These manuscripts, especially A-646 (F), in the case of other books, among them Kings, mainly have old and good texts; and 2) the translation technique and vocabulary are also old. A word-by-word comparison of GeII to all of the Greek textual forms (o' L La) revealed that: 1) It has similarities to all of them: a) having not only common parts of these three versions, b) but also containing those episodes, which are specific to only one of the three and are not found in the others. 2) It also has extra textual material, which mainly is not attested anywhere else and must reflect a source, which has not come down to us.

These facts lead us to conclude that the text of GeII is a contaminated one. The sources of contamination are o', L, GrLa (i.e. the Greek text which served as the Vorlage for the sources of the Vetus Latina) and GrX (i.e. a now-lost text-form not attested elsewhere), while the base-text is o' Esther. The purpose of my paper is to supply arguments in the following pages for these statements. For this, as I already mentioned above, I will discuss compositional and structural features of the so-called apocryphal sections and the peculiarities of their transmission in Georgian.

Apocryphal Sections in the Textual Traditions of Esther The apocryphal sections are not adapted by Greek and Old Latin textual forms of Esther in the same quantity and in the same way: some chapters are shared by all of them, while others are found only in one form of the book. 4  Abbreviations GeI and GeII are conditional and are not connected to the translation date or order. 5  I would like to express my deep gratitude to Bernard Outtier, who informed me of the Georgian fragments of Esther (Fr.14-1, 7125 d(4)r) that he catalogued in the Matenadaran collection of Georgian fly-leaves and fragments and supplied me with their codicological description and variant readings. In 2017, I visited to Matenadaran and checked the fragments myself. 6  Includes GeII revised according to the Latin (Vulgate). 7  In addition to these, there is also one more Old Georgian version of the Book of Esther—the translation of the Septuagint text-form (published in 1743 in Moscow) rendered from Slavonic.

323

The Old Georgian Version of the Book of Esther—All in One

The relation of GeII to the remaining tradition, based on the example of apocryphal material, is apparently outlined in the following table. Table 18 o'

A

L

La

GeII

A:1–11

A:1–11

Mordecai’s Dream

A:1–11

A:1–10

Conspiracy of the Eunuchs

A:12–17

A:11–18

A:12–17

King’s Edict Addressed to the Male Inhabitants of the Kingdom B

Haman’s Edict

H

Prayer of the Jews Prayer of Mordecai Preparation for Prayer

C

Prayer of Esther

The Lord’s Acts The Petition

I (between o' 1:22 and 2:1) B:1–7

3(B):14–18

B:1–7

B:1–7

H:1–5

H:1–5

C:1–11

4(C):12b–17

C:1–11

C:1–11

C:12–15

4(C):18– 19(15)

C:12–15

C:12–15

C:16 A Brief Mention

4(C):20(16) A Brief Mention

C:16 An Enu­ mer­ation

C:16 An Enu­ mer­ation

C:17–30

4(C):21(17)– 29(30)

C:17–30

C:17–30

8  Here and elsewhere, versification of the Greek texts follows the Göttingen edition. La is versified according to the Haelewyck’s edition, and GeII is versified similarly to the Greek texts. Table 1 does not note different verse-succession inside the episode. If an episode is missing from the textual form, cells are left empty.

324

Natia Mirotadze

o'

L

La

GeII

D

Esther’s Unsummoned Entrance before the King

D:1–16

5(D):1–12

D:1–16

D:1–16

E

The King’s Edict

E:1–24

7(E):22–32

E:1–24

E:1–24

Mordecai’s Letter

Spring, river

F

Inter­ preta­ tion of the Dream

After the Col­ophon (F:11)

7:33b–38

F:1–3

7(F):53–4a

F:1–3

Shouts, Confusion, Thunder, Earthquake and Chaos on the Earth Dragons, Sun, Light

The Colophon

F:1–3

F:3'

F:4–10

F:11

7(F):54a:59

F:4–10

F:4–10

F:11

As the table shows, GeII includes: 1) All apocryphal parts that are common for all three Greek text-forms: Mordecai’s dream, Haman’s edict, Mordecai’s prayer, Esther’s preparation for prayer, Esther’s petition to God, Esther’s unsummoned entrance before the king, the king’s edict, interpretation of the dream (in slightly different form); 2) Episodes which belong to only one particular version and are not found in the remaining ones: Colophon (= o' ); Mordecai’s Letter (= L); Prayer of the Jews (= La) and the extended Prayer of Esther (= La); and 3) Those, which now are found only in GeII: the king’s edict addressed to the male inhabitants of the kingdom and the interpretation of some parts of the dream.

The contaminated character of GeII is very clearly demonstrated by the apocryphal additions, as they are unevenly diffused in the Greek tradition while GeII

The Old Georgian Version of the Book of Esther—All in One

325

attests all of them. I will present examples demonstrating this peculiarity of GeII according to the storyline; firstly those, where GeII is supported by only one of the existing sources (I. GeII = La; II. GeII = L; III. GeII = o' ), and finally those where the Georgian stands alone (IV. GeII = GrX).

I. GeII = La I.1. H:1–5—Prayer of the Jews The context of the episode in o' L La is as follows: Haman’s edict was written down, and then sent out and promulgated in the peripheries as well as in Susa. From this point, the story continues differently in different textual forms: In the o', the edict’s promulgation has a different effect on the Jews on the one hand, and on the King and Haman, on the other. The first were crying while the others were feasting. The o' text focuses on the opposition of Haman and Mordecai, and the feast (a symbol of Haman’s Triumph) is followed by Mordecai’s mourning and efforts to cancel the edict. As a result, Jews are rescued from the anticipated tragedy and Haman is sentenced to death. La moves the spotlight from the Haman-Mordecai dichotomy and creates new contrast opposing feast and joy to the mourning of the entire nation, while both of them are caused by the same reason (annihilation of the Jews of the Persian Empire). This contrast is achieved by placing the Jews’ mourning and prayer after the feast scene. L keeps everything simple and the edict’s promulgation in Susa is directly followed by Mordecai’s acts to stop the forthcoming disaster. What occurs in GeII is a result of contamination from all these text-forms: GeII includes the edict’s promulgation in Susa twice, in both L and the o' forms. The first promulgation (L 3:19(15) καὶ ἐν Σούσοις ἐξετέθη τὸ πρόσταγμα τοῦτο) divides Haman’s edict into two parts, making its final verse (B:7) an announcement read in Susa. In B:7 the declaration that Jews must be killed is followed by La’s prayer of the Jews. After this, the plot continues as it is in the o': the edict is promulgated in the peripheries and in Susa, the people are terrified while the king and Haman are feasting, and Mordecai starts to act in order to stop Haman’s evil plot. The mourning of the Jews precedes the prayer in La, where these two episodes are also semantically connected and the prayer is the natural continuation of the mourning. In the o' and GeII it is included little bit below but not in the same place. As a result, in both of them the mourning of the Jews serves as a background highlighting the importance of Mordecai’s and Esther’s actions. For clarity, I have shown the storyline of all textual forms side-by-side in the table below.

326

Natia Mirotadze

Table 29 o'

La

1.

Haman’s Edict

1.

B:1–7

L

1.

GeII 1.

B:1–6

2.

L 3:19(15a)

3.

B:7

>

5.

+

3:19(15a)

6.

o' 3:15a

+

3(B):14–18

2.

Edict’s promulgation in the peripheries

2.

3:14

3.

Edict’s promulgation in Susa

3.

3:15a

4.

3:15b

>

7.

7.

4:3 (> 583)

>

11. o' 4:3

>

>

4.

H:1–5

4.

Lamentation of Jews and feast of the king and Haman

2.

5.

Mourning of Jews across the entire empire

6.

Prayer of the Jews (H:1–5)

7.

Mourning of Mordecai

5.

4:1

3.

4:1

8.

+

8.

Mordecai comes to the gates of the king’s palace

6.

4:2

4.

4:2

9.

+

8.

4:4

5.

4:3(4)

10. +

9.

Esther enquires of Mordecai what the matter is

9  Since the prayer of the Jews, besides the Georgian, is found in the Old Latin version and its text is the most complete, I used its sequence of the episodes as starting point for comparison. Episodes of other versions are numbered corresponding to the sequences that they have in each (o’ L GeII) of the textual forms. The Greek and Georgian texts are versified according to the Göttingen edition.

The Old Georgian Version of the Book of Esther—All in One

327

Despite the various locations of the prayer of the Jews in La and GeII, the prayer’s text itself is quite similar in both versions. The following table visibly illustrates the close relationship of GeII and the Old Latin sources of the prayer.10 Table 311 LaR

Lavar

GeII

H:1 et

ხოლო (δε)

invocabant

ჰხადოდეს 2

Iudaei

ჰურიანი 1

deum R-130

deum excelsum 130; excelsum deum IJF

მაღალსა ღმერთსა (= IJF)

patrum

მამათა

suorum

მათთასა

et

და

dicebant

იტყოდეს

domine

უფალო

deus

ღმერთო,

tu

შენ

solus

მხოლოჲ es IJF

ხარ (= IJF)

deus

ღმერთი

in caelo sursum

ცათა შინა

10  The editor of Vetus Latina Esther, Jean-Claude Haelewyck, distinguishes several stages throughout the textual history of the Old Latin translation of the Book. The oldest and closest text to the original translation is represented by the R type (accordingly, this version is treated as the third Greek text in the following discussion); the next better one is I. As per Haelewyck’s very own statement, the remaining groups show subsequent approximations to the Septuagint (J) and the L text (F). The gaps, which R (or the third Greek text) have in relation to the other Greek texts, are frequently filled within these (J F) text-types (J.-C. Haelewyck, “The Relevance”, 449). 11  The Old Latin and Old Georgian versions of the prayer are aligned word-for-word in columns. When the Georgian equals the text taken for comparison, it is written in the same row without translation. If variants of the Georgian version are found in Latin sources as well, common readings are displayed in the column of variants. If the Georgian stands alone, variants are translated back into Greek and written without accents in parenthesis.

328

Natia Mirotadze

LaR

Lavar

GeII და ქუეყანასა ზედა, (και επι της γης)

et

და

non

არავინ (ουδεις)

est

არს

alius deus შენსა გარეშე. 

praeter te H:2 si enim

უკუეთუმცა

fecissemus

ვყავთ

legem

სჯული

tuam

შენი

et

და

praecepta RI

praecepta tua JF

და-მცა-ვიმკჳდრეთ

habitassemus forsitan cum pace

R-130

მცნებანი შენნი (= JF)

cum securitate et pacem კრძალულებით და R130; cum securitate მშჳდობით (= Lavar) 123 et pace I F; cum securitatem et pacem I152

omne

ყოველსა

tempus

ჟამსა. 

vitae nostrae

>

nunc autem

და აწ (και νυν)

quoniam

რამეთუ

non

არა

fecimus

ვყვენით

praecepta

მცნებანი

tua

შენნი, ამისთჳს (δια τουτο)

supervenit R-130IF

venit 130

მოიწია

in (super 151 123) nos

ჩუენ ზედა (= 151 123)

omnis

>

The Old Georgian Version of the Book of Esther—All in One

LaR

Lavar

329

GeII

tribulatio

ჭირი

ista

ესე H:3

iustus es et tranquillus et excelsus et magnus domine et omnes viae tuae iudicia

>

H:4 და აწ

et nunc deus RF

domine I

უფალო ღმერთო (κυριος ο θεος)

noster F

ჩუენო (= F)

non

ნუ

des

მისცემ

filios

შვილთა

tuos RI-152F

nostros 152

ჩუენთა (= 152)

in captivitatem RI

captivitate F

ტყუედ

neque

და (και )

uxores

ცოლთა

nostras

ჩუენთა

in violationem RF

conviolationem I

შეგინებად და ნუცა

neque RI

ჭაბუკთა ჩუენთა (τους νεανισκους υμων) in perditionem

წარსაწყმედელად

qui

რომელი

factus es

გუექმენ

nobis

ჩუენ

propitius

მალხინებელ

ab Egypto

ეგჳპტითგან

et R usque nunc

> IF

> (= IF) ვიდრე მოაქამომდე,

330

Natia Mirotadze

LaR

Lavar

GeII

H:5 miserere

ნუ აღიჴოცებინ (μη εξαλειφθητω)

principali

დასაბამითგან

tuae parti R

parti tuae F

ნაწილი შენი (F)

et

და

non

ნუ

des

მისცემ

hereditatem

სამკჳდრებელსა

nostram RF

tuam I

შენსა (= I)

in infamiam RI

infamio F

საყუედრელად, რაჲთა

ut hostes dominentur nostri R-130I

dominentur nobis gentes 130; hostes nostri dominentur nobis F

მთავრობდენ მათ წარმართნი (αρξουσιν αυτων τα εθνη) (≈ 130)

As can be seen from the table, the Georgian quite closely follows LaR, which is considered by the editor to be the closest textual form to the initial Old Latin translation:12 H:4 qui factus es nobis propitius RI; რომელი გუექმენ ჩუენ მალხინებელ GeII] pr quia tu es F; > J| H:1 domine deus RF; უფალო ღმერთო GeII] domine domine IJ.

In some cases, variants of GeII are supported by Manuscript sources of La: H:2 cum pace R151] cum securitate et pacem R130; cum securitate et pace I123F; cum securitatem et pacem I152; კრძალულებით და მშჳდობით GeII; > J.

Occasionally GeII also witnesses readings absent in La sources: H:1 in caelo (saeculo 130) sursum; ცათა შინა GeII] + და ქუეყანასა ზედა (και επι της γης) GeII.

12  See n. 10.

The Old Georgian Version of the Book of Esther—All in One

331

I.2. C:16—The Extended Prayer of Esther Greek textual forms and daughter versions represent the prayer of Esther in two short (o' L GeI) and long (La GeII Arm) forms. The context of Esther’s prayer is as follows: Mordecai convinces Esther that for her people’s sake she is obliged to enter before the king unsummoned despite the fact that it is against the law and she will be sentenced to death. Therefore, she prepares to fulfill this impossible mission spiritually (to receive favor in the eyes of God) as well as physically (in order to impress the King with her beauty). Her first step of preparation is praying. She starts her prayer with admission that God is the only helper of the Jews in calamities and that He is the only one who can save Esther and help her to save her people. Esther puts her hope in God, as she knows about God’s special mercy toward the Jewish people, because she has heard since her birth that God chose Israel among other nations to be an everlasting inheritance. Instead of the words she has heard since her birth that God chose Israel among other nations to be an everlasting inheritance, which are found in the short versions (o' L GeI) of the prayer, the long versions of La and GeII present a listing of the miracles that God performed for the Jews. The Armenian has both short and long forms of the verse: Verse C:16 in the form of the o' (i.e. the short form) is followed by a listing, and these two parts are linked with the phrase: ‘and she (Esther) increasing her prayer said again in this way’.13 All three versions of the extended prayer of Esther are quite different from each other and once again prove how diverse the Book of Esther is. To illustrate how various forms of C:16 are related, they are displayed in Table 4.

13  The English translation of the verse in the Armenian version is thus: “I have heard since my birth from my paternal people that you Lord chose Israel among all nations and our fathers from their forbears as your everlasting inheritance and did for them all that you said. And [she] increasing her prayer said again in this way: God of Abraham, God of Isaak and God of Jacob you are blessed forever. I heard from the paternal books that you, Lord, changed from the earth into the heaven Enoch from remorse to eternal life. I heard from the paternal books, Lord, that you saved Noah from the water flood. I heard from the paternal books, Lord, that you gave glory to Abraham in front of his men among the Kings. I heard from the paternal books, Lord, that you delivered Jonah from the whale fish. I heard from the paternal books, Lord that you saved three young men from the fire of the furnace and saved Daniel from the pit of lions. I heard from the paternal books, Lord, that you saved Ezekias King of Judea, condemned for death, and had mercy on him [who was] standing praying near you for his life and granted [him] and increased his life by fifteen years. I heard from the paternal books, Lord, that you gave child—Great Samuel—to Anna [who was] asking you, Lord, with prayers and requests.” For a more precise translation of the verse, see table 4.

332

Natia Mirotadze

Table 414 o' 16ἐγὼ

L 20(16)

ἐγω

LaR

GeII GrX

ArmZoh

ego

მე

Ես

audivi

მასმიეს

լսէի

δὲ ἤκουον

ἤκουσα

უფალო (κυριε) ἐκ γενετῆς

μου

ნათესავისაგან πατρικῆς

მამათა

=

ჩუენთაჲსა (ημων)

ἐν φυλῇ πατριᾶς

μου

ի ծննդենէ

իմմէ

ազգէն 2 βίβλου

in libris paternis meis

მამულთაგან წიგნთა (= La)

ի հայրենի 1

իմմէ,

14  In the table, all existing versions of the prayer are displayed in columns side-by-side. The Armenian version is cited according to the Zohrapian edition. When daughter versions parallel any Greek, the text is written in the same row without translation. If daughter versions have variant readings, they are translated into Greek if it can be reconstructed (back-translations are written without accents), otherwise they are translated into English. Proper names in Georgian are transcribed, to differentiate the stem and case mark stem is written in bold. Textual segments of the long version of the prayer are numbered to show the relationship between the different versions.

333

The Old Georgian Version of the Book of Esther—All in One

o'

L

ὅτι σύ κύριε

LaR

GeII GrX

ArmZoh

domine quoniam

რამეთუ შენ

եթէ դու,

ხარ ღმერთი რომელმან (ει ο θεος ος)

Տէր, (=o')

ἔλαβες τὸν Ισραηλ ἐκ πάντων τῶν ἐθνῶν καὶ τοὺς πατέρας ἡμῶν ἐκ πάντων τῶν προγόνων αὐτῶν εἰς κληρονομίαν αἰώνιον

ὅτι ἐλυτρώσω τὸν Ισραηλ ἐκ πάντων τῶν ἐθνῶν καὶ τοὺς πατέρας αὐτῶν ἐκ τῶν προγόνων αὐτῶν ἐπιθέμενος αὐτοῖς Ισραηλ κληρονομίαν αἰώνιον

առեր զԻսրայէլ յամենայն ազգաց եւ զհարսն մեր յազգատոհմէն իւրեանց ի ժառանգութիւն (+ pեզ = ημων) յաւիտեան

see below

see below

եւ արարեր նոցա զամենայն զոր խօսեցար: (Greek equivalent see below) 1

Եւ դարձեալ եւս յաճախեալ յաղօթսն իւր այսպէս ասէր. (and [she] again increasing her prayer said this way)

334

Natia Mirotadze

o'

L

LaR

GeII GrX

ArmZoh

2

Աստուած Աբրահամու, Աստուած Իսահակայ եւ Աստուած Յակոբայ, դու ես օրհնեալ յաւիտեանս. (God of Abraham, God of Isaac and God of Jacob, you are blessed forever)

3

ես լսէի ի հայրենի գրոց, զի դու, Տէր, (I heard from the paternal books that you, Lord)

4

փոխեցեր յերկրէ յերկինս զԵնովք, յապաշխարո­ ւթենէ ի կեանսն յաւիտենից. (changed from the earth into the heaven Enoch, from remorse to eternal life)

5

ես լուայ ի հայրենի գրոց, Տէր, եթէ դու (I heard from the paternal books, Lord, that you)

335

The Old Georgian Version of the Book of Esther—All in One

o'

L

LaR

GeII GrX

ArmZoh

6

Noe in aqua diluvii conservasti

ნოე (Noe) წყლით რღუნისაგან იჴსენ

զՆոյ ի ջուրցն հեղեղէ ապրեցուցեր.

7

ego audivi in libris paternis meis domine quoniam tu

>

ես լուայ ի հայրենի գրոց, Տէր, եթէ դու (I heard from the paternal books, Lord, that you)

8

Abrahae (Abraham 130) in trecentis et decem octo viris

და აბრაამი (abraami) (= 130) სამას ათრვამეტითა მონითა

Աբրահամու` առաջի արանցն իւրոց (to Abraham in front of his men)

ზედამოს­ რულთა მათ (who came upon him)

ի մէջ (among)

novem

ხუთთა (five)

reges

მეფეთა

tradidisti

მძლედ ետուր փառս. გამოაჩინე (make him win) (gave glory)

թագաւորացն (the Kings)

336

Natia Mirotadze

o'

L

LaR ≈ 16

GeII GrX ეზეკია (ezekia) მეფე (Εζεκια ο βασιλευς)

ArmZoh ≈ 16

ლოცვისა მისისათჳს სნეულებისაგან იჴსენ (due to his prayer saved from illness) 9

ես լուայի հայրենի գրոց, Տէր, եթէ դու (I heard from the paternal books, Lord, that you)

ego audivi in libris paternis meis domine quoniam tu

10 Ionam de ventre ceti liberasti

იონა (Iona) მუცლისაგან ვეშაპისა იჴსენ

զՅովնան ի կէտ ձկանն ապրեցուցեր. (Jonah from the whale fish delivered)

და გამოიყვანე მესამესა დღესა (and took [him] out on the third day) 11 ego audivi in libris paternis meis domine quoniam tu

ես լուայ ի հայրենի իմոց գրոց, Տէր, եթէ դու (I heard from the paternal books, Lord, that you)

337

The Old Georgian Version of the Book of Esther—All in One

o'

L

LaR

GeII GrX

ArmZoh

12 Ananiam Azariam (et 130 123 F) Misahel de camino ignis liberasti

ანანია (anania) აზარია (azaria) და (= 130 123 F) მისაელ (misael) საჴუმილისაგან ცეცხლისა იჴსნენ,

զերիս մանկունսն ի հնոցէ հրոյն փրկեցեր, (three young men from the fire of furnace saved)

14 Danihel de დანიელ (daniel) lacu leonum ჯურღმულ­ eruisti ისაგან ლომთაჲსა აღმოიყვანე,

եւ զԴանիէլ ի գբոյ առիւծուցն ապրեցուցեր. (and Daniel from the pit of lions saved)

15 ego audivi in libris paternis meis domine quoniam tu

ես լուայ ի հայրենի իմոց գրոց, Տէր, եթէ դու (I heard from the paternal books, Lord, that you)

13 ego audivi in libris paternis meis domine quoniam tu

16 Ezechiae regi

Iudaeorum

ეზეკია (ezekia) მეფე (Εζεκια ο βασιλευς)

զԵզեկիա արքայն (king Ezekia) Հրէաստ­անի` (of Judea)

338

Natia Mirotadze

o'

L

LaR

GeII GrX

ArmZoh մահուն դատապար­ տեալ` (condemned for death)

morte damnato

փրկեցեր, (saved) et oranti

ლოცვისა მისისათჳს (due to his prayer)

եւ ի կալ նորա յաղօթս (and on him standing praying)

სნეულებისაგან იჴსენ (saved from illness) pro vita misertus es

առ քեզ վասն կենաց իւրոց` ողորմեցար, (near you for his life [you] had mercy)

et donasti

եւ շնորհեցեր (and granted)

ei եւ յաւելեր (and increased) vitae

ի կեանս նորա (his)

annos quindecim

ամս հնգետասան.

339

The Old Georgian Version of the Book of Esther—All in One

o'

L

LaR

GeII GrX

ես լուայ ի հայրենի իմոցգրոց, Տէր, եթէ դու (I heard from the paternal books, Lord, that you)

17 ego audivi in libris paternis meis domine quoniam tu

18 Annae petenti in desiderio animae filii generationem donasti

ანა (ana) გთხოვდა წადიერებითა გულისა მისისაჲთა და მოჰმადლე შვილი. (Ana was asking with longing of her heart and [you] gave [her] son)

19 ego audivi in libris paternis meis domine quoniam tu 20 complacentes tibi liberas domine usque in finem

ArmZoh

და ყოველნი სათნონი შენნი სრულიად იჴსნენ, (you completely saved all your favorites)

ի խնդրելն ի քէն, Տէր, Աննայի աղօթիւք եւ խնդրուածովք, զաւակ շնորհեցեր նմա զմեծն Սամուէլ: (asking to you, Lord, Anna with prayers and requests, gave her child Great Samuel)

340

Natia Mirotadze

o'

L

LaR

GeII GrX

21

რამეთუ ამის მადლისათჳს გამოიყვანე აბრაამი (abraami) ქუეყანით ქალდეველ­ თაჲთ (for this sake [you] took Abraham from the country of Chaldeans)

22

და იჴსნენ ისაკი (isak’i) და იაკობი (iak’obi) (and saved Isaac and Jacob)

23

და ყოველნი პირველნი მამანი ჩუენნი ქუეყანით ეგჳპტით (egwp’t’it) და სხუათაგან თესლთა, რომელნი განიყოფდეს მათა სამკჳდრებე­ ლად. (and all our first fathers from Egypt and from other nations who divided them to inhabit.)

ArmZoh

341

The Old Georgian Version of the Book of Esther—All in One

o'

L

καὶ ἐποίησας = αὐτοῖς ὅσα (] ἃ L) ἐλάλησας

LaR

GeII GrX და ჰყავ მათ თანა, რაოდენსაცა იტყოდე.

ArmZoh եւ արարեր նոցա (+ զամենայն all) զոր խօսեցար: (post յաւիտեան tr Arm)

αὐτοῖς καὶ παρέσχου ὅσα ᾔτησαν.

As the comparison of these three versions (Vetus Latina, GeII, Old Armenian) of the long prayer shows, each of them preserves different forms of the prayer: 1) Both the Armenian and Georgian are longer than Latin, but not in the same way: the Armenian adds material before the listing of miracles (episodes: 1, 2, 3, 4, 5), while the Georgian adds material after the listing (episodes 21, 22, 23). The contents of the added episodes are different as well. 2) Sometimes the Latin and Armenian agree against the Georgian: a) The repetition of Esther’s words: ego audivi... (episodes: 7, 9, 11, 15, 17); b) The sequence of listed miracles is the same in La and Arm (episode 16 is transferred between episodes 8 and 9 in Georgian); c) Contents of episodes 10 and 16 are more similar in La and Arm as well. 3) In some cases, the Georgian and Latin agree against the Armenian: The content of episodes 8 and 12 are similar in La and GeII; both versions contain episode 20. 4) GeII and Armenian never agree against La.

II. GeII = L The letter of Mordecai is found only in the L text where it has a certain place and function and fits in well in the surrounding context, which is quite different from that of remaining textual forms.15 The letter is included in GeII but in a different place and out of context, and as a result, it does not function in the composition in the same way as in L. To understand what the function of Mordecai’s letter is in

15  De Troyer, Rewriting, 70 and n. 42; Haelewyck, “The Relevance”, 462.

342

Natia Mirotadze

L and how it is incorporated in the context, it is necessary to examine the larger section of the text more closely. In all textual forms, the storyline develops more or less in one direction until the following events: Mordecai was rewarded in a manner that demonstrated to the entire kingdom how the king’s loyalists were honored; when Haman’s plot and evil intentions were revealed and Haman was punished, but also when the danger threatening Jewish people was still in effect since Haman’s edict remained and required annulment; and when Mordecai and Esther stood before the King. From here, the story continues in two different ways: I. In the o' and La, the king rewards Mordecai through Esther by giving them everything that once belonged to Haman; and Esther falls before king’s feet appealing for the annulment of Haman’s Edict. Her request is granted and the king’s edict, informing people of previous events and his decision, is issued and sent out. II. In the L text, the king addresses Mordecai and promises to fulfill his request. Mordecai appeals for the annulment of Haman’s edict. After the fulfillment of his request, the king appoints Mordecai to care for the affairs of the kingdom. For her part, Esther asks for the destruction of the enemy of the Jews and the execution of Haman’s sons. The king agrees and punishment is carried out. Simultaneously, the king’s edict, informing the inhabitants of Persia that the Jews are not guilty and must not be killed, is issued and sent out. The king orders Mordecai to write his own letter to notify his people. Mordecai obeys and writes the letter which is included in the text. III. Regarding these two scenarios, GeII follows the Septuagint by not eliminating Esther’s role in the annulment of Haman’s edict and in the deliverance of her people; it also includes Mordecai’s involvement in this great deed. This is achieved by a change in grammatical number, that is pronouns and verbs in the singular are converted into the plural. After Mordecai and Esther are rewarded with Haman’s possessions, they together fall before king’s feet, and together appeal for the annulment of Haman’s edict. Nothing else is changed, and the storyline continues as it does in the o', i.e. GeII lacks Esther’s request to execute Haman’s sons and the king does not permit Mordecai to write his own letter. Despite this, GeII still adopts Mordecai’s letter but places it at the end of the text after the colophon. Although the letter of Mordecai is redundant in GeII, textually it closely follows the Greek equivalent as shown in the following table.16 16  To demonstrate that inclusion of the letter of Mordecai in GeII is not a result of inner-Georgian contamination, in the table GeI is also represented together with Greek L. As comparison has shown, distinctions between GeI and GeII, on the one hand, are caused by different translation techniques (ἀναπέμψαι მოცემად GeII] მომიძღუანეთ GeI| εἰς ἀπώλειαν წარსაწყმედელად GeII] და წარვწყმიდნე იგინი GeI) and by the use of a different

The Old Georgian Version of the Book of Esther—All in One

343

Table 517

7(8):35–38 L 35ἡ

GeI δὲ

GeII

ხოლო

ἐπιστολή,

წიგნი იგი სამოციქულოჲ,

ἣν

რომელ

ἀπέστειλεν

მისცა

ὁ Μαρδοχαῖος,

მარდოქჱ, (Mardokē)

ἦν

იყო

მარდოქეოზ: (Mardokeoz)

ἔχουσα ταῦτα

ესრეთ: ნათესავსა ჰურიათასა გიხაროდენ! (People of Judea, rejoice!)

36Αμαν

ჰამან, (Haman)

ამან (Aman)

ეგერა, (there) ἀπέστειλεν

მიგიძღუანა

წარუძღუანა

ὑμῖν

თქუენ

თქუენთჳს, (for you) 2

γράμματα

წიგნები

წიგნები (books) 1

ἔχοντα οὕτως

და ესრე თქუა, ვითარმედ

წერილი (written) ესრეთ:

῎Εθνος

ნათესავი

=

Ἰουδαίων

ჰურიათაჲ

=

ἀπειθὲς

ურჩ

ურჩი,

არის. (is) σπουδάσατε

იწრაფეთ

ისწრაფეთ

Vorlage, on the other (ἦν ἔχουσα ταῦτα იყო ესრეთ GeI] > GeII| ნათესავსა ჰურიათასა გიხაროდენ] > L GeI). These variations indicate that GeII did not take this episode from GeI, while the Greek source must have been slightly different from L as well. 17  In the table, texts are aligned word-for-word. Proper names in Georgian are transcribed. To differentiate between stem and case mark, the stem is written in bold. Variant readings of versions are translated into English.

344

Natia Mirotadze

L

GeI

GeII

ταχέως

>

ადრე

და შეკრიბეთ, (and collect) ἀναπέμψαι

მომიძღუანეთ (send [it])

მოცემად (to give)

μοι

მე (to me)

ჩუენდა (to us)

εἰς ἀπώλειαν.

და წარვწყმიდნე იგინი. წარსაწყმედელად. (to (and [I] will annihilate [them]) annihilate)

37ἐγὼ

ხოლო მე,

=

ὁ Μαρδοχαῖος

მარდოქჱ (Mardokē)

მარდოქე (Mardoke),

μηνύω

გაუწყებ

=

ὑμῖν

თქუენ,

=

δὲ

მისთჳს: τὸν ταῦτα ἐργασάμενον πρὸς ταῖς Σούσων

რომელმან ეგე ზრახვაჲ (intention) ქმნა,

რომელმან ესე ქმნა,

ესერა (here)

>

სუსესა (Susesa)

შუშანისათა (Šušanisata) 2

ქალაქსა (City) πύλαις

ბჭეთა ზედა

ბჭეთა ზედა 1

κεκρεμάσθαι

დამოკიდებულ არს.

=

იგი (he) καὶ

და

τὸν οἶκον

სახლი

αὐτοῦ

მისი

διακεχειρίσθαι·

მონიჭებულ არს ჩემდა. (is given to me)

>

38οὗτος

რამეთუ მას

რამეთუ მან

ἐβούλετο

უნდა

ზრახა

ἀποκτεῖναι ἡμᾶς

წარწყმედაჲ

=

ჩუენი

=

γὰρ

ერთბამად ყოველთა თანა სახლეულთა მისთა. (together with his whole household)

τῇ τρίτῃ καὶ δεκάτῃ მეათცამეტჱსა

მეათცამეტესა

τοῦ μηνός,

თთუესა

თუესა

The Old Georgian Version of the Book of Esther—All in One

345

L

GeI

GeII

ὅς ἐστιν Αδαρ.

ადარსა, რომელ არს იგრიკაჲ. (Adar which is Igrika)

ადარისასა, რომელ არს იგრიკისაჲ. (of Adar which is of Igrika) ცოცხლებით იყვენით! (Be Alive!)

Insertion of the letter of Mordecai at the end of the text (after the colophon) is a curious exception to how the textual material is usually adopted in the contaminated text of GeII, especially if we consider that GeII adds Mordecai’s voice to Esther’s request and gives him a role in the deliverance of the Jews, but does not change the text similar to L. This could have been enough reason to include the letter in a more suitable context, especially if we consider how closely other parts of the same section are compared and interwoven: Table 618 o'

L

8:1

7:14–15

1 καὶ

La

GeI

GeII

et

>

ἐν αὐτῇ τῇ ἡμέρᾳ

ipsa die

მას დღესა შინა

ὁ βασιλεὺς

rex 3

>

Ἀρταξέρξης

Artarxerxes 2

ასჳრეოს 2 (Ασυρηος)

18  In the table all textual forms of Esther and both Georgian versions are displayed in columns side-by-side. When the Georgian versions represent the similar reading of any Greek texts or Old Latin version, all of them are written in the same row. Variant readings of the Georgian versions unattested elsewhere, are translated into Greek if it can be reconstructed (back-translations are written without accents), otherwise they are written in English. In cases of transposition, words are numbered. Proper names in Georgian are transcribed. To differentiate between stem and case mark, stem is written in bold. The table is divided into four sections, namely 1, 2, 1', and 2'. 1 and 2 denote the text of Septuagint Esther, and 1' and 2' designate parts of the L text.

346

Natia Mirotadze

o'

L

La

GeI

ἐδωρήσατο (ὁ βασ. Ἀρτ. ἐδωρήσατο] αρτ. ο βασ. εδωρ. 392 542; εδωρ. ο βασ. αρτ. (artarxerxes rex LaPK; om αρτ. 71' Arm) O a 248 249' LaPKM Arm = MT)

donavit 1

მიანიჭა 1

Εσθηρ

Hester + reginae

მარდოქეს (mardokes),

ὅσα

რაოდენიცა

ὑπῆρχεν

მიენიჭა

Αμαν

ამანს (amans),

τῷ διαβόλῳ (> S* 71' La-M: cf L (7:15))

>

და ყოველი მონაგები მისი (και παντα τα υπαρχοντα αυτου)

omnia quaecumque erant Aman

1ʹ

GeII

14καὶ

და

და

εἶπεν

ჰრქუა: 2

ჰრქუა

ὁ βασιλεὺς

>

მეფემან

τῇ Εσθηρ Καὶ Μαρδοχαῖον

ესთერს (esters) ესთერს და მარდოქჱს (esters) და (mardokēs) 1 მარდოქეს (mardokes):

ἐβουλεύσατο

უნდა + მას

მისი ეზრახა 5

The Old Georgian Version of the Book of Esther—All in One

o'

GeI

GeII

κρεμάσαι

დამოკიდებაჲ + მისი,

წარწყმედაჲ, 6

τὸν σῷσαντά

რომელმან მიჴსნა

რომელმან განმარინა 1

με

მე

მე 2

ἐκ χειρὸς

ჴელთაგან

ჴელთაგან 3

τῶν εὐνούχων;

საჭურისთა­ჲსა. მათ საჭურისთა­ ჲსა, 4

οὐκ

არამე

არა

ᾔδει

უწყოდაა,

იცოდა,

ὅτι

რამეთუ

=

πατρῷον αὐτοῦ

მამასა მისსა

მამაჲ ჩემი (πατηρ μου)

γένος

ნათესავ

>

ἐστὶν

ეყვის

არს.

ἡ Εσθηρ;

ესთერ. (ester)

>

15καὶ

და

=

Μαρδοχαῖος

τὸν Μαρδοχαῖον 3

მარდოქჱს (mardokēs) 3

მარდოქეს (mardokes)

προσεκλήθη

ἐκάλεσεν 1

მოუწოდა 2

მოუწოდა (εκαλεσεν)

ὑπὸ τοῦ βασιλέως

ὁ βασιλεὺς 2

მეფემან 1

მეფემან (ο βασιλευς)

2 καὶ

ὑπέδειξεν γὰρ

L

La

347

cognoverat autem

Εσθηρ

რამეთუ უთხრა ესთერ (ester)

+ τω βασιλει a Aeth-Q

rex

მეფესა, (to the King)

ὅτι

quod

ვითარმედ

Mardocheus ἐνοικείωται αὐτῇ

erat de genere reginae

თჳსი მისი არს (is her own kin)

348

Natia Mirotadze

o' 2ʹ

L

La

GeI

GeII

καὶ

და

ἐχαρίσατο

მისცა

αὐτῷ

მას

πάντα

ყოველი

τὰ

სახლი (τον οικον)

τοῦ Αμαν.

ჰამანისი (hamanisi)

As can be seen, GeII carefully compares these two forms of the same story and adopts only distinctive parts from L: 2ʹ repeats the same information as 1 (i.e. everything once belonging to Haman was given to Mordecai), so it is ignored; while 1ʹ, which contains textual material more or less different from the o', is completely interwoven into the text of GeII.

III. GeII = o' GeII incorporates an individual addition from Septuagint Esther as well, which is a colophon at the end of the Book informing the reader when and by whom the text was translated. Table 719

F:11 oʹ

GeII

Ἔτους

წელსა

τετάρτου

მეოთხესა

βασιλεύοντος

მეფობასა

19  In the table Greek and Georgian texts of the colophon are aligned word-for-word. Proper names in Georgian are transcribed. To differentiate stem and case mark, stem is written in bold. Variant readings of Georgian versions unattested elsewhere, are translated in Greek if it can be reconstructed (back-translations are written without accents), otherwise they are written in English.

The Old Georgian Version of the Book of Esther—All in One

oʹ

GeII

Πτολεμαίου

ფტოლემეოსისსა (pt’olemeosissa)

καὶ

და

Κλεοπάτρας

კლეოპატრაჲსსა (k’leop’at’rajssa)

εἰσήνεγκεν

შემოიღო

Δοσίθεος

დოსიმეოს (dosimeos),

ὃς

რომელი

ἔφη

იტყოდა

εἶναι

ყოფად თავსა თჳსსა (εαυτον)

ἱερεὺς

მღდელად

καὶ

და

Λευίτης

ლევიტელად (levit’elad)

καὶ

და

Πτολεμαῖος

ფტოლომეოს (pt’olomeos)

ὁ υἱὸς

ძემან

αὐτοῦ

მისმან,

τὴν προκειμένην

წინაშე ესე

ἐπιστολὴν

ებისტოლე

τῶν Φρουραι

ფრურეთაჲ (pruretaj), რომელ არს ნაწილთაჲ (which is of lots),

ἣν

რომელი

ἔφασαν

თქუეს

εἶναι καὶ

>

ἑρμηνευκέναι

თარგმანებულად

Λυσίμαχον

ლჳსიმაქოსისგან, (lwsimakosisgan)

Πτολεμαίου τῶν] τον 58 670

a–71 130 370

ἐν Ιερουσαλημ

> 46-98-248 55 318

რომელი (τον) + იყო (ην) იერუსალჱმს (ierusalēms)

349

350

Natia Mirotadze

Georgian faithfully resembles the Greek linguistically repeating its grammatical and lexical forms. This leads us to the conclusion that those few quantitative changes found between the texts can be attestations of different Vorlage. The Colophon is the final verse of the o' text. Obviously, it does not have any contextual connection to the narrative. As mentioned above, in GeII it is followed by the letter of Mordecai. Because of this sequence, in the contaminated text, not only the letter, but also the colophon (as paratextual element), are in the wrong places.

IV. GeII = GrX The king’s edict addressed to the male inhabitants of the kingdom, and some parts of Add. F that is an interpretation of Mordecai’s dream (Add. A), are additions of GeII that do not have direct equivalents in any existing sources.

IV.1 The King’s Edict Addressed to the Male Inhabitants of the Kingdom The King’s edict addressed to the male inhabitants of the kingdom is added at the end of chapter 1. The context of the addition is as follows: When Astin/ Ouastin (Αστιν [o'] / Ουαστιν [L]) refused to enter in front of the king, he was enraged and asked his advisors what to do. One of them, Muchaios/Bougaios (Μουχαιος [o'] / Βουγαῖος [L]), stated that all women throughout the kingdom would be encouraged by the queen’s behavior and could become disobedient to their husbands. Therefore, he suggested that the king issue a law ordering all women to submit to their husbands and all men must have foremost positions in their families. The king was pleased with the advice, issued the law, and sent it throughout the entire kingdom. In all existing textual forms of Esther, chapter 1 ends here, but GeII continues with the text of the edict composed in the first person which is linked to the preceding text with the phrase, ‘and this was written in the letter’: და ესე წერილ იყო მათ წიგნსა მას, რომელი მიუძღუნა: მეფე დიდი ასჳრეოს მთავართა და ჴელმწიფეთა, რომელნი ხართ ქუეშე სამეუფოსა ჩემსა ას ოცდაშჳდთა ნათესავთა ჰინდოეთითგან ვიდრე ეთიოპიადმდე მკჳდრთა. ამას იტყჳს მეფჱ: მთავრობდინ ცოლსა თჳსსა და სახლსა თჳსსა ყოველი მამაკაცი! And this was written in the letter, which was sent out: Assyeros the great king to the rulers and governors, you who are under my reign, of one hundred and twenty-

The Old Georgian Version of the Book of Esther—All in One

351

seven nations, inhabitants from India to Ethiopia. The king says as follows: Every man must be the ruler in his home and over his wife!

The edict under discussion was composed by applying the same model used in the case of other edicts and letters in the Book of Esther written by various officials for various purposes (i.e. B, E and Mordecai’s letter). Schematically, the model consists of the following elements: 1) briefly formulated content of the letter; + 2) the note that it was written down and sent out by couriers; + 3) the phrase, ‘This is a copy of the letter/this was written in the letter; + 4) the letter itself.

VI.2 Interpretation of Shouts/Confusion/Thunder/ Earthquake/Chaos on the Earth and Light in Add. F Add. F of GeII also has more textual material than in Greek text-forms and other daughter versions. To understand what is added and why, the content and function of Add. F must be discussed more closely. F is an interpretation of Mordecai’s prophetic dream—A:4–10 (L A:3–8). Therefore these two additional chapters are functionally connected. The dream placed in the beginning of the book is a symbolic depiction of the events which are described in the following chapters. Addition F, ending the book in turn, is an inner dialogue of Mordecai interpreting the dream to himself and to readers. Therefore, he in Add. F one by one explains the meaning of the elements of the dream (e.g. a small spring which became a big river is Esther who was made queen...). However, some details (the shouts, confusion, thunder, etc.) are not explained in the Greek texts and in the Old Latin version of the interpretation. Some of them are defined in additional material of Add. F in GeII, while others are interpreted in a different way, and some explanatory phrases are also added. I will discuss only part of this material to illustrate the character of F in GeII.

1) F:3

To gain a better understanding of how verses 3–5 are represented in various textual forms, I will display the text of these verses according to each textual form. Underlined parts designate that GeII corresponds to the o' text, and what are written in italics are additions. A translation of the whole episode of GeII is included in footnotes: o' F 3 ἡ μικρὰ πηγὴ, ἣ ἐγένετο ποταμός καὶ ἦν φῶς καὶ ἥλιος καὶ ὕδωρ πολύ· Εσθηρ ἐστιν ὁ ποταμός, ἣν ἐγάμησεν ὁ βασιλεὺς καὶ ἐποίησεν βασίλισσαν. 4 οἱ δὲ δύο δράκοντες ἐγώ εἰμι καὶ Αμαν 5 τὰ δὲ ἔθνη τὰ ἐπισυναχθέντα ἀπολέσαι τὸ ὄνομα τῶν Ἰουδαίων.

352

Natia Mirotadze

L 9(F) 54(3) Ἡ μικρὰ πηγὴ Εσθηρ ἐστίν, (4) καὶ οἱ δύο δράκοντες ἐγώ εἰμι καὶ Αμαν· (5) ποταμὸς τὰ ἔθνη τὰ συναχθέντα ἀπολέσαι τοὺς ᾿Ιουδαίους· ἥλιος καὶ φῶς ἣ ἐγένετο τοῖς ᾿Ιουδαίοις ἐπιφανεία τοῦ θεοῦ, τοῦτο τὸ κρίμα.—GeI 9(F) 54(3) მცირჱ იგი წყაროჲ ესთერ არს. (4) და ორნი იგი ვეშაპნი მე ვარ და ჰამანი. (5) და მდინარჱ იგი არიან წარმართნი, რომელნი შეკრბეს ჰურიათათჳს წარწყმედად. მზჱ იგი და ნათელი, რომელცა იყო ჰურიათა ზედა, ღმრთისა ესე შეწევნაჲ. და ესე არს განმარტება მისი. La F 3 pusillus fons Hester est, 4 duo enim dracones ego eram et Aman, 5 flumina autem gentes, quaecumque venerant perdere nomen Iudaeorum. sol et luna erant Iudaei hoc iudicium GeII F 3 რამეთუ მცირე იგი წყაროჲ, რომელი იქმნა მდინარე დიდ, ესთერ არს, რომელი შეირთო მეფემან. (a) ხოლო ჴმანი იგი და შფოთი, ქუხილნი და ძრვანი და შეძრწუნებანი ქუეყანასა ზედა (b) წარმართთანი იგი არიან, რომელნი შეკრებულ იყვნეს ჩუენ ზედა წარწყმედად ჩუენდა. (c) ხოლო ნათელი, რომელი გამობრწყინდა ჩუენ ზედა, და მდაბალთა შთანთქნეს დიდებულნი, (d) საშჯელი ღმრთისაჲ არს, რომელი მიყო ჩუენ საუკუნემან,20 5 ოდეს-იგი შეკრბეს ჩუენ ზედა წარწყმედად სახელისა ჩუენისა წარმართნი. 4 ხოლო ორნი იგი დევნი მე ვარ, მარდოქე, და ამან ნათესავისა ჩუენისა მბრძოლი, ოდეს-იგი ზრახა დამოკიდებაჲ ჩემი ძელსა, ხოლო ღმერთმან მიაქცია ბოროტი მისი მისავე ვიდრე აქა დღედმდე

1) In GeII, verse F:3 is shorter than in the o' (om ἦν φῶς καὶ ἥλιος καὶ ὕδωρ πολύ GeII; om ὁ ποταμός GeII; om καὶ ἐποίησεν βασίλισσαν 71 AethNP* GeII), but is still longer than in L and La. 2) The addition after verse 3 is composed using the material from different sources. (a) and (c) are details of the dream needing interpretation. They are taken from Add. A and repeat its sentences almost without change, mixing textual material from the o' and L. (b) and (d) are interpretations taken from Add. F of L, though (d) has some extra material as well. The comparison of (a) and (c) to the all existing forms (o' L La GeI and GeII) of A:4 and A:10 on the one hand, and of (b) and (d) to 7:54(5) of L La and GeI on the other, shows that the additional material of F:3 is not created by the Georgian editor by combining parts of the 20  For the small spring that became great river, is Esther whom the King married. And the shouts and confusion, the thunders, earthquakes, and the chaos on the Earth of the pagans are those, which had been gathered to destroy us. And the light which shone out upon us, and the lowly swallowed up the greats, is the judgment of the God which the Eternal One made for us when the pagans were gathered to destroy our name. And the two dragons are I, Mordecai, and Haman who was fighting against our nation, when he intended to crucify me on the pole and God returned his own evil on him until this day.

353

The Old Georgian Version of the Book of Esther—All in One

Georgian Esther. The additional material in GeII is not identical to the corresponding parts of the textual forms currently found in the o', L and La.21 Therefore, an entire additional fragment composed by combining the textual data of the o', L and GrX must have been inserted in the contaminated text from a nowlost text form of Esther. To represent clearly how the components ((a), (b), (c) and (d)) of this addition are related to the corresponding parts of all textual forms, they are aligned wordfor-word in the following tables: Table 8(a)22 oʹ—A

L—A

La—A

A καὶ

A3(4) =

et

ἰδοὺ

=

ecce

4

GeII—A

GeI—A

აჰა ესერა,

მუნქუეს ოდენ

GeII—F

ხოლო (δε) φωναὶ

φωνὴ

voces

καὶ

=

et

ჴმანი,

κραυγὴ θόρυβος

ჴმანი იგი (= oʹ)

და

=

ღაღადებაჲ

θορύβου tumultus

+ και 74ʹ 392 La Aeth Arm

ჴმაჲ

შფოთი

ამბოხისაჲ

შფოთი,

et

და ქუხილისაჲ ქუხილნი

βρονταὶ

=

tonitrua

καὶ

=

et

და

=

=

σεισμός (σεισμοι O–93)

=

terremotus

ძრვა

=

ძრვანი (=O–93)

+ και rel = L] > BS* V 967 55 108 249 318 542

καὶ

et

და

=

=

21  The Armenian version of Add F follows the Septuagint text form. 22  In Tables 8(a), 8(b), 8(c) and 8(d) texts in forms of Greek texts and Latin and Georgian versions are aligned word-for-word. Variant readings of Georgian versions unattested elsewhere, are translated in Greek if it can be reconstructed (back-translations are written without accents), otherwise they are rendered in English.

354

Natia Mirotadze

oʹ—A

L—A

τάραχος

=

+ μεγας 392 La-V Arm ἐπὶ τῆς γῆς

=

La—A

GeII—A

GeI—A

conturbatio შეძრწუნებაჲ = magna

დიდი

super terram

ქუეყანასა ზედა

=

GeII—F შეძრწუნება­ ნი (pl)

=

Table 8(b) oʹ—F

L—7 54(5)ποταμὸς

La

GeII—F

GeI—7

autem 2

და (και)

flumina 1

მდინარჱ იგი არიან (εισιν)

τὰ ἔθνη

gentes

წარმართთანი იგი (των εθνων)

წარმართნი,

არიან (εισιν) τὰ συναχθέντα

quaecumque რომელნი venerant შეკრებულ იყვნეს (= La)

რომელნი შეკრბეს

ჩუენ ზედა (εφ᾽ ημας) ἀπολέσαι τοὺς ᾿Ιουδαίους·

perdere

წარწყმედად

nomen

ჩუენდა (ημας)

Iudaeorum

წარწყმედად. 2 ჰურიათათჳს 1

355

The Old Georgian Version of the Book of Esther—All in One

Table 8(c) oʹ—A

L—A

La—A

GeII—A

GeI—A

GeII—F ხოლო (δε)

და 10 φῶς

8(10) =

lux

καὶ ὁ ἥλιος

ἥλιος

solis

ნათელი

=

და

=

მზე

მზჱ

=

რომელი (το) ἀνέτειλεν,

=

illuxit

გამობრწყინდა =

= ჩუენ ზედა (εφ᾽ ημας)

καὶ

et

და

=

=

οἱ ταπεινοὶ οἱ ποταμοὶ

humiles

მდაბალნი (= oʹ)

მდინარენი იგი (= L)

მდაბალთა (= oʹ La)

ὑψώθησαν

=

exaltati sunt

აღმაღლდეს

=

καὶ

=

et

და

=

κατέφαγον κατέπιον

comederunt

შეჭამნეს (=oʹ)

დაინთქნეს (= L)

τοὺς ἐνδόξους.

gloriosos

დიდებულნი

პატიოსანნი დიდებულნი იგი

=

=

შთანთქნეს (= L)

მათნი (αυτων)

Table 8(d) oʹ—F

L—7

La

54(5) τοῦ

ღმრთისაჲ 2

θεοῦ,

τοῦτο

GeII—F

ღმრთისა და ესე არს

hoc

τὸ κρίμα. iudicium

GeI—9

საშჯელი 1 არს (εστιν)

განმარტება მისი

356

Natia Mirotadze

oʹ—F

L—7

La

GeII—F

GeI—9

რომელი მიყო ჩუენ საუკუნემან, (which eternal did for us)

F:4

Addition from an unknown source is included after verse F:4, which on its turn is transposed after verse F:5. Table 923 oʹ—F 4 οἱ

δὲ

L—7

La

GeII—F

GeI—9

(4)καὶ

enim 2

ხოლო (= oʹ La)

და (= L)

δύο

οἱ δύο

duo 1

ორნი იგი

=

δράκοντες

=

dracones

დევნი

ვეშაპნი

ἐγώ

=

ego

მე

=

εἰμι

=

eram

ვარ,

=

მარდოქე (mardoke) καὶ

=

et

და

=

Αμαν·

=

Aman

ამან (aman)

ჰამან (haman)

ნათესავისა ჩუენისა მბრძოლი ოდეს-იგი ზრახა დამოკიდებაჲ ჩემი ძელსა, ხოლო ღმერთმან მიაქცია ბოროტი მისი მისავე ვიდრე აქა დღედმდე (who was fighting against our nation, when he intended to crucify me on the pole and God returned his own evil on him until this day)

23  In the table, texts in forms of Greek texts, and Latin and Georgian versions are aligned word-for-word. Proper names are transcribed, and variant readings of the Georgian versions unattested elsewhere are translated into English.

The Old Georgian Version of the Book of Esther—All in One

357

Conclusions 1. Textual Type of GeII: The text of GeII is a conglomerate created through contamination, full of doublets, and sometimes even grammatically incomprehensible. 2. Sources Used in Contamination: The sources used in contamination are a. o' Esther; b. L text; c. GrLa; d. GrX (now traceable due to passages of GeII, which are not found in any other surviving sources). The fact that some structural characteristics are maintained despite severe contamination, indicates that the base text for contamination was a text-type similar to the o' Esther. 3. Character of the Contamination of GeII: All existing textual forms that were at the editor’s disposal are interwoven. In some instances, the coherent character of the resulting text is maintained, while in others, parts of the various textual forms are mechanically attached to each other without any justification. In short, occasionally GeII does not represent a smooth-reading text. It is possible to formulate the only logic, which can be seen through this conglomerate as thus: collection and documentation. Even if the editor aimed to create a coherent text, from time to time he failed to achieve his goal. This is because he was trying to harmonize sometimes contextually incompatible material at his disposal, whilst aiming to document and preserve as much as possible.

Text Critical Signs and Abbreviations = Identical reading > omission + addition tr transposition * former reading of the manuscript ] mark of lemma | end of lemma r recto cf confer MT Masoretic Text om omisit—omission post after pr praemittit—addition in front of lemma o' // LXX – The Septuagint L // AT – The second Greek text, L text, Alpha Text GrLa – The third Greek version preserved in Old Latin Translation

358 GrX La

Natia Mirotadze

–  Unknown Greek version preserved through the additions of Georgian Translation –  Vetus Latina—Old Latin translation of the unpreserved Greek text

Vetus Latina R I J F

– 151 130 (partly) 155 and PS-AU – 123 and its satellites: 135 136 137 138 139 140 141 142 154 159 159B 159F 159G,130 (partly) 152, AM HI O H – 146 – 109 AM SUL and PROS

Georgian GeI – Georgian version of the Book of Esther, translation of the L Text source: O—Ath.-Iver.-1 (so-called Oshki Bible, 978) GeII – Georgian version of the Book of Esther, translation of the contaminated text source: M—M-1694.4, Fr.14-1, 7125 d(4)r (Matenadaran fly-leaves, XIII cent.) I—A-570 (1460) F—A-646 (XV—XVI) D—H-885 (XVII) S—A-51(Mtskheta Bible XVII—XVIII)

Bibliography De Troyer, K., “The Letter of the King and the Letter of Mordecai: An Analysis of MT & LXX 8.9–13 and AT 7.33–38”, in Textus 21 (2002) 175–207. De Troyer, K., Rewriting the Sacred Text: What the Old Greek Texts Tell Us about the Literary Growth of the Bible (TCSt 4; Leiden: Brill, 2003). Haelewyck, J.-C. (ed.), Hester, Band 7/3 (Freiburg: Herder, 2003–2008). Haelewyck, J.-C., “The Relevance of the Old Latin Version for the Septuagint, with Special Emphasis on the Book of Esther”, JTS (2006) 439–73. Hanhart, R. (ed.), Esther, in Septuaginta, Vetus Testamentum Graecum, Auctoritate Academiae Litterarum Gottingensis, editum vol. VIII/3 (Göttingen: Vandenhoeck & Ruprecht, 1966). Moore, C.A., Daniel, Esther and Jeremiah: The Additions (AB 44; Garden City, NY: Doubleday, 1977).

The Old Georgian Version of the Book of Esther—All in One

359

Moore, C.A., “On the Origin of the LXX Additions to the Book of Esther”, JBL 92 (1973) 382–93. Tov, E., “The ‘Lucianic’ Text of the Canonical and Apocryphal Sections of Esther: A Rewritten Biblical Book”, in The Greek and Hebrew Bible: Collected Essays on the Septuagint (VTSup 72; Leiden: Brill, 1999) 535–48.

Magda Mtchedlidze A Translation, Paraphrase, or Metaphrasis? Regarding Euthymius the Hagiorite’s Versions of the Orations by Gregory the Theologian The essence of a translation, its limits, and the notion of equivalence—these and other theoretical problems naturally face a medievalist when working on texts that have been translated from the Greek. The peculiarities of the version of Gregory the Theologian’s Orations by Euthymius the Hagiorite—an especially authoritative figure for the Georgians who was active during the late 10th and the first quarter of the 11th century—and the difficulties associated with defining the character of the Georgian text will be examined precisely in relation to these problems. Firstly, it should be noted that in the ancient Georgian language there were several words denoting ‘translating’, including ‘თარგმანება’1 (‘targmaneba’—a word of Aramaic origin), which meant a rendering of what had been said into another language, i. e. a translation proper. It also meant conveying the communication into a different form in the same language, in a simpler or more explanatory manner (i. e. paraphrase), as well as commentary, exegesis,2 or everything associated with grasping the essence of the text and its rendering.3 ‘Targmani’ (the noun form) and ‘targmaneba’ (the infinitive form) specifically correspond in ancient Georgian texts to the following Greek words: ἐξήγησις, ἑρμηνεία, ἀνάπτυξις, ἔμησις, ἔκδοσις, μεταφράζειν, σαφηνίζειν, δηλοῦν.4 It is known that the 1  Various terms corresponding to translation can be confirmed even with one and the same author. For example, these words are found in the works of Ioane Petritsi: ‘გარდმოღება’ (შ. ნუცუბიძე, ს. ყაუხჩიშვილი (გამ.), იოანე პეტრიწის შრომები [Sh. Nutsubidze/S. Kaukhchishvili (ed.), The Works of Ioane Petritsi] (2 vol.; Tbilisi: Tbilisi State University Press, 1937, 1940), 2. p. 219, 1–2; 227, 5–6, ‘გარდამოენება’ (Petritsi, 2. p. 108, 28), ‘თარგმანი’ (Petritsi, 2. p. 222, 27). 2  კ. კეკელიძე, ძველი ქართული ლიტერატურის ისტორია [K. Kekelidze, The History of the Old Georgian Literature] (2 vol.; Tbilisi: Metsniereba, 1980), 1.457–8. 3  ‘Targmani’ is certainly not the only term denoting translation, in which the understanding of a text and its rendering into another language are interconnected (Cf., for instance, the Latin term ‘interpretatio’ and its loan words in modern languages). 4  ს. ყაუხჩიშვილი, ლ. კვირიკაშვილი (გამ.), ბერძნულ-ქართული დოკუმენტირე­ ბული ლექსიკონი [S. Kaukhchishvili/L. Kvirikashvili (ed.), Greek-Georgian Documented Dictionary] (7 vol.; Tbilisi: Program “Logos”, 2002–2007), 6.320. The second basic meaning of “targmani” is also confirmed with Ioane Petritsi, whereas in one case, this word is mentioned two times in the same sentence with different meanings: “translation” and “explanation” (Petritsi, 2. p. 6, 25–6).

362

Magda Mtchedlidze

Greek term μεταφράζειν meant both translation into another language (interlingual translation) and conveying the meaning of the words into another form (intralingual translation). With such a relation between translation and exegesis, the translator’s hermeneutical theory and method may also define the character of his translation.5 Of course, it is not by chance that Ephrem Mtsire (11th century),6 who is known for his literal translations as well as for his theory in translation, was active on the Black Mountain close to Antioch.7 The tradition of the Antiochian school of exegesis, being distinguished for a faithfulness towards the text, must have certainly influenced translation theories. Ephrem not only shows a preference for literal translation, but also a desire for a clear differentiation between a text’s translation and its exegetical commentary. Ephrem Mtsire had also theoretically formulated (by giving reference to his own teachers and the Greek literary tradition) the stances toward various types of texts that had naturally become established in the praxis of Georgian translation throughout the course of time. According to him, it is impermissible to revise the writings of the Holy Fathers; they must be translated “simply, purely and without change”, or literally with the accompanying commentary being written separately on the margin. They cannot be approached in the same way that it is possible to 5  Clearly, the abilities of certain translators to convey the content of the text correctly in an understandable form must not be ruled out. Instances of literal, as well as free translations of biblical text into Georgian are already attested in the so-called Khanmeti texts from an early time period, which supposedly must be caused by this. For the old Georgian versions of the Bible, cf. ა. ხარანაული, “ბიბლიის ძველი ქართული თარგმანების ტექნიკა (ხანმეტი ესაიას, იერემიას და ეზრა I-ის ფრაგმენტების მაგალითზე)” [A. Kharanauli, “Translation Technique of the Old Georgian Bible (According to the Khanmeti Fragments of Isaiah, Jeremiah and Ezra I)”], in ნ. მახარაძე, მ. გიორგაძე (რედ.), ბიზანტოლოგია საქართველოში, ეძღვნება აკადემიკოს გრიგოლ წერეთლის ხსოვნას [N. Makharadze/M. Giorgadze (ed.), Byzantine Studies in Georgia 2. Dedicated to Academician Grigol Tsereteli] (Tbilisi: Program “Logos”, 2009) 938–68. 6  Ephrem Mtsire—a 11th century Georgian writer and translator of works by many important Byzantine authors, including Gregory the theologian, Pseudo-Dionysius the Areopagite, John of Damascus and others. He is considered to be the introducer of so-called Hellenophilism in Georgian literature. 7  At this time, as we know, the Black Mountain represented a kind of international center of knowledge and culture, where representatives of various nationalities (Greeks, Syrians, Armenians, Georgians) actively translated each other’s literary works within its monasteries. Thus a beneficial environment was created for the development of translation theories and the interchange of philological methods for studying the texts and the technical experience for their redaction. Particularly in regard to the translation methods of the Syrians and the manifestation of their tradition with Ephrem Mtsire, cf. ნ. დობორჯგინიძე, ლინგვისტურჰერმენევტიკული მეტატექსტები. პრაქტიკული გრამატიკა და ჰერმენევტიკა X– XIII საუკუნეების ქართულ წყაროებში [N. Doborjginidze, Linguistic and Hermeneutic Metatexts: Practical Grammar and Hermeneutics in Georgian Sources of the 10th–13th Centuries] (Tbilisi: Ilia State University Press, 2012), 157–60.

A Translation, Paraphrase, or Metaphrasis?

363

approach hagiographical or other narratives written by laymen, which have usually been verbally transformed or “embellished”, being called a metaphrasis. This is parallel to the Gospels and the Pauline Epistles, which no one dares to touch, even if they have been written in simple phrases, as well as the sort of works similar to Fountain of Knowledge by John of Damascus.8 The Antiochian literary school, whose representative was Ephrem Mtsire, was preceded by the school of Mount Athos, an especially important cultural center in terms of translators, with its enlightening literary activities being connected to the Georgian monastery of Iveron built there at the end of the 10th century. The Athonites were driven by the desire for all writings enjoying great authority in the Byzantine Church during this time period to be in the Georgian language. Thus they translated works that had heretofore been untranslated and also retranslated earlier versions in order to correct them according to the Greek originals. Euthymius the Hagiorite was the head of this school (approx. 955–1028).9 The works of Euthymius the Hagiorite and his translation methods have been well researched. Ketevan Bezarashvili has summarized on the whole the features peculiar to his translations of the orations by Gregory of Nazianzus, which are namely: textual additions or subtractions, an inclusion of his own commentary into the text, partial translation, compositional changes, interpolations, periphrases, epitomization, and compilations, “with some redactions sometimes

8 Ephrem, Foreword, in მ. რაფავა (გამ.), იოანე დამასკელის „დიალექტიკა“ [M. Raphava (ed.), John of Damascus. Dialectics (Tbilisi: Metsniereba, 1976) 67–68. In regard to the theoretical views of Ephrem Mtsire, cf. ქ. ბეზარაშვილი, რიტორიკისა და თარგმანის თეორია და პრაქტიკა გრიგოლ ღვთისმეტყველის თხზულებათა ქართული თარგმანების მიხედვით [K. Bezarashvili, Theory and Practice of Rhetoric and Translation: A Study of Georgian Translations of Gregory the Theologian’s Writings] (Tbilisi: Metsniereba, 2004), 123–46. 9  Euthymius the Hagiorite translated works of all the genres of spiritual literature. Apart from this, he has also translated the mythological commentary of Pseudo-Nonnus for the purpose of education. Euthymius has compiled a manual “The Guide of Orthodoxy” based on the works of John of Damascus, Anastasius of Sinai, Michael Syngellos, et. al. (cf. N. Chikvatia’s introduction in თ. ჭყონია, ნ. ჩიკვატია (გამ.), წმ. ეფთვიმე მთაწმინდელი. წინამძღუარი (სარწმუნოებისათვის) [T. Chkonia/N. Chikvatia (ed.), Euthymius the Athonite, Odegos] (Tbilisi: Artanuji, 2007), 136). Euthymius also translated from Georgian to Greek. According to some information by Giorgi the Hagiorite (which is corroborated by many Georgian and foreign researchers), a popular story about the spread of Christianity in India—The Life of Barlaam and Ioasaph (a re-worked version of the life of Buddha, as it is considered) had been translated into Greek by Euthymius the Hagiorite himself. The Georgian version supposedly must have been translated from Arabic in the 9th–10th centuries. Later, the work was translated from Greek into many languages (cf. the introductory letter in the book: ვ. გაფრინდაშვილი (მთარგ.), ვარლაამისა და იოასაფის ცხოვრება [V. Gaprindashvili (tr.) The Life of Barlaam and Ioasaph] (Tbilisi: Program “Logos”, 2005), 8–19.

364

Magda Mtchedlidze

becoming completely different from the original as a result”.10 Maia Rapava has defined Euthymius’ translations as “having an exegetical and metaphrastic quality”,11 whereas Edisher Chelidze has pointed out his original theological interpretations of statements by Gregory.12 Despite this, Georgian texts that have been converted from the Greek by Euthymius are defined as translations, specifically as free translations, expository translations,13 or translations of a dynamic equivalence type.14 These are frequently acceptable, since Euthymius had adapted the original text for educational purposes, adequately conveying Gregory’s aspirations and bringing across the main ideas of his works to Georgian readers, but in certain instances, the work of Euthymius seems to exceed the limits of translation, and thus it is necessary to define it even further. The peculiarities characteristic of Euthymius when conveying a Greek text into Georgian had already become well known to those who continued on with his work. Respect was accorded to his work, since his approach towards the text was attributed to the grace of the Holy Spirit,15 but it was considered impermissible to emulate him. Thus, the works that had been rendered by Euthymius were frequently translated a second time, with the aim of maximally conveying the original text in a literal manner.16 10 Bezarashvili, Rhetoric and Translation, 11; 44–48. A lot of research regarding the activities and translation methods of Euthymius have been referenced in this same work. 11  M. Raphava, “Les traductions Géorgienes du Discours 39 de Grégoire de Nazianze”, Collection. A. B. Schmidt (ed.) Studia Nazianzenica II (Corpus Christianorum Series Graeca 73; Corpus Nazianzenum 24; Turnhout-Leuven: Brepols, 2010) 491–517, on p. 516. 12  Cf. “Introduction” in H. Metreveli et alii (ed.), Gregorius Nazianzenus, Opera: versio Iberica, I: Orationes I, XLV, XLIV, XII (Corpus Christianorum Series Graeca 36; Corpus Nazianzenum 5; Turnhout-Leuven: Brepols, 1998), pp. XIX–XX. 13 Cf. თ. ოთხმეზური, კომენტარული ჟანრი შუა საუკუნეების ქართულ მთარგ­ მნელობით ტრადიციაში, ეფრემ მცირე და გრიგოლ ღვთისმეტყველის თხზულებათა კომენტარები [T. Otkhmezuri, The Commentarial Genre in the Medieval Georgian Translation Tradition, Ephrem Mtsire and Commentaries on the Sermons of Gregory the Theologian] (Tbilisi: Ilia State University Press, 2011), 77. 14 Bezarashvili, Theory and Practice, 48. 15  The work of Euthymius was evaluated in the following way: “He had authority through the Holy Spirit to add or reduce [the text] as well” (Ephrem Mtsire’s Introduction to his translation of John of Damascus, in Raphava, Dialectics, 67); “The grace of the Holy Spirit was abundantly stored up within him” (ე. გაბიძაშვილი, ი. აბულაძე (გამ.), ძველი ქართული აგიოგრაფიული ლიტერატურის ძეგლები, IV, სვინაქსარული რედაქციები (XI–XVIII სს.) [E. Gabidzashvili/I. Abuladze (ed.), Monuments of Old Georgian Hagiographic Literature, IV, Synaxarion Redactions (11th–18th centuries)] (Tbilisi: Metsniereba, 1968), 333. 16  Clearly, Euthymius has not been an exception in this regard. Such an attitude towards the text was also characteristic for Syriac translators and scribes, to say nothing of those Greek editors who used to supplement one or another of his Orations with excerpts from his other works (Bezarashvili, Theory and Practice, 30–7). The task of approximation to the parent text was also characteristic of Syriac translations at a certain time (See Doborjginidze, Metatexts, 157).

A Translation, Paraphrase, or Metaphrasis?

365

One moment has to be taken into account when analyzing Euthymius’ work. This is the development of metaphrastic literature in 10th century Byzantium.17 Yet, in addition to intellectual currents and hermeneutic approaches toward the text, it is apparent that the literary process also influences the translator as a man of letters (especially since conceptual motives and not only stylistic ones had also been the basis of the metaphrastic genre). It was precisely because of this reason that Ephrem Mtsire, as mentioned previously, considered it necessary to define the boundaries of metaphrasis for Georgian translators.18 Considering all that has been mentioned, it will be necessary to define in what sense the terms presented in the title are used in the paper before an examination of the Georgian texts can proceed.19 Translation—rendering a text from one language into another, where the content and the form, or only the content is equivalent to the source (in this case syntactic or lexical changes made by translator are exclusively conditioned by the difficulties of literal translation). Paraphrase—conveying the original text through simplifying it, reducing it, adding commentary, or making an interpretative expansion of what has been said.20 Paraphrase can be also regarded as a kind of adapted translation. Metaphrasis21—a creative reworking of the original text, a motivated change in the literary-stylistic and conceptual character (motivated generally by the translator’s critical stance towards the source material). It may be said that a metaphrast is a co-author of the text. Two passages from Gregory of Nazianzus’ Oratio 29 and Oratio 40 will be examined in the paper, which were rendered into Georgian by Euthymius the

17  The reworking of a certain type of literature in Byzantium had a place from an earlier period, but metaphrasis as the conceptualization of a genre is essentially connected to the name of Simeon in the 10th century (called the Metaphrast precisely because of this), who had edited hagiographic legends in correspondence with the theological thinking and literary tastes of his time. 18  It will be noted that metaphrasis is associated more with an expansion (the Byzantines denoted a reduction with the term “epitome”). See M. Hinterberger, “Byzantine Hagiography and its Literary Genres. Some Critical Observations”, in S. Efthymiadis (ed.), The Ashgate Research Companion to Byzantine Hagiography, Volume II: Genres and Contexts (Dorchester: Ashgate Publishing, 2014) 25–60, on p. 40. 19  Discussions concerning terminology are not examined here. I only remark that in my opinion, the terminology of literary theory and the terminology of translation studies ought to be reconciled. 20  The Greeks always favored the literary genre of paraphrase (paraphrases of the Gospel texts are known, a setting of John’s Gospel by Nonnus of Panopolis in dactylic hexameter for instance). Nikita of Paphlagonia’s (10th century) commentary for the poetry of Gregory the Theologian has been called a paraphrase, where Gregory’s text has been conveyed in an explanatory manner. 21  The form “metaphrasis” has been chosen in order to differentiate this term from “metaphrase”, denoting a literal translation in English-language scientific literature.

366

Magda Mtchedlidze

Hagiorite. Oratio 40 has also been translated by Ephrem Mtsire.22 Before the analysis of Euthymius’ translation is to commence, the versions of Euthymius and Ephrem will be presented in parallel fashion, with the comparison highlighting two types of stances towards Gregory’s text or the text of an author who is especially honored by Eastern Christianity and is cited and commented upon like the Holy Scriptures. Changes of the source text in the Georgian version will be denoted with a line: One line denotes the syntactic or lexical freedom of translation, which has been caused by a consideration of the peculiarities of the Georgian language; paraphrases (including exegetical expansions) taking into consideration the background of target text readers have been denoted with two lines; metaphrases— conceptual additions or differences—have been marked with a broken line. Words left out in Euthymius’ text have been highlighted in the Greek text with a dotted line. Gr. Naz.,Oratio 40, 223

Version by Euthymius24

Version by Ephrem25

Τρισσὴν γέννησιν ἡμῖν οἶδεν ὁ λόγος· τὴν ἐκ σωμάτων, τὴν ἐκ βαπτίσματος, τὴν ἐξ ἀναστάσεως.

სამნი შობანი უწყებულ არიან ჩუენდა: ერთი იგი ჴორცთაჲ და მეორე— ნათლისღებისაჲ და მესამე—აღდგომისაჲ. ამათგანი ერთი იგი ბნელისაჲ არს და მონაჲ და ვნებული; ხოლო მეორე – ნათლისაჲ და აზნაური, დამჴსნელი ვნებათაჲ, რომელი ყოველსავე შობითგანსა ბიწსა

სამნი შობანი უწყნის ჩუენდა სიტყუამან: ჴორცთაგანი, ნათლისღებისაგანი, აღდგომისაგანი.

Τούτων δὲ, ἡ μὲν νυκτερινή τέ ἐστι καὶ δούλη καὶ ἐμπαθής· ἡ δὲ ἡμερινὴ καὶ ἐλευθέρα καὶ λυτικὴ παθῶν, πᾶν τὸ ἀπὸ γενέσεως κάλυμμα περιτέμνουσα, καὶ πρὸς τὴν ἄνω ζωὴν ἐπανάγουσα·

ამათგანი პირველი ღამისაჲ არს და მონაჲ და ვნებული; ხოლო მეორე— დღისაჲ და აზნაური და დამჴსნელი ვნებათაჲ, ყოვლისავე შობითგანისა საბურველისა

22  The Georgian versions of the Orations by Gregory the Theologian have been prepared for publication thanks to cooperation between the Institute of Georgian Manuscripts and Catholic University of Louvain. 23  C. Moreschini (ed.), P. Galley (tr.), Gregoire de Nazianze. Discours 38–41. Introduction, texte critique et notes (SC 358; Paris: Cerf, 1990), 198–200. 24 B. Coulie/H. Metreveli et alii (ed.), S. Gregorii Nazianzeni opera. Versio Iberica, V. Orationes XXXIX, XL (CCSG 58; Corpus Nazianzenum 20; Turnhout-Leuven: Brepols, 2007), 122–6. 25 Coulie/Metreveli, S. Gregorii Nazianzeni opera, 123–5.

A Translation, Paraphrase, or Metaphrasis?

Gr. Naz.,Oratio 40, 223

ἡ δὲ φοβερωτέρα καὶ συντομωτέρα, πᾶν τὸ πλάσμα συνάγουσα ἐν βραχεῖ τῷ πλάστῃ παραστησόμενον καὶ λόγον ὑφέξον τῆς ἐνταῦθα δουλείας καὶ πολιτείας, εἴτε τῇ σαρκὶ μόνον ἐπηκολούθησεν εἴτε τῷ Πνεύματι συνανῆλθε, καὶ τὴν χάριν ᾐδέσθη τῆς ἀναπλάσεως.

Ταύτας δὴ τὰς γεννήσεις ἁπάσας παρ’ ἑαυτοῦ τιμήσας ὁ ἐμὸς Χριστὸς φαίνεται τὴν μὲν, τῷ ἐμφυσήματι τῷ πρώτῳ καὶ ζωτικῷ·

367

Version by Euthymius24

Version by Ephrem25

განსწმედს და ცხორებისა მიმართ ზეცათაჲსა წარგჳძღჳს; ხოლო მესამე იგი უსაშინლესი არს და უმაღლესი, რომელი ყოველსავე დაბადებულსა შეჰკრებს მეყსა შინა წარდგომად წინაშე დამბადებელისა და სიტყჳს მიცემად მოქალაქობისა მისთჳს, რომელი ამას სოფელსა შინა აღასრულა თითოეულმან, გინათუ ჴორცთა ნებასა ოდენ დაემორჩილა, ანუ თუ სულსა შეუდგა და მადლისა მისგან მეორედ შობისა შეიკდიმა. ამათ სამთავე შობათა თავისა თჳსისა მიერ პატივსცა ქრისტე, უფალმან ჩემმან: ერთსა მას განკაცებითა თჳსითა, რომელმანიგი შეიმოსა ყოვლითურთ კაცებაჲ და სული იგი, რომელი პირველ შთაჰბერა ადამს (დაბ. 2.7), მიიღო მან თავადმან და ჴორცნი იგი მიწისაგანნი შეიმოსნა; და მეორესა მას— ნათლის-ღებითა, რომლითა ნათელ-იღო მან;

მომკუეთელი და ზენაჲს ცხორებისა მიმართ აღმყვანებელი; ხოლო მესამე— უსაშინლესი და უშემოკლებულესი, წუთსა შინა შემკრებელი ყოვლის დაბადებულისაჲ წარდგინებად დამბადებელისა და სიტყჳსმიმჴდელი აქაჲსათჳს მონებისა და მოქალაქობისა, გინა თუ მხოლოდ ჴორცთა ვინმე შეუდგა, ანუ სულსა თანა-წარჰყვა და შეიკდიმა მადლისაგან აღმოდაბადებისა.

ამათ ყოველთა შობათა თავისა მიერ თჳსისა პატივისმცემელად გამოჩნდების ქრისტე ჩემი: პირველსა მას – შთაბერვითა პირველითა და განმაცხოველებელითა (დაბ. 2.7);

368

Magda Mtchedlidze

Gr. Naz.,Oratio 40, 223

Version by Euthymius24

Version by Ephrem25

τὴν δὲ, τῇ σαρκώσει καὶ τῷ βαπτίσματι, ὅπερ αὐτὸς ἐβαπτίσατο· τὴν δὲ, τῇ ἀναστάσει, ἧς αὐτὸς ἀπήρξατο· ὡς ἐγένετο πρωτότοκος ἐν πολλοῖς ἀδελφοῖς (Rom 8.29), οὕτω καὶ πρωτότοκος ἐκ νεκρῶν γενέσθαι καταξιώσας (Col 1.18).

ხოლო მესამესა მას –აღდგომითა, რომლისა იქმნა დასაბამ; რომელიიგი ვითარცა იქმნა პირმშო მრავალთა ძმათა შორის (რომ. 8.29), ეგრეთვე თავს-იდვა პირმშო შესუენებულთა ყოფად (კოლ. 1.18).

ხოლო მეორესა— განჴორციელებითა და ნათლისღებითა, რომელმან ნათელიღო; ხოლო მესამესა— აღდგომითა, რომელმან დაიწყო; ვითარცა პირმშო-ყოფაჲ მრავალთა შორის ძმათა (რომ.8. 29), ეგრეთვე პირმშო-ყოფაჲ მკუდრეთით ღირსიჩინა (კოლ. 1.18).

The divergence between the two versions is obvious, appearing in the very first sentence: Τρισσὴν γέννησιν ἡμῖν οἶδεν ὁ λόγος· τὴν ἐκ σωμάτων, τὴν ἐκ βαπτίσματος, τὴν ἐξ ἀναστάσεως.26 სამნი შობანი უწყებულ არიან ჩუენდა: ერთი იგი ჴორცთაჲ და მეორე— ნათლისღებისაჲ და მესამე—აღდგომისაჲ [Three births have been announced to us: the one of the flesh, the second of baptism, and the third of the Resurrection]. (Euthymius)27 სამნი შობანი უწყნის ჩუენდა სიტყუამან: ჴორცთაგანი, ნათლისღებისაგანი, აღდგომისაგანი [The Word knows three births for us: [one] in the flesh, [one] by baptism, [one] by the Resurrection]. (Ephrem)28

It can be seen that Ephrem does a literal translation of the Greek phrase, retaining also the active construction of the source text in the Georgian. Euthymius however, has used a passive construction, leaving out ὁ λόγος (understood here by translators as being the Scriptures29).The phrase “have been announced” (used by Euthymius for οἶδεν ὁ λόγος) means that there exists knowledge that has been 26  Gr. Naz., Or. 40, 2, 1–2 (Moreschini/Galley, 198). 27  Iber. 40, 2, 4 (Coulie/Metreveli, 122). 28  Iber. 40, 2, 4 (Coulie/Metreveli, 123). 29  See, for instance, a French translation of Gregory’s text (Moreschini/Galley, 199).

A Translation, Paraphrase, or Metaphrasis?

369

received from another, whereas the source from where it has been received is made apparent by the context. Ephrem’s translation “The Word knows” is not a good enough phrase in Georgian.30 Also in other cases Ephrem has tried to construct Georgian phrases in emulation of the Greek syntax, with an active form being used for an active form, or a passive one for a passive form. Ephrem’s translation has been weighed down a bit by using participle forms in the Georgian through mimicking the participle forms of the Greek.31 In such a case however, Euthymius frequently shows a preference for using finite verb forms. An example of this is seen in the following passage: Τούτων δὲ, ἡ μὲν νυκτερινή τέ ἐστι καὶ δούλη καὶ ἐμπαθής· ἡ δὲ ἡμερινὴ καὶ ἐλευθέρα καὶ λυτικὴ παθῶν, πᾶν τὸ ἀπὸ γενέσεως κάλυμμα περιτέμνουσα, καὶ πρὸς τὴν ἄνω ζωὴν ἐπανάγουσα·32 ამათგანი ერთი იგი ბნელისაჲ არს და მონაჲ და ვნებული; ხოლო მეორე– ნათლისაჲ და აზნაური, დამჴსნელი ვნებათაჲ, რომელი ყოველსავე შობითგანსა ბიწსა განსწმედს და ცხორებისა მიმართ ზეცათაჲსა წარგჳძღჳს [Of these the one is that of darkness, and is slavish, and subject to affectivity; but the second is that of light, and is free, destructive of passions, which purifies all the stain existing from birth and leads us to the heavenly life].(Euthymius)33 ამათგანი პირველი ღამისაჲ არს და მონაჲ და ვნებული; ხოლო მეორე— დღისაჲ და აზნაური და დამჴსნელი ვნებათაჲ, ყოვლისავე შობითგანისა საბურველისა მომკუეთელი და ზენაჲს ცხორებისა მიმართ აღმყვანებელი [Of these the first is that of night, and is slavish and subject to affectivity; but the second is that of day, and is free, destructive of passions, cutting off the entire veil that exists from birth, and leading us to the higher life]. (Ephrem)34

With Euthymius and Ephrem, the Greek ἡ μέν... ἡ δέ here also, as well as in the first phrase, has been conveyed naturally through the abilities of the Georgian language: “the one (or the first)... the second...” It must also be noted that here 30  The differences between Euthymius’ and Ephrem’s versions, manifested in Euthymius’ use of syntactic constructions and expressions natural to Georgian and in the maximal imitation of the original by Ephrem, are also noted by researchers. Cf. for example, ნ. მელიქიშვილი, „რამდენიმე დაკვირვება ეფთვიმე ათონელისა და ეფრემ მცირის მთარგმნელობით მეთოდზე“ [N. Melikishvili, “Some Observations Regarding the Translation Method of Euthymius the Hagiorite and Ephrem Mtsire”], Matsne, SLL 4 (1987) 119–28, on pp. 120, 125. 31  Unlike the Greek, the participle has a less verbal character in Georgian. 32  Gr. Naz., Or. 40, 2, 3–6 (Moreschini/Galley, 200). 33  Iber. 40, 2, 4 (Coulie/Metreveli, 122). 34  Iber. 40, 2, 4 (Coulie/Metreveli, 123).

370

Magda Mtchedlidze

and elsewhere as well Euthymius has transferred over the Greek article (“ერთი იგი”, “მესამე იგი”, “მოქალაქეობისა მისთჳს”, “მადლისა მისგან”), which has been ignored by Ephrem.35 But first of all, the present passage is interesting in light of the interpretation by Euthymius. Specifically, in Gregory’s text πᾶν τὸ ἀπὸ γενέσεως κάλυμμα περιτέμνουσα is a complex metaphorical statement. On one hand, it refers to 2 Cor 3:15–18, where it is stated that the understanding of God has been hidden from the Jews by a veil (τὸ κάλυμμα), which is stripped away (περιαιρεῖται) upon turning towards Christ.36At the same time however, replacing the term περιαιρεῖται with the term περιτέμνουσα Gregory has also alluded to Col. 2:11–12, where the discourse centers upon spiritual circumcision, a removal of the pollution caused by the flesh through baptism.37 In light of the context of Gregory’s Or. 40, 2, τὸ κάλυμμα (‘a veil’) is understood to be the innate sinfulness which obscures God. Thus Gregory has taken two well-known metaphors: περιαιρεῖται τὸ κάλυμμα (“stripping away the veil”, in which seeing the Truth is implied as a result of turning towards the Lord) from 2 Cor. 3:16 and περιτομὴ ἀχειροποίητος (“the circumcision made without hands”, conveying the sense of a spiritual circumcision with Christ) from Col. 2:11–12, creating a new metaphor for a second birth through baptism—κάλυμμα περιτέμνουσα (“cutting away the veil”). This

35  Such forms have not been used in the Georgian language for some time, which has no articles. 36  It has also been stated here that there is freedom wherever the Spirit of God is, through which we also resemble the Lord. See 2 Cor 3:12–18, particularly 2 Cor 3:15–18: “ἀλλ’ ἕως σήμερον ἡνίκα ἂν ἀναγινώσκηται Μωϋσής κάλυμμα ἐπὶ τὴν καρδίαν αὐτῶν κεῖται· ἡνίκα δὲ ἐὰν ἐπιστρέψῃ πρὸς κύριον, περιαιρεῖται τὸ κάλυμμα. ὁ δὲ κύριος τὸ πνεῦμά ἐστιν· οὗ δὲ τὸ πνεῦμα κυρίου, ἐλευθερία. ἡμεῖς δὲ πάντες ἀνακεκαλυμμένῳ πρωσώπῳ τὴν δόξαν κυρίου κατοπτριζόμενοι τὴν αὐτὴν εἰκόνα μεταμορφούμεθα ἀπὸ δόξης εἰς δόξαν, καθάπερ ἀπὸ κυρίου πνεύματος.” The understanding of κάλυμμα as words of the Holy Scriptures obscuring their spiritual significance has also been accepted in Patristics (G.W.H. Lampe, A Patristic Greek Lexicon [Oxford, 1961], 699). Specifically 2 Cor 3:16 is attested when justifying an allegorical interpretation of the Holy Scriptures: with Origen (Or., De Pr. IV, 1, 6: 172–6), when he speaks about the light revealed by Christ who had cut the veil (περιαιρεθέντος τοῦ καλύμματος), hiding it in the Old Covenant (H. Crouzel/M. Simonetti [ed.; tr.], Origène,Traité des Principes, t. III, Livre III et IV [SC 268; Paris, 1980], 282), and also with Gregory of Nyssa in the Introduction to “Song of Songs” (Gr. Nyss. In Cant., Prol.), who states that the Apostle has called the transcendence from corporeal things to things of an intelligible nature a return to the Lord and a cutting of a veil (καλύμματος περιαίρεσιν) (F. Dünzl [ed.; tr.], Gregor von Nyssa, In canticum canticorum homiliae [3 vol.; Freiburg: Herder, 1994], 1.100, 17–19). 37  Cf. Col 2:11–12: “ἐν ᾧ καὶ περιετμήθητε περιτομῇ ἀχειροποιήτῳ ἐν τῇ ἀπεκδύσει τοῦ σώματος τῆς σαρκός, ἐν τῇ περιτομῇ τοῦ Χριστοῦ, συνταφέντες αὐτῷ ἐν τῷ βαπτισμῷ, ἐν ᾧ καὶ συνηγέρθητε διὰ τῆς πίστεως τῆς ἐνεργείας τοῦ θεοῦ τοῦ ἐγείραντος αὐτὸν ἐκ νεκρῶν·” (Cf. Rom 2:29).

A Translation, Paraphrase, or Metaphrasis?

371

metaphor signifies a purification from the innate sin preventing a person from perceiving the Truth, the Word of God, through the spiritual birth by baptism.38 Euthymius has translated “πᾶν τὸ ἀπὸ γενέσεως κάλυμμα περιτέμνουσα” as “which purifies all the stain existing from birth”, with Ephrem rendering it as “cutting off the entire veil that exists from birth”. It is as if Ephrem’s literal translation provides the means for comprehending all the nuances of Gregory’s metaphor, but the question is how intelligible would it have been for a reader during Euthymius’ time? Especially since the term ‘მოკვეთა’ (‘cut’), corresponding in Ephrem’s translation to περιτέμνειν, is not used in the Georgian language for the concept of circumcision and thus there is no association of “cutting off the entire veil” (“საბურველის მოკვეთა”) with spiritual circumcision. Euthymius has avoided Gregory’s complex metaphor, with the objective of presenting the concept to the reader in a simple manner. As a result of this, the understanding of an immanent sin as something obscuring the truth from the human mind has not been reflected in his translation, but instead a person’s emancipation from inherent impurities through the mercy of baptism has been clearly revealed, which is the main point of this text by Gregory. An avoidance of metaphors certainly signifies that the style of the author has not been conveyed, although it is possible to say that the primary accent of the original text, and appropriately, the equivalence of its contents have been preserved in Euthymius’ paraphrase. One more thing must be highlighted in the fragment cited. According to Gregory, the first birth is associated with the night (νυκτερινή), with the second birth being associated with the day (ἡμερινή). In Euthymius’ translation however, it is “darkness and light” instead of “night and day”. The gist is that he takes the semantics of Georgian words into full account,39 with the Georgian language having not adopted “day and night” as a metaphor for good and bad (despite the 38  It must be mentioned here that Ephrem Mtsire (11th c.) has himself used the metaphor of 2 Cor 3:12–18 to denote Euthymius’ feat in enlightening the Georgians, when he called him a “light” which “cut away the veil of ignorance from our minds” (ი. აბულაძე (რედ.), ძველი ქართული აგიოგრაფიული ლიტერატურის ძეგლები, II (XI–XV სს.) [I. Abuladze (ed.), Monuments of Old Georgian Hagiographic Literature, II (11th–15th Centuries)] (Tbilisi, 1968), 108. 39  It is apparent from the translation of another passage from the 40th Oration, that Euthymius has thought a lot about using ‘darkness and light’ for rendering νυκτερινή—ἡμερινή, where the discourse is in regard to illusory light: “τοῦτο καὶ σκοτία ἐστι, καὶ μεσημβρία δοκεῖ, τὸ τοῦ φωτὸς ἀκμαιότατον. Οὕτως ἀκούω τοῖς διαπαντὸς ἐν μεσημβρινῇ σκοτίᾳ φεύγουσι (Is 16:3) τοῦτο καὶ νύξ ἐστι, καὶ φωτισμὸς νομίζεται τοῖς ὑπὸ τῆς τρυφῆς διεφθαρμένοις” (Gr. Naz., Or. 40, 37, 5–9 in Moreschini/Galley, Gregoire de Nazianze. Discours 38–40, 282–284). “იგი ბნელ არს და ნათლად ჩანს დიდად ბრწყინვალედ. ესე ღამე არს და დღედ იჩემებს და ეჩუენების განჴრწნილთა მათ შუებისა მიერ” [It is darkness and appears as a greatly shining light. This is night and claims to be day, and it seems [such] to those who have been corrupted by luxury.]. (Eutymius, Iber. 40, 37, 169, inCoulie/Metreveli, 272).

372

Magda Mtchedlidze

fact that it is possible to make an inference to this in translated works). Euthymius preferred the usual the Georgian expression, although here, the literal translation would be completely recognizable for any Georgian reader. Despite that, in using the terms ‘light’ and ‘darkness’ while juxtaposing corporeal and spiritual birth, Euthymius has taken general views accepted in Christianity into particular account, especially a passage from Gregory’s 39th Oration where the discourse centers around darkness being analogous to a carnal life in this world, with its antithesis, light, being likened to a proximity with God.40 It seems that he has also considered a comment by Basil Minimus, who directly states that Gregory also calls the flesh “darkness”.41 As is well known, basing the exegesis of an author his own words is a principle that was recognized in Antiquity and the Middle Ages as well.42 Thus, a use of Gregory’s other works, as well as the extant commentary devoted to him is natural for Euthymius’ hermeneutical translation. In order to convey the text to the reader in as simple a format as possible, changes that have been incorporated by Euthymius are also encountered in the following passage from Or. 40, 2:43 ἡ δὲ φοβερωτέρα καὶ συντομωτέρα, πᾶν τὸ πλάσμα συνάγουσα ἐν βραχεῖ τῷ πλάστῃ παραστησόμενον καὶ λόγον ὑφέξοντῆς ἐνταῦθα δουλείας.καὶ πολιτείας, εἴτε τῇ σαρκὶ μόνον ἐπηκολούθησεν εἴτε τῷ Πνεύματι συνανῆλθε, καὶ τὴν χάριν ᾐδέσθη τῆς ἀναπλάσεως.44 ხოლო მესამე იგი უსაშინლესი არს და უმაღლესი, რომელი ყოველსავე დაბადებულსა შეჰკრებს მეყსა შინა წარდგომად წინაშე დამბადებელისა და სიტყჳს მიცემად მოქალაქობისა მისთჳს, რომელი ამას სოფელსა შინა აღასრულა თითოეულმან გინათუ ჴორცთა ნებასა ოდენ დაემორჩილა, ანუ თუ სულსა შეუდგა და მადლისა მისგან მეორედ შობისა შეიკდიმა [and the third is more terrible and superior, which in a moment brings together all creatures to stand before the Creator, and to give an account of his citizenship Euthymius has introduced the following changes into Gregory’s text: In order to use ‘light’ (‘ნათელი’) as the antithesis for ‘darkness’ (‘ბნელი’, σκοτία), he has left ‘midday’ (‘შუადღე’, μεσημβρία) of the original text and the citation of Isaiah along with it. In the following phrase, he mentions ‘day’ (‘დღე’) in opposition to ‘night’ (‘ღამე’, νύξ), when in that case, φωτισμὸς is in the parent text, creating a kind of symmetry: darkness–light, day–night. 40  Gr. Naz. Or. 39, 2 (Moreschini/Galley, 151–3). 41  Ephrem’s translation of Basil’s text has been published by Tamar Otkhmezuri. Cf. Otkhmezuri, The Commentarial Genre, 248. Tamar Otkhmezuri has also given reference to examples of how Basil’s commentary is reflected in Euthymius’ translation. 42  Homer is elucidated by using Homer himself, Plato through Plato himself, the Holy Scriptures by the Holy Scriptures, etc. 43  Since it follows the Greek text almost verbatim in every case, Ephrem’s translation will no longer be presented and attention will be focused solely on Euthymius’ translation. 44  Gr. Naz. Or. 40, 2, 6–10 (Moreschini/Galley, 200).

A Translation, Paraphrase, or Metaphrasis?

373

that everyone has carried out in this world, whether he only submitted to the desire of the flesh, or he followed the spirit and respected the grace of its second birth]. (Euthymius)45

As can be seen, apart from a change in the syntactic structure, συντομωτέρα has been interpreted in Euthymius’ translation as ‘superior’ (‘უმაღლესი’), ἀνάπλασις has been translated as the ‘second birth’ (‘მეორედ შობა’), which is correct in this context (Ephrem however, has translated it as ‘აღმოდაბადება’, which is a calque of the Greek word, being totally unnatural and incomprehensible in Georgian).46 The word δουλεία has been omitted; πολιτείαις accompanied by the commentary “that everyone has carried out in this world” (ἐνταῦθαις interpreted by Eutymius as ‘this world’); and τῇ σαρκὶ... ἐπηκολούθησεν has been translated explicatively: “he submitted to the desires of the flesh.” Thus, aside from the interpretations, this time Euthymius has not reduced the text; on the contrary, he has resorted to an expansion of the original text in order to convey what Gregory has stated. It must be noted that Ephrem Mtsire has directly stated in one of his colophons that he refused to have Euthymius’ additions in his translation.47 Despite everything—the interpretations, an expansion or reduction of the original text—the contents of Euthymius’ text up until now have not departed substantially from the original. It is hard to make a similar statement for the following passage: Ταύτας δὴ τὰς γεννήσεις ἁπάσας παρ’ ἑαυτοῦ τιμήσας ὁ ἐμὸς Χριστὸς φαίνεται· τὴν μὲν, τῷ ἐμφυσήματι τῷ πρώτῳ καὶ ζωτικῷ· τὴν δὲ, τῇ σαρκώσει καὶ τῷ βαπτίσματι, ὅπερ αὐτὸς ἐβαπτίσατο· τὴν δὲ, τῇ ἀναστάσει, ἧς αὐτὸς ἀπήρξατο· ὡς ἐγένετο πρωτότοκος ἐν πολλοῖς ἀδελφοῖς (Rom 8:29), οὕτω καὶ πρωτότοκος ἐκ νεκρῶν γενέσθαι καταξιώσας (Col 1:18).48 ამათ სამთავე შობათა თავისა თჳსისა მიერ პატივსცა ქრისტე, უფალმან ჩემმან: ერთსა მას განკაცებითა თჳსითა, რომელმან-იგი შეიმოსა ყოვლითურთ კაცებაჲ და სული იგი, რომელი პირველ შთაჰბერა ადამს, მიიღო მან თავადმან (Gen 2:7) და ჴორცნი იგი მიწისაგანნი შეიმოსნა; და მეორესა მას—ნათლისღებითა, რომლითა ნათელ-იღო მან (Luke 3:21); 45  Iber. Or. 40, 2, 5 (Coulie/Metreveli, 122–4). 46  According to the translations, it is possible to retrace the formation of a Georgian literary language: as one translator under the influence of the Greek uses syntactic constructions, calques are made when reconciling words. After this however, the other translator rejects that which the language was unable to assimilate and uses that which is acceptable for the language. 47  თ. ბრეგაძე, გრიგოლ ნაზიანზელის თხზულებათა შემცველ ქართულ ხელნაწერთა აღწერილობა [T. Bregadze, A Description of Georgian Manuscripts Containing Compositions By Gregory of Nazianzus] (Tbilisi, 1998), 149. 48  Gr. Naz. Or. 40, 2, 11–17 (Moreschini/Galley, 200).

374

Magda Mtchedlidze

ხოლო მესამესა მას—აღდგომითა (Mark 16:6), რომლისა იქმნა დასაბამ, რომელი-იგი ვითარცა იქმნა პირმშო მრავალთა ძმათა შორის (Rom 8:29), ეგრეთვე თავს-იდვა პირმშო შესუენებელთა ყოფად (Col 1:18) [Christ, my Lord, honoured all these births in His own Person; one, by becoming a man [sc. by his Incarnation], Who put on human nature entirely and Himself assumed the soul that He first breathed into Adam (Gen 2:7) and put on the earthly body; the second by the Baptism wherewith He was baptized (Luke 3:21); and the third by the Resurrection (Mark 16:6), the beginning of which He became; as He became the Firstborn among many brothers (Rom 8:29), so also He took upon Himself to be the Firstborn from the dead (Col 1:18)]. (Euthymius)49

Σάρκωσις, as we see, corresponds here to ‘განკაცება’ (becoming a man), which is the precise equivalent of the term ἐνανθρώπησις (In this case, Ephrem remains faithful to his principle by writing ‘განჴორციელება’). But, however much ‘განკაცება’ has mostly become established in the Georgian language to express the concept of the Incarnation, this term and other details in this passage will no longer be dealt with; instead the focus will only be in regard to what casts doubt upon the equivalence of Euthymius’ Georgian text with the Greek original. In Gregory’s text regarding the three births (bodily, through baptism, and through resurrection), it has been stated that all three forms have been honored by Christ himself: “One—through the first life-giving Inbreathing, the second one through the Incarnation and baptism, by which He Himself was baptised, whereas the third one was through the Resurrection, which He Himself had initiated.”50 The phrase “by which He Himself was baptised” regarding baptism, and the phrase “which He Himself had initiated” regarding the Resurrection, in Gregory’s text must mean that the first inbreathing, having a relation to physical birth, implies the creation of man and the infusion of his soul, the common activity of the Trinity.51 The Incarnation and baptism, belonging to the same oikonomia, are associated with the second birth. Gregory’s text has been changed by Euthymius, however, with the Incarnation being thought of as a way to honor the birth in the flesh. The inbreathing has also been associated with this through the inclusion of a comment by him, as far as becoming a human being is presupposed through the reception of a soul and body. It is true that the Lord had put on an earthly body, but He himself had also received the same soul that He had 49  Iber. Or. 40, 2, 6 (Coulie/Metreveli, 124–6). 50  Gr. Naz. Or. 40, 2, 11–14 (Moreschini/Galley, 200). 51  It is true that the first human being had been created (γένεσις, πλάσις) in the primary sense and not born (γέννησις), but since the subsequent birth of human beings in the flesh has occurred precisely through the merit of the creation, our birth in the flesh, or the first birth, clearly must be ascribed to the mercy of God. The second birth (ἀναγέννησις, Δευτέρα γέννησις) is also called a ‘re-making’, ‘formation anew’ (ἀνάπλασις), which Euthymius has translated as ‘the second birth’ (‘მეორედ შობა’).

A Translation, Paraphrase, or Metaphrasis?

375

previously breathed into Adam. He honored the second birth, however, through His own baptism.52 It is interesting that Euthymius’ alteration of Gregory’s text finds a parallel in a comment made by Michael Psellos (11th century) regarding the same passage in Or. 40.53 Psellos has distinguished in birth two moments—animating (ἐμψύχωσις) and embodying (σωμάτωσις).54 In his opinion, Gregory has associated Christ’s reception of a soul with the first birth, considering the reception of the body together along with baptism. According to his statement, we have not sprung forth from a sinless Adam, but from one who had fallen into sin, but Christ had not received such a soul at His Incarnation (σαρκούμενος), receiving instead a soul like the one He breathed into the first ancestor (γενάρχης), who had been created. Thus He honored birth in the flesh with the first, life-giving inbreathing, in the sense that He was animated both spiritually and in purity.55 Regarding the association of the Incarnation and baptism with the second birth, in Psellos’ opinion, the notion of being purified through baptism has been stressed by Gregory in relation to God, because in comparison to us, the humanity assumed by the Lord in a manner which is beyond understanding had not been fouled by the impurities of birth. Psellos brings forth a comparison to explain the essence of honoring baptism: A grimy pearl is cleansed with water, but a pure, white pearl makes the water even more brilliant.56 It is true that Psellos has ascribed the reception of the human’s body and soul by the Lord to the various births that have been classified by Gregory, Euthymius however, does not separate them from each other. The common denominator between their commentaries concerns the relationship between Adam and Christ when interpreting the inbreathing of the soul. It is clear that both of them follow the patristic tradition in this regard.57 It must be mentioned that Psellos has pointed out in his own exegesis the difficulty of commenting upon how Gregory considers the Incarnation to be together with Baptism in connection with the second birth. Thus, Euthymius prefers not to translate the text as it is in the original, but at the same time to convey in his own fashion the complete information given in Gregory’s text. It is true that the 52  Iber. Or. 40, 2, 6 (Coulie/Metreveli, 122). 53 Psell. Opusc. 25, in P. Gautier (ed.), Michaelis Pselli Theologica (Lipsiae, 1989), 98–102. 54 Psell. Opusc. 25, 75–76 (Gautier, 101). 55 Psell. Opusc. 25, 57–72 (Gautier, 100). 56 Psell. Opusc. 25, 113–24(Gautier, 102). When commenting upon “the First born of all creation”, Psellos has considered baptism along with the Incarnation in another treatise written concerning the same cited passage in Gregory’s 40th Oration: the Lord raised up distorted human nature from sin once again through the acquisition of human nature, thereby leading the way to the second birth, which is imparted through baptism and the virtues (Psell. Opusc. 48, 22–43 in Gautier, 182–3). 57  It is possible to consider the comment by Basil Minimus as one of the immediate sources.

376

Magda Mtchedlidze

Incarnation has been named in the Georgian text as giving honor to the first birth, instead of the inbreathing mentioned in the Greek original, but the inbreathing of Adam’s soul by the Lord has been represented in the commentary. Nevertheless, it seems that the Georgian text exceeds the boundaries of translation here; Gregory’s text has been purposefully changed by Euthymius with the appropriate explication accompanying it as a result of studying the problem and deeply reflecting upon it. Ephrem Mtsire,58 with his translation here being extremely literal (sometimes it is so literal, it is difficult to make sense of it),59 has declared in the introduction appended to John of Damascus’ Fountain of Knowledge: “There are some things I don’t understand, but through a comparison with the Greek, I have not cut anything out by God’s mercy.”60 It seems that such an approach is unacceptable for Euthymius, with him only writing what he himself has fully comprehended. Now here is a look at the beginning of a translation by Euthymius the Hagiorite for Gregory the Theologian’s 29th Oration.61 Oratio 29, 1–262 1. Ἃ μὲν οὖν εἴποι τις ἂν ἐπικόπτων τὴν περὶ τὸν λόγον αὐτῶν ἑτοιμότητα καὶ ταχύτητα, καὶ τὸ τοῦ τάχους ἐπισφαλὲς ἐν πᾶσι μὲν πράγμασι, μάλιστα δὲ ἐν τοῖς

Version by Euthymius63 1. რაჲ-იგი ჯერ-იყო თქუმად დასაყენებელად მრავალმეტყუელებისა მის ამაოდ მეტყუელთაჲს და

58  It seems that he had in mind precisely such passages when asserting: Thus it is written in the Greek, as the Theologian has stated. Don’t bother me about it and address him instead (Bregadze, A Description, 149). 59  It is precisely because of this, that in some cases with Ephrem, there is not an equivalence, but ratherformal correspondence. 60 Raphava, Dialectics, 66. Apparently this does not mean that Ephrem had not studied the text to be translated in lesser depth. “I tried as much as I was able to appropriately convey the Greek and for this I addressed many knowledgeable Greek and Georgian advisors with questions”, he states in the Introduction to the exegesis for the Catholic Epistles (Kochlamazashvili, Comments, 56). Regarding specific examples of research that had been carried out by Euthymius on the text to be translated, cf. Otkhmezuri, The Commentarial Genre, 200–5. 61  It must be noted that this theological homily has been preserved in the Georgian only through the version by Euthymius. 62  P. Galley (ed.; tr.), Gregoire de Nazianze. Discours 27–31. Introduction, texte critique, traduction et notes (SC 250) (Paris, 1979), 178–80. 63  The Georgian text from ms 68 (Athos, Iviron) has been prepared for publication by Tsiala Kurtsikidze, and is preserved in the Georgian National Center of Manuscripts. This researcher has dedicated some publications to the peculiarities characteristic of Euthymius’ translations. The following article is specially in regard to the translation of Gregory’s theological works: ც. ქურციკიძე, „ექვთიმე ათონელის მთარგმნელობითი მეთოდისათვის“,

A Translation, Paraphrase, or Metaphrasis?

377

Oratio 29, 1–262

Version by Euthymius63

περὶ θεοῦ λόγοις, ταῦτά ἐστιν. ἐπεὶ δὲ τὸ μὲν ἐπιτιμᾷν οὐ μέγα· ῥᾷστον γὰρ καὶ τοῦ βουλομένου παντὸς· τὸ δὲ ἀντεισάγειν τὴν ἑαυτοῦ γνώμην ἀνδρὸς εὐσεβοῦς καὶ νοῦν ἔχοντος·

გამოცხადებად, ვითარმედ საცთურ არს წარმდებებით მეტყუელებაჲ ყოველთავე საქმეთა შინა? და, უფროჲს ყოვლისა, ღმრთისათჳს თქუმულთა სიტყუათა შინა? ზემოწერილსა მას თავსა შინა მცირედ რაჲმე წარმოვთქჳ, ხოლო ვინაჲთგან მხილებაჲ ოდენ და შერისხვაჲ წინააღმდგომთაჲ არა დიდ არს, არამედ ადვილ არს, ვისცა ენებოს, ხოლო ჭეშმარიტისა სიტყჳსა გამოცხადებაჲ საქმე არს კაცისა გონიერისა და მართლმორწმუნისაჲ. მოვედით და შევიწიოთ სული წმიდაჲ, რომელსა იგინი გმობენ, ხოლო ჩუენ თაყუანის-ვსცემთ და აღსაარებაჲ მართლმორწმუნეობისა ჩუენისაჲ, ვითარცა ნაყოფი შუენიერი და მწიფჱ, ნათლად გამოვიღოთ და ჴმამაღლად ვქადაგოთ, რომელი-ესე არცა სხუასა სადა ჟამსა დაგჳდუმებიეს, რამეთუ ამას ხოლო საქმესა ზედა ახოვან ვართ და მაღალ გონებითაცა და სიტყჳთა.

φέρε, τῷ ἁγίῳ θαρρήσαντες πνεύματι, τῷ παρ’ αὐτῶν μὲν ἀτιμαζομένῳ, παρ’ ἡμῶν δὲ προσκυνουμένῳ, τὰς ἡμετέρας περὶ τῆς θεότητος ὑπολήψεις, αἵ τινές ποτέ εἰσιν, ὥσπερ τινὰ τόκον εὐγενῆ τε καὶ ὥριμον εἰς φῶς προενέγκωμεν· οὐδὲ ἄλλο τε μὲν σιωπήσαντες, τοῦτο γὰρ μόνον ἡμεῖς νεανικοί τε καὶ μεγαλόφρονες,

კრებულში: დ. ჩიტუნაშვილი (რედ.), კავკასია აღმოსავლეთსა და დასავლეთს შორის, ისტორიულ-ფილოლოგიური ძიებანი, მიძღვნილი ზაზა ალექსიძის დაბადების 75 წლისთავისათვის [Ts. Kurtsikidze, “Regarding the Translation Method of Euthymius the Athonite”, in D. Chitunashvili (ed.), The Caucasus Between The East and the West. Historical and Philological Researches. Dedicated to the 75th Anniversary of Zaza Alexidze] (Tbilisi: National Center of Manuscripts, 2012) 297–303.

378

Magda Mtchedlidze

Oratio 29, 1–262

Version by Euthymius63

νῦν δὲ καὶ μᾶλλον παρρησιαζόμενοι τὴν ἀλήθειαν· ἵνα μὴ τῇ ὑποστολῇ, καθὼς γέγραπται, τὸ μὴ εὐδοκεῖσθαι κατακριθῶμεν. διττοῦ δὲ ὄντος λόγου παντός, τοῦ μὲν τὸ οἰκεῖον κατασκευάζοντος, τοῦ δὲ τὸ ἀντίπαλον ἀνατρέποντος, καὶ ἡμεῖς τὸν οἰκεῖον ἐκθέμενοι πρότερον οὕτε τὰ τῶν ἐναντίων ἀνατρέψαι πειρασόμεθα·

ხოლო აწ უმეტესად კადნიერ ვიქმნებით ჭეშმარიტებისა ქადაგებად, რაჲთა არა ვდუმნეთ, თუ დაგუესაჯოს უდებებაჲ. ხოლო ვინაჲთგან ორსახე არს ყოველი სიტყუაჲ—ერთი იგი, რაჲთა თჳსსა თქუმულსა დაამტკიცებდეს და მეორე, რაჲთა წინააღმდგომისასა უკუ-არღუევდეს, ჩუენცა პირველად თჳსი წარმოვთქუათ და მერმე წინააღმდგომთა მათ ზრახვანი ჴელ-ვყვნეთ დაჴსნად და განქარვებად. და ორივე ესე შემოკლებულად, რაჲთა ადვილ იყვნენ სასწავლელად სიტყუანი ჩუენნი, რაჲთა არა სიგრძისაგან სიტყუათაჲსა მოუძლურდებოდის და განიბნეოდის გონებაჲ მსმენელთაჲ, (ვითარცა-იგი მზაკუვართა მათ მწვალებელთა მოიპოვეს შემზადებაჲ მოკლისა სიტყჳსაჲ მასწავებელისა წვალებისა მათისაჲ საცთურად უსწავლელთა, გინათუ უგუნურთა კაცთა) ესრეთ უკუე ვიტყჳ, ვითარმედ ჟამთა მათგან პირველთა სამთა შჯულთა მოძღურებანი შემოვიდეს სოფლად: უმთავრობაჲ, ესე იგი არს უღმრთოებაჲ, და მრავალმთავრობაჲ, ესე იგი არს მრავალღმრთეებაჲ, და ერთმთავრობაჲ, რომელ არს ერთღმრთეებისა ქადაგებაჲ და მსახურებაჲ.

καὶ ἀμφότερα ὡς οἷόν τε διὰ βραχέων, ἵν’ εὐσύνοπτα γένηται τὰ λεγόμενα, ὥσπερ ὃν αὐτοὶ λόγον εἰσαγωγικὸν ἐπενόησαν πρὸς ἐξαπάτην τῶν ἁπλουστέρων ἢ εὐηθεστέρων, καὶ μὴ τῷ μήκει τοῦ λόγου διαχεθῇ τὰ νοούμενα, καθάπερ ὕδωρ οὐ σωλῆνι σφιγγόμενον, ἀλλὰ κατὰ πεδίου χεόμενον καὶ λυόμενεον.

2. Τρεῖς αἱ ἀνωτάτω δόξαι περὶ θεοῦ, ἀναρχία, καὶ πολυαρχία, καὶ μοναρχία.

A Translation, Paraphrase, or Metaphrasis?

379

Oratio 29, 1–262

Version by Euthymius63

αἱ μὲν οὖν δύο παισὶν Ἑλλήνων ἐπαίχθησαν, καὶ παιζέσθωσαν. τό τε γὰρ ἄναρχον ἄτακτον· τό τε πολύαρχον στασιῶδες, καὶ οὕτως ἄναρχον, καὶ οὕτως ἄτακτον. εἰς ταὐτὸν γὰρ ἀμφότερα φέρει, τὴν ἀταξίαν, ἡ δὲ εἰς λύσιν· ἀταξία γὰρ μελέτη λύσεως.

იგი უკუე ორნი—უღმრთოებაჲ და მრავალღმრთეებაჲ—წარმართთა მოიპოვნეს საცთურად მათდა და საკიცხელად, რამეთუ უღმრთოებაჲ უწესოებაჲ არს და უშჯულოებაჲ და მრავალღმრთეებაჲ შფოთთა და ბრძოლათა მომატყუებელ და ეგრეთვე უმთავრო და უწესო, რამეთუ ორივე იგი უშჯულოებაჲ არს და უწესოებაჲ. და უწესოებაჲ ყოველი მიზეზი არს სრულიადისა დარღუევისა და განქარვებისა. ხოლო ჩუენ ერთსა მთავრობასა ერთღმრთეებისასა პატივვსცემთ და ვქადაგებთ, ხოლო ერთმთავრობასა არა ესრეთ ვქადაგებთ, რაჲთამცა ერთპირად შემოიწერებოდა, ვითარ-იგი იტყჳან გონებაცთომილნი ჰურიანი, რამეთუ ერთპირად და ერთგუამად თქუმაჲ წინააღმდგომ არს ჭეშმარიტებისა, არამედ რომელსა-იგი ბუნებისა ერთპატივობაჲ სრულ-ჰყოფს და განუყოფელობაჲ ნებისაჲ და ერთობაჲ არსებისაჲ და ერთისა მის მიმართ მისგანთაჲ მათ უცვალებელი ერთობაჲ (რომელ-ესე დაბადებულთა შორის შეუძლებელ არს), თუმცა რიცხჳთა განყოფილნი არსებითა და ნებითა არა განიყოფებოდეს, ხოლო საღმრთოჲ ქადაგებაჲ ესრეთ დამტკიცებულ არს.

ἡμῖν δὲ μοναρχία τὸ τιμώμενον· μοναρχία δὲ, οὐχ ἣν ἓν περιγράφει πρόσωπον· ἔστι γὰρ καὶ τὸ ἓν στασιάζον πρὸς ἑαυτὸ πολλὰ καθίστασθαι·

ἀλλ’ ἣν φύσεως ὁμοτιμία συνίστησι, καὶ γνώμης σύμπνοια, καὶ ταυτότης κίνησεως, καὶ πρὸς τὸ ἓν τῶν ἐξ αὐτοῦ σύννευσις, ὅπερ ἀμήχανον ἐπὶ τῆς γεννητῆς φύσεως, ὥστε κἂν ἀριθμῷ διαφέρῃ, τῇ γὲ οὐσίᾳ64μὴ τέμνεσθαι.

64 v.l. τῇ ἐξουσίᾳ (Cf. Galley, Gregoire de Nazianze, 178). Euthymius’ translation “არსებითა და ნებითა” might possibly represent a reflection of two different redactions that he had.

380

Magda Mtchedlidze

Oratio 29, 1–262 διὰ τοῦτο μονὰς ἀπ’ ἀρχῆς εἰς δυάδα κινηθεῖσα, μέχρι τριάδος ἔστη. καὶ τοῦτό ἐστιν ἡμῖν ὁ πατήρ, καὶ ὁ υἱός, καὶ τὸ ἅγιον πνεῦμα· ὁ μὲν γεννήτωρ καὶ προβολεύς, λέγω δὲ ἀπαθῶς, καὶ ἀχρόνως, καὶ ἀσωμάτως· τῶν δὲ, τὸ μὲν γέννημα, τὸ δὲ πρόβλημα,

ἢ οὐκ οἶδ’ ὅπως ἄν τις ταῦτα καλέσειεν, ἀφελὼν πάντῃ τῶν ὁρωμένων. οὐ γὰρ δὴ ὑπέρχυσιν ἀγαθότητος εἰπεῖν θαρρήσομεν, ὃ τῶν παρ’ Ἕλλησι φιλοσοφησάντων εἰπεῖν τις ἐτόλμησεν, οἷον κρατήρ τις ὑπερερρύῃ, σαφῶς οὑτωσὶ λέγων, ἐν οἷς περὶ πρώτου αἰτίου καὶ δευτέρου φιλοσοφεῖ· μή ποτε ἀκούσιον τὴν γέννησιν εἰσαγάγωμεν, καὶ οἷον περίττωμά τι φυσικὸν καὶ δυσκάθεκτον, ἥκιστα ταῖς περὶ θεότητος ὑπονοίαις πρέπον.

διὰ τοῦτο ἐπὶ τῶν ἡμετέρων ὅρων ἱστάμενοι τὸ ἀγέννητον εἰσάγομεν, καὶ τὸ γεννητόν, καὶ τὸ ἐκ τοῦ πατρὸς ἐκπορευόμενον, ὥς πού φησιν αὐτὸς ὁ θεὸς καὶ λόγος.

Version by Euthymius63 ამისთჳს ერთგუამოვნებაჲ, პირველითგან ორგუამოვნებად წარმატებული, სამგუამოვნებად სრულ იქმნა. და ესე არს ჩუენდა მამაჲ და ძჱ და სული წმიდაჲ. მამაჲ—მშობელი და გამომავლინებელი უვნებლად და უჟამოდ და უჴორცოდ, ძჱ— მამისაგან შობილი უწინარეს ყოველთა საუკუნეთა და სული წმიდაჲ—მამისაგან გამომავალი, გინათუ არა უწყი, ვითარ ვინ სახელ-სდვას ამათ, რამეთუ უზეშთაეს არს სიტყჳსა და უაღრეს გონებათა საიდუმლოთა ამათ მიუწთომელობაჲ, დაღაცათუ ვიეთმე უგუნურთა და ცუდად ფილაფოზ წოდებულთა ამაოჲ და უჯეროჲ მრავლისმეტყუელებაჲ თქუეს, ვითარმცა რასმე მეცნიერ იყვნეს. არამედ ზრახვანი მათნი უცხო არიან წესთაგან ღმრთისმეტყუელებისათა, ვითარცა იტყჳს წმიდაჲ მოციქული, ვითარმედ: „ამაო იქმნეს გულისზრახვითა მათითა და დაუბნელდა უგულისჴმოჲ გონებაჲ მათი. იტყოდეს თავთა თჳსთა ბრძენ და განცოფნეს“ (რომ. 1. 21–22). ხოლო ჩუენ წესიერებასა ზედა ჩუენსა და საზღვართა შინა მართლმადიდებლობისათა მტკიცედ მდგომარე ვართ და გურწამს მამაჲ უშობელი, ძჱ მამისაგან შობილი და სული წმიდაჲ მამისაგან გამომავალი, ვითარცა თქუა თავადმან ღმრთისა სიტყუამან, ჭეშმარიტმან ღმერთმან, ვითარმედ: „სული იგი ჭეშმარიტებისაჲ, რომელი მამისაგან გამოვალს“ (იოანე 15.26).

A Translation, Paraphrase, or Metaphrasis?

381

I think the portion of free translations, paraphrases, and metaphrases in the Georgian translation is most apparent through the material presented. The passage is especially interesting from the standpoint of the translator’s attitude towards philosophy and the theological style of Gregory. It must be noted that Or. 29, 2 is remarkable for its philosophical character, which is not reflected in the Georgian text. Here are some examples: διὰ τοῦτο μονὰς ἀπ’ ἀρχῆς εἰς δυάδα κινηθεῖσα, μέχρι τριάδος ἔστη. καὶ τοῦτό ἐστιν ἡμῖν ὁ πατήρ, καὶ ὁ υἱός, καὶ τὸ ἅγιον πνεῦμα· ὁ μὲν γεννήτωρ καὶ προβολεύς, λέγω δὲ ἀπαθῶς, καὶ ἀχρόνως, καὶ ἀσωμάτως· τῶν δὲ, τὸ μὲν γέννημα, τὸ δὲ πρόβλημα65 ამისთჳს ერთგუამოვნებაჲ, პირველითგან ორგუამოვნებად წარმატებული, სამგუამოვნებად სრულ იქმნა. და ესე არს ჩუენდა მამაჲ და ძჱ და სული წმიდაჲ. მამაჲ—მშობელი და გამომავლინებელი უვნებლად და უჟამოდ და უჴორცოდ, ძჱ—მამისაგან შობილი უწინარეს ყოველთა საუკუნეთა და სული წმიდაჲ—მამისაგან გამომავალი [Therefore the being as one hypostasis, having from the very beginning proceeded to become two hypostases, came to completeness as three hypostases.This is for us the Father, the Son, and the Holy Spirit: the Father is the Begetter and the Emitter without passion, without time and incorporeally, the Son is Begotten by the Father before all ages, and the Holy Spirit proceeds from the Father].

“Therefore the Monad having moved towards the Duality from the very beginning, came to rest at the Triad” (διὰ τοῦτο μονὰς ἀπ’ ἀρχῆς εἰς δυάδα κινηθεῖσα, μέχρι τριάδος ἔστη). This is probably Gregory’s most problematic statement, with many commentaries being written regarding it.66 Euthymius, as can be seen, has preferred an interpretive translation of Gregory’s statement instead of commenting upon it: “Therefore the being as one hypostasis, having from the very beginning proceeded to become two hypostases, came to completeness as three hypostases.” Euthymius’ source in understanding Gregory’s statement must be 65  Gr. Naz. Or. 29, 2, 13–17 (Galley, 180). 66  A special interest towards this statement is manifested with intellectuals in the 11th–12th centuries. For example, Michael Psellos throws the Orthodoxy of the holy Father’s statement into doubt for rhetorical reasons in one of his comments, though subsequently justifying him (Psell. Opusc. 20, in Gautier, 76–80). In his commentary regarding the triad, Psellos has highlighted the differences between Gregory and Neoplatonic concepts (Psell. Opusc. 23, 90–138, in Gautier, 90–91). On the contrary, the Georgian philosopher Ioane Petritsi (12th century) makes an allusion to this same statement when trying to prove a certain commonality between Platonic and Christian views (Petritsi, pp. 209–10). Nicholas of Methone (12th century) holds an opposing view contrasting the monism of Proclus with this statement of Gregory expressing the truth regarding the Divinity (A. D. Angelou [ed.], Nicholas of Methone. Refutation of Proclus’ Elements of Theology [Athènes/Leiden, 1984], 5).

382

Magda Mtchedlidze

John of Damascus, particularly the following passage from his “On the Thriceholy Hymn”: Μονὰς γὰρ ἀπ’ ἀρχῆς εἰς δυάδα κινηθεῖσα, μέχρι τριάδος ἔστη... [λέγομεν] ἐπὶ θεοῦ μία ὑπόστασις πατρὸς, δύο ὑπόστασεις πατρὸς καὶ υἱοῦ, τρεῖς ὑπόστασεις πατρὸς καὶ υἱοῦ καὶ ἁγίου πνεύματος.67

In the same passage, after Gregory had mentioned the Father, the Son, and the Holy Spirit, he uses pronouns in the next phrase, as we can see. He states that the first one is the Begetter and the Emitter (ὁ μὲν γεννήτωρ καὶ προβολεύς), with one of the two others being the Begotten, and the other the Procession (τὸ μὲν γέννημα, τὸ δὲ πρόβλημα).68 Corresponding to this characteristic of the member of the Trinity at the end of 29:2, this has been stated in the form of a summarizing conclusion for the discourse represented throughout the entire passage, that regarding the Trinity we introduce the notions of unbegotten, begotten, and that which proceeds from the Father.69 The members of the Trinity have been directly named by Euthymius in the first case: “the Father is the Begetter and the Emitter without passion, without time, and incorporeal, the Son is Begotten by the Father before all ages, and the Holy Spirit proceeds from the Father.” This time Euthymius avoids the use of locutions that are equivalents to the pronouns ὁ μέν... ὁ δέ (“The First, the Second...”, with this example having been used at the spot in the 40th Oration where the different forms of genesis have been listed). All the more, at the end of the passage, terminology denoting generalized concepts (τὸ ἀγέννητον εἰσάγομεν, τὸ γεννητὸν, τὸ ἐκ τοῦ πατρὸς ἐκπορευόμενον) have been personalized in the Georgian text as “The Unbegotten Father” (“მამაჲ უშობელი”), “The Son Begotten of the Father” (“ძჱ მამისაგან შობილი”), and “the Holy Spirit Who proceeds from the Father” (“სული წმიდაჲ მამისაგან გამომავალი”). As in the first case, Gregory’s text has also been expanded here with additional exegesis, followed by a citation from the Holy Scriptures as well: διὰ τοῦτο ἐπὶ τῶν ἡμετέρων ὅρων ἱστάμενοι τὸ ἀγέννητον εἰσάγομεν, καὶ τὸ γεννητόν, καὶ τὸ ἐκ τοῦ πατρὸς ἐκπορευόμενον, ὥς πού φησιν αὐτὸς ὁ θεὸς καὶ λόγος.70

67  Io. Dam., De hymno trisagio, 28 in J.-P. Migne, Patrologiae cursus completus, Series Graeca 95 (1864), col. 60, A 7-B 11. It must be noted that Euthymius has used this work by John of Damascus when compiling the dogmatic manual “Hodegos” (See Introduction in Chkonia/ Chikvatia, Odegos, 31, 38). 68  Gr. Naz., Or. 29, 2, 15–17 (Galley, 180). 69  Gr. Naz., Or. 29, 2, 24–27 (Galley, 180). 70  Gr. Naz., Or. 29, 2, 24–27 (Galley, 180).

A Translation, Paraphrase, or Metaphrasis?

383

ხოლო ჩუენ წესიერებასა ზედა ჩუენსა და საზღვართა შინა მართლმადიდებლობისათა მტკიცედ მდგომარე ვართ და გურწამს მამაჲ უშობელი, ძჱ მამისაგან შობილი და სული წმიდაჲ მამისაგან გამომავალი, ვითარცა თქუა თავადმან ღმრთისა სიტყუამან, ჭეშმარიტმან ღმერთმან, ვითარმედ: „სული იგი ჭეშმარიტებისაჲ, რომელი მამისაგან გამოვალს“ (იოანე 15.26).[But we stand firmly on our system and within the limits of orthodoxy, and we believe in the Unbegotten Father, the Begotten Son, and the Holy Spirit proceeding from the Father, as the Word of God, the true God Himself said: “the Spirit of Truth who proceeds from the Father” (John 15.26)].

Euthymius has also conveyed some terminology denoting general concepts in Gregory’s following statements with commentary directly included in the text: Τρεῖς αἱ ἀνωτάτω δόξαι περὶ θεοῦ, ἀναρχία, καὶ πολυαρχία, καὶ μοναρχία.71 ესრეთ უკუე ვიტყჳ, ვითარმედ ჟამთა მათგან პირველთა სამთა შჯულთა მოძღურებანი შემოვიდეს სოფლად: უმთავრობაჲ, ესე იგი არს უღმრთოებაჲ და მრავალმთავრობაჲ, ესე იგი არს მრავალღმრთეებაჲ, და ერთმთავრობაჲ, რომელ არს ერთღმრთეებისა ქადაგებაჲ და მსახურებაჲ. [I say, that the three doctrines on laws came into this world from ancient times: anarchia that is atheism, polyarchia that is polytheism, and monarchia that is preaching and serving monotheism]

Euthymius, as we see, has rendered δόξαι περὶ θεοῦ as “doctrines on laws” (“რჯულთა სწავლებანი”), but he uses the word God when defining ἀναρχία, πολυαρχία, μοναρχία in the very text as atheism, polytheism, and monotheism.72 Euthymius candidly reveals his own attitude toward vain philosophers when he exchanges Gregory’s text for his own in one place: ἢ οὐκ οἶδ’ ὅπως ἄν τις ταῦτα καλέσειεν, ἀφελὼν πάντῃ τῶν ὁρωμένων. οὐ γὰρ δὴ ὑπέρχυσιν ἀγαθότητος εἰπεῖν θαρρήσομεν, ὃ τῶν παρ’ Ἕλλησι φιλοσοφησάντων εἰπεῖν τις ἐτόλμησεν, οἷον κρατήρ τις ὑπερερρύῃ, σαφῶς οὑτωσὶ λέγων, ἐν οἷς περὶ πρώτου αἰτίου καὶ δευτέρου φιλοσοφεῖ· μή ποτε ἀκούσιον τὴν γέννησιν

71  Gr. Naz., Or. 29, 2, 1–2 (Galley, 178). 72  Perhaps in order to prevent such interpretations that were attested later with Michael Psellos and his disciple, John Italos. This statement of Gregory gives them occasion to philosophize about the Principle and appropriately (as in the case of the previous statement), have a more expansive discourse regarding the theology of movement and number. See for instance, Italos, Opusc. 68 in ნ. კეჭაღმაძე, იოანე იტალოსის შრომები [G. Tsereteli/N. Kechagmadze (ed.), Joannis Itali Opera] (Tbilisi: Mecniereba, 1966), 185–8.

384

Magda Mtchedlidze

εἰσαγάγωμεν, καὶ οἷον περίττωμά τι φυσικὸν καὶ δυσκάθεκτον, ἥκιστα ταῖς περὶ θεότητος ὑπονοίαις πρέπον.73 გინათუ არა უწყი, ვითარ ვინ სახელ-სდვას ამათ, რამეთუ უზეშთაეს არს სიტყჳსა და უაღრეს გონებათა საიდუმლოთა ამათ მიუწთომელობაჲ, დაღაცათუ ვიეთმე უგუნურთა და ცუდად ფილაფოზ წოდებულთა ამაოჲ და უჯეროჲ მრავლისმეტყუელებაჲ თქუეს, ვითარმცა რასმე მეცნიერ იყვნეს, არამედ ზრახვანი მათნი უცხო არიან წესთაგან ღმრთისმეტყუელებისათა, ვითარცა იტყჳს წმიდაჲ მოციქული, ვითარმედ: ,,ამაო იქმნეს გულისზრახვითა მათითა და დაუბნელდა უგულისჴმოჲ გონებაჲ მათი. იტყოდეს თავთა თჳსთა ბრძენ და განცოფნეს“ (რომ. 1:21– 22) [Or, I do not know how someone would name them, because the mystery concerning them exceeds the word, and it is absolutely inaccessible to the mind. Nevertheless some fools, wrongly called philosophers, spoke much in vain and in an improper manner, as if they knew something. But their discourse is alien to the rules of theology, as the holy Apostle said: “They became futile through their thoughts and their unwise minds74 darkened. They called themselves sages and they became fools” (Rom 1:21–22)]

As is seen, Gregory has stated that he cannot dare to be like a Hellene, who speaks about “the overflowing of the bowl”75 when philosophizing about the First and Second Causes, and even mention “the overflow of goodness” (the intra-Triadic movement is implied), in order to not admit to an involuntary birth, being completely inappropriate for the meaning of Deity. Euthymius has not translated this discourse, choosing instead to develop a critique of false philosophers who speak loquaciously in an improper and vain manner, attesting the Apostle as well in denouncing these self-proclaimed wise men.76 Despite the fact that such a position is not foreign to Gregory (he holds a view in opposition to false philosophers in his work directed against Julianus), this particular passage does not express any such opposition. On the contrary, Gregory has indirectly cited the Greek philosopher in a positive manner, although he is not emboldened to accept the statement in full out of the fear of an erroneous interpretation.

73  Gr. Naz., Or. 29, 2, 18–24 (Galley, 180). 74  The text here in the Greek is καρδία. 75  Plotinus has been indicated as Gregory’s source with contemporary editors, specifically: Enn. 5, 1, 6 (See Gr. Naz., Or. 29, 2, in Galley, 181, Note 4). 76  In regard to Euthymius’ attitude of philosophy, cf. ქ. ბეზარაშვილი, ბიზანტიური და ძველი ქართული რიტორიკის თეორიისა და ლექსთწყობის საკითხები ანტიოქიური კოლოფონების მიხედვით: ეფრემ მცირე [K. Bezarashvili, The Problems of Byzantine and Old Georgian Rhetorical Theory and Versification according to the Antiochian Colophons: Ephrem Mtsire (3 Vol.; Tbilisi: 2011), 1.187–90.

A Translation, Paraphrase, or Metaphrasis?

385

One trope characteristic of Gregory’s style has also been overlooked in a represented section of the 29th Oration in the text by Euthymius. Specifically, in order to say that following an extended discourse is mentally difficult, Gregory has compared loquaciousness to water that has not been contained within a canal, thereby spilling out into the field and dissolving.77 Euthymius has not translated this comparison, but instead, without any sort of symbolism, speaks directly about the vanity of verboseness in his denouncement of false philosophers.78 As it is known, style has a great importance in understanding any authors, especially authors like Gregory. His theological thought has been strictly formulated in a terminological and literary manner. He intentionally uses philosophical terms and literary tropes which are not just rhetorical adornments (although, as far as the Orations go, rhetoric is also quite important), but they precisely define the content, they bear profound theological ideas, and are characteristic of his theological style. His thought was considered as a model in Byzantium in view of its content as well as its form. This is precisely the reason why extensive commentary was written not only in regard to Gregory’s theology, but in regard to his rhetorical methods as well.79 Despite the fact that Euthymius is quite familiar with these commentaries, due to his own educational goals to convey theology in a maximally comprehensive format by taking into account the readiness of his own reader, he has reserved himself the right to rework even the text of Gregory the Theologian, whose authority was nearly on par with the Holy Scriptures. It is possible to state that Euthymius comes across as being Gregory the Theologian’s co-author in the passages that have been examined.80 This is why the Greeks were discontent in regard to such an attitude toward the original text, despite having an especial esteem for Euthymius. Those who continued his work resorted to different methods of translation. Ephrem Mtsire, for example, tries to preserve the nuances characteristic of Gregory’s style in the Georgian text, along with a literal translation.81 It is interesting that he has noted 77  Gr. Naz., Or. 29, 1, 22–23 (Galley, 178). 78  Scholars have focused on the problem of losses of philosophical and rhetorical modes when analyzing the peculiarities of Euthymius’ translations. See, for instance, ლ. ხოფერია, „მაქსიმე აღმსარებლის „პიროსთან სიტყვისგების“ ეფთვიმე ათონელისეული თარგმანი“ [L. Khoperia, “The Translation of Euthymius the Hagiorite for Maximus the Confessor’s ‘A Response to Piros’”], რელიგია 1–2–3 (1996) 82–101, on pp. 84–9. 79  Michael Psellos has also devoted a special treatise to Gregory’s theological style, paying special attention to the forms of expressing a thought when commenting upon some of his Orations and statements. 80  If Maximus the Confessor had dedicated a separate work to the exegesis of Gregory’s passages that were difficult to understand, Euthymius, as has been seen, presents the text in an elucidated manner. On the other hand, he leaves whatever is difficult to perceive for an unprepared reader and exchanges it for his own discourse. 81  Regarding Ephrem’s consideration of the rhetorical commentary of Basil Minimus when translating Gregory’s Orations, cf. Otkhmezuri, The Commentarial Genre, 83–93.

386

Magda Mtchedlidze

the neglect of Gregory’s “verbal brevity and depth” and the philosophical aspect of his Orations in Euthymius’ translations.82 The fact is that not only are Gregory’s and Euthymius’ addressees representatives of different cultures, there is also a disparity between Euthymius’ and Ephrem Mtsire’s readers. There is already a noticeably increased circle of Georgians who had assimilated the thoughts and literary traditions of Antiquity during Ephrem’s time,83 i.e. the traditions adopted by the philosophical and theological thought of Byzantium.84 Thus, Euthymius is oriented toward the reader, caring more for their theological Christian education than familiarizing them with Gregory the Theologian as an author and preserving the individualistic and cultural nuances of his text. It can be said that the transmission of a certain text belonging to a certain author has the same purpose for Euthymius as the composition of a compilation serving as a manual and a source of information. At the same time, it is apparent that he has considered translation as an activity containing everything associated with making the text understandable. Thus, by taking into account the character of the adaptation, how should the presented texts of Gregory the Theologian, rendered into Georgian from the Greek by Euthymius the Hagiorite, be defined? A translation (a free translation, a type of dynamic equivalence)? A paraphrase? A metaphrasis? How correct is it 82 Bregadze, A Description, 149. Also cf. მ. რაფავა, “თარგმანთაგანისა” და “ზედადართულის” გაგებისათვის გრიგოლ ნაზიანზელის ჰომილიათა ეფთვიმე მთაწმინდელის თარგმანების მიხედვით”, კრებული: ელ. მეტრეველი, მ. მამაცაშვილი (რედ.), მრავალთავი, ფილოლოგიურ-ისტორიული ძიებანი [M. Raphava, “Regarding How to Understand ‘From the Translations’ and ‘An Extra Addendum’ According to Euthymius the Hagiorite’s Translations of the Homilies of Gregory of Nazianzus”, in E. Metreveli/M. Mamatsashvili (ed.), Mravaltavi, Historical and Philological Researches 17 (1992) 85–92. 83 Euthymius himself had translated the mythological commentary of Pseudo-Nonnus, which Ephrem subsequently expanded. See ა. გამყრელიძე, თ. ოთხმეზური (გამ.), ფსევდონონეს მითოლოგიურ კომენტართა ქართული თარგმანები [A. Gamkrelidze/T. Otkhmezuri (ed.), Georgian Translations of the Mythological Commentary of Pseudo-Nonnus] (Tbilisi: Metsniereba, 1989). The compendiums of ancient philosophers, works of logic also appear in the Georgian language. The number of Georgian monasteries founded in the cultural and educational centers of Byzantium and the Georgians active there also increased. 84 Despite the fact that Hexaemeral compositions, as well as patristic texts containing dogmatic, polemical material had been translated into Georgian during an early time period (adapted texts of Orations by Gregory the Theologian have been included in the “Mravaltavi”), their numbers did not appear to be sufficient for an elaboration of a Georgian philosophical and theological language corresponding to the Greek. This process was favoured by an exceptional increase in translation activities starting in the 10th century. In regard to John Petritsi in the 12th century, despite him also complaining about the lack of philosophical terms in the Georgian language when translating Proclus Diadokhos, there is no longer any difficulty in conveying the terms of Gregory the Theologian’s phrase from Or. 29, 2: Monad... came to rest at the Triad (Petritsi, p. 210 Nutsubidze/Kaukhchishvili).

A Translation, Paraphrase, or Metaphrasis?

387

to define as a translation a work that has been rendered from another language and is such a combination of periphrases, interpretations, and exegesis? How justified is it to speak of a version’s equivalence to the source, when the translator’s purposeful, so to say, creative intervention in the parent text is apparent, being characteristic of a paraphrastic and metaphrastic work and implying, in a sense, the creation of a new text, even if he keeps within the limits of the author’s doctrine and manages to convey the main information of the source text despite all the changes?

Bibliography Abuladze, I. (ed.), Monuments of Old Georgian Hagiographic Literature, II, (11th–15th Centuries) [ი. აბულაძე (რედ.), ძველი ქართული აგიოგრაფიული ლიტერატურის ძეგლები, II, (XI–XV სს.)] (Tbilisi, 1968). Angelou, A.D. (ed.), Nicholas of Methone. Refutation of Proclus’ Elements of Theology (Athènes/ Leiden, 1984). Bezarashvili, K., Theory and Practice of Rhetoric and Translation. A Study of Georgian Translations of Gregory the Theologian’s Writings [ქ. ბეზარაშვილი, რიტორიკისა და თარგმანის თეორია და პრაქტიკა გრიგოლ ღვთისმეტყველის თხზულებათა ქართული თარგმანების მიხედვით] (Tbilisi: Metsniereba, 2004). Bezarashvili, K., The Problems of Byzantine and Old Georgian Rhetorical Theory and Versification according to the Antiochian Colophons: Ephrem Mtsire [ქ. ბეზარაშვილი, ბიზანტიური და ძველი ქართული რიტორიკის თეორიისა და ლექსთწყობის საკითხები ანტიოქიური კოლოფონების მიხედვით: ეფრემ მცირე] (3 Vol.; Tbilisi, 2011). Bregadze, T., A Description of Georgian Manuscripts Containing Compositions By Gregory of Nazianzus [თ. ბრეგაძე, გრიგოლ ნაზიანზელის თხზულებათა შემცველ ქართულ ხელნაწერთა აღწერილობა] (Tbilisi, 1998). Chkonia, T. & Chikvatia, N. (ed.), Euthymius the Athonite, Odegos [თ. ჭყონია, ნ. ჩიკვატია, წმ. ეფთვიმე მთაწმინდელი. წინამძღუარი (სარწმუნოებისათვის)] (Tbilisi: Artanuji, 2007). Coulie, B., Metreveli, H. et alii (ed.), S. Gregorii Nazianzeni opera. Versio Iberica, V. Orationes XXXIX, XL (CCSG 58; Corpus Nazianzenum 20; Turnhout-Leuven: Brepols, 2007). Crouzel, H. & Simonetti, M. (ed.; tr.), Origène, Traité des Principes, t. III, Livre III et IV (SC 268; Paris, 1980). Doborjginidze, N., Linguistic and Hermeneutic Metatexts. Practical Grammar and Hermeneutics in Georgian Sources of the 10th–13th Centuries [ნ. დობორჯგინიძე, ლინგვისტურჰერმენევტიკული მეტატექსტები. პრაქტიკული გრამატიკა და ჰერმენევტიკა X– XIII საუკუნეების ქართულ წყაროებში] (Tbilisi: Ilia State University Press, 2012). Dünzl, F. (ed.; tr.), Gregor von Nyssa, In canticum canticorum homiliae (3 vol.; Freiburg: Herder, 1994).

388

Magda Mtchedlidze

Gabidzashvili, E. & Abuladze, I. (ed.), Monuments of Old Georgian Hagiographic Literature, IV, Synaxarion Redactions (11th–18th Centuries) [ე. გაბიძაშვილი, ი. აბულაძე (გამ.), ძველი ქართული აგიოგრაფიული ლიტერატურის ძეგლები, IV, სვინაქსარული რედაქციები (XI–XVIII სს.)] (Tbilisi: Metsniereba, 1968). Galley, P. (ed.; tr.), Gregoire de Nazianze. Discours 27–31. Introduction, texte critique, traduction et notes (SC 250; Paris, 1979). Gamkrelidze, A. & Otkhmezuri, T. (ed.), Georgian Translations of the Mythological Commentary of Pseudo-Nonnus [ა. გამყრელიძე, თ. ოთხმეზური (გამომც.), ფსევდონონეს მითოლოგიურ კომენტართა ქართული თარგმანები] (Tbilisi: Metsniereba, 1989). Gaprindashvili, V. (tr.) The Life of Barlaam and Ioasaph [ვ. გაფრინდაშვილი (მთარგ.), ვარლაამისა და იოასაფის ცხოვრება] (Tbilisi: Program “Logos”, 2005). Gautier, P. (ed.), Michaelis Pselli Theologica (Lipsiae, 1989). Hinterberger, M., “Byzantine Hagiography and Its Literary Genres. Some Critical Observations”, in S. Efthymiadis (ed.), The Ashgate Research Companion to Byzantine Hagiography, Volume II: Genres and Contexts (Dorchester: Ashgate Publishing, 2014) 25–60. Kaukhchishvili, S. & Kvirikashvili, L., Greek-Georgian Documented Dictionary [ს. ყაუხჩიშვილი, ლ. კვირიკაშვილი (გამ.), ბერძნულ-ქართული დოკუმენტირებული ლექსიკონი] (7 vol.; Tbilisi: Program “Logos”, 2002–2007). Kekelidze, K., The History of the Old Georgian Literature [კ. კეკელიძე, ძველი ქართული ლიტერატურის ისტორია] (2 vol.; Tbilisi: Metsniereba, 1980). Kharanauli A., “Translation Technique of the Old Georgian Bible (According to the Khanmeti Fragments of Isaiah, Jeremiah and Ezra I)” [ა. ხარანაული, “ბიბლიის ძველი ქართული თარგმანების ტექნიკა (ხანმეტი ესაიას, იერემიას და ეზრა I-ის ფრაგმენტების მაგალითზე)”], in N. Makharadze/M. Giorgadze (ed.), Byzantine Studies in Georgia 2. Dedicated to Academician Grigol Tsereteli [ნ. მახარაძე, მ. გიორგაძე (რედ.), ბიზანტოლოგია საქართველოში, ეძღვნება აკადემიკოს გრიგოლ წერეთლის ხსოვნას] (Tbilisi: Program “Logos”, 2009) 938–68. Khoperia, L., “The Translation of Euthymius the Hagiorite for Maximus the Confessor’s ‘A Response to Piros’” [ლ. ხოფერია, „მაქსიმე აღმსარებლის „პიროსთან სიტყვისგების“ ეფთვიმე ათონელისეული თარგმანი“], Religion 1–2–3 (1996) 82–101. Kochlamazashvili, E., Comments on the Apostolicum, According to the Works of John Chrysostom and Other Holy Fathers, Translated by Ephrem Mtsire [ე. კოჭლამაზაშვილი (გამ.), სამოციქულოს განმარტება, გამოკრებული იოვანე ოქროპირისა და სხვა წმინდა მამათა თხზულებებიდან, თარგმნილი ეფრემ მცირის მიერ] (Tbilisi: P.H. Alilo, 2011). Lampe, G.W.H., A Patristic Greek Lexicon (Oxford, 1961). Kurtsikidze, Ts., “Regarding the Translation Method of Euthymius the Athonite”, in D. Chitunashvili (ed.), The Caucasus Between The East and the West. Historical and Philological Researches. Dedicated to the 75th Anniversary of Zaza Alexidze [ც. ქურციკიძე, „ექვთიმე ათონელის მთარგმნელობითი მეთოდისათვის“, კრებულში: დ. ჩიტუნაშვილი (რედ.), კავკასია აღმოსავლეთსა და დასავლეთს შორის,

A Translation, Paraphrase, or Metaphrasis?

389

ისტორიულ-ფილოლოგიური ძიებანი, მიძღვნილი ზაზა ალექსიძის დაბადების 75 წლისთავისათვის] (Tbilisi: National Center of Manuscripts, 2012) 297–303. Lortkipanidze, K. & Shanidze, A. (ed), The Georgian Versions of the Catholic Epistles According to Manuscripts of the 10th–14th Centuries [ქ. ლორთქიფანიძე (გამ.), ა. შანიძე (რედ.), კათოლიკე ეპისტოლეთა ქართული ვერსიები X–XIV საუკუნეთა ხელნაწერების მიხედვით] (Tbilisi: The Georgian SSR Academy of Sciences Press, 1956). Melikishvili, N., “Some Observations Regarding the Translation Method of Euthymius the Hagiorite and Ephrem Mtsire” [ნ. მელიქიშვილი, „რამდენიმე დაკვირვება ეფთვიმე ათონელისა და ეფრემ მცირის მთარგმნელობით მეთოდზე“], Matsne, SLL 4 (1987) 119–28. Metreveli, H. et alii (ed.), Gregorius Nazianzenus, Opera: versio Iberica, I: Orationes I, XLV, XLIV, XII (Corpus Christianorum Series Graeca 36; Corpus Nazianzenum 5; TurnhoutLeuven: Brepols, 1998). Migne, J.-P. (ed.), Ioannis Damasceni De hymno trisagio, 28 in Patrologiae cursus completus, Series Graeca 95, col. 21–62. Moreschini, C. (ed.); Galley, P. (tr.), Gregoire de Nazianze. Discours 38–41. Introduction, texte critique et notes (SC 358; Paris: Cerf, 1990). Nutsubidze, Sh. & Kaukhchishvili, S. (ed), The Works of Ioane Petritsi [შ. ნუცუბიძე, ს. ყაუხჩიშვილი (გამ.), იოანე პეტრიწის შრომები] (2 vol.; Tbilisi: Tbilisi State University Press, 1937, 1940). Otkhmezuri, T., The Commentarial Genre in the Medieval Georgian Translation Tradition, Ephrem Mtsire and Commentaries on the Sermons of Gregory the Theologian [თ. ოთხმეზური, კომენტარული ჟანრი შუა საუკუნეების ქართულ მთარგმნელობით ტრადიციაში, ეფრემ მცირე და გრიგოლ ღვთისმეტყველის თხზულებათა კომენტარები] (Tbilisi: Ilia State University Press, 2011). Raphava, M. (ed.), John of Damascus. Dialectics [მ. რაფავა (გამ.), იოანე დამასკელის „დიალექტიკა“] (Tbilisi: Metsniereba, 1976). Raphava, M., “Regarding How to Understand ‘From the Translations’ and ‘An Extra Addendum’ According to Euthymius the Hagiorite’s Translations of the Homilies of Gregory of Nazianzus” [მ. რაფავა, “თარგმანთაგანისა” და “ზედადართულის” გაგებისათვის გრიგოლ ნაზიანზელის ჰომილიათა ეფთვიმე მთაწმინდელის თარგმანების მიხედვით”], in E. Metreveli/M. Mamatsashvili (ed.), Mravaltavi, Historical and Philological Researches [ელ. მეტრეველი, მ. მამაცაშვილი (რედ.), მრავალთავი, ფილოლოგიურ-ისტორიული ძიებანი] 17 (1992), 85–92. Raphava, M., “Les traductions Géorgienes du Discours 39 de Grégoire de Nazianze”, კრებული A.B. Schmidt (ed.) Studia Nazianzenica II (Corpus Christianorum Series Graeca 73; Corpus Nazianzenum 24; Turnhout-Leuven: Brepols, 2010), 491–517. Tsereteli, G. & Kechagmadze, N. (ed.), Joannis Itali Opera [ნ. კეჭაღმაძე, იოანე იტალოსის შრომები] (Tbilisi: Metsniereba, 1966).

Anna Kharanauli Septuagint Text Types in the Georgian Translations What is Holy Scripture? A stone that has fallen from the sky, being unchangeable? A text with sacredness attached to its letters? Or living words that can be changed by those who bear them—the body in which they live and function? What makes a text holy? Is there a critical line beyond which the holiness is lost? An answer to these questions strays outside the sphere of science—an objective discussion. It can only be described as to what sort of attitude those who considered the text of the Bible as the Holy Scriptures had towards it and how much this attitude differed in various epochs and religious circles. It is a fact that “the Septuagint”, as a translation, is in a sense, a modified form of the Holy Scriptures and, at the same time, a text that has been perceived and accepted as Holy Scripture. It is also a fact that this “Septuagint” was always changing, yet it remained the Holy Scriptures of the Christian Church in its varied forms. How much was the Old Greek translation altered, and what forms did these alterations take when it was translated into Georgian? The situation is different in the case of each particular book. Therefore, while answering this question I will choose the books where, first of all, the origin of the Georgian translation is indisputable, and secondly, there is a sharp distinction between the text types in the Greek textual transmission. I will skip the Georgian translations of Esther, Tobit and Sirach1—extraordinary evidences of the text-formation process of the Old Testament books. These translations demonstrate that there existed many more literary editions of the Old Testament narratives in Greek than we would suppose according to the known sources; moreover, Georgian translations of the mentioned books show how these editions influenced each other and how many ways of their compilation or simplification existed. In the present article I will only discuss the changes of the texts, which already achieved one, completed form, and were translated into Greek and recognized as authoritative by the Jewish and Christian communities.

1  For the Georgian translations of these books see A. Kharanauli/N. Chantladze, “Oi Alloi or Lost Anonymous Literary Editions of Sirach Preserved in the Georgian and Armenian Translations”, in L. DiTommaso/M. Henze/W. Adler (ed.), The Embroidered Bible: Studies in Biblical Apocrypha and Pseudepigrapha in Honour of Michael E. Stone (SVTP; Leiden: Brill, 2017), 588–601 and two articles in the present volume: N. Mirotadze, “The Old Georgian Version of the Book of Esther—All in One” and N. Dundua, “What Can the Georgian Translation of the Book of Tobit Tell about GIII”.

392

Anna Kharanauli

Thus, how did the history of the Greek Old Testament, called Septuagint by its users, develop, and how can the Georgian translations contribute to clarifying these questions? It seems that there was continuous development especially in the initial stages of the Septuagint’s textual history. The Septuagint was altered and, interestingly, the sources of these alterations were mainly the texts, which were not accepted and not used as texts of Holy Scripture. Such texts were numerous, however we are able to identify only translations of Aquila, Symmachus and Theodotion thanks to the marginal scholia of the manuscripts. Intrusion of Jewish translators’ readings, closely approximated to MT, into the text points out that the textual history of the Septuagint is oriented toward the text considered to be the original.2 Consequently it is supposed that the purpose of the textual changes in the Septuagint is to keep its original from being changed. However, protection of the original reflected in the translation from changes implies changes in the divinely inspired text of LXX itself (and the Septuagint was indeed regarded as such text by its users). The representation of the textual history of the Septuagint ignores the existence of this dilemma. Moreover, the sources are not taken into account, which convey the attitudes of Jews and Christians towards the Septuagint.3 The second problem—rather practical than theoretical and related to the textual data—that also requires further clarification is how this textual change can be described. Was it the result of an intentional, one-time correction of the text based on regularly applied and recognizable principles, or was it the result of a gradual process? In the latter case, was it intended, sporadic or both? It is known that incorporation of readings closely approximated to MT in the Septuagint has been accomplished, let us say, in two stages: pre- and postOrigenian stages. In the Georgian sources both of these stages are reflected. I will approach the characterization of these stages exactly in relation with these sources.

2  However, the fact that for approximation to MT, variants of Theodotion, closest to the Septuagint, was used instead of variants of Aquila, most closely approximated to the Hebrew, does not support this conclusion. 3  This problem concerning Hieronymus has been addressed by Eva Schulz-Flügel; see E. Schulz-Flügel, “Hieronymus, Feind und Überwinder der Septuagint? Untersuchungen anhand der Arbeiten an den Psalmen”, in A. Aejmelaeus/U. Quast (ed.), Der Septuaginta-Psalter und seine Tochterübersetzungen (MSU 24; Göttingen: Vandenhoeck & Ruprecht, 2000) 33–50.

Septuagint Text Types in the Georgian Translations

393

The Pre-Origenian Changes in the Septuagint and the Georgian Translation Ezekiel4 The textual changes in the case of Greek Ezekiel are especially evident. Here the influences of the Hexapla are quite vivid: changes marked with asterisks and obeli are already attested in the Greek majuscules: Pap. Ra 922 (Oxford, Bodl. libr., Gr. bibl. d.4(P), between ca. 250 and 350 c.e.)5 and Codex Marchalianus ([= Q], Rom, Bibl. Vat., Vat. gr. 2125, 5th cent.), in their text as well as on the margins. The Old Georgian translation6 is almost free from these influences, and because of this typical feature, it joins the 9677 (late 2nd or early 3rd

4  Among the other Georgian translations of Old Testament books whose origins have been studied thus far, the translations of the Minor Prophets and Ecclesiastes also contain pre-­ Origenian text forms. The Georgian translation of the Minor Prophets follows the Greek majuscules, although it does not bear a particular proximity to any of them. In regard to Ecclesiastes, P. Gentry and N. Dundua have assigned the Georgian translation to the so-called Egyptian text group, see N. Dundua, “The Textual Value of the Old Georgian Version of Ecclesiastes”, in W. Kraus/M. van der Meer/M. Meiser (ed.), XV Congress of the International Organization for Septuagint and Cognate Studies, Munich, 2013 (BIOSCS XV; Atlanta: SBL, 2016) 231–9; P.J. Gentry (ed.), Septuaginta. Vetus Testamentum Graecum, XVI.1 (Gottingen: Vandenhoeck & Ruprecht, 2019), 33. 5  See F. Schironi, “P. Genf. 1.5, Origen, and the Scriptorium of Caesarea”, BASP 52 (2015) 181–223, on p. 217. 6  The oldest Georgian translation (= GeoOJ) of Ezekiel that has come down to us has been preserved in 10th–11th cent. manuscripts: Ath. 1 of Athos, Iviron monastery [= O] and Jer. 11 of Jerusalem Library of the Greek Patriarchate [= J]. In the 12th century GeoOJ was revised according to a Greek catenae manuscript which contained L texts (like Greek catenae mss. 36 and 51). This hellenophile revision, the so-called Gelati Bible revision (GeoG), is included in the catenae manuscript A 1108 Tbilisi NCM (12th cent.), written by the catenarist himself. Below, shared readings of GeoOJ and GeoG are marked with Geo. 7 Selected literature and editions: A. Rahlfs/D. Fraenkel, Verzeichnis der griechischen Handschriften des Alten Testaments, Bd. 1: Die Überlieferung bis zum VIII Jahrhundert (Septuaginta. Vetus Testamentum Graecum Auctoritate Academiae Scientiarum Gottingensis editum. Supplementum 1/1; Göttingen: Vandenhoeck & Ruprecht, 2004), 98–103; J. Ziegler (ed.), Ezechiel, Septuaginta. Vetus Testamentum Graecum, XVI.1 (3. Auflage, mit einem Nachtrag von Detlef Fraenkel; Göttingen: Vandenhoeck & Ruprecht, 2006); J. Ziegler, “Die Bedeutung des Chester Beatty-Scheide Papyrus 967 für die Textüberlieferung der Ezechiel-Septuaginta”, in J. Ziegler, Sylloge: gesammelte Aufsätze zur Septuaginta (Göttingen: Vandenhoeck & Ruprecht, 1971) 321–39; F.G. Kenyon (ed.), The Chester Beatty Biblical Papyri Descriptions and Texts of Twelve Manuscripts on Papyrus of the Greek Bible (2 Vols; London: Emery Walker 1937–1938); A.C. Johnson/H.S. Gehman/E.H. Kase (ed.), The John H. Scheide Biblical papyri. Ezekiel (Princeton University Studies in Papyrology 3, Princeton: Princeton University Press, 1938); L.G. Jahn (ed.), Der griechische Text des Buches Ezechiel: nach dem Kölner Teil des Papyrus 967 (Bonn, 1972); J. Barton Payne, “The Relationship of the Chester Beatty Papyri of Ezekiel to

394

Anna Kharanauli

cent.)8 and Codex Vaticanus (4th cent.), which are considered as the main witnesses to the OG.9 Generally, the “hexaplaric influences” are first of all evident in large additions marked with asterisks in post-Origenian sources. Pap. Ra 967, B, the Georgian translation and, in a number of cases, the Old Latin sources10 and Coptic, are against the entire transmission in not having these additions.11 But along with this, B and Ra 967 together and independently from each other contain minor, stylistic changes and readings corresponding to the MT, being sometimes marked with asterisks and obeli in post-Origenian sources. I could provide additions of pronouns having various functions as examples. In some of those cases, pluses attested in both B and 967 are reflected by the Georgian as well, e.g.: 47:12 καρπός 1°] + (※Q) αυτου B 967 A O-62´ L´’-46 verss. Tht. Hi. Geo = MT α' θ' Qtxt.

The fact that secondary readings corresponding to the MT appear in the witnesses of the Old Greek text is usually explained as a demonstration of a particular interest toward the Hebrew original that existed even before Origen. I will attempt to fill out the data of the papyrus, B and the Georgian translation with each other and bring this general opinion into question based on typical examples. The key to solve the problem of how the text was changing, from my point of view, lies in the manuscripts themselves. In the manuscripts readings aligned to MT are often placed in the margins, as e.g. in the following case, where the Geek equivalent of Hebrew has been written in the margin of the papyrus by the scribe’s own hand. B and GeoOJ however preserve the initial text: 20:40 ὑψηλοῦ B 967txt A′’-403′ 62′ 490mg Co Aeth GeoOJ] Ισραηλ 763; + (÷ pro※88) Ισραηλ rel (967mg ისრაჱლისა [israēlisa] GeoG) = MT. Codex Vaticanus”, JBL 68 (1949) 251–65; I.E. Lilly, Two Books of Ezekiel: Papyrus 967 and the Masoretic Text as Variant Literary Editions (VTSup 150; Leiden: Brill, 2012). 8  For a summary of the considerations about the date of the Papyrus, see Lilly, Two Books of Ezekiel, 227–8. 9  See Ziegler, Ezechiel, 24–8. Concerning the text-critical value of Pap. 967, see Lilly, Two Books of Ezekiel, 18–25. 10  See P.M. Bogaert, “Le témoignage de la Vetus Latina dans l’étude de la tradition des Septante. Ézéchiel et Daniel dans le papyrus 967”, Bib 59 (1978) 384–95. 11  Typical examples: 13:2 Ισραηλ B 967 Sa Hi.test GeoOJ] + (※O) τους προφητευοντας και ερεις τοις προφηταις τοις προφητευουσιν απο καρδιας αυτων rel (GeoG) = MT α', θ', Qtxt; 13:7 fin. B 967 Sa Hi.test GeoOJ] + (※O) και ελεγετε φησιν κυριος και εγω ουκ ελαλησα rel (GeoG) = MT Qtxt, cf σ'; 26:17 ἐπαινετὴ B 967 LaCW Bo Arab Tyc. GeoOJ] + (※O) ητις εγεν(ν)ηθη ισχυρα εν θαλασση αυτη και οι κατοικουντες αυτην rel (GeoG) = MT θ', Qtxt. Cf. the list of asterisked additions from the Köln part of Papyrus 967 (Jahn, Der griechische Text des Buches Ezechiel, 130–41).

Septuagint Text Types in the Georgian Translations

395

The marginal readings are indications of philological work, of an ekdosis, which means a collation of the copy not only with the model manuscript (the goal of such a collation is to correct scribal errors), but with different manuscripts containing different types of text (which is already a work of professional philologists). Alternative readings resulting from this collation were placed on the free spaces of the manuscript page—in the margins and between the lines. By all appearances, not all the variant readings did occur in the margins of the ekdosis, instead they were chosen at the interest of a particular editor or according to particular needs.12 A mechanical transfer of variants from the free spaces into the text when copying the manuscript is a well-known cause of textual changes.13 An example of such change is 21:3(8). Here the reading corresponding to MT is placed in the margin of B, while in other sources and among them in 967, Q, La Co and Geo, it occurs in the text: 21:3(8) Ισραηλ Btxt 106] θ′ + (※Q) ταδε λεγει κυριος κυριος (/αδωναι κυριος/ κυριος ο θεος) Bmg 967 Qtxt La Co ამას იტყჳს უფალი უფალი Geo et alii.

Another well-known fact is that marginal readings are not always put in the proper (that is, corresponding to MT) place in the text. Such examples can be found in sources containing Ezekiel’s pre-Origenian texts as well. Here also parallels with MT occur in different places—before the lemma and after it: 27:2 υἱὲ ἀνθρώπου LaW Arab Arm GeoOJ] + και συ B; pr (※O-Syh) και συ rel (967 GeoG) = MT14 ex 12:3; 47:9 ὁ ποταμός 2° Geo] pr ※ (88) εκει 967 O-Syh-62 = MT; + (※Syh) εκει B A′-106 Syh lII LaC Arab.

It seems that in the previous examples pluses, which coincide with MT and occur in 967 and B had not always been attested in the text of the Greek Vorlage of the Georgian translation. In the following case an equivalent of the MT has been reflected in the Georgian as well, however it is in the proper place (like B) in contrast to the papyrus, e.g.:

12  See A. Kharanauli, “Origen and Lucian in the Light of Ancient Editorial Techniques” in the present volume. 13  For illustrative examples, see, e.g., A. Schenker “Der Platz der altlateinischen Randlesarten des Kodex von León und der Valvanera-Bibel in der Biblischen Textgeschichte (1–4Kgt)”, in S. Kreuzer/M. Sigismund (ed.), Der Antiochenische Text der Septuaginta in seiner Bezeugung und seiner Bedeutung (DSI 4; Göttingen: Vandenhoeck & Ruprecht, 2013) 199–210. 14  For Ziegler’s comment on the given example, see Ziegler, Ezechiel, 41.

396

Anna Kharanauli

32:4 τῆς γῆς 147 L´’ 534 410 Aeth Arm Tht. Geo] + πασης 967; pr (※Q) πασης rel (B ყოვლისა Geo Tyc.) = MT.

To sum up, through the examples given above and similar cases here and elsewhere it becomes obvious that the readings, which coincide with MT firstly appear in the margins of the manuscript and then are transmitted in the text. Furthermore, the marginal variants did not become established in the text as the result of a one-time editorial work, but were inserted into various different manuscripts independently from each other. In most of the cases this process seems to be rather more mechanical than intentional and it does not imply that, while including marginal variants in texts, scribes consulted either translations of the Three (their continuous texts) or the Hebrew text. Margins are intended not only for those variant readings, which are missing from the text, but also to denote lexical differences. In the copying process, these alternative lexical units are either substituted for the lemma (when a scribe considers the variant reading denoted on margin of the Vorlage as a supplement for the lemma) or are added to it, thus creating a doublet (when a scribe considers the marginal reading of its Vorlage as an omitted segment of the text). A doublet is attested in the Georgian as well. It consists of an original reading preserved in B and an equivalent of the MT preserved in the papyrus and also in the manuscript group L: 30:5 μου ჩემისანი] μετ’ αυτων 967 = MT; + მათ თანა GeoOJ, cf 30:5 fin.] + μετ’ αυτων L´’ მათ თანა GeoG Tyc.

A doublet, attested in the GeoOJ and the addition of L, attested at the end of the verse, again confirms that in 967 the reading of the OG had not been replaced by a lexical equivalent of the MT with the aim of approximating it to the Hebrew; instead the scribe erroneously substituted lemmas with the marginal readings. The next argument against an assumption that approximation to the Hebrew was intended, being thus purposely followed by the establishment of a “correct” text, is the irregular occurrence of readings corresponding to MT in the manuscripts. Papyrus 967, Vaticanus, and the Georgian sometimes depart from each other in respect to variants which correspond the Hebrew; equivalents to the MT are sometimes in one and sometimes in another. 967 is the one containing the most parallels with MT among pre-Origenian sources. Additions according to MT have very rarely been preserved in the Georgian translation, however. Yet one example can be cited (for the time being), when the Georgian diverges from B and 967 and contains this sort of plus: 25:8 Μωαβ B 967 lI 544 Lamg Bo Arab Hi.test] + (※ O Hi.) και Σηειρ და სეირ GeoOJ (and Seir); და სიირმან (and Siir) GeoG et alii = MT.

Septuagint Text Types in the Georgian Translations

397

In another case Geo stands beside 967: 24:25 ἰσχὺν B Q-62’ L´106 Hi.] + (※O-Q) αυτων rel (967 Geo) = MT;

or beside B, e.g.: 26:17 ἐροῦσι 967] + (※O-Q) σοι B A´’ O-62 L´-36 46 LaCW Tyc. Hi. Geo = MT α' θ'.15

Apart from asterisked additions in the post-Origenian witnesses, there are some correspondences to the MT in 967 (mostly) and B that are not marked by Origenian signs in the extant manuscripts. At the same time, it is not rare when the Georgian and codex Vaticanus together are opposed to a large majority of manuscripts by not having such MT parallels, e.g.:16 47:1 τοῦ αἰθρίου B 407 Geo] + του οικου 967 სახლისასა (of house) GeoG et alii = MT.

Considering this, the presence of MT equivalents in the Georgian (in contrast to B and 967) should be noted: 32:4 ἐμπλήσω B 967 Bo Tyc. განვაძღნე] + εκ σου (შენგან Geo) rel = MT; 24:13 ἐμπλήσω τὸν θυμόν μου B 967 L´ LaW აღ-ვე-ვავსო (აღვავსო GeoG) გულისწყრომაჲ ჩემი Geo] + in ea Bo Arm; + εν σοι rel (შენ შორის GeoG) = MT; + შენგან (εκ σου) GeoOJ.

Both additions create an unnatural construction for Georgian that clearly points to the Greek Vorlage. Therefore, these cases when the Georgian, unlike B and 967, coincides with MT are more proofs of the irregularity of MT correspondences in the pre-Origenian text. Besides the additions, there are other types of variants in B and 967 that also attest to similarity with MT and coincide with the post-Origenian texts, such as for example, omissions having an obelus in post-Origenian texts: 45:17 ἔσονται] ÷ Q; > B 967 O’-Q L´’-36-613 Arm Geo Hi. = MT.

15  Cf. Ziegler, Ezechiel, 39–41, where these sorts of variants have been listed and analyzed. 16  The fact that some of the additions according to the MT have been marked with asterisks and others have not, attests that the manuscripts which served as Vorlagen for Origen and the post-Origenian editors had only partially contained these variants. This, in itself, still shows that the intent of creating a “correct” text closely approximated to the Hebrew had not existed at least before Origen.

398

Anna Kharanauli

The papyrus still stands out by omitting this type of readings.17 In quite a number of cases neither B nor the Georgian share these omissions, e.g.: 26:16 ἐκ τῶν ἐθνῶν ნათესავთა Geo] ÷ O; > 967 62 LaW Tyc. Hi. = MT.

Nevertheless, there is an example, when the minus of the papyrus is supported by the Georgian, whereas B keeps the initial reading, e.g.: 44:28 ὅτι რამეთუ GeoG] ÷ O; > 967 GeoOJ = MT.

To sum up, here also, as in the cases of additions, irregularity of occurrences of omissions according to MT is one more argument against the existence of intentionally “corrected” texts, which served as a parent text for one particular textual tradition. Parallels with the MT in the papyrus are even manifested in the vocabulary.18 As in the previous cases, the Georgian minimally contains these parallels, as for example in the following case: 28:15 ἐγενήθης იყავ GeoG] επορευθης 967 L´’ LaC ხჳდოდე GeoOJ Tht. Tyc. Or.lat VIII 326 Aug. Arm. = MT.

The source of such approximation towards the MT was apparently different from the Three: α' εμπεριεπατησας (επεριπατ. 91) σ' ανεστρεφου θ' ητοιμασθης 87–91. The Georgian translation mostly preserves the OG readings, as it is demonstrated in the following cases: 25:15 ἐξαλεῖψαι ἕως αἰῶνος აღჴოცად ვიდრე უკუნისამდე GeoOJ] εξαλειψαι εχθραν αιωνιον 967 აღჴოცაჲ და მტერობაჲ საუკუნოჲ GeoG = MT; cf διαφθειραι εκ ψυχης (om εκ ψ. 46) κατ εχθραν αιωνιον (-νιαν 46) Qmg lII Tht. = MT and reading of Symmachus: εις διαφθοραν κατ εχθραν αιωνιον 86 (om εις διαφθ.) Syh.19

If we take into account existence of the variant reading on the margin of Q in the last example (25:15) and what I already said above about the doublets, we could reconstruct the process of the textual changes in the following way: a reading of a text was erroneously substituted by a scribe of 967 or its Vorlage with a variant from an alternative source, denoted on a margin of an archetype of 967. While, 17  See Ziegler, Die Bedeutung des Chester Beatty-Scheide Papyrus 967, 321–2; Jahn, Der griechische Text des Buches Ezechiel, 142–4. 18 Ziegler, Die Bedeutung des Chester Beatty-Scheide Papyrus 967, 325. 19  Interestingly, there is another lexical change in B (supposedly caused by writing under dictation), which also has not been reflected in Georgian: αἰῶνος] ενος B L´ Bo.

Septuagint Text Types in the Georgian Translations

399

apparently there was no such variant reading in the edition (ekdosis) which was an archetype of B and Geo’s Greek Vorlage, or there was one, but the scribes had not made a mistake—they did consider it neither as an omitted word, nor as an emendation of the rejected text. From what I have said, therefore, the textual history of the Septuagint in the period before Origen must have been related to the traditional, Alexandrian form of ekdosis, in which the texts of the Three along with other Greek sources had been used and cited in the form of marginal notes.20 The use of the text of the Three does not necessarily mean at all a desire for approximating the text to the Hebrew (it is remarkable that Pap. 967 has a Christian provenance and was created probably in Alexandria).21 This must be an indicator of the general interest toward all kinds of the alternative variants (and the variants of the Three among them) required by this or that particular editor for understanding and commenting upon the text. The alternative readings that had been placed in the scholia thereafter became the cause of the textual variety of different Septuagint manuscripts.

Origenian and Post-Origenian Changes in the Septuagint and the Georgian Translations The post-Origenian history of the Septuagint does not typologically differ from its pre-Origenian stage. In fact, it is a continuation (even though more dynamic) of the same tradition of philological work on the text. The Georgian manuscripts containing the post-Origenian texts are much older than the Georgian ones containing translations of the pre-Origenian text type. The most ancient Georgian manuscripts—the lower layers of 5th–7th century palimpsests—attest the O and L text types of the Septuagint. If in the case of late-medieval (10th cent. and later) Georgian manuscripts it is theoretically possible to suppose that their texts have been revised according to some foreign language source or another Georgian translation, it has been completely ruled out in the case of these fragments. Thus, their texts must be examined as texts focusing on the post-Origenian 4th-century Greek Vorlagen of the Georgian translation.

20  For me it has not been ruled out that the signs and markings used in the papyrus are a reference to the form of such an ekdosis in its Vorlage. Cf. Lilly, Two Books of Ezekiel, 243–57. 21  See Lilly, Two Books of Ezekiel, 257–61.

400

Anna Kharanauli

Jeremiah The text of the Georgian fragments of Jeremiah from the Cairo Genizah (GeoX, 7th cent.)22 follows the O text type, represented by only two sources: 88 (10th cent.) and Syrohexapla (Syh). Thus, the Georgian fragment is quite an ancient and rare witness to O. It reflects the asterisked additions of 88 and Syh encountered in one part of the L text type (so called Lucianic manuscript group) as well. 18:6 ὁ πηλός თიჴაჲ Geo] + (※O) εν χειρι (ταις χερσι Syh 538 Arm GeoO) O L-62538-613 Arm ჴელსა (-თა O) შინა Geo Tht. Hi. = MT α' 18:7 καὶ τοῦ ἀπολλύειν და წარწყმედად Geo] pr (※O) και καθελειν O L-62-538 (om και) Arm და დარღუევად (+ მათ(დ)ა GeoOJ) GeoXOJ Tht. Hi. = MT σ’ 20:12 κύριε] + (※O) των δυναμεων O-233 l-198 ძალთაო Geo = MT; κυριος (pr et Arm = MT) των δυναμεων L Arm Or. θ' 18:6 ὑμεῖς ἐστε] pr (※O) ουτως O-233 62–198 239 538 Bo Aeth Arm ეგრე(თ)ცა GeoXJO ეგრეთ GeoG Or. Tht. PsAmbr. Hi.cit = MT α' σ' θ'.23

At the same time, like O, it does not reflect the same type of addition attested in L, e.g. 18:5 λέγων] + ειπε αυτοις Qmg(sub※)-86mg(sub※) L-62-538-613 არქუ მათ Geo-X.

Neither is there the doublet of L (attested in later Georgian sources as well), containing readings of OG and the Three: 20:9 φλέγον შემწუელი GeoX] φλεγομενον και συνεχομενον L´-198 (om φλεγ.) შემწუველი და შემყენებელი (შეყენებული GeoO) GeoOJ Chr. (φλογιζομενον pro φλεγ.) Tht. cf. α' σ' συνεχομενον 86 (sub σ') Syh; მოტყინარე და შემყენებელი GeoG.24 22  The fragments are kept in Bodleian Library (Oxford, Ms. Georgian C 1[P]) and Cambridge University Library (Taylor-Schechter Ms. 12,183: Ms. Georgian 1 recto-verso, 2 recto-verso). It contains the following parts of the text: 12:10–16; 17:26–27; 18:2–8; and partly 20:9–10,12,14–16. The editions and studies: A. Shanidze (ed.), “Xanmet’i ieremias k’embriǮuli nac’q’vet’ebi [The Cambridge Fragments of the Khanmeti Jeremia]”, Enimk’is moambe [Bulletin of Language, History and Material Culture Institute] II(1) (1937) 29–42 (in Georgian); R.P. Blake (ed.), “Khanmeti Palimpsest Fragments of the Old Georgian Version of Jeremiah”, HTR 25 (1932) 225–72; A. Kharanauli, “Das Chanmeti-Fragment aus Jeremia—Fragen seiner Entstehung und seiner Übersetzungstechnik”, OrChr 85 (2001) 204–36. Along with the khanmeti fragment, there is another different Georgian translation of Jeremia, preserved in Athonite and Jerusalem manuscripts (GeoOJ) (see above, n.8). The hellenophile recension of the latter has been included in the Gelati Bible (GeoG). 23  More examples see in Kharanauli, “Das Chanmeti-Fragment aus Jeremia”, 210–11. 24  Cf. Kharanauli, “Das Chanmeti-Fragment aus Jeremia”, 212.

Septuagint Text Types in the Georgian Translations

401

Apart from agreements with O listed above, there is a case of significant difference from extant O sources: the Georgian transmits a doublet absent from O and extant in other sources related and not related with the Origenian tradition: 17:26 καὶ θύματα Zi.] και θυσιαν και θυμιαματα B-S-130 A-410 36-62 Q-V-26-4686´-198-534-544 c-239-613 LaSg Arab (και θυσιαν in 86 sub ÷) შესაწირავი და საკუმეველი GeoOJ; მსხუერპლთა და საკუმეველთა GeoG;25 საკუმეველი და შესაწირავი (θυμιαμα και θυσιαν) GeoX;26 και θυμιαματα O-233 L-36-407-538 106 Arm Tyc.; θυσιας C Bo Aeth.

Thus, all the Georgian manuscripts, including the khanmeti fragment, contain a pre-Origenian doublet that has been preserved in the oldest sources of the Septuagint and is attested only in some manuscripts of the O L C groups (36-6227 and c-239-613). This doublet was created as a result of comparison of two Greek manuscripts containing the Septuagint and seemingly, the second part of it was not yet adopted by the Vorlagen of extant O L and C sources from the marginal notes.28 The mentioned examples first of all confirm that the absence of asterisked pluses in the O sources and in Georgian reflects the old form of the O text tradition. Secondly they demonstrate that manuscripts containing post-Origenian texts (i.e. manuscripts of O and L groups) have different traditions of adopting the readings corresponding to the MT.29 In other words, the Georgian fragment stresses the fact that the post-Origenian manuscripts which contain additions corresponding to the MT are not based on one and the same constant textual form (a “hexaplaric” recension), they rather represent the result of the work that was competed by philologists who used the same sources, applied the same methodology, but had different interests and worked independently from each other.

25  Cf. J. Ziegler, Beiträge zur Ieremias-Septuaginta (Göttingen: Vandenhoeck & Ruprecht, 1958), 103. 26  It is interesting that later Georgian manuscripts, which in Jeremiah generally follow the L manuscript tradition, also contain this doublet, although in various forms. 27  Regarding the readings of OG in the manuscripts 36–62 cf. J. Ziegler (ed.), Ieremias. Baruch. Threni. Epistula Ieremiae, Septuaginta. Vetus Testamentum graecum XV (2nd edn; Göttingen: Vandenhoeck & Ruprecht, 1976), 81–2. 28  The second possibility, though I do not agree with it, would be that the Georgian reading reflects a variant of O’s substrate, subsequently becoming simplified in the extant sources as the result of the obelisation of και θυσιαν. 29  Cf. e.g., J.W. Wevers, “The L Text of Ezekiel”, in J.W. Wevers/D. Frankel, Studies in the Text Histories of Deuteronomy and Ezekiel (Göttingen: Vandenhoeck & Ruprecht, 2003), 115.

402

Anna Kharanauli

Isaiah A 5th-century Georgian fragment of Isaiah also reflects a post-Origen text.30 It basically follows the L manuscripts of the Septuagint,31 the oldest of which dates back to the 10th cent. And therefore, this Georgian fragment is the oldest source for the investigation of the textual history of L. Additions corresponding to the MT are reflected in the fragment: 1) Those which are attested in O and in one part of the L manuscripts, e.g.: 8:2 Ουρίαν] + (※V 22-48-96) τον ιερεα 26-106 V-oII L´`-233-456 (+ τον μεγαν)-764mg C 403´449´544 მღდელი Geo Eus. Chr. Tht. Arm = MT οἱ γ‘; 2) Those extant in L and not in O, e.g.: 8:7 τὴν δόξαν] pr πασαν 8832 L´’`-46-233-456 393 Bo ყოველი GeoXOJ Chr. Tht. = MT; 3) Those which are present in O and are absent in the L: 8:5 λαλῆσαί μοι ἔτι] + (※) λεγων V-oII C 456 Arm და თქუა Geo = MT σ' θ'.

As I have already mentioned, the text preserved in the Georgian fragments is not contaminated. Therefore, the existence of two different traditions—the O and the L—in the fragment can be only explained by the fact that these two textual traditions had not yet been established when Isaiah was being translated into Georgian, i.e. during an early stage of the post-Origen history of the Greek Isaiah. According to Geo, this stage of the textual history is also characterized by the absence of large asterisked additions, which are characteristic of O and L manuscripts, e.g.:

30  The palimpsest A-844 of the Tbilisi Kekelidze National Centre of Manuscripts contains 54 folios with Isaiah as the underlying text. Only 9 folios are deciphered yet by L. Kajaia, who has kindly given me opportunity to study an unpublished transcription. The deciphered text contains the following parts of Isaiah: 7:20–23, 8:1–8, 40:4–12, 46:6–7, 47:2–6, 49:3–6. Concerning the palimpsest and its text, see L. Kajaia, “Khanmeti Palimpsests”, in A.D. Liublinskaia et al. (ed.), Problemy paleografii i kodikologii v SSSR [Problems of Paleography and Codicology in the Soviet Union] (Moscow: Nauka 1974) 409–27 (in Russian); A. Kharanauli, “The Georgian Translation of the Book of Isaiah and Aporiai of the Lucianic Recension”, in K. de Troyer/T.M. Law/M. Liljestrom (ed.), In the Footsteps of Sherlock Holmes. Studies in the Biblical Text in Honour of Anneli Aejmelaeus (Leuven: Peeters, 2014) 417–36. 31  Late Georgian manuscripts Ath. 1(= O) and the Jer. 7 (= J), being very closely approximate to the text preserved in the fragment, confirm this conclusion in the case of the remaining part of the text. 32  In the case of Isaiah, Greek manuscript 88 frequently contains variants of L.

Septuagint Text Types in the Georgian Translations

403

7:22 γάλα] + (※22-48-96) φαγεται βουτυρον (βουτυρον φαγεται 93-233-456) οτι (> oII) V-oII L´’`-46-233-456 C 403´ Arm Syh Eus. Chr. Tht. = MT π'; 40:7 ἐξέπεσε = Petr] + (※22-48-231-96) οτι πνευμα κυριου επνευσεν εις αυτο αληθως ομοιος χορτω ο λαος (8) εξηρανθη ο χορτος το ανθος εξεπεσε(ν) V-oII L´’`-46456 403´538 Arm Sypa b clim Eus. Tht. = MT σ' θ'; 40:11 ἄρνας] + (※ 88 L-51-96) και (> 88) εν (> lII) τω (> oII) κολπω αυτου βαστασει V-88-oII L´’`-233 403´ 770 Arm Syp Tht. = MT σ' θ'.

The absence of asterisked pluses in the manuscript-groups O or L (as well as in C) or within individual manuscripts of these respective groups are usually explained as an intentional omission of the readings marked with asterisk (i.e. text not belonging to the Septuagint) or as due to their loss during textual transmission. As in the case of Jeremiah I doubt that that was possible for such a large amount of essential readings, which had already become a part of the text. Besides, it remains unclear as well, firstly, why only one part of the asterisked pluses was lost, and secondly, why some of such pluses were lost in one part of the sources and some in another. Later Georgian as well as Greek manuscripts also confirm that the absence of asterisked additions in the Isaiah fragment is not an individual characteristic of the latter; instead it is an indication of a well-expressed trend.33 Thus, I think that our understanding of the textual history of the O and L traditions must be revised. The text-forms preserved in the extant O and L manuscripts do not represent “recensions” (the sound texts of Origen and Lucian having had the goal of approximating the Septuagint text to the Hebrew) which afterwards—in various distinct manuscripts and in varying amounts—in the process of copying the text, gradually and sporadically came to lack the Hexaplaric readings. On the contrary, they represent text forms which are the result of a gradual and sporadic filling in with the alternative readings (with readings from the ekdoseis of the Three among them) to their own substrate.34

1 Ezra The main criteria for identification of “Hexaplaric” and “Lucianic recensions” of the Septuagint are the parallels with the MT. But what about the books where the Hebrew original was apparently unavailable already at least in Origen’s time? J. Ziegler and R. Hanhart suppose the existence of “Hexaplaric” and “Lucianic recensions” for the books not belonging to the Hebrew Canon as well. However here, in contrast to “canonical” books, there are other media (“Mittelglied”) which 33  Half of the large asterisked additions attested in O and/or L have not been reflected in GeoOJ. 34  See A. Kharanauli, “Origen and Lucian in the Light of Ancient Editorial Techniques” in the present volume.

404

Anna Kharanauli

serve as criteria for emendations. In 1 Ezra, for example, Hanhart assumes that such media are the canonical books 2 Ezra and 2 Chronicles,35 considering the parallels with these books as the main recensional characteristic of the L text. This characteristic is further stressed by the Georgian palimpsest fragment of 1 Esdras dated back to the 5th–6th cent. The translation presented by the fragment follows L very closely, having survived only in two Greek manuscripts—19 (12th cent.) and 108 (13th cent.). Consequently, once again, the Georgian manuscript is an oldest witness of the textual tradition it belongs to.36 It reflects all those changes of the L manuscripts of 1 Ezra, which have parallels in canonical books, such as: Lexical changes: 9:14 Λευὶς] οι λευιται (λεβ. 108) L ლევიტელნი GeoX = 9:48, cf Esdr II 10:15 ὁ Λευίτης 9:22 ᾿Ωκίδηλος] ιωζαβαδ L იოზაბათ (iozabat) = Esdr II 10:22 𝔊, cf Esdr II ιαζαβαδ 19; ιζαβαδ 93-108 9:22 Σάλθας] ελασας coni Ted.; ηλασα L ჱლასა (ēlasa) = Esdr II 10:22 et ιλασα 236 46; 9:23 ᾿Ιωανάς B 370? 119 Aeth: cf 1] ελιεζερ L cf ელიაზარ (eliazar) = Esdr II 10:23 𝔊 et ελιαζαρ S 9:35 Μαζιτίας] και μαθθαθιας (mathathias LaV; mathatias LaC) L La cf და მატთია (et mat’tia) = Esdr II 10:43 𝔊, cf ματθαθιας L 248 Compl

Transcription of the Hebrew proper names:

9:14 ᾿Ιεζίας] ιωζιας 19; ιαζιας 108 Lag. იაზია (iazia) = Esdr II 10:15 9:21 ᾿Ιερεήλ] ιειηλ L Ra. იეიჱლ (ieiēl) = 9:27 L et Esdr II 10:21 9:23 ᾿Ωούδας] ιουδας L 71-236 Sy Arm Sixt = Esdr II 10:23 იუდა (iuda) 9:26 ᾿Ιερμάς] ραμιας L = Esdr IIL 10:25 cf. რამია (ramia) = Esdr II 10:25 Ῥαμιὰ 9:27 ᾿Ιερεμώθ] ιεριμωθ L Arm იერიმოთ (ierimot) = Esdr II 10:26 𝔊L

35  R. Hanhart, Text und Textgeschichte des 1. Esrabuches (MSU 12; Gottingen: Vandenhoeck & Ruprecht, 1974), 20. 36  The fragment is a part of Codex Georgicus 2 of the Austrian National Library. It includes the following parts of the text: 1:46–48; 2:4–7; 9:14–23; 9:25–35. For the editions, see: J.N. Birdsall, “Palimpsest Fragments of a Khanmeti Georgian Version of I Esdras”, Mus 85 (1972) 97–105; J. Gippert/Z. Sarjveladze/L. Kajaia (ed.), The Old Georgian Palimpsest: Codex Vindobonensis georgicus 2, Ms. IV: Ezdras Zorobabel. Volume I (Monumenta Palaeographica Medii Aevi, Series Ibero-Caucasica; Turnhout: Brepols, 2007) 4–11–14. About the textual characteristics of the fragment see A. Kharanauli, “Ein Chanmeti-Fragment der georgischen Übersetzung von Esdra I (Fragen der Authentizität, Vorlage und Übersetzungstechnik)”, Mus 116 (2003) 181–216.

Septuagint Text Types in the Georgian Translations

405

Additions: 9:34 Ζαμβρίς, ᾿Ιώσηπος] και ιωσηφ και σελλουμ και αμαριας L და იოსჱფ და სელუმ და ამარია (da iosēp da selum da amaria): cf Esdr II 10:42 (καὶ B L et allii) Σαλουμ (καὶ L et allii) Αμαρια (+ καὶ L et alii) Ιωσηφ 9:35 Νοομά] ναβαυ ιεϊηλ L ნაბავისთა იეიჱლ (nabavista ieiēl) = Esdr II 10:43

Omissions: 2:5 om εἰ L Sy GeoX = Par II 36:23 Esdr II 1:3

Syntactic changes: 9:14 Λευὶς και Σαββαταῖος] σαβεθεος (σαββ. 108) οι λευιται (λεβ. 108) L საბათეოს ლევიტელნი (sabateos levit’elni) GeoX, cf Esdr II 10:15)

Above all these impressive amounts of parallels there are readings in the Georgian fragments which are not attested in the Greek manuscripts of 1 Ezra, but which occur in the Greek sources of 2 Ezra and 2 Par., e.g., alterations of proper names, quantitative changes and transposition: 9:20 αὐτῶν 2°] + κριθεν L: ex κριον ვერ