Collected Essays on the Greek Bible and Greek Lexicography (Contributions to Biblical Exegesis & Theology, 112) 9789042948068, 9789042948075, 904294806X

Citation preview

C O N T R I BU T I O N S TO

BIBLICAL EXEGESIS & THEOLOGY

112

John A. L. Lee

Collected Essays on the Greek Bible and Greek Lexicography

PEETERS

COLLECTED ESSAYS ON THE GREEK BIBLE AND GREEK LEXICOGRAPHY

CONTRIBUTIONS TO BIBLICAL EXEGESIS AND THEOLOGY

SERIES EDITORS K. De Troyer (Salzburg) G. Van Oyen (Louvain-la-Neuve)

ADVISORY BOARD Reimund Bieringer (Leuven) Lutz Doering (Münster) Mark Goodacre (Duke) Bas ter Haar Romeny (Amsterdam) Annette Merz (Groningen) Madhavi Nevader (St Andrews) Thomas Römer (Lausanne) Jack Sasson (Nashville) Tammi Schneider (Claremont)

John A. L. LEE

COLLECTED ESSAYS ON THE GREEK BIBLE AND GREEK LEXICOGRAPHY

PEETERS LEUVEN – PARIS – BRISTOL, 2022

CT

A catalogue record for this book is available from the Library of Congress. © 2022 — Peeters, Bondgenotenlaan 153, B-3000 Leuven ISBN 978-90-429-4806-8 eISBN 978-90-429-4807-5 D/2022/0602/33 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the publisher.

PREFACE Whether one ought to publish a collection of one’s essays is a question not always to be answered in the affirmative. The author may not be the best person to judge. Everything seems ‘important.’ The old standby, ‘at the urging of many friends,’ would not serve in this case (if it ever did): only Jim Aitken urged me to do it; Trevor Evans was supportive. I thank them both. Later, I was greatly encouraged by the welcoming response of Kristin De Troyer as editor of the CBET series, and of Peeters of Leuven. I have had to disappoint some of my darlings. Not everything could be included anyway, for space and copyright reasons, and others were out of date or too slight to be worth reviving. But others were worth including and I am sorry to lose them; perhaps a second collection might be possible. I see that one essay, number 11, was almost a monograph in itself; yet it is only the beginning of what awaits in that inviting field of study. The papers may seem ποικίλα, but through them runs a single thread: Greek language. Not lexicography alone, not the Septuagint and the New Testament as such, nor Renaissance Greek, but the study of Greek itself and what it brings to other areas of scholarship. The papers are published unrevised, except for formating and style changes and occasional improvements in wording. I have added small updates here and there in square brackets, and longer postscripts at the end of some essays. My best thanks are due to Anneli Aejmelaeus for agreeing to write an introduction, both an honour to me and a token of friendship; to Elisabeth Hernitscheck of Peeters for her meticulous help with preparation and submission of copy; and to my wife Diane, who has always understood and encouraged my need to keep at work. John Lee University of Sydney October, 2021

TABLE OF CONTENTS PREFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

V

ABBREVIATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . .

IX

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . .

XV

1 A Note on Septuagint Material in the Supplement to Liddell and Scott . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Aposkeuê in the Septuagint . . . . . . . . . . . . . . . . . . . 3 Equivocal and Stereotyped Renderings in the LXX . . . . . . 4 The Future of zên in Late Greek . . . . . . . . . . . . . . . . 5 Some Features of the Speech of Jesus in Mark’s Gospel. . . . 6 A Non-Aramaism in Luke 6:7 . . . . . . . . . . . . . . . . . 7 The United Bible Societies’ Lexicon and its Analysis of Meanings 8 A Lexical Study Thirty Years On, with Observations on ‘Order’ Words in the LXX Pentateuch . . . . . . . . . . . . . . . . . 9 The Present State of Lexicography of Ancient Greek . . . . . 10 Review of Muraoka, A Greek-English Lexicon of the Septuagint: Chiefly of the Pentateuch and the Twelve Prophets (2002) . . 11 Dimitrios Doukas and the Accentuation of the New Testament Text of the Complutensian Polyglot. . . . . . . . . . . . . . . 12 Exapostellô . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 A Lexicographical Database for Greek: Can It Be Far Off? The Case of ἄμφοδον . . . . . . . . . . . . . . . . . . . . . . . . 14 The Complutensian Polyglot, the Text of Sirach, and a Lost Greek Word . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Review of Muraoka, A Greek-English Lexicon of the Septuagint (2009) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Auxiliary thelô . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Releasing LSJ From Its Past. . . . . . . . . . . . . . . . . . . 18 Why Didn’t St Basil Write in New Testament Greek? . . . . . 19 The Atticist Grammarians . . . . . . . . . . . . . . . . . . . . 20 Etymological Follies: Three Recent Lexicons of the New Testament . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 The onoma Rule . . . . . . . . . . . . . . . . . . . . . . . . .

1 11 21 37 47 71 79 105 117 129 143 183 199 207 223 237 257 279 293 319 339

VIII

TABLE OF CONTENTS

22 The Literary Greek of Septuagint Isaiah . . . . . . . . . . . . 23 The Sayings of the Desert Fathers: Their Evidence for Late Koine Greek . . . . . . . . . . . . . . . . . . . . . . . . . . 24 The Vocabulary of the LXX and Documentary Evidence . . . 25 The Puzzle of John 21:15–17: A Formality Solution . . . . . 26 The Brill Dictionary of Ancient Greek: New Clothes for an Old Emperor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Greek Idiom in the LXX-Pentateuch: The Preposition para . . 28 Back to the Question of Greek Idiom . . . . . . . . . . . . . .

351 373 389 403 407 427 441

BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455 NOTICE OF ORIGINAL PUBLICATION . . . . . . . . . . . . . . . . . . 479 INDEX OF GREEK WORDS . . . . . . . . . . . . . . . . . . . . . . . 483 INDEX OF HEBREW WORDS . . . . . . . . . . . . . . . . . . . . . . 489 INDEX OF BIBLICAL REFERENCES . . . . . . . . . . . . . . . . . . . 491 INDEX OF MODERN AUTHORS . . . . . . . . . . . . . . . . . . . . . 497

ABBREVIATIONS Abbreviations for Greek authors and works are those used by LSJ or an expanded version thereof. Papyrological publications follow Oates et al., Checklist, and epigraphical publications follow Horsley and Lee, ‘A Preliminary Checklist.’ References to text numbers in PCairZen follow LSJ’s practice and omit the first 59(00). All works not listed below are referred to by author and short title (see bibliography). Abbreviations of biblical books, standard works, and series titles follow SBLHBS. BAG = W. Bauer, trans. W. F. Arndt and F. W. Gingrich, Greek-English Lexicon of the New Testament (Chicago and Cambridge, 1957). BAGD = W. Bauer, W. F. Arndt, F. W. Gingrich, and F. W. Danker, A GreekEnglish Lexicon of the New Testament and Other Early Christian Literature (2nd ed.; Chicago, 1979). Bauer2 (1928) = W. Bauer, Griechisch-Deutsches Wörterbuch zu den Schriften des Neuen Testaments (2nd ed.; Gießen, 1928). BDAG = W. Bauer, F. W. Danker, W. F. Arndt, and F. W. Gingrich, A GreekEnglish Lexicon of the New Testament and Other Early Christian Literature (3rd ed.; Chicago, 2000). BDB = F. Brown, S. R. Driver, and C. A. Briggs (eds.), A Hebrew and English Lexicon of the Old Testament (Oxford, 1907). BDF = F. Blass and A. Debrunner, A Greek Grammar of the New Testament and Other Early Christian Literature (trans. and rev. R. W. Funk; Cambridge, 1961). BDR = F. Blass and A. Debrunner, Grammatik des neutestamentlichen Griechisch (rev. F. Rehkopf; Göttingen, 1975). Beekes, EDG = R. Beekes, with L. van Beek, Etymological Dictionary of Greek (2 vols.; Leiden, 2010). BHS = Biblia Hebraica Stuttgartensia (Stuttgart, 1983). Biblia Sacra = Biblia Sacra: Iuxta Vulgatam Versionem, ed. R. Weber et al. (3rd ed.; Stuttgart, 1983). BM = A. E. Brooke and N. McLean, The Old Testament in Greek (3 vols.; Cambridge, 1906–1940). BRAA = W. Bauer, Griechisch-Deutsches Wörterbuch zu den Schriften des Neuen Testaments (6th ed.; rev. V. Reichmann, K. Aland, and B. Aland, Berlin, 1988). Brenton = [L. C. L. Brenton], The Septuagint Version of the Old Testament, with an English Translation (London, [1844]). CGL = J. Diggle et al. (eds.), The Cambridge Greek Lexicon (2 vols.; Cambridge, 2021).

X

ABBREVIATIONS

CHB = S. L. Greenslade (ed.), The Cambridge History of the Bible, Vol. 3: The West from the Reformation to the Present Day (Cambridge, 1963). Chantraine, DELG = P. Chantraine, Dictionnaire étymologique de la langue grecque: histoire des mots (Paris, 1968–1980; new ed. with supplement 2009). Class. = Classical. DDBDP = Duke Databank of Documentary Papyri. DGE = F. R. Adrados et al. (eds.), Diccionario Griego-Español (7 vols. to date; Madrid, 1980–2009). DGENT = J. Mateos et al. (eds.), Diccionario Griego-Español del Nuevo Testamento (5 fascs. to date; Córdoba, 2000–2012). Dizionario Biografico = A. M. Ghisalberti (ed.), Dizionario Biografico degli Italiani (Rome, 1960–2020). DMLBS = R. E. Latham, then D. R. Howlett et al. (eds.), Dictionary of Medieval Latin from British Sources (London, 1975–2013). Engl. = English. EDNT = H. Balz and G. Schneider (eds.), Exegetical Dictionary of the New Testament (3 vols.; Grand Rapids, 1990–1993). Fraser, LGPN, IIIA = P. M. Fraser, E. Matthews et al. (eds.), A Lexicon of Greek Personal Names (vol. IIIA; Oxford, 1997). Frisk, GEW = H. Frisk, Griechisches Etymologisches Wörterbuch (3 vols.; Heidelberg, 1960–1972). GE = F. Montanari, et al. (eds.), The Brill Dictionary of Ancient Greek (Leiden, 2015). GI = F. Montanari, Vocabolario della Lingua Greca (GI) (1st ed.; Torino, 1995; 2nd ed. 2004; 3rd ed. 2013). Göttingen ed. / Gött. = Septuaginta Vetus Testamentum Graecum: Auctoritate Academiae Scientiarum Gottingensis editum. HALOT = L. Koehler and W. Baumgartner, The Hebrew and Aramaic Lexicon of the Old Testament (rev. W. Baumgartner, J. J. Stamm et al., trans. M. E. J. Richardson; 2 vols.; Leiden, 2001). HDM, La Bible grecque = M. Harl, G. Dorival, and O. Munnich, La Bible grecque des Septante: Du judaïsme hellénistique au christianisme ancien (Paris, 1988).1 Heb. = Hebrew HR = E. Hatch and H. A. Redpath, A Concordance to the Septuagint and the Other Greek Versions of the Old Testament (Oxford, 1897). HR Suppl. = HR Supplement, by H. A. Redpath (Oxford, 1906), Part II: Concordance to Portions of Ecclesiasticus with Hebrew Equivalents. IE = Indo-European. KB, Lexicon = L. Koehler and W. Baumgartner, Lexicon in veteris testamenti libros (Leiden, 1958). KB, Grammatik, I = R. Kühner and F. Blass, Ausführliche Grammatik der griechischen Sprache, I.1, 2 Elementar- und Formenlehre (Hannover, 1890, 1892). KG, Grammatik, II = R. Kühner and B. Gerth, Ausführliche Grammatik der griechischen Sprache, II.1, 2 Satzlehre (Hannover and Leipzig, 1898, 1904). 1

The title page has the names in the order Dorival, Harl, Munnich, but the front cover has Harl, Dorival, Munnich. I have chosen the latter τιμῆς ἕνεκα Marguerite Harl.

ABBREVIATIONS

XI

KJV = King James Version (1611). Kriaras, Λεξικό = E. Kriaras, Λεξικὸ τῆς μεσαιωνικῆς Ἑλληνικῆς δημώδους γραμματείας 1100–1669 (14 vols. to date; Thessaloniki, 1968–1997). Lampe, PGL = G. W. H. Lampe, (ed.), A Patristic Greek Lexicon (Oxford, [1968]). Lat. = Latin. LBG = E. Trapp et al. (eds.), Lexikon zur byzantinischen Gräzität: Besonders des 9.–12. Jahrhunderts (8 fascs.; Vienna, 2017). LEH, Lexicon = J. Lust, E. Eynikel, and K. Hauspie, Greek-English Lexicon of the Septuagint: Revised Edition (2 vols.; Stuttgart, 2003). Lewis and Short = C. T. Lewis and C. Short, A Latin Dictionary (Oxford, 1879). Lex. Acad. = Λεξικὸν τῆς Ἑλληνικῆς Γλώσσης. Αʹ. Ἱστορικὸν Λεξικὸν τῆς Νέας Ἑλληνικῆς, τῆς τε κοινῶς ὁμιλουμένης καὶ τῶν ἰδιωμάτων (5 vols.; Athens, 1933–1989). Lex. Proia = Λεξικόν της Ελληνικής Γλώσσης (Athens, 1933). LN = J. P. Louw and E. A. Nida et al., Greek-English Lexicon of the New Testament: Based on Semantic Domains (2 vols.; New York, 1988). [In paper 7] LS1 (1843) = H. G. Liddell and R. Scott, A Greek-English Lexicon: Based on the German work of Francis Passow (Oxford, 1843). LS1 Abridged = [G. Marshall], A Lexicon Chiefly for the Use of Schools Abridged from the Greek-English Lexicon of H. G. Liddell, M.A., and R. Scott, M.A. (Oxford, 1843; 28th ed. 1903). LSJ = H. G. Liddell and R. Scott, A Greek-English Lexicon (rev. and augm. H. Stuart Jones; Oxford, 1940). LSJ Suppl. (1968) = H. G. Liddell, R. Scott, and H. Stuart Jones, Greek-English Lexicon: A Supplement, ed. E. A. Barber et al. (Oxford, 1968). LSJ Suppl. (1996) = H. G. Liddell and R. Scott, A Greek-English Lexicon: Revised Supplement, ed. P. G. W. Glare, with assist. of A. A. Thompson (Oxford, 1996). LXX.D = W. Kraus and M. Karrer (eds.), Septuaginta Deutsch: Das griechische Alte Testament in deutscher Übersetzung (Stuttgart, 2009). LXX.D Komm. = M. Karrer and W. Kraus (eds.), Septuaginta Deutsch: Erläuterungen und Kommentare zum griechischen Alten Testament (2 vols.; Stuttgart, 2011). LXX.H, 3 = E. Bons and J. Joosten (eds.), Handbuch zur Septuaginta: Handbook of the Septuagint, vol. 3 Die Sprache der Septuaginta: The Language of the Septuagint (Gütersloh, 2016). Mega Lexikon = I. S. Zervos (ed.), Μέγα Λεξικόν της Ελληνικής Γλώσσης (9 vols.; Athens, 1953). Moulton, Grammar, I = J. H. Moulton, A Grammar of New Testament Greek, vol. I Prolegomena (Edinburgh, 1908). MH, Grammar, II = J. H. Moulton and W. F. Howard, A Grammar of New Testament Greek, vol. II Accidence and Word-Formation (Edinburgh, 1919– 1929). MHT, Grammar, III = J. H. Moulton, A Grammar of New Testament Greek, vol. III Syntax, by N. Turner (Edinburgh, 1963). MHT, Style = J. H. Moulton, A Grammar of New Testament Greek, vol. IV Style, by N. Turner (Edinburgh, 1976).

XII

ABBREVIATIONS

Migne, PG = J.-P. Migne, (ed.), Patrologiae cursus completus: Series Graeca (162 vols.; Paris, 1857–1866). Migne, PL = J.-P. Migne, (ed.), Patrologiae cursus completus: Series Latina (217 vols.; Paris, 1844–1864). MM = J. H. Moulton and G. Milligan, The Vocabulary of the Greek Testament Illustrated from the Papyri and Other Non-Literary Sources (5 vols. 1914– 1929; one-vol. edition London, 1930). MT = Masoretic Text. NA28 = B. and K. Aland et al. (eds.), Novum Testamentum Graece (28th ed.; Stuttgart, 2012). NETS = A. Pietersma and B. G. Wright (eds.), A New English Translation of the Septuagint and other Greek Translations Traditionally Included under that Title (New York/Oxford, 2007). New Documents = G. H. R. Horsley et al. (eds.), New Documents Illustrating Early Christianity (vols. 1–6; Macquarie University, 1981–1992; vol. 7 ed. S. R. Llewelyn, 1994). OED = Oxford English Dictionary (2nd ed.; 20 vols.; Oxford, 1989). OGELD = D. N. Stavropoulos (ed.), Oxford Greek-English Learner’s Dictionary (Oxford, 1988). OLD = P. G. W. Glare (ed.), Oxford Latin Dictionary (Oxford, 1968–1982). Pauly-Wissowa, RE = Paulys Real-Encyclopädie der classischen Altertumswissenschaft (50 vols.; Stuttgart, 1894–1980). Pent. = Pentateuch. Preisigke, Wörterbuch = F. Preisigke and E. Kiessling, Wörterbuch der griechischen Papyrusurkunden (vols. 1–3 by Preisigke, Berlin, 1925–31; vol. 4 by Kiessling et al., Berlin/ Marburg/ Wiesbaden, 1944–1993). Rahlfs / Ra. = A. Rahlfs, Septuaginta: Id est Vetus Testamentum graece iuxta LXX interpretes (rev. R. Hanhart; Stuttgart, 2006). RBLG = P. Boned Colera et al. (eds.), Repertorio Bibliográfico de la Lexicografía Griega (Madrid, 1998). SBLHS = The SBL Handbook of Style (2nd ed.; Atlanta, 2014). Schwyzer, Grammatik, I = E. Schwyzer, Griechische Grammatik, I. Allgemeiner Teil. Lautlehre. Wortbildung; Flexion (Munich, 1938). Schwyzer, Grammatik, II = E. Schwyzer, and A. Debrunner, Griechische Grammatik, II. Syntax und syntaktische Stilistik (Munich, 1950). Shipp, MGE = G. P. Shipp, Modern Greek Evidence for the Ancient Greek Vocabulary (Sydney, 1979). Th. = Theodotion. Thomson-Muses = C. Thomson, The Septuagint Bible (rev. C. A. Muses; 2nd ed.; Indian Hills, 1960). TLG = Thesaurus Linguae Graecae (online at ). TLG Canon = L. Berkowitz, and K. A. Squitier, Thesaurus Linguae Graecae: Canon of Greek Authors and Works (3rd ed.; New York, 1990). TLL = Thesaurus Linguae Latinae (Leipzig, 1900–). Vetus Latina: Sirach = Vetus Latina: Die Reste der Altlateinischen Bibel, Band 11/2 Sirach (Ecclesiasticus), 9. Lieferung Sir 23,7–24,47; Register, ed. Walter Thiele (Freiburg, 2005).

ABBREVIATIONS

Note abbreviations of the following biblical books: 1 2 1 2

Sam = 1 Samuel (MT) = 1 Kgdms = 1 Kingdoms (LXX) Sam = 2 Samuel (MT) = 2 Kgdms = 2 Kingdoms (LXX) Kgs = 1 Kings (MT) = 3 Kgdms = 3 Kingdoms (LXX) Kgs = 2 Kings (MT) = 4 Kgdms = 4 Kingdoms (LXX)

Siglum ~ = corresponds to

XIII

INTRODUCTION By Anneli AEJMELAEUS

Reading through the essays of this collection has been for me like an adventure through well-familiar terrain but taking new paths, going through thickets and rocky ground, and seeing everything from a fresh angle. Ever since I got my hands on the freshly printed book titled A Lexical Study of the Septuagint Version of the Pentateuch (1983), by a scholar named J.A.L. Lee, I have been a great fan of this scholar (whose first name I learned only decades later), who elucidated the language used by the Septuagint translators so impressively from contemporary sources, either newly discovered or at long last thoroughly examined. It was a major opening in Septuagint scholarship, and for me, it was like a revelation that gave a different perspective on the Septuagint translators and their milieu in Hellenistic Alexandria, very different from my own translation-technical approach that concentrated on the Hebrew-Greek correspondence of syntactical features. Reading these essays gave me a vivid recollection of that early eye-opening experience. This collection contains essays written over a period of five decades, organized in their order of appearance. It displays a kind of cross-section of a long and successful scholarly career and finely complements the bookform production of the author. The essays also reflect the great changes that have taken place in the conditions and practices of scholarly work during this long career. The themes discussed in the collection range from classical Greek to koine and biblical Greek, from lexicography to the creation of databases, from the analysis and history of single words to assessments of stylistic levels of expressions, and from Atticism to the mysteries of the Complutensian polyglot. The strength of John Lee is that nothing Greek is without interest for him. Whatever the starting point, his unparalleled knowledge of the Greek language, his thorough familiarity with texts from different periods, as well as his perseverance in going through all the relevant material allow him to draw conclusions on the development of the language through centuries, to place linguistic phenomena on that continuum, and to elucidate the evolving meanings and uses of expressions against that background.

XVI

INTRODUCTION

There is no need to give a summary of the contents of the essays at hand in this introduction. Each essay begins with an abstract introducing its aims and results. At the end of each essay, the results and further conclusions are once more collected in a summary. To some essays, the author has added actualizing comments concerning discussions that were sparked by the essay or interesting details about the process of writing. At this point, it is befitting to highlight the major strands in the collection and in John Lee’s scholarly work. The collection opens and closes with essays on Septuagint Greek, which seems to have been a central object of John Lee’s enthusiasm and scholarship from early on and throughout his career. His specialty is to look at the language produced by the Septuagint translators as Greek representative of contemporary koine Greek usage. His ambition is to match words and expressions that have been labelled as special Septuagint language or translationese with parallels from genuine Greek texts, contemporary documentary sources as well as literature. At the same time, unlike many others who interest themselves in the Septuagint per se, he always pays due attention to the Hebrew source text in order to be able to differentiate between interference from the underlying Hebrew and expressions of the translators’ Greek usage and competence. He works like a true detective, and the results are both convincing and amazing. The availability of contemporary koine texts is naturally a decisive prerequisite for this kind of research, but it takes a scholar like John Lee not to be satisfied with mere sampling but to pursue the study as far towards perfection as possible. Of course, the computers that arrived on the scene midway through his career and the evergrowing number of texts in electronic searchable form have had a tremendous effect on this project. Moreover, the results of this kind of thorough investigation can be made fruitful in different ways: in the linguistic and lexical study of the Septuagint as well as for using the Septuagint as evidence for contemporary koine Greek, in the relative dating of the various parts of the Septuagint based on the development of the language, in the assessment of the influence of the Septuagint on New Testament Greek, in describing the development of the meaning of vocabulary in focus throughout the history of the Greek language, not to speak of the benefits to numerous other areas of biblical study, not least of all textual criticism. The essays at hand can be warmly recommended to scholars and students of the Greek Bible whatever their area of study. Mutatis mutandis, the principles that govern research on Septuagint Greek apply to the linguistic study of the New Testament as well. In each case, the relationship between biblical Greek and contemporary Greek usage is

INTRODUCTION

XVII

in principle the same, just with a difference in time period. Corresponding to the possible interference from the underlying Hebrew in the Septuagint, the New Testament carries the possibility of influence from the Septuagint, which was a sacred text for New Testament writers, a text that was frequently studied as well as quoted, alluded to, and imitated. In both cases, John Lee underlines the priority of the criterion of whether similar usage can be detected in contemporary sources. Old inherited explanations of features in New Testament Greek, and corresponding definitions of meaning, that claim Semitic/Aramaic or Septuagint influence are repeatedly found to be in need of correction in the light of contemporary parallels. An enormous amount of time and energy has been invested by John Lee in examining various lexica of New Testament and Septuagint Greek. In the case of the Septuagint, the history of lexicography is much shorter than in the case of the New Testament, to which John Lee has dedicated a whole book. Several essays in this collection deal with various dictionaries of the Septuagint, of the New Testament, as well as of ancient Greek in general. Writing a review article on a lexicon is not a task for anybody, but John Lee is matchless in this regard. He is thorough and critical in his scrutiny of each lexicon and strict in his argumentation, illustrating his points by rich examples. He brings to light all the shortcomings and inconsistences in the presentation of the material as well as in adherence to the principles of definition and glossing. He can be very sharp in his criticism, but he can also express it in a gentle tone, reminding the reader again and again that he does not wish so much to criticize or blame as to help improve the lexicon for a new edition and to enhance the level of ancient Greek lexicography. Every student of languages has a high regard for dictionaries, and they are no doubt an important tool for working especially on texts in classical languages. But to what extent can we rely on them? Discussions on the various dictionaries reveal the complexities of producing a dictionary as well as the very human nature of most of the available products with which we need to get along as students and scholars. It is striking to see how most dictionaries are dependent on their predecessors, repeating the glosses and quotations given in them and taking for granted questionable explanations of rare cases and even downright erroneous definitions of meaning. It is important for a dictionary-user to be aware of such problems, to know what to expect from a dictionary and where to suspect deficiencies. While reviewing existing lexica, John Lee proceeds in a most pedagogical way, giving ample evidence and clear arguments for each of his points, offering sample entries and explaining the principles to be followed, so that the reader actually receives a complete introduction into lexicography.

XVIII

INTRODUCTION

At the same time, a wide range of language phenomena are investigated. Theory goes hand in hand with praxis. The essays at hand reveal a deep concern for the future of ancient Greek lexicography. Dictionaries in book-form have obviously come to the end of their time, at least as comprehensive treasuries of the Greek language. When envisioning the future of Greek lexicography, John Lee suggests that a database be founded for the collection of all relevant material, pieces of research on Greek vocabulary scattered in different publications and information on the occurrence and usage in the various texts of different periods of the Greek language. In fact, he already proposed this a quarter of a century ago and has repeated it at regular intervals. Such a database would be the only sensible solution for the advancement of ancient Greek lexicography, but it might take some time to find an agent powerful enough to organize such an immense endeavour. In the meantime, the work could be launched on a smaller scale limited to certain periods or literatures, and the various databases that come about this way could ultimately add up to a real “Thesaurus Linguae Graece.” As for the area of Septuagint Greek, John Lee emphasizes that he believes this to be the new desideratum of Septuagint studies. Even a database that only compiles all the existing material for the Greek language of the Septuagint is a huge task. Nevertheless, this is no doubt the sole correct way to proceed. One major obstacle to the realization of this proposal might be the present-day science policy that does not favour longterm projects. This is really a paradox. Now that we have the technology to handle all that huge amount of data much faster and more effectively than the previous generations that have worked from one century to another on book-form dictionaries almost beyond the size that can be handled by human hands, modern scholars are rather expected to concentrate on shortterm projects, and such fundamental research is no longer in fashion. And yet, the need for a new approach to ancient Greek lexicography, or any part of it, is obvious, as demonstrated by John Lee. Let us hope that this collection of essays finds its way into the hands of young scholars who get inspired by it and are bold enough to begin to realize this proposal. It remains for me to congratulate John Lee on his long and successful career – half a century so far – as well as on his forthcoming milestone birthday. I wish him many happy returns and hope that we shall all soon see his dream come true. Anneli Aejmelaeus

1 A NOTE ON SEPTUAGINT MATERIAL IN THE SUPPLEMENT TO LIDDELL AND SCOTT 1969

Abstract In 1968 a Supplement to LSJ was published, containing 153 pages of ‘corrections and additions’ to LSJ (1940) that had accumulated in the hands of Oxford University Press up to that time. Among them were many insertions and changes related to the LXX. This review examines (some of) the new LXX material. A number of methodological faults are detected and demonstrated by examples. The most notable is that of attributing to a Greek word, without adequate justification, a new meaning derived from the meaning of the Hebrew word it translates.

I In the recently published Supplement to Liddell and Scott’s GreekEnglish Lexicon there are a large number of entries introducing new material from the LXX. Below are discussed certain of these which are open to criticism, less in points of detail than as regards the method and assumptions underlying the treatment of the words concerned. It is a basic principle of lexicography that in order to establish the existence of a new sense of a given word incontrovertible examples of that sense must be found. So long as the word can be understood in one of its established senses without undue strain, it ought to be classified under that sense. Only if the new sense is clearly demanded by the context can it be regarded as definitely established. The authors of the Supplement have, for particular reasons, introduced unjustified new senses for a number of words in neglect of this principle. Thus under ψυχή an additional sense is given which arises from the error of including as part of the indicator value of a word a notion expressed by the context in which it is used.

2

COLLECTED ESSAYS ON THE GREEK BIBLE

Add ‘VIII. slave, πᾶσαν ψυχήν LXX Ge. 12.5, 1Ma. 10.33; dub. sens., ψυχὰς ἀνθρώπων Apoc. 18.13.’

The two LXX passages are as follows: Gen 12:5 καὶ ἔλαβεν Ἀβρὰμ τὴν Σάραν γυναῖκα αὐτοῦ καὶ τὸν Λὼτ υἱὸν τοῦ ἀδελφοῦ αὐτοῦ καὶ πάντα τὰ ὑπάρχοντα αὐτῶν, ὅσα ἐκτήσαντο, καὶ πᾶσαν ψυχήν, ἣν ἐκτήσαντο ἐν Χαρράν, καὶ ἐξήλθοσαν. 1 Macc 10:33 (Ra.) καὶ πᾶσαν ψυχὴν Ιουδαίων τὴν αἰχμαλωτισθεῖσαν ἀπὸ γῆς Ιουδα εἰς πᾶσαν βασιλείαν μου ἀφίημι ἐλευθέραν δωρεάν.

‘Slave’ is a possible translation of ψυχήν in these contexts, but it does not follow that ψυχή had itself developed a new sense ‘slave.’ Such a development would not be impossible (cf. σῶμα in late Greek), and it could have originated in transitional contexts such as these, but there is no evidence whatever for it. What ψυχή refers to [emphasis added] in these two places (household slaves; captives in war) is indicated by other words in the sentence, not by ψυχή itself, which can be readily translated ‘soul’ or ‘person.’ Unless an example of ψυχή meaning ‘slave’ quite apart from the indications given by other words in the context can be found, the listing of these instances under a new heading is unjustified; they belong with the other examples of ψυχή referring to persons, in section IV.1. Moreover, in Gen 12:5 ψυχή is, as very frequently throughout the LXX, simply a literal translation of the Hebrew word ‫נפשׁ‬, ‘soul,’ ‘person.’ The same is probably true of the example in 1 Macc, the Hebrew original of which is now lost. The inclusion of the difficult Rev 18:13 is misleading. ψυχάς cannot have the sense ‘slaves’ there. It is the whole phrase ψυχὰς ἀνθρώπων that may mean ‘slaves.’ The same underlying error is seen in the following entries. S.v. ἐκκαίω: Add ‘IV. ἐ. ὀπίσω τινός pursue with fierce enmity, LXX 3 Ki. 20(21).21.’

The passage is: 3 Kgdms 20(21):21 (Ra.) ἰδοὺ ἐγὼ ἐπάγω ἐπὶ σὲ κακὰ καὶ ἐκκαύσω ὀπίσω σου (‫ )ובערתי אחריך‬καὶ ἐξολεθρεύσω.

The rendering ‘pursue with fierce enmity,’ though no doubt a possible translation of the Greek, obscures the actual value of ἐκκαύσω, which is clearly intended to be a literal rendering of the Hebrew word, meaning ‘burn,’ etc.1 If there is any idea of pursuit it is conveyed by the combining of ἐκκαύσω with ὀπίσω σου, not by ἐκκαύσω itself; and the idea of 1

‫ בער‬piel, ‘kindle, burn, consume, devastate,’ BDB.

A NOTE ON SEPTUAGINT MATERIAL

3

enmity is suggested by the general tone of the passage. Admittedly the construction in which ἐκκαίω is here used is unusual for Greek, but that is a result of the word-for-word rendering of the original. There is no new sense of ἐκκαίω here.2 S.v. ἐπίπεμπτος. For the present section II we are to substitute: ἐπίπεμπτον, τό, one-fifth of the votes in a trial, Ar. Fr. 201, cf. Eup. 65. 2. a fifth in addition, LXX Le. 5.16; τὸ ἐ. a fifth as a fine, IG 7.3073.1 (Lebadea, ii B.C.).

Lev 5:16 reads as follows: καὶ ὃ ἥμαρτεν ἀπὸ τῶν ἁγίων, ἀποτείσαι, καὶ τὸ ἐπίπεμπτον προσθήσει ἐπ’ αὐτὸ καὶ δώσει αὐτὸ τῷ ἱερεῖ.

That the fifth is additional is indicated by προσθήσει. The use of ἐπίπεμπτον here is no different from that in 1. and ought to be listed there. So also the IG example, which, as the Supplement’s own translation shows, does not differ from sense 1.

II A somewhat different error of method, that of supposing without adequate reason that the Greek word has the same meaning as the Hebrew word it renders, has produced unwarranted new meanings in other entries in the Supplement. A typical example is found under κόλπος: III, after ‘6.’ insert ‘bowl or dish, LXX Pr. 19.21(24), 26.15.’

The passages are: Prov 19:21(24) (Ra.) ὁ ἐγκρύπτων εἰς τὸν κόλπον αὐτοῦ χεῖρας ἀδίκως, οὐδὲ τῷ στόματι οὐ μὴ προσαγάγῃ αὐτάς. Prov 26:15 (Ra.) κρύψας ὀκνηρὸς τὴν χεῖρα ἐν τῷ κόλπῳ αὐτοῦ οὐ δυνήσεται ἐπενεγκεῖν ἐπὶ τὸ στόμα.

The proposed meaning is based solely on the meaning of the Hebrew word to which κόλπος corresponds in these two places, i.e., ‫‘( צלחת‬dish,’ BDB; ‘pan,’ KB, Lexicon). There is no support for such a sense in the known uses of κόλπος; if it had developed this sense it is hardly to be doubted that we should have other examples of it. Nor are there any grounds 2

Cf. the uses of ἐκκαίω in its fifty or so other occurrences in the LXX: (1) act. ‘kindle,’ ‘set alight,’ pass. ‘be kindled,’ lit. and fig. (ὀργὴ κυρίου etc.); pass. ‘be stoked up’ (Dan 3:22 κάμινος); (2) act. ‘burn’ (trans.); ‘scorch’; pass. ‘be burnt.’

4

COLLECTED ESSAYS ON THE GREEK BIBLE

for assuming that κόλπος here is an exact equivalent of the word it translates. There is no point of connexion between the root ‫ צלח‬and κόλπος which would suggest that the translation is an etymologising one of the kind quite often found in the LXX, and the context of the Greek does not demand that κόλπος should have the same meaning as the Hebrew. Translating κόλπος in these two passages as ‘bosom,’ ‘pocket,’ gives just as good a sense as translating it ‘bowl,’ since precisely what the sayings as a whole are intended to convey in the Greek is something we can only guess at. There is no difficulty in supposing the translator did not know the meaning of the uncommon word ‫צלחת‬.3 There is moreover a certain improbability in the suggestion that the Proverbs translator, whose rendering Thackeray classified as ‘literary,’ under ‘paraphrases and free renderings,’4 would use the word κόλπος in a very unusual sense, unlike his use of it elsewhere,5 in a context where its normal sense would seem most natural. It is not as if no other words for ‘bowl,’ ‘dish,’ were available. That it must not be assumed that a word in the LXX will automatically have the same sense as the Hebrew word it renders is a principle which is not likely to be doubted by anyone.6 But this is not to say that the LXX can always be understood purely as Greek. Words are in fact very frequently used in senses as far as we know abnormal for Greek, and their meaning can be established only from the Hebrew original rendered in each instance. The cases, however, where a word can be assumed to have a sense unlike its known senses and the same as that of its Hebrew original are quite special ones to which certain conditions apply. These conditions are, I suggest, somewhat as follows: (1) if a Greek word whose main sense coincides more or less with what is felt to be the main sense of a certain Hebrew word is used to render that Hebrew word 3

4 5

6

Elsewhere only 2 Kgs (= LXX 4 Kgdms) 21:13 (rendered ἀλάβαστρος), and perhaps 2 Chr 35:13 ‫צ לחות‬, ֵ taken as pl. of ‫צלחת‬, KB, Lexicon (LXX rendering not clear). That the LXX translators at times display a limited knowledge of Heb. is generally accepted: cf. Swete, Introduction, 319, 330; Jellicoe, Septuagint, 329. It is interesting to note, however, that according to Toy, Proverbs, on 19:24, ‫ צלחת‬was taken as ‘bosom’ by some medieval Jewish commentators; as ‘armpit’ by Aq., Sym., Targ., Lat., and ‘Rashi reports a rendering slit in a garment (= bosom).’ It would seem possible then that the LXX rendering is not just a mistake, but represents a traditional understanding of the word in these passages. Thackeray, Grammar, 13. See Prov 6:27; 16:33; 17:23; 30:4. There and elsewhere in the LXX, about 35 times, κόλπος is used in its normal senses, except perhaps in 3 Kgdms 22:35 bis ὁ κόλπος τοῦ ἅρματος (κόλπος 2º ~ ‫קיח‬, ‘bosom,’ etc.). There is similarly no sign of a meaning ‘bowl’ in any of its NT occurrences. See, e.g., Deissmann, Bible Studies, 73–4.

A NOTE ON SEPTUAGINT MATERIAL

5

in an instance where the Hebrew word has a sense unparalleled in the Greek word (‘Hebraism,’ e.g. δίδωμι rendering ‫ נתן‬both when it means ‘give’ and when it means ‘put,’ ‘make’);7 or, what is virtually the same, if a Hebrew word is rendered by its literal Greek equivalent in a context (e.g., an idiomatic expression) which is unusual for the Greek word (‘literalism,’ e.g., υἱός rendering ‫ בן‬in Gen 11:10 Σὴμ υἱὸς ἑκατὸν ἐτῶν, ‘Sem was a hundred years old’); (2) if the Greek word appears to copy the etymological connexions, real or fancied, of the Hebrew word (‘etymologising,’ e.g., Cant 4:13 ἀποστολαί σου for ‫שׁלחיך‬, cf. Suppl. s.v.); (3) if the context of the Greek, preferably in more than one place, is such that it is impossible to understand the word in any other sense than that of its Hebrew original (e.g., κῶλον rendering ‫פגר‬, ‘corpse,’ Num 14:29, etc.).8 This last type is rare and it may be that further evidence of the Greek word would show the use to be normal for Greek. Without an indication like one of these there can be no certainty that the LXX rendering in a given instance is an exact equivalent of the Hebrew original, and it is precarious to suggest for it on the basis of the Hebrew a meaning which differs considerably from the known uses.9 Another entry affected by the same faulty procedure is that under βάσανος: III. 2, add ‘; … shame, disgrace, LXX Ez. 16.52, al.’ Ezek 16:52 (Ra.) καὶ σὺ κόμισαι βάσανόν σου, ἐν ᾗ ἔφθειρας τὰς ἀδελφάς σου ἐν ταῖς ἁμαρτίαις σου, αἷς ἠνόμησας ὑπὲρ αὐτὰς καὶ ἐδικαίωσας αὐτὰς ὑπὲρ σεαυτήν.

Here and in Ezek 16:54; 32:24, 30 βάσανος corresponds to MT ‫כלמה‬, (‘insult,’ ‘reproach,’ ‘ignominy,’ BDB). It is difficult to explain why the Hebrew was rendered in this way, since it is clear that the translators of Ezekiel were familiar with the root ‫כלם‬: cf., e.g., in this same verse 16:52 the rendering of ‫ כלמה‬by ἀτιμία, and in 34:29 etc. by ὀνειδισμός. It is nevertheless dubious to assume that βάσανος was intended to be an exact 7

8

9

Cf. Weinreich, Languages in Contact, 48 on one of the major types of lexical interference possible in the speech of bilingual speakers: ‘If two languages have semantemes, or units of context, which are partly similar, the interference consists in the identification and adjustment of the semantemes to fuller congruence.’ [In ‘Context is King,’ 172–4 Pietersma has challenged the meaning ‘corpse’ for κῶλον, arguing that it has its usual Greek sense of ‘limb,’ ‘member’ (of the body). I think he is right. In Gen 15:11 the translator renders ‫ פגרים‬by σώματα then explicates it by adding τὰ διχοτομήματα αὐτῶν. The meaning of ‫ פגר‬itself could be looked into and refined.] I leave aside of course the difficulty that it is impossible to be certain that the translator had the same Hebrew as found now in MT. It is usual to assume that it was the same unless there are strong indications to the contrary.

6

COLLECTED ESSAYS ON THE GREEK BIBLE

equivalent of ‫כלמה‬. The evidence we have for the meaning of βάσανος in Greek is quite clear: ‘touch-stone,’ etc.; ‘torture,’ ‘torment,’ are its usual senses.10 In the four places where βάσανος renders ‫ כלמה‬it can be readily translated ‘torment.’ Cf. also its three other occurrences in Ezekiel, at 3:20; 7:19 ~ MT ‫מכשׁול‬, ‘stumbling-block’; 12:18 ~ MT ‫רגזה‬, ‘quivering,’ ‘quaking.’ Are we to suppose that βάσανος has the same meaning as the Hebrew in those places also? The meanings proposed in the following cases are I believe open to suspicion on the same grounds: ἀδόκιμος ‘ἀ. ἀργύριον dross,’ Prov 25:4; Isa 1:22.11 ἀθετέω ‘treat despotically,’ Deut 21:14. ἀντιπίπτω ‘dispute,’ ‘strive,’ Num 27:14. ἐγκολλάω ‘reach as far as,’ Zech 14:5. ἐκκρούω Pass., ‘to be thrust out,’ Deut 19:5. ἐμπυρισμός ‘rust’ (on plants), 3 Kgdms 8:37. πνοή ‘gap,’ ‘vent,’ Neh 6:1. πύρωσις ‘rust’ (on plants), Amos 4:9. σωτήριος ‘peace-offering,’ Lev 3:6; 7:11, al.12

III In certain entries we find not only that the Greek word has been wrongly assumed to have the same sense as the word it translates but also that the meaning of the Hebrew word is itself not generally agreed upon. Thus the meaning given for the Greek word derives from only one of the possible meanings of the Hebrew word, selected by the authors of the Supplement as the right one. For instance, s.v. κάνθαρος: Add ‘VII. = Lat. cantherius, spar, rafter, LXX Hab. 2.11.’ Hab 2:11 διότι λίθος ἐκ τοίχου βοήσεται, καὶ κάνθαρος (~ ‫ )כפיס‬ἐκ ξύλου φθέγξεται αὐτά.

The meaning of ‫( כפיס‬found only here in MT) upon which the proposed meaning of κάνθαρος is clearly based is given by BDB with some hesitation, ‘appar. (si vera l.) term. tech. for some beam in a house, perhaps 10 11

12

See LSJ; BDAG; and for examples contemporary with the LXX, Preisigke, Wörterbuch. Apart from the main objection to which this entry is open there is the additional point that the phrase ἀδόκιμον ἀργύριον does not in fact occur in Isa 1:22. The text is: τὸ ἀργύριον ὑμῶν ἀδόκιμον· οἱ κάπηλοί σου μίσγουσι τὸν οἶνον ὕδατι (‘your money is not genuine …’). For a full discussion of this term in the LXX, giving no support to the Supplement’s meaning, see Daniel, Recherches, 272–81. [Cf. now Lee, Greek of the Pentateuch, 197.]

A NOTE ON SEPTUAGINT MATERIAL

7

rafter, or girder,’ and does not appear at all in KB, Lexicon, who give the meaning ‘stucco-work (laid on panelling).’ What κάνθαρος here was in fact intended to mean cannot be known with certainty, at least with our present evidence. It is usually ‘beetle’ and various objects (a type of cup, boat, fish, ornament, and the mark under the tongue of the Apis-bull, Hdt. 3.28) apparently so called from similarity to a beetle. The sense ‘beetle’ may be possible in Hab 2:11, but seems unlikely. Perhaps the meaning is ‘knot,’ suggested by Schleusner.13 This, though unsupported by a parallel, would give a satisfactory sense and be more in keeping with the known uses than ‘rafter.’ (κάνθαρος occurs only here in the LXX.) The identification of κάνθαρος with cantherius is mistaken and the Latin word therefore irrelevant. κάνθαρος is to be identified with cantharus,14 and cantherius probably with κανθήλιος (-ον).15 S.v. σκελίζω: Delete ‘LXX Je. 10.18’, [i.e. from its present place in 1.] and add ‘2. used for translating Hebr. qāla (uprooted), LXX Je. 10.18.’ Jer 10:18 ἰδοὺ ἐγὼ σκελίζω τοὺς κατοικοῦντας τὴν γῆν ταύτην ἐν θλίψει, ὅπως εὑρεθῇ ἡ πληγή σου.

We note first of all the minor point that the Hebrew word translated by σκελίζω is in fact ‫( קולע‬participle of ‫)קלע‬. The usual meaning given for this word, in accordance with its use elsewhere (twice of slinging stones), is ‘sling forth’ (so BDB; KB, Lexicon). The meaning ‘uproot’ for it in this passage derives from a proposal by G. R. Driver.16 I would not presume to criticize this proposal;17 the point is merely that unless the meaning of the Hebrew word can be regarded as definitely established it is at least doubtful to use the word as evidence for the meaning of the Greek word. There is no reason to suppose that this Greek word has the same meaning as ‫ קלע‬here, as the classification of the example under a new heading would suggest and as is stated explicitly in the entry under ὑποσκελίζω 13 14 15

16

17

Schleusner, Lexicon in LXX (‘nodus’). So TLL, s.v. ‘Cant(h)erius a gr. κανθήλιος tractum esse videtur; cf. κανθήλια, κάνθων,’ TLL. For κανθήλιον, τό, ‘rafter’ see LSJ, s.v. κανθήλια III. Driver, ‘Linguistic and Textual Problems,’ 107. Driver does not suggest that σκελίζω means ‘uproot,’ though his footnote 42 (‘The LXX’s σκελίζω “I trip up, supplant” is not far from this, the true meaning’ [i.e., ‘uproot’]) seems to be a hankering after using the Greek in support of the meaning ‘uproot.’ In fact ‘trip up’ is not really much closer to ‘uproot’ than to ‘sling forth.’ For a comment on it see Barr, Comparative Philology, 108.

8

COLLECTED ESSAYS ON THE GREEK BIBLE

below. The evidence at present available to us indicates clearly that σκελίζω usually meant ‘trip up,’ ‘upset’: cf. the examples in LSJ, and also the much better attested ὑποσκελίζω, which has the same meaning. Note especially: Jer 23:12 γενέσθω ἡ ὁδὸς αὐτῶν αὐτοῖς εἰς ὀλίσθημα ἐν γνόφῳ, καὶ ὑποσκελισθήσονται καὶ πεσοῦνται ἐν αὐτῇ.

A similar entry, relating to the same passage (Jer 10:18), is found under ὑποσκελίζω: 2. add ‘b. uproot, used in translating Heb. qāla (uprooted) LXX Je. 10.18.’

I am however unable to discover any sign of a reading ὑποσκελίζω for this place. The entry under παρατείνω, although it does not propose a new category for the LXX example, is of a similar kind, involving the use of a Hebrew word of uncertain meaning as the basis for the meaning of the Greek: II. 3, add ‘; ὅλην τὴν παρατείνουσαν (sc. ἡμέραν) all the lengthening day, i.e. throughout the morning, LXX 2 Ki. 2.29.’ 2 Kgdms 2:29 (Ra.) καὶ Αβεννηρ καὶ οἱ ἄνδρες αὐτοῦ ἀπῆλθον εἰς δυσμὰς ὅλην τὴν νύκτα ἐκείνην καὶ διέβαιναν τὸν Ιορδάνην καὶ ἐπορεύθησαν ὅλην τὴν παρατείνουσαν καὶ ἔρχονται εἰς τὴν παρεμβολήν. (ὅλην τὴν παρατείνουσαν ~ MT ‫כל־הבתרון‬.)

The meaning of ‫בתרון‬, a word found only here, must be regarded as still under discussion. BDB take it as the proper name of a place, while KB give ‘commonly: gully,’ noting forenoon proposed by Arnold, and also (in Suppl.) Batruna the name of a place mentioned in the El-Amarna tablets. The Supplement’s entry gives the impression of being an attempt to force the Greek to agree with the meaning ‘morning’ for ‫ בתרון‬at any price.18 To understand παρατείνουσαν in this way is very far-fetched. Granted παρατείνω could have the sense ‘become longer,’ but how ‘the lengthening day’ can be understood to mean ‘the morning’ is not clear, and whether it is a likely sense of the Greek here seems not to have been considered. A satisfactory and more natural sense is obtained by taking παρατείνω in the sense of ‘stretch, lie, beside’ (LSJ II.l): ‘they travelled across all the adjacent country (sc. χώραν or γῆν).19 18

19

Neither W. R. Arnold, who first proposed this meaning for ‫‘( בתרון‬The Meaning of ‫בתרון‬,’ 274–83) nor Eissfeldt, supporting it (‘Ein gescheiterter Versuch,’ 117 n. 9), makes any attempt to use the Greek translation as evidence for that sense. Arnold thought the LXX rendering to be ‘obviously based on Aramaic ‫בתר‬, after, or one of its derivatives’ (275). So Schleusner, Lexicon in LXX, s.v. παρατείνω. Cf. Ezek 27:13 ἡ Ἑλλὰς καὶ ἡ σύμπασα καὶ τὰ παρατείνοντα, οὗτοι ἐνεπορεύοντό σοι. … The acc. after ἐπορεύθησαν is a

A NOTE ON SEPTUAGINT MATERIAL

9

IV In another type of entry the Supplement classifies certain occurrences of words in the LXX under new headings because the Greek is a ‘mistranslation’ of the Hebrew original. Thus s.v. ὀπωροφυλάκιον: Add ‘2. mistranslation in LXX Mi. 3.12 (cf. Ps. 78[79].1, Je. 33[26].18) of Hebr. ‘îyîn ‘heap of ruins’ (perh. from resemblance to Syr. ‘ay(y)en ‘guarded’).’

The separation of the example under a new heading, suggesting that because it is a mistranslation it somehow differs from other examples, is unjustified. In fact the information that the translator understood the Hebrew wrongly and the reason why he did so have no relevance at all to establishing the meaning of the word here. We should not know there had been mistranslation if we did not already know what the Greek word could be expected to mean (i.e., ‘garden-watcher’s hut’) and feel it to be most improbable that it meant ‘heap of ruins.’ Similar remarks apply to the entry under πλατύνω: Add … ‘6. in LXX Pr. 24.43(28) a mistranslation of Hebr. pittāh ‘deceived’ in the sense of the Syr. pattî ‘widened’.’

In conclusion we may notice the Supplement’s entry s.v. ὑφοράω: Add ‘II. overlook, disregard, LXX De. 3.26, Ps. 77.59, 62.’

In the three places cited the text has ὑπερεῖδεν (< ὑπεροράω). Postscript The opening observations about the need for ‘incontrovertible’ examples were written with input from John Chadwick. The term ‘indicator value’ was derived from Barr but is not one I would use now. Chadwick wrote in 1994 (‘Replacing Liddell and Scott,’ 7–8) that his ‘suspicions [of the 1968 Supplement] were first aroused by J. A. L. Lee, then a research student at Cambridge, who was working on the Greek of the Septuagint.’ He went on to say, ‘I decided that the most probable explanation of these faults was that the editor had received a series of notes from some eminent Hebrew scholar, whose advice he had accepted without bothering to check it.’ The LXX material in the first Supplement was said to have been contributed by G. R. Driver (preface, vi). consequence of literal representation of the Heb., but in any case can be paralleled elsewhere in Greek, e.g., Arr., An. 6.23.1.

10

COLLECTED ESSAYS ON THE GREEK BIBLE

All the entries commented on in this paper were dropped or rewritten in the Revised Supplement (1996), though there is no mention in the preface. Further examination of the LXX entries in both LSJ Supplements (1968 and 1996) may be found in Hauspie, ‘The LXX Quotations’ (2004). In 2015 Al Pietersma, in ‘Context is King,’ 165, wrote (to my surprise): ‘I like to date the beginning of a new era [in LXX lexicography] with the publication of John Lee’s ‘A Note on Septuagint Material.’

2 ΑΠΟΣΚΕΥΗ IN THE SEPTUAGINT 1972

Abstract This paper examines the word ἀποσκευή, a third century BC formation that shows a surprising semantic development, leading to the meaning ‘military dependants, family.’ The non-LXX evidence establishes this and other meanings, and the LXX occurrences are shown to have the same meanings, including ‘dependants, family.’ Contemporary usage is seen to be indispensible to understanding the word in the LXX. This evidence is a corrective to the false attribution of the meaning ‘children’ to ἀποσκευή on the basis of Hebrew ‫טף‬, which it often renders. Further, ‫ טף‬itself sometimes has a meaning close to ‘dependants, family.’

I In early Koine Greek ἀποσκευή, a new formation attested first in the third century BC, undergoes a remarkable semantic development. The evidence for this in papyri and Koine authors has previously been observed, but the LXX examples of the word, which are quite numerous, have received little attention. My purpose here is to examine the word in the LXX in the light of the contemporary usage. It will be seen that a knowledge of the latter helps considerably in establishing the meaning in the LXX. This word provides a good illustration of the importance of investigating the LXX vocabulary in conjunction with the vocabulary of its time. The evidence of the word in non-biblical Greek is extensive. It occurs frequently in Koine authors, especially Polybius and Diodorus Siculus, and a number of times in the papyri. These examples have been discussed at some length by others.1 Here I merely summarise what has been established.

1

Holleaux, Études, 3:15–26; Kiessling, ‘Die Aposkeuai’; Wilcken, ‘III. Referate,’ 88. See also Preisigke, Wörterbuch, 4, s.v. ἀποσκευή, and discussion by the editors of PHal 1.

12

COLLECTED ESSAYS ON THE GREEK BIBLE

The primary sense is, as the etymology leads us to expect, ‘movable property,’ ‘baggage’ (both in the sing. as a collective, and in the pl.).2 This is seen, for example, in D.S. 19.32.1: τὰ βαρύτατα τῆς ἀποσκευῆς προαπέστειλεν εἴς τινα τῶν πλησίον πόλεων. The heaviest of the baggage he sent on ahead to one of the nearby cities.

Similarly Plb. 1.68.3; 2.70.5; 2.3.8, and probably in PCairZen 1.93.9 (257 BC). The interesting step in meaning is that the word came to include persons as well as inanimate objects. The context in which this development took place was, as Holleaux shows (18 ff.), a military one. Along with each army went its ἀποσκευή, its ‘baggage-train,’ which included not only baggage proper, but also the soldiers’ wives and children and all the other persons attached to the army.3 This is clearly seen, for example, in Polyaen. 4.6.13: Ἀντίγονος δὲ τῶν Εὐμενείων στρατιωτῶν εἰδὼς ἑπομένην τὴν ἀποσκευήν, ἐν ᾗ γυναῖκες ἦσαν αὐτῶν καὶ τέκνα καὶ παλλακαὶ καὶ οἰκέται καὶ χρυσὸς καὶ ἄργυρος καὶ ὅσα ἄλλα ἐκτήσαντο. … Antigonos, knowing that the baggage-train of the Eumenian soldiers was following, in which were wives, children, concubines, household slaves, gold, silver, and everything else they had acquired. …

Other examples of ἀποσκευή in the sense of ‘baggage-train’ are found in Plb. 2.3.7; D.S. 19.42.2; 43.7. Moreover, each soldier had his own individual ἀποσκευή, comprising his family and all his worldly possessions. A good example of this use is found in the following passage in Polybius 1.66.7–9. The Carthaginians request their mercenary troops to leave Carthage and withdraw to another town until they can be paid off. The mercenaries agree, but wish to leave behind τὰς ἀποσκευάς (‘their families’). At this the Carthaginians demur, fearing … μήποτε διὰ χρόνου παραγεγονότες, καί τινες μὲν τέκνων, ἔνιοι δὲ καὶ γυναικῶν ἱμείροντες, οἱ μὲν οὐκ ἐκπορευθῶσι τὸ παράπαν, οἱ δ᾽ ἐκπορευθέντες αὖθις ἀνακάμπτωσι πρὸς ταῦτα. 2

3

Cf. ἀποσκευάζομαι, ‘pack up and carry off’ (see LSJ; Holleaux, Études, 3:23 n. 6). ἀποσκευή basically = ‘all the personal property that one can pack up and take away.’ See also Parke, Soldiers, 207, on the contents of the ἀποσκευή and its importance to a Hellenistic army. The ἀποσκευή naturally included animals: see D.S. 19.84.7; Holleaux, Études, 3:20 n. 1.

ΑΠΟΣΚΕΥΗ IN THE SEPTUAGINT

13

… lest, longing to be with their wives or children after their recent protracted absence, they might in many cases refuse to leave, or, if they did, would come back again to their families. (Paton, Loeb ed.)

Consequently the Carthaginians compel the mercenaries to take τὰς ἀποσκευάς with them. The ἀποσκευαί clearly include wives and children.4 Other similar examples are found in D.S. 18.30.2, 20; 47.4. In a number of examples in Egyptian papyri we find a further development in meaning, though a slight one. Here ἀποσκευή is used exclusively of persons, namely, the family left behind by a soldier on active service. Most of the occurrences are in legal contexts: in Ptolemaic Egypt the families of soldiers in the field were accorded special privileges in legal proceedings.5 Thus in PHal 1.128–44 (III BC) the word is found six times in regulations concerning the treatment of cases involving soldiers’ families, e.g., 134–6: ἐὰν δέ τινες φάσκωσιν εἶναι τῆς ἀποσκευῆς, οἱ δικασταὶ περὶ τούτου διαγινωσκέτωσαν καὶ ἐὰν γνωσθῶσιν ὄντες τῆς ἀποσκευῆς. … If any claim to be (in the category) of military family, the dikastai are to make a decision on the matter and if they judged to be of military family. …

Compare PBaden 4.48.9 (126 BC), a letter in which a wife writes to her husband describing the action of an adversary in a law-suit: ἔλεγε γὰρ μήτε σε στρατεύεσθαι μήτ᾽ ἐμὲ εἶναι ἀποσκευήν. For he said that neither were you on campaign nor was I military family.

Similarly in UPZ 1.110.90; 2.199, 206 (164 BC); SB 5.8009.3 (I BC). The senses of the word, then, in non-biblical Greek may be classified as follows: 1. 2. 3. 4.

movable property the baggage-train of an army a soldier’s personal property and family a soldier’s family

It is to be noticed that in senses 3 and 4 ἀποσκευή refers specifically to a soldier’s (property and) family. Whether it could be used more generally of any man’s (property and) family is not indicated by our evidence. 4 5

On this passage cf. Walbank, Commentary, ad loc. See esp. Kiessling, ‘Die Aposkeuai.’

14

COLLECTED ESSAYS ON THE GREEK BIBLE

II In the LXX ἀποσκευή occurs over thirty times.6 We have, first of all, the use in the original sense of ‘movable property.’7 This is found commonly, especially outside the Pentateuch. Examples are: Gen 15:14 μετὰ δὲ ταῦτα ἐξελεύσονται ὧδε μετὰ ἀποσκευῆς (‫)רכושׁ‬ πολλῆς. After that they will come out to here with much property. 2 Chr 20:25 (Ra.) εὗρον κτήνη πολλὰ καὶ ἀποσκευὴν (‫ )רכושׁ‬καὶ σκῦλα καὶ σκεύη ἐπιθυμητά. They found many cattle and (much) property and booty and desirable objects.

In other passages, however, it is clear that we must recognise a sense similar to sense 4 above, for example: Exod 12:37 ἀπῆραν δὲ οἱ υἱοὶ Ἰσραὴλ ἐκ Ῥαμασσὴ εἰς Σοκχώθα, εἰς ἑξακοσίας χιλιάδας πεζῶν οἱ ἄνδρες πλὴν τῆς ἀποσκευῆς (‫)טף‬, (38) καὶ ἐπίμικτος πολὺς συνανέβη αὐτοῖς, καὶ πρόβατα καὶ βόες καὶ κτήνη πολλὰ σφόδρα. The sons of Israel set out from Ramasse to Sokkhotha, up to 600,000 footsoldiers, (that is) the men apart from their families, and a large mixed crowd went up with them, both sheep and beef animals and a great many cattle.

Here the word plainly refers to all persons apart from the full-grown men; that is, the men’s wives and children and all the other persons attached to them. (The singular is used collectively.) So too in Gen 43:8 the reference is to living persons, the families of Judah and his brothers. Judah appeals to his father to let him take Benjamin and go to Egypt in order to buy food: Gen 43:8 ἵνα ζῶμεν καὶ μὴ ἀποθάνωμεν καὶ ἡμεῖς καὶ σὺ καὶ ἡ ἀποσκευὴ ἡμῶν. So that we might and live and not die, both we and you and our families.

Two further examples may be seen in Exod 10:8–11. Pharaoh agrees to let the Israelites go and offer worship, and asks who are to go (vs. 8). Moses replies that they wish to take young and old, sons and daughters, sheep and cattle (vs. 9). Pharaoh objects, saying: 6 7

See below for the full classified list. The word is not in the NT. This is a definition to cover all cases; the translation may vary from case to case depending on the context (‘baggage,’ ‘property,’ ‘goods,’ etc.).

ΑΠΟΣΚΕΥΗ IN THE SEPTUAGINT

15

Exod 10:10 καθότι ἀποστέλλω ὑμᾶς, μὴ καὶ τὴν ἀποσκευὴν ὑμῶν; … (11) μὴ οὕτως· πορευέσθωσαν δὲ οἱ ἄνδρες, καὶ λατρεύσατε τῷ θεῷ. As I let you go, am I also to let your families go? … (11) Oh no. Let the men go and worship your God.

The plague of locusts follows, then the three days of darkness. Pharaoh relents, and says: Exod 10:24 βαδίζετε λατρεύσατε κυρίῳ τῷ θεῷ ὑμῶν· πλὴν τῶν προβάτων καὶ τῶν βοῶν ὑπολείπεσθε· καὶ ἡ ἀποσκευὴ ὑμῶν ἀποτρεχέτω μεθ᾽ ὑμῶν. Go, perform your cultic service to the Lord your God, except you must leave behind the sheep and beef animals; but your families may leave with you.

In the first instance ἀποσκευή may well include the cattle besides the men’s dependants – the vagueness of the word is apparent – but in the second it clearly means the men’s families alone. In the four examples just quoted ἀποσκευή renders Hebrew ‫טף‬, on which some comment is necessary. ‘Children,’ ‘little ones,’ the meaning given by BDB, is one of the senses of ‫ טף‬but not the only one. BDB themselves note in their Addenda et Corrigenda that ‘the word includes (or implies) women as well as children,’ citing Gen 47:12; Exod 12:37; 10:10, 24; Num. 32:16, 17.8 Even this does not go far enough. It seems clear that in fact ‫ טף‬had, in addition to ‘children,’ a wider sense similar to that of ἀποσκευή, namely, ‘a man’s family, including wife, children, and other dependants.’9 Moreover, this sense of ‫ טף‬is probably to be recognised not only in the places cited by BDB but elsewhere as well, for example: 2 Sam 15:22 ‫ויעבר אתי הגתי וכל־אנשׁיו וכל־הטף אשׁר אתו‬ And Ittai the Gittite went past and all his men and all his dependants who were his.10

The LXX translators’ renderings of ‫ טף‬also indicate that they recognised this general sense of ‫טף‬. Apart from ἀποσκευή, and words meaning ‘children,’11 we find the following renderings: οἰκίαι (Gen 50:21), πανοικία (Judg 18:21A), συγγένεια (Gen 50:8), λαός, (Judg 21:10A), ὄχλος 8 9 10

11

Similarly Skinner, Genesis on Gen 47:12. Cf. NEB, in which ‫ טף‬is almost invariably rendered as ‘dependants.’ Cf. Smith, Samuel, note ad loc.: ‘he marched by with a train which embraced his men and their families.’ (‫ ~ טף‬LXX ὄχλος.) I would see this sense of ‫ טף‬also in Gen 43:8; 50:21; Num 32:24; Judg 18:21; Ezra 8:21. Renderings of ‫ טף‬meaning ‘children’: παιδία (Gen 45:19; Num 14:3, 31; Deut 1:39; 3:6; Josh 1:14; 9:2f (MT 8:35); 2 Chr 20:13); τέκνα (Deut 2:34; 3:19; Judg 18:21B;

16

COLLECTED ESSAYS ON THE GREEK BIBLE

(2 Sam 15:22), κατὰ σῶμα (Gen 47:12 ~ ‫)לפי הטף‬, τὰ λοιπά (Jer 48(41):16); also ἀπαρτία (Num 31:17, 18).12 There is therefore no reason to suppose that in the four passages just considered (Gen 43:8; Exod 10:10, 24; 12:37) ἀποσκευή is intended to have the meaning ‘children.’ Other passages where the same sense of ‘family’ appears certain are Num 32:16, 17, 24 (‫ טף‬in all), and Judith 7:2. The widest sense of ἀποσκευή is probably to be discerned in: Gen 14:12 ἔλαβον δὲ καὶ τὸν Λὼτ υἱὸν τοῦ ἀδελφοῦ Ἀβρὰμ καὶ τὴν ἀποσκευὴν (‫ )רכשׁ‬αὐτοῦ καὶ ἀπῴχοντο. And they also took Lot, the son of Abram’s brother, and his property and went away.

To judge from the context, the word seems to cover both goods and persons. Later, when Lot is rescued (vs. 16), mention is made of the recovery also of τὰ ὑπάρχοντα αὐτοῦ καὶ τὰς γυναῖκας καὶ τὸν λαόν. The last-mentioned would presumably be the miscellaneous crowd of relatives and slaves which formed part of Lot’s ἀποσκευή. On the other hand, the fact that ἀποσκευή here renders ‫ רכשׁ‬means that we cannot entirely rule out the sense of ‘movable property.’ ἀποσκευή may have been intended simply as a literal rendering of the Hebrew word, without regard to what follows in the narrative.13

III A number of LXX passages raise a special problem. They are those in which ἀποσκευή, translating ‫טף‬, occurs side by side with γυναῖκες, for example: Deut 20:13–14 καὶ πατάξεις πᾶν ἀρσενικὸν αὐτῆς ἐν φόνῳ μαχαίρας, (14) πλὴν τῶν γυναικῶν καὶ τῆς ἀποσκευῆς καὶ πάντα τὰ κτήνη. And you shall strike every male in it (the city) with slaughter of the knife, except the women and the families and all the cattle.

What is the meaning of our word here? ‘Baggage,’ ‘possessions’ will not fit the context and can be eliminated. Is it ‘children,’ as the immediate context and the Hebrew original seem to indicate? Clearly this is a possible

12 13

2 Esd 8:21); ἔκγονα (Deut 29:10; 31:12); νήπια (Jer 50(43):6; Ezek 9:6); ἐπιγονή (2 Chr 31:18). [I omit a lengthy note on ἀπαρτία included in the original article.] Similarly 2 Macc 12:21 προεξαπέστειλεν ὁ Τιμόθεος τὰς γυναῖκας καὶ τὰ τέκνα καὶ τὴν ἄλλην ἀποσκευήν (καὶ τὰ τέκνα om. A).

ΑΠΟΣΚΕΥΗ IN THE SEPTUAGINT

17

sense. In my view, however, there is no compelling reason for seeing a new sense of ἀποσκευή here. The Hebrew original is an unsure guide to the meaning because: (a) assuming ‫ טף‬means ‘children’ here, there is no certainty that the translators intended their rendering to have the same sense;14 and (2) it has not been established beyond doubt that ‫ טף‬itself has the specific sense of ‘children’ here. A less precise meaning is possible. I suggest that it is possible to take ἀποσκευή tautologically, in the same wide sense that has been established for other passages, namely ‘family’ (including wife, children, and other dependants). The word has a vague meaning, and we ought not to give it such a precise one as ‘children’ unless the context overwhelmingly demands it. This suggestion is strongly supported by a further example where ἀποσκευή is plainly tautological: Num 16:27 καὶ Δαθὰν καὶ Ἀβιρὼν ἐξῆλθον, καὶ εἱστήκεισαν παρὰ τὰς θύρας τῶν σκηνῶν αὐτῶν, καὶ αἱ γυναῖκες αὐτῶν καὶ τὰ τέκνα αὐτῶν καὶ ἡ ἀποσκευὴ (‫ )טף‬αὐτῶν. And Dathan and Abiron came out and stood beside the doors of their tents, and their wives and their children and their dependants.

Here the word must be vague and general, and include the women and children just mentioned and other persons in these families. (‘Baggage’ is ruled out, as is ‘children.’) Several other passages in the Pentateuch are similar to Deut 20:14 above in having ἀποσκευή and γυναῖκες together, with the added complication that the immediate context will permit ἀποσκευή to be taken in the sense of ‘baggage,’ ‘property,’ as well as ‘family.’ Although one cannot say for certain, the latter seems the more probable in all these places.15 I take two examples, in the first of which the mention of property in the next verse (6) strongly suggests that ἀποσκευή refers to persons: Gen 46:5 καὶ ἀνέλαβον οἱ υἱοὶ Ἰακὼβ Ἰσραὴλ τὸν πατέρα αὐτῶν καὶ τὴν ἀποσκευὴν (‫ )טף‬καὶ τὰς γυναῖκας αὐτῶν ἐπὶ τὰς ἁμάξας … (6) καὶ ἀναλαβόντες τὰ ὑπάρχοντα αὐτῶν καὶ πᾶσαν τὴν κτῆσιν, ἣν ἐκτήσαντο ἐν γῇ Χανάαν. … And the sons of Jacob took up their father Israel and their families and their wives on the wagons … (6) and taking up their property and everything they had acquired in the land of Canaan. …16 14 15

16

Cf. my remarks on this subject in ‘Note on Septuagint Material,’ 237–8 [= Essay 1]. They are: Gen 34:29; 46:5; Num 31:9; 32:26, 30. ἀποσκευή ~ ‫ טף‬in all except Num 32:30, where there is no equivalent in MT. Cf. LSJ Suppl. (1968), s.v. ἀποσκευή: ‘children, little ones, LXX Ge. 46. 5, al.’ This is plainly based on the supposed meaning of the Heb. word (note ‘little ones’).

18

COLLECTED ESSAYS ON THE GREEK BIBLE

The second is: Num 32:26 ἡ ἀποσκευὴ (‫ )טף‬ἡμῶν καὶ αἱ γυναῖκες ἡμῶν καὶ πάντα τὰ κτήνη ἡμῶν ἔσονται ἐν ταῖς πόλεσιν Γαλαάδ. … Our households and our wives and all our cattle shall be in the cities of Galaad. …

Here the sense is likely to be the same as in the nearby verses 32:16, 17, 24, all of which concern the Reubenites’ and Gadites’ intention to leave their families and cattle secure in Gilead while they themselves cross the Jordan to aid the other tribes in their entry into Canaan. The meaning ‘dependants,’ ‘family’ seems certain in these.17 A textual question arises in one instance, where the reading adopted by Rahlfs is questionable: Exod 39:22 (Ra.) ὅσα συνέταξεν κύριος τῷ Μωυσῇ, οὕτως ἐποίησαν οἱ υἱοὶ Ισραηλ πᾶσαν τὴν ἀποσκευήν [of the tabernacle]. αποσκευην Bahqu: κατασκευην FbMegjnsvwz: παρασκευην AF*ia? rell. [BM]

Ἀποσκευή is hardly possible in this context. If read it will have to bear an unparalleled sense: none of the established senses is suitable. On the other hand, either of the alternative readings will give a satisfactory sense (κατασκευή ‘fittings,’ ‘furniture’; παρασκευή ‘preparation,’ or ‘equipment’).18 IV To sum up, I give a full list of the LXX examples, classified according to the senses I propose (omitting Exod 39:22): 1. movable property (including animals): Gen 15:14; 31:18; 1 Chr 5:21; 2 Chr 20:25; 32:29bis; 2 Esd 1:4, 6; 1 Macc 9:35, 39: probably 2 Chr 21:14, 17; 1 Macc 5:13, 45. (‫ רכושׁ‬in all except 1 Chr 5:21; 2 Chr 32:29[1°] ‫מקנה‬.) 2. movable property and persons comprising a man’s family: 2 Macc 12:21; probably Gen 14:12 (‫)רכושׁ‬. 3. the group of persons comprising a man’s family: Gen 34:29, 43:8; 46:5; Exod 10:10, 24; 12:37; Num 16:27; 31:9; 32:16, 17, 24, 26, 30; Deut 20:14; Judith 7:2. (‫ טף‬in all except Num 32:30: no Heb. equivalent.) 17

18

Outside Pent., there is a similar uncertainty about deciding between ‘property’ and ‘family’ in: 2 Chr 21:14, 17; 1 Macc 5:13, 45: I take all as probably ‘property.’ Support for κατασκευήν may be seen in Exod 35:24, where it is used of the tabernacle fittings and translates ‫עבדה‬. [Göttingen ed. now reads παρασκευήν.]

ΑΠΟΣΚΕΥΗ IN THE SEPTUAGINT

19

V We have seen, then, that the usage of ἀποσκευή in the LXX is closely linked with that in the Greek of the time. In particular, the use of the word in reference to persons has now been established in the LXX. There is, however, one point of difference between LXX and contemporary usage. As was noticed earlier, according to our present evidence the word is used outside the LXX only in reference to soldiers’ families. In one part of the LXX, the Pentateuch, however, it is frequently used in a more general way of any man’s family. It is difficult to tell whether this was an innovation in the translators’ own Greek. The extension is a slight one and could easily have occurred already in the Greek of the time. It can hardly be due to Hebraism. Nevertheless, there is a possibility that the translators themselves extended the usage of this convenient term. The extension would have been helped by the fact that many of the contexts in which such a word is needed are quasi-military ones. In particular, the Israelite host in its journey from Egypt into Canaan is very like an army on the march, and is readily described in military language.19 Outside the Pentateuch ἀποσκευή in reference to persons is confined to military contexts (Judith 7:2; 2 Macc 12:21). Another point of difference between books of the LXX is also to be noticed, though it is difficult to tell whether it is significant. The use of ἀποσκευή in reference to persons is practically restricted to the Pentateuch. Outside the Pentateuch it is found only in the two instances just mentioned; otherwise the word is used of inanimate property and animals, mostly translating ‫רכושׁ‬. Linked with this is the fact that ἀποσκευή as a rendering of ‫טף‬ is found only in the Pentateuch; in the other books ‫ טף‬is most often rendered by words for ‘children.’ In this the later translators do not depart from the example of the Pentateuch: ‫ טף‬is several times so taken there [and rendered παιδία etc. (see n. 11)]. But in four instances where they take ‫ טף‬in a wider sense the later translators use words of a quite different kind from ἀποσκευή.20 It seems possible therefore that they consciously avoided ἀποσκευή in reference to persons. If so, we might see in this a deliberate rejection of the use because of the pejorative implication it might seem to carry. On the other hand, the passages where ἀποσκευή might have been expected are few, and the choice of another word may have been accidental. 19

20

See esp. Num 31:9, where ἀποσκευή occurs in the context of a battle against the Midianites, and Exod 12:37, where in the same verse as ἀποσκευή the men are called πεζοί. πανοικία (Judg 18:21A); λαός (Judg 21:10A); ὄχλος (2 Kgdms 15:22); τὰ λοιπά (Jer 48:16).

20

COLLECTED ESSAYS ON THE GREEK BIBLE

Postscript This paper supersedes what was first presented in my 1970 dissertation, and was subsequently published without change in Lexical Study, 101–7 (1983). The suggestion made in the final paragraph of the above paper, that ἀποσκευή in reference to persons may have been consciously avoided in later books of the LXX, was included at the behest of the JTS reader; I was doubtful then and still am. There is nothing to show that ἀποσκευή was felt to be pejorative. The Pentateuch examples seem quite neutral (e.g., Exod 12:37; Gen 43:8 quoted above), and the documentary examples (e.g., PHal and PBaden above) are clearly not pejorative. There may be contextual reasons why ἀποσκευή was not chosen in the later LXX passages. In O’Connor and Lee, ‘The Case of Hebrew tap and Greek aposkeuê’ (2007), the semantic developments in ‫ טף‬and ἀποσκευή are set out and compared. The authors note the approaches to the words in recent commentaries and elsewhere, remarking that ‘for these two words confusion and lack of acquaintance with recent developments have lasted too long in certain quarters’ (409). Attachment to the meaning ‘children’ for ‫ טף‬to the exclusion of others persists (see, e.g., DCH), as does reluctance to accept the new meaning of ἀποσκευή (see Wevers, Numbers, 512).

3 EQUIVOCAL AND STEREOTYPED RENDERINGS IN THE LXX 1980

Abstract This paper takes up some methodological issues raised by Tov’s article ‘Three Dimensions’ (1976). While agreeing with the suggested two dimensions of meaning of a LXX rendering, that is, (a) if read simply as Greek, and (b) if read with reference to the original Hebrew whereby the translator’s intention becomes clear, I put forward some instances in which the translator may have been playing with the two levels intentionally, creating an ‘equivocal’ rendering. The paper then turns to stereotyped renderings and the question whether they can be said to have a lexical meaning identical to that of the word they translate, as Tov proposes in some cases. I argue against this, particularly in the case of Tov’s example παντοκράτωρ, and in the case of παραπικραίνω, which Walters regarded as having the meaning of ‫מרה‬, ‘rebel,’ the word it translates in the Psalms.

I In a recent article Emanuel Tov has tackled some basic questions of method in LXX lexicography, and made several valuable suggestions.1 In concluding he remarks that ‘it will be to the benefit of LXX lexicography if attention be directed not only to single words and semantic fields, hut also to methodological questions, even if the answers provided are partial and tentative, as in the present paper.’ It is in the same spirit that I offer here some further observations on the subject, prompted by Tov’s discussion. Tov points out that what was intended by the LXX translator in one way may subsequently be understood in another. Actual instances of this in the later versions and the Church Fathers are given by Tov, and of course in our own reading of the LXX today we often meet with the same dilemma between reading a passage simply as Greek and interpreting it 1

Tov, ‘Three Dimensions.’ [A revised version has since appeared in Tov, Collected Essays (1999), 85–94.]

22

COLLECTED ESSAYS ON THE GREEK BIBLE

with reference to the original Hebrew, by means of which the translator’s real intention becomes clear.2 There are, then, potentially two dimensions, or levels, of understanding of any LXX rendering.3 Tov stresses that the meaning intended by the translator is the dimension that has priority in LXX lexicography. This is clearly right, and the establishing of this principle is, I believe, an important step forward. It is what the translator intended his rendering to mean at the time of translation that the lexicographer must try to recover, not what a subsequent reader might take it to mean.4

II I want to suggest, however, that there are instances in which the translator himself may have been playing a double game. That is to say, in some cases it looks as if the translator himself, at the time of translation, was aware of, and was making use of, the possibility of these same two levels of understanding, when he gave a rendering that read one way as Greek and another way in the light of the Hebrew. Though this seems to be uncommon, it is a phenomenon to be reckoned with. I put forward the following three examples for consideration. (1) Gen 39:10 ἡνίκα δὲ ἐλάλει τῷ Ἰωσὴφ ἡμέραν ἐξ ἡμέρας, καὶ οὐχ ὑπήκουσεν αὐτῇ καθεύδειν μετ᾽ αὐτῆς τοῦ συγγενέσθαι αὐτῇ. :‫ויהי כדברה אל־יוסף יום יום ולא־שׁמע אליה לשׁכב אצלה להיות עמהּ‬

If we read the Greek on its own, the meaning of συγγενέσθαι is unmistakable. We have been plainly told in the earlier part of the narrative what Potiphar’s wife had in mind, so it is natural to take it that Joseph refused ‘to go to bed with her in order to have sexual intercourse with her.’ This is a well-established sense of συγγί(γ)νομαι, attested in Classical and later Greek.5 Of the several possible senses of the word, this is clearly the one 2

3

4

5

Cf. Ottley, Handbook, 167–9 for an earlier appreciation of the problem, leaving it unresolved. [Tov’s ‘third dimension’ refers to (new) meanings attached to LXX words in the ‘postSeptuagintal period.’] Tov, ‘Three Dimensions,’ 529–32. Though the argument is not affected, Tov’s third illustration, κύριος τῶν δυνάμεων, is open to doubt: it is hardly certain that the translators intended no more than the sense of ‘army’ for δύναμις here. This principle, that what the author intended his/her words to mean is the meaning, has always been followed in reading original Greek texts, but there it has not been necessary to formulate it. Examples form Xenophon to Plutarch in LSJ, s.v. II.3. The sexual sense is plain also in Jdt 12:16 and Sus (Th.) 11. Though this use originates in euphemism, the reference, in

EQUIVOCAL AND STEREOTYPED RENDERINGS IN THE LXX

23

that suits the context best. But when we look at the Hebrew a new facet of the rendering shows itself. We now see that the words τοῦ συγγενέσθαι αὐτῇ are open to being interpreted as a rendering of the literal meaning of the original, ‘to be with her,’ with no further connotations.6 Is this the meaning the translator intended?7 It should be noticed that συγγίνομαι is not normally used in this basic etymological sense but rather in senses that are developments of the primary meaning (‘associate with,’ ‘meet,’ ‘converse with,’ etc.),8 and that a really literal rendering would have been (τοῦ) εἶναι μετ᾽ αὐτῆς, as found elsewhere.9 Nevertheless it is clear enough that the translator has availed himself of the possibility that συγγενέσθαι can be understood in its etymological sense, and that he intended to represent the Hebrew words literally.10 But is that all? I do not think it can be. I suggest that it is highly likely that the translator was also well aware of the final effect of the rendering, read as Greek and in the context of the story so far. He has managed both to be faithful to the original and to produce a meaning natural to the context. The existence of a Greek word with the appropriate ambiguity made this possible.. Whether he deliberately sought out such a word can hardly be known for certain, but it seems significant that the strictly literal rendering that was available was not used. It is safe to say, I think, that the translator was at least conscious of the double significance. The translator’s intention, then, in itself had two dimensions.11

6

7

8

9

10

11

the right context, is quite definite. For parallel semantic development cf., e.g., σύνειμι, συνουσία, ὁμιλῶ, -ία, and Engl. intercourse. For τοῦ + infin. in a final sense see, e.g., MHT, Grammar, III, 141. The exact significance of the Heb. is debatable. The confused statements in BDB and KB, Lexicon (s.v. ‫ היה‬III.4.d(a) and 7) are difficult to interpret, but most translations and commentaries take it simply in its literal meaning. To me it seems more likely that this expression meant ‘have sexual intercourse with’ (cf. esp. 2 Sam 13:20); simil. Weinfeld, Genesis, ad loc. (I am grateful to Mr Brian Parker for this reference.) If this was the meaning of the Heb., it might complicate but not vitiate my argument. Here and throughout I use ‘translator’ in the sing. for practical convenience, without meaning to exclude the possibility of more than one translator. See LSJ, s.v. The first sense in LSJ is in fact to be born with, not ‘be with,’ which does not appear. E.g., Gen 26:3; 31:3; 48:21. συγγίνομαι is used to render ‫ היה עם‬only here in Gen 39:10. Cf. Tov’s classification (‘Compound Words,’ 211) of this example along with other compounds described as ‘more or less literal renderings.’ My argument is nicely illustrated by the divergence in the renderings given by the two main Engl. translations of the Greek: ‘… to sleep with her, so as to be with her’ (Brenton); ‘… to lie down with her to have commerce with her’ (Thomson-Muses). [NETS: ‘… to sleep with her in order to have relations with her.’]

24

COLLECTED ESSAYS ON THE GREEK BIBLE

My next example is: (2) Exod 36:2 καὶ ἐκάλεσεν Μωυσῆς τὸν Βεσελεὴλ καὶ Ἐλιὰβ … καὶ πάντας τοὺς βουλομένους προσπορεύεσθαι πρὸς τὰ ἔργα ὥστε συντελεῖν αὐτά. ‫ויקרא משׁה אל־בצלאל ואל־אהליאב … כל אשׁר נשׂאו לבו לקרבה אל־המלאכה‬ :‫לעשׂת אתה‬

Of course προσπορεύεσθαι is a more or less literal equivalent of ‫לקרבה‬. The sense of ‘go to,’ ‘draw near to’ (physically, in space) is attested for the Greek word since Aristotle, and occurs quite often elsewhere in the LXX, as, e.g., Exod 30:20 ὅταν προσπορεύωνται πρὸς τὸ θυσιαστήριον λειτουργεῖν … (MT ‫)בגשׁתם‬. But if we look more closely into the usage of the word we find reason to think that the translator has again succeeded in killing two birds with one stone. Just as English words such as go to, approach, show a tendency to move away from the purely spatial sense and to develop additional, figurative, uses,12 so too προσπορεύομαι shows a number of developed senses, attested especially in papyri contemporary with the Pentateuch translators. One of these is ‘apply oneself to,’ ‘proceed with’ (an activity). This, and not the merely spatial sense, is clearly required, e.g., in PCairZen 1.60.6 (257 BC) προσπορεύεται δὲ καὶ πρὸς [ταῦτα] καὶ πρὸς τὰ λοιπὰ μαθήματα, ‘he is proceeding with this and his other studies.’13 If we look again at Exod 26:2, it is evident that such a sense would be most appropriate there. Moreover, there is enough evidence to suggest that the very phrase found in Exodus was a stock expression in the Greek of the time: προσπορεύομαι πρὸς τὰ ἔργα (something like ‘get on with the job’) is attested more than once in documents of the third century BC.14 Could our translator have been unaware of all this? It seems hardly likely. On the contrary, I think it probable the translator welcomed, if he did not actually seek, the opportunity afforded by this word to be both literal and idiomatic. A final supporting detail is that if a merely literal rendering had been wanted, ἐγγίζω was available: it is used 8 times in the Pentateuch as a rendering of ‫קרב‬ qal. 12 13

14

Similarly, e.g., Lat. adeo, accedo. Similarly PCairZen 1.132.4 (256 BC). Note also that προσέρχομαι develops the same meaning: see LSJ, s.v. I.6. Cf. also προσχωρῶ in Pl., Rep. 539a.1. PSI 4.352.8; PCairZen 3.531.2, 16. Though both the last two are partly restored, one fills the gap in the other in such a way that the restorations are hard to doubt. Helbing (Kasussyntax, 300) noted the parallel between the PSI ex. and Exod 36:2, translating the latter as ‘sich heranmachen an.’ It seems significant that the plural ἔργα in Exod is not required by the original. [I.e., it is influenced by contemporary usage.]

EQUIVOCAL AND STEREOTYPED RENDERINGS IN THE LXX

25

A third example is: (3) Num 14:16 παρὰ τὸ μὴ δύνασθαι κύριον εἰσαγαγεῖν τὸν λαὸν τοῦτον εἰς τὴν γῆν, ἣν ὤμοσεν αὐτοῖς, κατέστρωσεν αὐτοὺς ἐν τῇ ἐρήμῳ.

In MT the word corresponding to κατέστρωσεν αὐτοὺς is ‫וישׁחטם‬ (i.e., ‫‘ שׁחט‬slaughter’). Seeing that the basic sense of στρώννυμι is ‘spread,’ it is reasonably certain that the translator, whether because of misreading or because he actually had such a form in front of him, has rendered a form of ‫‘ שׁטח‬spread.’ But what exactly did he intend his rendering to mean? Two senses of καταστρώννυμι present themselves. The primary sense of ‘spread out,’ ‘strew’ obviously corresponds literally to the Hebrew word the translator was trying to translate. Taking into account the original as the translator saw it, then, we should be led to give καταστρώννυμι some such meaning as ‘scatter’ here. On the face of it this would seem an acceptable understanding of what the translator intended. But there is also a well-established sense of the word that takes a quite different direction: καταστρώννυμι is attested in both Classical and later Greek in the sense of ‘lay low,’ with more than a hint of the idea of killing.15 There are examples of this sense in the LXX in Judith and 2 Maccabees and a very similar use turns up in a papyrus of the second century BC.16 So there is no reason to doubt that it was known to our translator. This latter meaning makes equally good sense; in fact, much better sense, if we look at the whole context. The sentence occurs in words spoken by Moses as he tries to persuade God to change his mind. Moses imagines aloud what other nations will say if God exterminates the Israelites, as he has just threatened to do. That the threat is to wipe them out, and not something milder, is clear from what precedes: in v. 12 God’s words are πατάξω αὐτοὺς θανάτῳ καὶ ἀπολῶ αὐτούς, and in v. 15 Moses, referring to this threat, says καὶ ἐκτρίψεις (‫ )והמתה‬τὸν λαὸν τοῦτον. A word meaning ‘slaughter’ or something similar is therefore expected here in v. 16.17 15

16

17

See BAG, s.v. 1; LSJ, s.v. III and s.v. καταστορέννυμι III. See also LSJ, s.v. στόρνυμι I.2.b. Jdt 8:14, 25; 14:4; 2 Macc 5:26; 11:11; 12:28; 15:27; UPZ 77.II.28 of throwing a cow down on the ground (but without any suggestion of killing). It may be wondered if we have not come full circle and proved that the translator was, after all, trying to render the same text as MT, without any confusion with ‫שׁטח‬. This seems unlikely: the standard rendering of ‫ שׁחט‬in Pent. is σφάζω (qal × 41; niph × 3); and if σφάζω was felt unsuitable to apply to human slaughter, ἀποκτείνω (e.g.) was

26

COLLECTED ESSAYS ON THE GREEK BIBLE

Moreover, a NT passage provides evidence that κατέστρωσεν was understood in just this way by at least some subsequent readers. In 1 Cor 10:5 we find κατεστρώθησαν ἐν τῇ ἐρήμῳ in reference to the destruction of some of the Israelites, as the context as a whole makes clear, and the wording plainly derives from Num 14:16. The same understanding of the word is shown by the Old Latin rendering contriuit.18 It will already be clear that I am suggesting that both meanings were in the translator’s mind. His rendering κατέστρωσεν was faithful to the Hebrew as he saw it, but also open to an interpretation better suited to the context. In effect such a rendering puts the responsibility for the final decision elsewhere. The reader must decide, ideally with reference to the original Hebrew.19 For his part the translator can honestly claim that he has not misrepresented his original. III The three examples considered so far follow the same pattern: the Greek rendering is both a literal representation of the Hebrew and at the same time capable of a different meaning equally suited to the context. My next (and last) example, while still an illustration of the phenomenon we are considering, shows a somewhat different pattern. (4) Gen 19:4–5 καὶ οἱ ἄνδρες τῆς πόλεως οἱ Σοδομῖται περιεκύκλωσαν τὴν οἰκίαν … (5) καὶ ἐξεκαλοῦντο τὸν Λώτ, καὶ ἔλεγον πρὸς αὐτόν Ποῦ εἰσιν οἱ ἄνδρες οἱ εἰσελθόντες πρὸς σὲ τὴν νύκτα; ἐξάγαγε αὐτοὺς πρὸς ἡμᾶς, ἵνα συγγενώμεθα αὐτοῖς.

The word of interest is once again συγγίνομαι. But here we find a neat reversal of what we saw before. This time the sexual meaning is not clear in the Greek but explicit in the Hebrew. Let us begin with the Greek on its own. I submit that if the story is read through without knowledge of the underlying Hebrew, there is nothing

18 19

available. καταστρώννυμι is unusual (only here in Pent.) and seems an unmistakable clue that ‫ שׁטח‬was involved. This is taken for granted by BHS and Gray, Numbers, ad loc. Note also that ‫( שׁטח‬an uncommon word: × 6 in MT) is rendered once elsewhere by καταστρώννυμι, in Job 12:23. I refrain from defending the Num translator at length against the imputation in Gray’s note: ‘G foolishly ‫וישׁטחם‬.’ I would say only that far from folly, we may be looking at an intelligent attempt to overcome a difficulty created by a faulty text. As reported by Brooke-McLean. One is reminded here of Brock’s observation that ‘the free translation aims at bringing the original to the reader, while the literal one the reader to the original’: Brock, ‘Phenomenon,’ 17; cf. 28.

EQUIVOCAL AND STEREOTYPED RENDERINGS IN THE LXX

27

to lead one to understand συγγενώμεθα in anything but a vague sense of ‘meet,’ or ‘converse with.’20 This is difficult to prove, depending as it does on personal impression, but it must at least be agreed that the vaguer meaning is possible in the context. The Sodomites could well be represented as merely asking to meet the strangers, without stating plainly their real objective. The fact that their real objective is known to Lot, and may be guessed from his subsequent offer (v. 8), does not alter this. Nor does the likelihood that most readers would already know the Sodomites’ reputation.21 Turning to the original, we find that συγγενώμεθα αὐτοῖς corresponds to ‫ונדעה אתם‬. At once it is clear that a quite different implication of the Greek word is possible. But is it the full story to say that since the original has the meaning ‘know sexually,’ that or something close to it was the meaning the translator intended? The translator’s purpose appears to me to have been more subtle. Avoiding the literal rendering with γινώσκω, employed regularly elsewhere,22 he has chosen a word capable of an innocuous interpretation in the context. It is hardly fanciful to suppose that the motive behind this was a sense of delicacy. At the same time, however, the meaning of the Hebrew is there in his rendering for anyone who can recognise it, whether by reference to the original or in some other way.23

IV Enough has been said to show the existence in the LXX of the phenomenon of equivocal rendering.24 If my analysis is right, in such cases we are dealing with a double intention on the part of the translator. The lexicographer has therefore to deal with a double meaning. It must be 20

21

22

23

24

To the ample Class. attestation of these senses noted by LSJ we can add from the papyri SB 3.6300.10 (I BC) and probably PGrenf 1.35.1 (II BC). Not, however, from reading other parts of the OT: this incident is the only indication that the Sodomites’ offence was the one now associated with the name. Elsewhere their crime is vague or, indeed, different: see Speiser, Genesis, 142. Six times in Gen alone: 4:1, 17, 25; 24:16; 19:8, the last just three verses away from our example. This too is the rendering in the closely parallel incident in Judg 19:22AB. συγγίνομαι as a rendering of ‫ ידע‬is unique in Gen 19:5 and clearly calls for explanation. Boling (Judges, 276) suggests that ‫( ידע‬in the parallel Judg 19:22) is ‘deliberately ambiguous.’ This does not seem necessary, but if it was accepted, and we assumed also that the Gen translator recognised it, then the ambiguity of his rendering would be a deliberate reflection of the Heb. Two further examples I have noted are προσπορευέσθωσαν in Exod 24:14 and ὕπανδρος/ ὑπ᾽ ἀνδρός in Num 5:20, 29. [See Lee, ‘Accuracy and Idiom,’ 96–7 on the text in Num.]

28

COLLECTED ESSAYS ON THE GREEK BIBLE

stressed that his task is not simply to choose between the two; he must come to terms with the fact that both are present and both have equal claim on his attention. In practice this means that a lexicon of the LXX will need to record in one way or another that the word exhibits a combination of senses in the place concerned. As to the frequency of this type of rendering, it can be said only that it does not appear to be common. Nevertheless the lexicographer will be wise to keep the possibility of it in mind. I have not so far noticed any example outside the Pentateuch, but that is probably accidental. Further study may however show that some translators were more prone to this habit than others, and perhaps that it does not occur at all in some books. It may be useful to put forward some terminology. I suggest that, where a double significance is possible, we apply the terms exterior and interior meaning to the two interpretations involved. The exterior meaning is the sense the word appears to have when read in the context of the Greek without reference to the original. The interior meaning is the probable sense when the word is considered as a representation of the corresponding Hebrew. When there is reason to suppose that the translator consciously intended both, we have an equivocal rendering. Most of the time, of course, the question of equivocal rendering does not arise, because the exterior and interior meanings are one and the same.

One final point must be made to complete the picture. Of course not all the renderings that we find to be open to two interpretations need have been produced deliberately. There are at least some, and probably many cases where we may reasonably deduce that the translator did not intend the double meaning. I give one example where this seems to be so: (5) Gen 44:32 ὁ γὰρ παῖς σου ἐκδέδεκται τὸ παιδίον παρὰ τοῦ πατρὸς λέγων Ἐὰν μὴ ἀγάγω αὐτὸν πρὸς σὲ καὶ στήσω αὐτὸν ἐναντίον σου, ἡμαρτηκὼς ἔσομαι πρὸς τὸν πατέρα πάσας τὰς ἡμέρας. ... ‫כי עבדך ערב את־הנער מעם אבי לאמר‬

The construction in the Greek is misleading: παρὰ τοῦ πατρός seems to require us to take ἐκδέδεκται as ‘received,’ a possible sense of the word. But reference to the original makes it clear that here, as in the earlier occurrence in the same story,25 ἐκδέχομαι has the particular sense of ‘stand surety for,’ which is well attested in third century BC papyri.26 The misleading construction in the Greek has come about through word-for-word adherence to the Hebrew. The translator apparently did not feel able, or did not take the trouble, to abandon the Hebrew construction with ‫מעם‬ 25 26

Gen 43:9 (MT ‫)ערב‬. The speaker in 44:32 is recounting this. See Lee, Lexical Study, 59–60.

EQUIVOCAL AND STEREOTYPED RENDERINGS IN THE LXX

29

in favour of one more natural to Greek. There is, however, nothing here to suggest that he was trying for an equivocal rendering. The exterior meaning ‘received’ has no advantage in the context, nor can we see any other motive for wanting the word to be open to that interpretation. We may conclude therefore that the interior meaning, and no other, was what the translator intended, and consequently that it alone is to be reported by the lexicon.

V I turn now to a different aspect of the problem of determining the translators’ intentions, namely stereotyped renderings. On this subject, too, Tov has offered a valuable discussion, which is my starting-point.27 While in agreement with him at many points, I wish to raise objections in one area involving a fundamental question of principle. Though it is difficult to prescribe exact conditions for recognising one, stereotyped renderings are a well-known feature of LXX translation methods. The problem for the lexicographer is to arrive at satisfactory principles for determining their meaning. In particular, difficulty arises when stereotyped representation leads a translator into using a word in contexts where its Hebrew counterpart has a sense outside the normal semantic range of the Greek word. How are we to define the meaning of the Greek word in these cases? In accordance with the principle enunciated by Tov, the answer lies in trying to discover the meaning intended by the translator. Tov’s discussion of εἰρήνη in this connexion (535–6) seems to me to establish a sound basis for approaching this difficult question. The use of εἰρήνη as a standard equivalent of ‫ שׁלום‬leads to a number of occurrences of the Hebraism ἐρωτῆσαί τινα εἰς εἰρήνην (and the like). Tov’s conclusions regarding this use of εἰρήνη are (in part) as follows: [I]t appears that several translators did not use εἰρήνη in accordance with ordinary Greek usage, and that they did not have a definable meaning of εἰρήνη in mind. They simply equated ‫ שׁלום‬with εἰρήνη on a practical level. Consequently, one might say that for many of the translators εἰρήνη was merely a symbol representing ‫שׁלום‬. If this is correct, a Greek lexicon should not create a new meaning of εἰρήνη on the basis of the LXX alone, claiming, as it were, that the translators enlarged the semantic range of εἰρήνη. 27

Tov, ‘Three Dimensions,’ 532–42.

30

COLLECTED ESSAYS ON THE GREEK BIBLE

Whether or not our final view of this particular example will need to be modified, these conclusions appear to me to be right as far as they go, and to offer a valuable starting-point for dealing with all stereotyped renderings. There are, I think, three essential points here: (a) The Greek word is equated in a mechanical fashion, for the sake of practical convenience, with the Hebrew word, and may be no more than a symbol for it.28 (b) It is not certain that the translator has a definable meaning in mind for the Greek word. (c) The Greek word, as such, does not automatically acquire a new sense from being used in this way. Tov follows the same lines in analysing the example of προσήλυτος (537–8), and, further, some Hebraistic uses of δικαίωμα, which, whether or not they are classed as stereotyped renderings, raise the same questions (539–40). In both cases Tov concludes that the translators used the words mechanically, and recognition of new senses is not called for. Tov’s view of προσήλυτος seems to me unsoundly based, however. First the record must be put straight regarding the mistaken reference to Apollonius Rhodius I, 834 in BAG and BDAG. This derives from the correct observation of Hort (see MM) that the word occurs in the Scholia on that line (see edition of A. Wellauer, 1828). We have there two examples, both clearly in the sense of ‘stranger,’ not ‘proselyte,’ in a non-Jewish source of uncertain, but not necessarily late, date (see Fränkel, Einleitung, 105f.). There are also Patristic examples of the same sense cited by Lampe, PGL. ‘Stranger,’ ‘newcomer,’ is, then, a possible sense; but even more, as etymology and the parallel formations ἔπηλυς, etc., suggest, it is likely to be the original one (so Chantraine, DELG, s.v. ἐλεύσομαι). The difficulty in Exod 22:20, etc. then disappears: ‘sojourner’ was the meaning intended there (and often elsewhere). J. A. Loader’s thesis, that προσήλυτος is derived from προσέρχομαι in the technical sense of ‘be accepted as a convert to Judaism’ (‘An Explanation,’ 270–7), would be convincing if he could cite clear LXX examples of the verb in that sense, but there are none. (I am grateful to Professor G.P. Shipp and Dr A.W. James for assistance with this note.)

VI But now Tov takes a further step, tentatively at first, but with increasing confidence. Having remarked (538) that ‘if a certain Greek word represents a given Hebrew word in most of its occurrences, almost by implication it has become a mere symbol for that Hebrew word in the translation,’ he continues: ‘In other words, if a lexeme as πατήρ represents ‫ אב‬in most of its occurrences, its lexicographical description could be identical with 28

John Strugnell has orally suggested: YHWH read Adonaï or melek read Shah in Persian. [Note added by RB editor.]

EQUIVOCAL AND STEREOTYPED RENDERINGS IN THE LXX

31

that of the Hebrew word, because it follows its Hebrew equivalent in all its meanings and usages.’ When we reach the discussion of παντοκράτωρ (541), this possibility has become a certainty: ‘the lexical meaning of παντοκράτωρ in Jer and the Minor Prophets is ‫ )ה(צבאות‬because it reflects only this Hebrew phrase …,’ and further: ‘The second level or dimension refers to the stage when the word came to be used as a stereotyped equivalent for ‫ )ה(צבאות‬in Jer and the Minor Prophets. At this stage the meaning of the Greek has to be expressed as ‫ )ה(צבאות‬since it represented that word in all its usages.’29 This step seems to me hazardous. We have moved from regarding the Greek word as a symbol, whose meaning in the translator’s mind may be hard to define, to the quite different position that its lexicographical description is, beyond a doubt, exactly that of the Hebrew word. To do this is to lose sight of our guiding principle that the translator’s intention determines the meaning. What in fact can be said about the translator’s intention in a case of stereotyped rendering? At the outset we must recognise the limitations to our knowledge. What was in the translator’s mind cannot be known for certain. Care must he taken therefore not to assume anything without great caution, and to do so only if there is some indication on which to build. Now if a translator uses a stereotyped rendering, even with complete consistency, does it automatically follow that he had full knowledge of, and consciously intended to reproduce, the usage of the Hebrew word as it would have been known to the original speakers and writers of the language? It does not. It must surely be recognised that this method of rendering is consistent with ignorance as much as knowledge. The translator’s comprehension of the true meaning of the Hebrew word may be anywhere between completely accurate and completely inaccurate. It cannot be assumed that it was the former simply on the basis that the rendering is stereotyped. Therefore the lexicographical description of the Greek word must not automatically be equated with that of the Hebrew word. The impetus towards making this equation is strong, for reasons not difficult to see. The distribution of the Greek word in the text is identical with that of the Hebrew word; it is hardly surprising, then, that in each context the former appears, to one familiar with Hebrew, to have the same meaning as the latter. But this is a matter of what one might call physical 29

Similarly on διαθήκη (‘Three Dimensions,’ 542): ‘the second dimension of διαθήκη is fully identical with ‫ברית‬.’

32

COLLECTED ESSAYS ON THE GREEK BIBLE

location; it does not of itself tell us what meaning was in the translator’s mind, and this, we have agreed, is to be the basis of the lexicographical description. These remarks are of course aimed at trying to establish the general principles that must underlie our approach. I am not suggesting that a Greek word used as a stereotyped rendering will never be described lexicographically in terms that coincide more or less fully with the description of its Hebrew counterpart. There are likely to be a good many such examples (e.g., νύξ ~ ‫)לילה‬. But the description will be arrived at primarily on the basis of other observations than that the rendering is stereotyped. Each word will be examined on its own merits and every indication utilised, but the fact that there is stereotyped rendering will not be taken necessarily to imply full knowledge of the sense of the Hebrew word.

VII Let us look again at παντοκράτωρ. Tov, as we have seen, asserts that ‘the lexical meaning of παντοκράτωρ in Jer and the Minor Prophets is ‫)ה(צבאות‬.’ The point I am making is that this is true only if we can assume that the translator(s) of those books had in mind the true meaning of ‫צבאות‬ and intended παντοκράτωρ to convey it. If there is any doubt about that assumption we cannot say ‘the meaning of παντοκράτωρ is ‫צבאות‬.’ I think most would agree that there is doubt in this case. The meaning of ‫צבאות‬ continues to be a problem today.30 It is very likely that the translator’s conception of it did not correspond exactly with the true meaning. This is to give the best estimate of his knowledge. At worst, it is possible that he had no clear idea at all of it, but having adopted παντοκράτωρ for whatever reason as his standard equivalent, used it mechanically throughout. What meaning of παντοκράτωρ he had in his thoughts in that case is of course hard to say, especially as it need not have been clearly defined at all. But I would suggest that if he was conscious of any it is most likely to have been something close to the one current in Greek, since after all, as Tov says (541), ‘the Greek word had a meaning of its own before it was used in the LXX.’ In short, we cannot simply define the meaning of παντοκράτωρ as ‫צבאות‬, but must try, however hard it is, to define the word in terms of what was in the translator’s mind as he used it. 30

[The entry in HALOT, s.v. ‫ צבא‬B, presents two columns of data, lists seven proposed meanings, and makes no decision; DCH s.v. ‫ צבא‬11: ‘the meaning of ‫ צבאות‬may be (1) military … (2) celestial … (3) … intensive abstract plural.’]

EQUIVOCAL AND STEREOTYPED RENDERINGS IN THE LXX

33

The same argument applies to a well-known problematic case occurring in the Psalms (and occasionally elsewhere), namely, παραπικραίνω, which would be expected to mean ‘embitter,’ provoke.’31 No non-biblical examples of this word are known apart from Philo, but the indications of its meaning are clear: the simplex πικραίνω is well attested (Class. Greek and later, including LXX) in the meaning ‘embitter,’ ‘irritate.’ ἐκπικραίνω in Deut 32:16 (~ ‫ כעס‬hiph, parallel to παροξύνω) has a similar meaning. In some LXX examples παραπικραίνω itself renders ‫ כעס‬hiph (Jer 39:29 +): ‘provoke’ is clearly the sense intended. Such is the meaning also in the two Philo examples (see BDAG). Finally, Hesychius glosses it παροργίζω. Whether or not this compound is a coinage of the LXX, I suggest it is formed on the analogy of the old word παροξύνω. Compare the compounds παραζηλῶ, παροργίζω, παρενοχλῶ. παραπικραίνω is the Psalms translator’s stock equivalent of ‫‘ מרה‬rebel.’ It is possible, but it does not at once follow, that παραπικραίνω has the sense of ‘rebel.’ The step of supposing that the translator intended the word in that sense should be taken with care, and only after all available clues have been weighed. Though it is difficult to feel confident of any conclusion, the indications seem to me to point away from the meaning ‘rebel.’ It is generally accepted that this rendering has arisen from associating ‫ מרה‬with the root ‫ מרר‬and especially the adjective ‫מר‬, ‘bitter.’ It follows that its originator supposed the notion of bitterness to be present in the Hebrew word and intended this notion to be conveyed by the Greek. In other words παραπικραίνω was intended in something like its normal meaning ‘embitter’ rather than in the sense of ‘rebel,’ which the translator failed to recognise in the Hebrew word. If this line of argument is denied, it becomes hard to explain why a word containing πικρ- was used at all. παραπικραίνω can only have been used because it has the idea of ‘embitter,’ which the translator saw in ‫מרה‬. After all, if a translator understood ‫ מרה‬to mean ‘embitter,’ ‘provoke,’ how else would he indicate it than by using a Greek word that would normally mean that? Of course it may be, in fact it is likely, that the Psalms translator’s ideas were less clearly thought out than this, and that παραπικραίνω was a makeshift rendering, with an ill-defined meaning in his mind. But the result is the same: παραπικραίνω cannot be equated with ‫מרה‬. This is still true if, following Flashar and Walters, we see this equivalent as having been simply lifted by the Psalms translator from Deut 31:27. Walters himself 31

For details, with references to earlier discussions, see Walters, Text, 150–3.

34

COLLECTED ESSAYS ON THE GREEK BIBLE

says ‘the characteristic feature of these borrowed equivalents … is that they betray a lack of understanding’ (152). It is precisely this lack of understanding that would make it impossible to equate παραπικραίνω with ‫מרה‬. If the translator does not understand the true meaning of ‫ מרה‬then it cannot be said that he intends his rendering to convey it.32 Walters reached the opposite conclusion to mine, namely, that the translation of παραπικραίνω ought to be ‘rebel’ in all cases (153). I have tried to show that his own premises point the other way. Walters himself is seen to go back on his conclusion when later on the same page he discusses renderings of this type generally and says that if the lexicon records such blunders it must do so ‘without ascribing the meaning of the Hebrew to the Greek word, which in fact reflects its homonym.’ It should also be noticed that there are a few instances in which the root ‫ מרה‬is rendered by other, better known, words for ‘provoke,’ ‘irritate,’ namely παροξύνω and ἐρεθίζω with its derivatives.33 If παραπικραίνω is to be taken as ‘rebel’ on the basis of the Hebrew, then why not these too? But this cannot be seriously maintained. Rather, these examples show that some of the translators did actually believe ‫ מרה‬means ‘provoke.’ It is fair to suppose that among them was the translator of Deut 31:27, who in the one verse renders ‫ מרי‬by ἐρεθισμός and ‫ מרה‬hiph by παραπικραίνω.34

VIII Finally I return to the expression ἐρωτῆσαί τινα εἰς εἰρήνην and variations of it.35 As I have indicated above, Tov’s conclusions appear to me sound as far as they go, and perhaps we should go no further. But it may be that here is a case in which the further step of giving the Greek the same meaning as the Hebrew is justified. What can be said about the probable state of the translators’ understanding here? This is a fairly common 32

33

34

35

Note that ‘rebel’ could have been expressed by ἀπειθῶ or ἀφίσταμαι, both of which are found in the LXX, sometimes even as renderings of ‫מרה‬. Num 20:24 παροξύνω, Deut 21:18 ἐρεθιστής, 20 ἐρεθίζω, (all these ~ ‫)מרה‬, 31:27 ἐρεθισμός (~ ‫)מרי‬. Walters notes the last three (Text, 150). Cf. Tov’s discussion of the rendering of ‫ שׁדי‬by ἱκανός (‘Three Dimensions,’ 540). ‘We must … ascribe to ἱκανός that meaning of the Hebrew word which the translator had in mind and not the one which we ascribe to the Hebrew word.’ The same principle applies to these unexpected renderings of ‫מרה‬. List of examples in Thackeray, Grammar, 40–1. In some instances τά is added before εἰς, which looks like an attempt to make the expression more intelligible as Greek.

EQUIVOCAL AND STEREOTYPED RENDERINGS IN THE LXX

35

Hebrew idiom, of a kind that even a translator with a poor knowledge of Hebrew might be expected to know. Some of the translators certainly had a good grasp of it, as idiomatic renderings elsewhere show.36 In any case it is not difficult to interpret: its meaning is fairly obvious in context, and it is readily linked with the well-known use of ‫ שׁלום‬in greetings and the like. In the light of these considerations it might be argued that the translators who used this rendering were well aware of the meaning of the Hebrew expression and did intend the Greek in the same sense. If so, the consequences for the LXX lexicographer are clear: the Greek expression is to be given the meaning of the Hebrew one.

IX These few remarks on stereotyped renderings do not go very far into a highly complex subject. The enormous range of individual examples, no two exactly alike, makes generalisation difficult and uncertain. Principles will always need to be open to modification in the light of practical experience. But it is clear, I think, that in cases of stereotyped rendering we cannot proceed at once to give the Greek word exactly the meaning of the Hebrew word, but must take the more difficult course of trying to establish what meaning the translator actually intended. If the result is that the lexicographer cannot give an exact definition of the word, this may often be an accurate reflection of the fact that the translator himself had no definable meaning in mind.

Postscript The first topic discussed, ‘equivocal’ renderings, which gives its name to the paper, can be seen to connect with an emerging perspective on LXX translation technique that sees the potential for ‘multicausality’ in producing renderings. This approach has recently been introduced from translation studies by James Aitken and elaborated further by Marieke Dhont.37 Extensive discussion of how to determine the meaning of a Greek rendering in the LXX has taken place since 1980, much of it generated by the NETS project. Our understanding of the translators’ methods, bilingual interference, the ‘text as produced,’ the role of context and the Hebrew original, and more, has advanced 36

37

E.g., ἠσπάσαντο ἀλλήλους/ αὐτόν Exod 18:7; Judg 18:15A; ἠρώτησεν αὐτούς Πῶς ἔχετε; Gen 43:27. Aitken, ‘Origins of καί γε,’ 37; Dhont, ‘Multicausality.’

36

COLLECTED ESSAYS ON THE GREEK BIBLE

greatly. Even so, the above observations on stereotypes retain their validity and usefulness. The examples παντοκράτωρ and παραπικραίνω are prototypical cases. I continue to hold to the principle enunciated by Tov, that the text means what the translator intended it to mean. It is important as our ultimate guide, even if determining intention is problematic and the translator’s intention can never be fully known. This too has been the subject of continued debate. See further Essay 15.

4 THE FUTURE OF ZHN IN LATE GREEK 1980

Abstract The paper takes as its starting-point the interpretation by G. D. Kilpatrick of a statement in the Antiatticista about the forms of the future of ζῆν. I argue that he has misunderstood the lemma and taken it in the opposite way to what was intended, in view of the usual purpose of the Antiatticist’s remarks. The evidence for the forms ζήσω, ζήσομαι, and βιώσομαι is then collected from surviving texts and an interpretation of the historical development is offered, taking account of the significant role of Atticism. Conclusions about NT usage are offered but not decisions on the NT text.

I The form of the future of ζῆν in the NT has been the subject of some discussion since G. D. Kilpatrick’s important paper ‘Atticism and the Text of the Greek New Testament.’ Kilpatrick used it to illustrate his thesis that where the NT textual tradition shows a Koine and an Atticising variant the former is to be regarded as original, since a change from Koine to Atticistic under the influence of Atticism is more likely than a change in the opposite direction.1 He argued in favour of reading the middle ζήσομαι in John and elsewhere on the premise that ζήσομαι was the Koine form and ζήσω the Atticistic. There is however a fundamental error in the argument by which Kilpatrick established this premise. This error remains undetected in the most recent discussion of the example, by Fee.2 The necessary correction has been suggested by Martini, but tentatively and without much supporting detail.3 It seems worthwhile therefore to discuss 1 2

3

Kilpatrick, ‘Atticism and the Text,’ 132–3. Fee, ‘Reasoned Eclecticism,’ 188–9. Fee opposes Kilpatrick’s view on other grounds; similarly in his earlier brief discussion in Fee, Papyrus Bodmer II, 49. Kilpatrick’s premise is accepted at face value also by Kieffer, Au dela des recensions?, 160–1 and Elliott, ‘The UBS Greek NT,’ 299. Martini, ‘Eclecticism,’ 154–5. [Martini correctly explains the import of the Antiatticist’s lemma.]

38

COLLECTED ESSAYS ON THE GREEK BIBLE

the point fully and establish that Martini is correct and that Kilpatrick’s view of this example rests on a misunderstanding. In addition there is need for a fuller examination of the attestation of the two forms outside the NT. All discussions so far have relied on the material in the standard lexicons and grammars. As we shall see, this hardly gives an adequate picture. Until these points have been cleared up, the debate over the NT readings cannot proceed on a sound footing. Kilpatrick quotes the following observation of the ancient author known as the Antiatticista: ζήσει· Πλάτων Πολιτείας ἕκτῳ, οὐ ζήσεται.4

He continues: ζήσει is the correct form supported by a reference to Plato’s Republic and ζήσεται is condemned. This is confirmed by the evidence of the authors. ζήσω is used by Attic writers and the Atticisers, ζήσομαι (except for [Dem.] xxv.82) is the form current among non-Attic and later writers. On this evidence we would expect the New Testament writers to use ζήσομαι.

What needs to be recognised is that the Antiatticista’s remark implies the opposite of what Kilpatrick supposes. The Antiatticista was a rebel against the Atticising fashion of his time, which looked back to Classical Attic as the model of correct Greek and condemned anything in the living language that did not conform to it. He did not, however, simply reject all such appeal to the past as mistaken, but sought to refute the Atticists by applying their own methods against them. Where they stigmatised a feature as unacceptable, he produced an occurrence in a Classical author, thus proving them wrong by their own criterion. His purpose and method can be clearly seen from an entry such as the following: γέμειν τὴν ναῦν μόνον φασὶ δεῖν λέγειν, τὰ δ᾽ ἄλλα μεστὰ λέγειν. ἐλέγχει δ᾽ αὐτοὺς Εὔβουλος Εἰρήνῃ.5 4

5

Bekker, Anecdota Graeca, 1:97 [Valente, Antiatticist, ζ 2]. Πλάτων Πολιτείας ἕκτῳ = Plato, Rep. 9.591c.7. [Additional notes by NovT ed.: The text of the entry has not been commented on by Guil. F. A. van Dam, Observationes in Lexica Segueriana (thesis Leiden; Roterodami, 1873), nor by L. J. Sicking, Annotationes ad Antiatticistam (thesis Leiden; Amstelodami, 1883). Photius’ Lexicon (ed. S. A. Naber, Leiden, 1864, 247) contains the entry ζήσει· ἀντὶ τοῦ ζήσεται, which seems to mean: where the unusual ζήσει occurs, it stands for the more normal ζήσεται.] Bekker, Anecdota Graeca, 1:86. For similar entries see, e.g., γοργός (86), εὐειδής (92), κομψόν (102).

THE FUTURE OF ZHN IN LATE GREEK

39

They say one ought to say only ‘the ship is full’ (γέμειν), but with reference to other things (one ought) to say (they are) ‘filled’ (μεστά). Eubulus in Peace refutes them.

The Atticists (‘they’) condemned the current use of γέμειν with reference to objects other than ships, and advocated instead the use of the adjective μεστός. An occurrence of the condemned feature in Eubulus proves its respectability.6 The Antiatticista expresses the point in a variety of ways. Many entries (at least in their surviving form) are much more abbreviated than the one just quoted. The extreme is reached with those like διάφορος· ἐχθρός (88), and βόλβιτον· Ἱππῶναξ (86), where nearly everything is taken for granted. A little less cryptic are those on the pattern of ἀποκριθῆναι· οὐκ ἀποκρίνασθαι.7 This remark is to be understood as follows: ‘ἀποκριθῆναι, the Koine form condemned by the Atticists, is in fact found in Attic, instead of ἀποκρίνασθαι, the standard Attic form approved by the Atticists.’8 This type may be expanded by the addition of a reference to an Attic occurrence, as for example ἀρτύματα· οὐχ ἡδύσματα. Σοφοκλῆς Φαίαξιν.9 The meaning is: ‘Sophocles in Phaiakes has ἀρτύματα, not the recommended ἡδύσματα.’ With this we reach the closest available parallel to ζήσει· Πλάτων Πολιτείας ἕκτῳ, οὐ ζήσεται.10 In any case it is quite plain that however it is expressed, the Antiatticista’s point is always the same: a Classical example is cited (or it is implied that one is available) to prove the acceptability of a Koine feature rejected by the Atticists. The correct interpretation of the remark on ζήσει, then, is clear: the active was in use in the Koine of the writer’s time, the second century AD;11 the Atticists objected to it and advocated the middle. The fact that the Antiatticista cites Plato in support of the active makes this certain. This interpretation, moreover, agrees with what we should expect on general grounds. It is well known that the tendency of the Koine was to 6

7 8

9 10

11

That is, Eubulus Comicus (IV BC). The play is not otherwise extant. For other Class. exx. see LSJ, s.v. γέμω 2. Bekker, Anecdota Graeca, 1:77. Similarly ἀλήθειν (78), γαγγαλίζειν, γραία (87). By ‘Attic’ is meant strictly the dialect of Classical Athens. This should be carefully distinguished from what the Atticists thought was Attic (= ‘Atticistic’). Bekker, Anecdota Graeca, 1:82. Cf. αὐτοσχεδιάζειν (83), ἐκδημία (93). It is worth noting in addition that the censured form or word is almost invariably put first. Exceptions are at once made clear by the words that follow: e.g., γενέθλιον ἡμέραν ἀξιοῦσιν ἀεὶ λέγειν, οὐ γενέθλια (86). [Thus ζήσει is likely to be the censured form.] For the date see Debrunner–Scherer, Geschichte, 2:17.

40

COLLECTED ESSAYS ON THE GREEK BIBLE

replace middle futures of active verbs by the active.12 This was a slow process and not all verbs were affected at the same time, but there can be no doubt of the existence of the trend, and that it was always in the same direction. It is most unlikely that any Atticist would have wished to, or needed to, advocate ζήσω, when the tendency in the spoken language was to introduce this form anyway. Rather, we can expect the Atticists to have tried to defend the threatened middle, which they naturally assumed to be good Attic.13 II Let us now consider the evidence of the surviving texts. I begin by setting out such attestation as I have been able to find, arranged chronologically. Completeness cannot be claimed, but the list goes well beyond existing collections, and is sufficient to allow us to reconstruct with some confidence the history of ζήσω and ζήσομαι. Besides these two forms, there is another to be taken into account, namely βιώσομαι (-σω).14 2.1. ζήσω Attic of V–IV BC: Ar., Plut. 263 -ειν, Fr. dub. 899a -εις. Pl., Rep. 5.465d.2 -ουσι, 9.591c.7 -ει (var. causa: βιώσεται precedes), Leg. 7.792e.7 διαζήσει. Of uncertain date and origin: [Epich.] 267 -ων.15 ‘Menander,’ Γνῶμαι μονόστιχοι, 186 -εις.16 12

13

14

15

16

Moulton, Grammar, I, 154–5; BDF, §77; Thackeray, Grammar, 231–3; Bauer in BAGD, xiv; Mandilaras, Verb, §367. Cf. Browning, Greek, 52: ‘Atticising writers continuously fail in their purpose. Either they admit Koine forms censured by the grammarians … or they overcompensate and produce false Atticisms. … The literature of this period is full of middle voices where Attic uses in fact the active, of wrongly used datives. …’ Martini’s interpretation (‘Eclecticism,’ 154–5) agrees in essentials with mine. Sources of information were LSJ; Mandilaras, Verb; Schmid, Atticismus; Stephanus, Thesaurus; Thackeray, Grammar; Veitch, Greek Verbs; and esp. the standard author indexes. The following authors are fully covered, thanks to the indexes noted: Aristoph. (Todd); Demosth. (Preuss); Epict. (Schenkl); Eur. (Allen–Italie); Joseph. (Rengstorf); Philo (Mayer); Plato (Brandwood); Plb. (Mauersberger: no exx.). Coverage is extensive if not complete for: Arist. (Bonitz); Dio Cass. (Boissevain); Lucian (Jacobitz); Plut. (Wyttenbach); Xenoph. (Sturz). That is, attributed to the Doric poet Epicharmus Comicus (V BC). Perhaps to be read ζησῶν (= Doric fut.): so KB, Grammatik, I.2, 436; but ζήσων Kaibel. The collection known as the ‘Maxims of Menander’ continued to be augmented right up until Byz. times, and contains much that Men. ‘could not have written’: Edmonds, Fragments, IIIB, 901–3.

THE FUTURE OF ZHN IN LATE GREEK

41

LXX MS tradition: Prov 9:11 ζήσεις B etc. (-η, -ει AV); Sir 37:26 ζήσει A (-εται B etc.); Amos 5:6 ζήσετε A (-εσθε L, ζῆτε B etc., ζήσατε Rahlfs). All three of these are likely to be secondary.17 It is hard to know when they entered the tradition. A fourth example is of special interest: at 4 Kgdms 18:32 the majority reading is ζήσετε, while bdhjoprvxyc2e2 have ζήσεσθε. ζήσετε is thus the reading of the Kaige recension, and ζήσεσθε in all probability the original LXX.18 We may therefore roughly date this example of the active to the first half of I AD.19 (The two occurrences of the active at Ps 137:7 and 142:11 (ζήσεις με) are not relevant, as ζήσεις is causative, a quite different use which requires the active.) I AD: Philo, Quis Heres 258, Congr. 176 both ζήσεις, in quotations of Gen 20:7 and 27:40 respectively (but LXX ζήσῃ).20 Joseph., AJ 12.315 ζήσουσι, 16.119 -ομεν, BJ 7.394 ἐπιζήσουσι. I–II AD: Epict. 3.26.24 -εις, 4.1.104 -οντα, Plut., Mor. 194a καταζήσεις, in a quotation of a saying of Epaminondas, reported also in Aelian, VH 9.9, in almost identical words. Both authors, then are quoting a source. II AD: Aelian, NA 7.2 καταζήσουσι, VH 11.9 καταζήσεις (see above). Artem., Onir. 200.7 -εις, 200.18 -εις, 201.1 -εις, -ει, 201.4 -εις.21 Barn., Ep. 6.17 -ομεν. Lucian, Alex. 34 -εις, in a quotation of an oracle of Alexander the false prophet. II–III AD: Dio Cass. 53.9.2 -ειν, 58.27.3 -ων, 58.28.2 -ων. Here may be added the example in Cramer, Anecdota Graeca Oxon. II, 372: ζῶ· τὸ ἐπὶ τῆς ζωῆς, ὁ μέλλων ζήσω ἐστίν.22

17

18

19 20

21 22

Rahlfs accepts act. in Prov 9:11, without compelling reason. Arguments in favour of mid. are: (a) this is the usual LXX form; (b) Prov itself has mid. in 9:6; 28:16; (c) Prov is in ‘literary’ Greek. There are other, weakly attested exx. of the act. to be found, as, e.g., 4 Kgdms 20:1 ζήσεις v, ζήσῃ rell. Barthélemy’s discovery (Les devanciers) concerning Kgdms βγ and γδ (our example is in the latter) is well known. The Kaige recension of Kgdms γδ has the mid. otherwise (4 Kgdms 1:2; 4:7; 7:4, etc.; likewise 2 Kgdm 12:22 (βγ), and Hab 2:4 in Barthélemy’s Dodecaproph. fragments (Les devanciers, 175). The act. in 4 Kgdms 18:32 is, then, an accidental change, not one of the deliberate changes that characterise this recension. See Barthélemy, Les devanciers, 144, 148; Cross and Talmon, Qumran, 313. Were these readings already in Philo’s LXX, or did they enter later in the MS tradition of Philo? (No v.ll. are reported by Cohn-Wendland.) They appear to have no connexion with the recension identified by Katz in Philo’s Bible. Refs. to p. and l. number in Pack (Teubner ed., 1963). It is hard to know what to make of this, esp. without a date; it is presumably not earlier than II AD, but could be much later. The fact that the writer states the form of the fut. (there is no mention of other tenses) suggests there was difficulty about it.

42

COLLECTED ESSAYS ON THE GREEK BIBLE

2.2. ζήσομαι Ionic, probably V–IV BC: Hippocr., Nat. Puer. 30.8 -εται.23 Late Attic-Hellenistic, IV BC: Arist., Pol. 7.1327b.5 -εται. Probably to be included here is [Demosth.] 25.82 -εται. LXX: the middle is usual throughout (c.102 times). I AD: Philo, Leg. All. 2.81 -εται, Somn. 2.31 -εσθαι, Ioseph. 264 -εται (+ 6 exx. in LXX quotations). Joseph., AJ 1.193 -εται, 4.302 -ομένους, 9.75 -όμενοι, 12.281 -εσθε. Antiphilus in Anth. Gr. VII 175 -εσθε. I–II AD: Plut., Mor. 79e -εται, 1082b -εσθαι, c -εσθαι, -εται.24 II AD: Apostolic Fathers 43 times (+ 4 exx. in quotations).25 Artem., Onir. 199.8 -εται, 201.8 -εται. Lucian, Paras. 112 -εται × 3, [Lucian] Asin. 27 -εται. II–III AD: Dio Cass. 53.34.2 -ῃ. Philostr., VA 1.7 -ομαι, 37 -εται. III AD: Diog. L. 6.44 -οιτο. POxy 6.907.23 (276 AD) -εται. 2.3. βιώσομαι (-σω) Attic of V–IV BC: e.g., Ar., Equit. 699. Eur. Alc. 784. Isoc., Areop. 78, Pax 35. Xenoph., Ages. 9.4, Cyn. 7.4, Cyr. 5.4.33, Mem. 1.7.2; 4.8.8; 2.1.23 (δια-). Plato 18 times (incl. 2 in cmpds., δια-, συμ-).26 Late Attic-Hellenistic, IV(–III) BC: Arist., Eth. Nic. 1101a.17, 1127b.27, Magn. Mor. 1212b.31 (συμ-), 1213a.28 (συμ-), Pol. 1267b.36. Menand., Epitr. 694, Peric. 977 (Sandbach), Fr. 581.6 (Körte). ‘Menand.,’ Γνῶμαι 270 (act.). LXX: Sir 13:5 (συμ-); Prov 7:2; Job 29.18; 4 Macc 6:20 (ἐπι-) (last three act.). I BC: Philodem., Oec. 13.27. I AD: Philo, Leg. ad G. 85, Virt. 47, 127, Spec. Leg. 3.154, Dec. 49, Vita M. 2.48, Somn. 2.31, Quod Det. Pot. 62, 154, Opif. M. 172 (total 10). Joseph. 11 times (twice act.: AJ 1.46, 8.337).27 23

24

25 26 27

Ed. Joly (Budé ed., vol. XI, 1970) = Littré VII, 536. Joly (p. 23) accepts the traditional dating at the end of V BC for this; Pauly-Wissowa, RE, VIII.2, col. 1828, say 375 BC at latest. In Plut., Mor. 1082b, c ζήσομαι, instead of Plut.’s usual βιώσομαι, is more or less demanded by the context: … εἰ … ὁ βαδίζων ἐβάδισε καὶ βαδιεῖται; τὸ δεινῶν, φασί, δεινότατον, εἰ τῷ ζῶντι τὸ ἐζηκέναι καὶ ζήσεσθαι συμβέβηκεν …, and so on throughout. For the full list of references see Goodspeed, Index Patristicus. See Brandwood, Index to Plato. See Rengstorf, Concordance.

THE FUTURE OF ZHN IN LATE GREEK

43

I–II AD: Epict. 1.4.27, Ench. 24.1, Fr. 8.12 (δια-). Plut., Mor. 100c, 101d, 141d (συμ-), 160b, c, 187e, 563d, 1043c x 2 (one συμ-), 1119d, Demetr. 1.3, 47.4 (κατα-), Publ. 3.1 (δια-), Titus Flam. 20.3 (κατα-) (total 14 in Plut.). II AD: Lucian, Charon 20, Dial. Meretr. 15.1, Herm. 6.77, Nav. 26, Par. 12, Tox. 37, VH 1.12 (κατα-) (total 8). Appian, BC 4.119 (act.). II–III AD: Dio Cass. 52.18.6, 27.5, 39.2; 58.23.2, 27.3, 28.1; 69.17.1 (total 7). III AD: Diog. L. 2.68.

III This evidence, though it appears unwieldy at first sight, in fact yields a clear picture of the history of these forms. I suggest the following interpretation. The usual Attic future was neither ζήσω nor ζήσομαι, but, as everyone who has learnt their ‘principal parts’ knows, βιώσομαι. The occurrences noted are sufficient to confirm the traditional statement.28 Plato’s 18 examples, compared with only three of ζήσω, one of which is introduced for a special reason, are conclusive on this point. It is also significant that βιώσομαι is preferred by Xenophon, the ‘forerunner of the Koine,’ and is still to be found in Aristotle and Menander, on the threshold of late Greek. Thereafter it declines in the spoken language. Alongside βιώσομαι, Attic occasionally used ζήσω, but it did not last beyond Plato’s time. The examples attributed to Epicharmus and Menander are too untrustworthy to affect this conclusion. ζήσομαι cannot be traced back to Classical Attic. Its attestation, first in Ionic, next in late Attic, then in the Koine, follows a familiar pattern. It is an innovation of the Koine, originating in Ionic.29 Its appearance on the scene is linked with the obsolescence of βιώσομαι, on which it is modelled.30 This explains why the middle takes over so completely from the already existing active, a development that is puzzling otherwise. 28

29

30

See, e.g., Cobet, Variae lectiones, 610; Schwyzer, Grammatik, I, 675; Smyth, Grammar, 698. This interpretation is strongly supported by the fact that the aor. ἔζησα, the late Greek equivalent of Attic ἐβίων, also originates in Ion.: see BDF, §75; Bauer in BAGD, xiv; LSJ, s.v. ζῶ. On the Ionic element in Koine Greek generally see, e.g., Thumb, Hellenismus, 68, 209. Schwyzer, Grammatik, I, 782 n. 5: ‘Hippokr. ζήσομαι (: ἔζησα) für älteres βιώσομαι.’ A parallel is φάγομαι on the analogy of ἔδομαι, the old form that φάγομαι replaces.

44

COLLECTED ESSAYS ON THE GREEK BIBLE

Aristotle represents the transitional stage when ζήσομαι and βιώσομαι were in use side by side. ζήσομαι continued as the reigning form throughout the early Koine, as the LXX attests. Then early in the first century AD, perhaps somewhat earlier, the form was affected by the current which we have already noticed, the trend towards replacement of future middles by the active.31 ζήσω now begins to appear in our evidence, alongside the middle. The καίγε recension provides the earliest datable post-Classical example; then we find it sporadically in writers from the first century onwards. We may be sure that it was the usual form in the spoken language. This was a new development, not a revival of the old Attic ζήσω. Close on the heels of the appearance of ζήσω came the Atticist reaction against it. This was part of their general stand against the change that was affecting future middles. The active was condemned and ζήσομαι promoted as the correct form. The Antiatticista testifies to this. His evidence relates to the second century, but the reaction probably began earlier.32 It does not matter that the middle was not the true Attic form; the important thing is that it was believed to be. The case of ζήσω/-ομαι was thought to be just like all the others (ἀκούσω/-ομαι, etc.). Besides ζήσομαι, genuine Attic βιώσομαι was revived and encouraged by the Atticists, as we can deduce from its frequency in Atticising authors. This was quite artificial, and could have had little impact on ordinary language. All this is likely to have led to considerable uncertainty, if not confusion, among writers of the first century and later. (The efforts of the Antiatticista in support of ζήσω would only have added to it.) That this was indeed the case seems clear from our evidence. All three forms are liberally attested, and hardly any author keeps consistently to one or the other. A general pattern can nevertheless be seen: ζήσω tends to be avoided in favour of ζήσομαι or βιώσομαι. This is what we expect, given that the bulk of the evidence is more or less ‘literary.’ This is true even of the POxy example (see below). Among the authors noticed, Epictetus and Artemidorus are generally thought to be the nearest to the ordinary Koine. Let us look more closely at the practice of individual writers. Philo, Plutarch, and Lucian have ζήσομαι occasionally (3 or 4 times each), but βιώσομαι predominates and is evidently their usual word (10, 14, and 31

32

The fact that some verbs are affected much earlier than this (e.g., ἀκούσω appears at least as early as III BC) makes no difference. Each verb has a separate history. As is well known, Atticism shows itself as early as Dionysius of Halicarnassus, in I BC, and its roots can be traced even earlier: see, e.g., Thumb, Hellenismus, 252; Schwyzer, Grammatik, I, 130.

THE FUTURE OF ZHN IN LATE GREEK

45

8 times respectively). ζήσω occurs, but only in quotations: this is someone else’s usage, not theirs. Josephus is much the same (ζήσομαι 4 times, βιώσομαι, -σω, 11), but does not succeed in avoiding ζήσω altogether (3 times). The Apostolic Fathers represent as it were the intermediate level. They keep to ζήσομαι with one exception, but do not aspire to βιώσομαι. Epictetus and Artemidorus accord with expectation in having more examples of ζήσω than the middle, but Epictetus surprises us by also using βιώσομαι (3 times; the evidence is incomplete for Artemidorus). Dio Cassius, a little later, also seems uncertain of himself. βιώσομαι is his usual word (7 times), but he has ζήσω more often than the middle (3:1). For the remaining authors the evidence is too fragmentary to make much of. As to the example of ζήσομαι in POxy 6.907.23, this is not, as one might be tempted to suppose, evidence of popular Greek. The document is a will, likely to be cast in formal language.33 Moreover its literary level is clearly marked by τε and βούλομαι, to look no further than the immediate context.

IV Rather a lot of ground has had to be covered in order to reach our goal, but we are now in a position to draw some conclusions for the NT. One thing is abundantly clear. The usage of the first century AD was unsettled. We cannot say with certainty which out of ζήσω and ζήσομαι a given writer would have used, or whether he would have been consistent. The most that can be said, in my opinion, is that the closer the writer’s language to everyday Koine the more likely it is that he would have used ζήσω. On this basis we should expect ζήσω in John, unless he was quoting or wished to achieve a special effect. (The same would apply to Mark and Revelation.) The remaining books might be expected to show a tendency to prefer ζήσομαι. Subsequent influences on the MS tradition are equally difficult to reduce to absolutes. It is quite likely that there was a tendency to alter the active to the ‘correct’ middle in accordance with Atticist precepts. The reverse change, though less likely, is also possible (cf. the LXX). Where this happened it would be a case of an occasional, unintentional alteration in the direction of the scribe’s own speech. 33

καταλείπω δοθῆναί τε βούλομαι κατ᾽ ἔτος ἐφ᾽ ὃν ζήσεται χρόνον. …

46

COLLECTED ESSAYS ON THE GREEK BIBLE

The text of the NT has been purposely left out of this enquiry, so that the linguistic background could be established independently. I must leave it to others to examine the MS readings in the light of the results. Apart from its importance for the NT, the case we have studied also has some interest as a specimen of what happened when the Atticists tried to turn the clock back.

Postscript Kilpatrick responded in ‘Atticism and the Future of ζῆν’ (1983). He did not concede that he had taken the Antiatticist’s statement the wrong way, and marshalled many arguments in support of the correctness of his interpretation. Among them he put forward the suggestion that ‘abbreviation or corruption [may] both have caused the inconsistency’ (147), thus admitting that his interpretation was inconsistent with the usual purpose of the Antiatticist. It is hard to see what corruption has occurred or what emendation would be plausible to reverse the import of the lemma, when the text of Plato confirms the occurrence of the active ζήσει claimed by the Antiatticist. Other discussions are found in: Kilpatrick, ‘Eclecticism and Atticism’ (1977), 109–11; De Jonge, ‘De nieuwe Nestle’ (1980), 312–3. There is further description of the Antiatticist’s methods in Lee, ‘Atticist Grammarians’ [= Essay 19].

5 SOME FEATURES OF THE SPEECH OF JESUS IN MARK’S GOSPEL 1985

Abstract All five examples of the obsolescent particle μέν in Mark’s Gospel (leaving aside 16:19) occur in Sayings of Jesus. It is argued in this paper that the location is significant: Jesus uses a prestige feature appropriate to his status. The connexion with insights from sociolinguistics and the phenomenon of ‘formality’ are elaborated. Then four further features of a similar kind in Mark are presented in detail (εὖ, φάγοι, ὦ, οὐ μή), and several others are noticed briefly. Parallels in the LXX and elsewhere in the NT are noted. Brief remarks on the implications for the history of the Sayings are offered.

I In the Gospel of Mark the particle μέν seldom occurs. ‘Surprisingly rare,’ Taylor calls it, but does not pursue the matter.1 There is, however, something noticeable about the five occurrences in Mark (leaving aside one in the Longer Ending): they are all in words spoken by Jesus. This fact must have been noticed before, but as far as I am aware no significance has been attached to it. I wish to suggest, however, that this distribution is not accidental, but is in fact the key to understanding the use of the word in Mark. Furthermore, I shall argue in this paper that we have here only one instance of an important but overlooked phenomenon in New Testament Greek. I begin with μέν, the feature first observed, which also provides a good starting-point here. It is well known that in post-Classical Greek there is a general decline in the particles.2 That is to say, the wide and subtle range of Classical particles is sharply curtailed, leaving a few to do most of the work; at the 1 2

Taylor, Mark, on 4:4. See Blomqvist, Greek Particles, 132, with references to the main discussions; also Thrall, Greek Particles, 3–6, showing the decrease in combinations of particles.

48

COLLECTED ESSAYS ON THE GREEK BIBLE

same time there is a long-term tendency for all post-positives to be eliminated in favour of words that come first in the clause. Though there is room for further study of the details, the general trend is incontestable. It is clear that μέν (…δέ) shares in this decline. In the Ptolemaic period there is already a perceptible falling-off in its frequency.3 But this is not all. There is now a new factor, one which makes all the difference for our purpose. When μέν (…δέ) does occur, it tends to be found, Mayser tells us, in educated, official writing rather than private letters, which are our best evidence for everyday language. This indicates that we are dealing with one more instance of a familiar phenomenon in post-Classical Greek, the tendency to look back to the earlier language as the model of good Greek and to maintain its idioms after they had become obsolete in the spoken language. This reaches its peak in the second century AD, but the trend is discernible long before. It seems clear, then, that quite early in the Koine μέν (…δέ) was no longer the living idiom that it undoubtedly had been in Classical Greek, but had become, on the whole, a sign of an attempt at more educated Greek, a prestige feature consciously used. The evidence of the LXX points in the same direction. To begin with, the total number of occurrences, 215, is not high.4 Then if we distinguish between the literary books and the rest, following Thackeray’s classification,5 the former are found to have 181 examples (of which 2–4 Macc account for no less than 117), while all the rest together have only 34. In the Pentateuch, which in level of language is a fair specimen of unpretentious Greek of the third century BC, μέν is used very sparingly: there are only 19 examples.6 The contrast with 2–4 Macc is striking, especially when one bears in mind that the Pentateuch is more than four times the length of 2–4 Macc.7 To what extent is the rarity of μέν in the translated books due to the absence of any Hebrew equivalent? There is no way of knowing for certain, but I would argue that this is not the main factor. If it was, we might 3

4 5 6

7

Mayser, Grammatik, II.3, 128; sample figures 125. An indication of frequency of μέν in Class. Greek is given by figures for these authors (all approx. except Hdt.): Ar. 630; Hdt. 2567 (Powell); Pl. 9000; D. 4070; Th. 1750; X. 1420 (last two from Jannaris, Grammar, 400). Figures based on HR and Rahlfs’s text. [The ‘Accordance’ tally is close: 218 (Ra.).] Thackeray, Grammar, 13. Seven of these are μέν solitarium. A curious detail is that a high proportion of the exx. (12) are alike in that μέν follows εἰ (ἐάν), sometimes with, sometimes without εἰ (ἐάν) δέ in the next clause. Perhaps we should think of the obsolescence of μέν in the living language as proceeding at a different rate for different kinds of use, the use in εἰ-clauses being one of the last to go. Pages of Rahlfs’s text: Pent. 354; 2–4 Macc 85.

SOME FEATURES OF THE SPEECH OF JESUS IN MARK’S GOSPEL

49

expect μέν not to occur at all. And the comparison with δέ is suggestive: δέ is used with great frequency in some of these books, despite the availability of the more literal equivalent καί.8 I conclude that the translators were not necessarily constrained by the original in their use of particles, just as in general they show readiness, especially in the Pentateuch, to employ idiomatic Greek were appropriate. It seems probable, therefore, that the LXX reflects fairly well the status of μέν in the language generally. The history of μέν in the post-Ptolemaic Koine has not yet been fully studied, but there is no reason to doubt that the earlier trend continues, and the NT itself confirms this. Nevertheless our conclusions will be more secure if we can obtain further evidence of how μέν was being used in this period. I have therefore made a survey of 18 volumes of The Oxyrhynchus Papyri, covering a large sample of texts of the first to fifth centuries AD.9 The result agrees convincingly with what we have seen already. The total number of texts surveyed was 659, of which 90 are private letters, and 30 or so domestic ephemera (lists, private accounts, and the like). The rest are official and private documents, all of a more or less formal nature. These include such texts as a letter of the Emperor, records of court proceedings, leases, and contracts. The occurrences of μέν total 100, of which 81 are in the formal documents, and 19 in the private letters. The figure for the former group causes no surprise. Examination of the latter brings to light some significant points. (a) It is at once noticeable that μέν tends to occur in stilted, formulaic language, especially at the beginning of a letter. One particular phrase, πρὸ μὲν πάντων (παντός) recurs 8 times as part of an opening formula, usually the well-known πρὸ μὲν πάντων εὔχομαί σε ὑγιαίνειν.10 (b) In nearly half the examples (9 to be exact), μέν is not followed by δέ or other correlative. (c) Some of the letters in which μέν occurs show definite signs of a higher level of education. An unmistakable instance is POxy 17.2109 (late I AD), with two examples of μέν (lines 17, 40), both followed by correct δέ. Other features include γε (50), the learned-sounding word ἕξις (16), and a potential optative with ἄν (47). It is no accident that the writer is a student in Alexandria writing home to his father – no doubt in his best Greek.11 (d) This is the only letter using 8

9

10

11

[‘Accordance’ figures for δέ in Pent. (Gött. ed.): Gen 853; Exod 413; Lev 128; Num 74; Deut 100. Μέν 19 times in Pent.] POxy vols. 17–46 except those containing only literary texts, and vol. 22 with unreliable index. The few texts of VI AD and later are omitted from consideration. For a full treatment of this formula, and of others in which πρὸ μὲν πάντων appears, see Steen, ‘Les clichés,’ 155–6; Exler, Ancient Greek Letter, 107–13. This one phrase must account for a large number of the occurrences of μέν in the papyri. The only illiterate letter with μὲν ... δέ, 42.3087.11–14 (III–IV AD), interestingly has εἰ μὲν ... εἰ δέ (cf. note 6 above).

50

COLLECTED ESSAYS ON THE GREEK BIBLE

μέν that dates from before the second century. (The sample contains 10 private letters of I AD.) We may suspect that in the second century greater efforts were made to revive μέν. All this points to the conclusion that in the later as in the earlier Koine μέν (…δέ) was used with considerable artificiality, and that both user and reader (or hearer) felt it to be indicative of formality and correctness. The increased use of μέν without a following δέ (or its equivalent) is in keeping with this.12 In Classical Greek this use of μέν was idiomatic within certain limits;13 it was hardly common compared with the normal use. It is plain that later writers who indulge in it have little concern for the semantic value of μέν, but simply insert it, or a fixed phrase containing it, to add a touch of style.14 Turning now to the NT, we find first that μέν (…δέ), as has long been remarked, is not very common.15 Moreover, the distribution is significant. The books closest to the level of everyday language, Mark, John, and Revelation, have only 14 examples out of a total of 180: Mark has 6 (including 16:19), John 8, and Revelation 0. In contrast Acts, for example, has 48, and 1 Corinthians and Hebrews have 20 each.16 Μέν thus occurs predominantly in the group of books in which it would be expected. But even here it is noticeable that some books have it rarely or not at all, and that Luke, whose concern for good style is well known, uses it little in his Gospel (10 examples). In frequency and overall distribution, then, μέν in the NT reflects the usage in the language generally. So we come to Mark, where a more subtle reflection of current usage is to be seen. The examples are as follows (16:19 aside):17 4:4 καὶ ἐγένετο ἐν τῷ σπείρειν ὃ μὲν ἔπεσεν παρὰ τὴν ὁδόν … (5) καὶ ἄλλο ἔπεσεν ἐπὶ τὸ πετρῶδες … (7) καὶ ἄλλο ἔπεσεν εἰς τὰς ἀκάνθας. 12

13 14

15 16

17

For further indications of its incidence see Mayser, Grammatik, II.3, 129–30; MHT, Grammar, III, 332 (but Turner’s NT figures do not separate μέν without any correlative from μέν correlated with καί, ἀλλά, or πλήν). See KG, Grammatik, II.2, 271–2; Denniston, Greek Particles, 380–4. According to Mr George Harris, a member of one of my classes in 1976, it is still the practice in some Mod. Greek writing to insert a μέν, preferably early in the piece, ‘to show that you know how to use particles.’ Cadbury (Style of Luke, 145–6) seems to me rather to miss the point when he plays down μέν (οὖν) solit. as a test of style because it is ‘of doubtful literary excellence.’ μέν is still a sign of a conscious attempt at ‘better’ Greek even if the user cannot handle it in the Classical manner. See BDF, §447; MHT, Grammar, III, 331; BAGD, s.v. μέν. Figures from Aland, Konkordanz, II ‘Wortstatistik.’ There is of course considerable MS variation. NT text is NA28 (2012). Variant readings in Mark depend on Legg, NTG: Marcum.

SOME FEATURES OF THE SPEECH OF JESUS IN MARK’S GOSPEL

51

9:12 Ἠλίας μὲν ἐλθὼν πρῶτον ἀποκαθιστάνει πάντα· καὶ πῶς …; (13) ἀλλὰ λέγω ὑμῖν ὅτι. … 12:5 κἀκεῖνον ἀπέκτειναν, καὶ πολλοὺς ἄλλους, οὓς μὲν δέροντες, οὓς δὲ ἀποκτέννοντες. 14:21 ὅτι ὁ μὲν υἱὸς τοῦ ἀνθρώπου ὑπάγει καθὼς γέγραπται περὶ αὐτοῦ, οὐαὶ δὲ τῷ ἀνθρώπῳ ἐκείνῳ. 14:38 τὸ μὲν πνεῦμα πρόθυμον ἡ δὲ σὰρξ ἀσθενής.

The noticeable point, as I have already indicated, is that all five examples are found in words of Jesus. We may also note that μέν is three times correlated with δέ in the best manner (12:5; 14:21, 38), and is not without correlative in the other two places.18 As to the text, there is uncertainty only in 9:12, where a minority of MSS omit μέν.19 One less example, however, would make no difference to my argument. Moreover, if μέν in 9:12 is secondary, it is of interest, and may be significant, that it has been introduced into the words of Jesus.20 The conclusion to which my argument is leading will already be clear. Μέν, we have seen, is a prestige feature associated with formal and educated Greek. In Mark it appears only in the speech of Jesus. The natural conclusion is that as regards this feature at least, Jesus speaks ‘better,’ more formal Greek than is spoken by those around him and used by Mark in the narrative. It may be difficult to establish how this came about, but the essential reason for it is readily seen: Jesus speaks in a manner appropriate to his status. What of 16:19? It is interesting, to say the least, that the use of μέν here differs markedly from that elsewhere in the Gospel. Not only is it used in the combination μὲν οὖν, not found in the rest of Mark, but it occurs in narrative.21 This reinforces the commonly-held view of the Longer Ending, and highlights the special character of the use of μέν elsewhere in Mark. 18

19

20

21

The correlation with μέν in 4:4–7 is described by BAGD (s.v. μέν 2.d) as ‘an inexact usage.’ In 9:12 the correlative is ἀλλά, a stronger equivalent of δέ (BDF, §447.6); this would be quite acceptable Class. usage: see Denniston, Greek Particles, 5–6. DLWΨ fam.1 (exc. 118) 28.565.892. If μέν is not original here, the parallel Matt 17:11 could be the source (so Torrey, Gospels, 57–8). But Matt’s change of ἀλλά to δέ (vs. 12) is perhaps a clue that μέν was already present in Mark, otherwise ἀλλά could have been retained: Matt is always careful to correlate μέν with δέ. In the other places in Mark μέν is either unanimously attested (4:4; 12:5) or omitted only in some versions (14:21, 38). A search through Legg’s apparatus brings to light five more examples of μέν in the MS tradition of Mark: 1:8; 4:28; 8:28; 10:27, 39. Of these, three are in sayings of Jesus (4:28; 10:27, 39). The first point is noted (e.g.) by Farmer, Last Twelve Verses, 99–100, and Elliott, ‘Text and Language,’ 261, but the second seems to have been missed. Μὲν οὖν, a favourite of Luke in Acts, has a pretentious sound to it. This and a number of other features suggest that the Longer Ending is on a different level of language from the rest of Mark.

52

COLLECTED ESSAYS ON THE GREEK BIBLE

It is natural to ask if Mark’s Gospel is alone in this use of μέν or whether anything similar is to be found in the other Gospels. A brief answer must suffice. In Matthew there is a clear link between μέν and the speech of Jesus: out of 20 examples only two appear in other contexts (one is used by John the Baptist, the other by the disciples). Luke and John, however, do not show anything comparable.

II At this point we may consider more fully the nature of what has been found. It seems to me that the use of μέν in Mark is best understood as an instance of a general linguistic phenomenon that is well known to all of us, though we are often not conscious of it. In recent years it has received increased attention from linguists within the relatively new discipline of sociolinguistics.22 Language varies with the occasion. The way something is said (or written) is adjusted according to who is speaking to whom, the circumstances, the subject-matter, and so on (the ‘social context’). An important ingredient in this adjustment is the degree of ‘formality.’ The different varieties (‘registers,’ ‘styles’) appropriate in different situations show a gradation from informal to formal, from ‘slang to ceremony,’ in G. W. Turner’s phrase.23 This appears in the level of formality of the individual features used. Every feature or ‘variable,’ in all spheres of the language (pronunciation, morphology, syntax, and vocabulary), is related in some way to the scale of formality. Formality is a complex thing, difficult to define. It involves many elements, such as propriety, technicality, politeness, seriousness, and correctness, each of which has as it were a scale of its own.24 The last is especially to be noticed. Correctness seems always to play a part in formality. The use of a feature felt to be ‘correct’ brings a formal tone into any context, while formal speech (or writing) to be formal must be on the whole 22

23

24

I cannot claim any expertise in this subject, or more than slight acquaintance with its extensive literature. I have made use especially of: Bolinger, Aspects; Ellis and Ure, ‘Varieties,’ 251–9; Gregory and Carroll, Language and Situation; Trudgill, Sociolinguistics; G. W. Turner, Stylistics. Stylistic levels of language in connexion with translation of the Bible are usefully discussed in Nida, ‘Varieties of Language,’ 316–22. Turner, Stylistics, 185. Terminology is still under discussion. For a full system of classification see Gregory, ‘Aspects.’ In Gregory’s system formality is comprised under ‘tenor of discourse.’ Cf. Trudgill, Sociolinguistics, 110.

SOME FEATURES OF THE SPEECH OF JESUS IN MARK’S GOSPEL

53

‘correct.’ Whatever the difficulties of definition, in practice we recognise at once the level of formality of any feature in our own language, and can learn to do so in others, though hardly with the full appreciation of the native speaker. It would not be easy to describe in full the ‘social context’ of Jesus’s utterances. There is no need to do so, however, in order to see the most important element in the situation, the high status of Jesus as the Messiah and Son of God, a being acting with authority and dignity and deserving of respect. It would be consistent with this if features of a higher level of formality occurred in his speech. Μέν is just such a feature. It is another matter to decide whether such features could have originated with Jesus himself or whether they were incorporated into his sayings by others who felt them to be appropriate. Whatever the answer to that question, it seems to me that the rationale behind the phenomenon remains the same. We have, then, what appears to be a secure example of greater formality in the speech of Jesus as recorded by Mark. The next step is obvious: are there any other features of the same kind as μέν? I have found that there are, and the clearest of them will be set out in the remainder of this paper. Attention will be concentrated on Mark, in order to establish the case as securely as possible for that Gospel. It is apparent, however, that the same thing is at work in the other Gospels, and their evidence will be noticed for some features. Before we proceed, my general approach needs to be more fully explained.

III Greek in the first century AD undoubtedly allowed for variation in degree of formality. The kind of study undertaken here depends on our being able to form an idea of the place of a given feature on the formality scale. What means are available for doing so? We have already seen most of them employed in dealing with μέν, but it will be useful to gather them together and state more explicitly what they are. A pervading element in post-Classical Greek is the scale of ‘literariness,’ or gradation in quality of Greek. Something of the kind is found in most languages, but Greek in this period is especially affected, for this was the time when there first emerged the diglossia that has been a characteristic of the language ever since.25 We are still a long way from a satisfactory 25

See, e.g., Browning, Greek, 49.

54

COLLECTED ESSAYS ON THE GREEK BIBLE

analysis and description of this scale, as of the post-Classical language generally. Recent works have reopened the discussion, without clear result.26 Our terms remain ill-defined, and exact criteria for categorising the level of a given text or feature are lacking. Nevertheless it is certain that varying degrees of literary refinement can be recognised in the texts. Lucian writes ‘better,’ more literary Greek than Theon the schoolboy (POxy 1.119 [II–III AD]). Formality in post-Classical Greek is bound to be intimately connected with this scale: formality and correctness go hand in hand. Therefore what we can learn about the literary level of a feature will also give an indication of the degree of formality it was felt to have. The use of the papyri for this purpose requires care. As is now coming to be recognised, they do not provide a homogeneous body of evidence for popular Greek.27 The occurrence of a feature in the papyri does not automatically make it colloquial or ‘vulgar.’ There are not only differences of level between documents; we have also to remember that stereotyped expressions, which often contain obsolete features, occur frequently, and that even a poorly-educated writer may have his literary pretensions. Μέν has provided an illustration of these points. In dealing with the NT, it is important to recognise that the books differ in literary level. This is fundamental to the present enquiry. Though we still await an adequate treatment of this subject,28 I take it as generally accepted that on the scale from vernacular to literary Mark, John, and Revelation belong to the lower levels, Revelation being lowest of the three;29 Matthew, Luke-Acts, and the Epistles are decidedly higher, though with differences between them, the high point being marked by Hebrews.30 None of the books reaches the level of the Atticists proper; but we should 26

27 28

29

30

Rydbeck, Fachprosa; Frösén, Prolegomena. For criticisms of Rydbeck’s proposed ‘Zwischenschichtsprosa’ see Frösén, 25–6, 50–1, 92–3, and reviews by Kilpatrick and Thesleff. In my view the language is best regarded as a continuum. Any divisions we make are artificial and do not indicate discrete varieties. Atticistic Greek, when it becomes the fashion, is not isolated but shows mutual interference with lower levels. Cf. the Mod. Greek situation: though Katharevousa and Demotic have separate identities, there is in practice much interference between the two: see Browning, Greek, 112–3; Sotiropoulos, ‘Diglossia,’ 7. Cf. esp. Rydbeck, Fachprosa, 26, 127, 176, 195. N. Turner’s recent continuation of Moulton’s Grammar (MHT, Style), is disappointing in this respect, thanks to his preoccupation with the question of Semitic influence. The older discussions in MH, Grammar, II, 5–11 and BDF, §3 leave a good deal to be desired. Moule’s remarks (Idiom Book, 2–3), though brief, are, as always, valuable. Kilpatrick’s attempt to add the Pastorals to this group seems to me to fail (‘What John Tells Us,’ 79–81). Some of Kilpatrick’s tests, especially παρά + acc., and σύν, are uncertain, and there are indications contrary to his thesis, such as μέν (2 Tim 1:10; 2:20; 4:4), οὖν (1 Tim 2:1, 8; 3:2; 5:14; 2 Tim 1:8; 2:1, 21), βούλομαι (1 Tim 2:8; 5:14; 6:9; Titus 3:8), ἡμέτερος (2 Tim 4:15; Titus 3:14), and ὦ (1 Tim 6:11, 20). Bauer in BAGD, xii.

SOME FEATURES OF THE SPEECH OF JESUS IN MARK’S GOSPEL

55

be misled if we therefore placed Matthew, Luke-Acts, and the Epistles low on the scale. It is not true, as is sometimes believed, that the NT was written in popular Greek pure and simple.31 We are of course speaking of the general character of the books. There is undoubtedly unevenness within books (some of it is the subject of this paper). It should also be noticed that level of language, which we are talking of here, is quite distinct from individual style. Mark and John, for example, are similar in level of language but distinguishable in style.32 Semitic influence is yet another matter. There is no simple correlation between incidence of Semitic idiom and literary level. The books of the LXX likewise vary in level of language. We have already seen in the case of μέν how this may help to establish the tone of a feature. We must also keep in mind that there is a gap of two to three hundred years between the LXX and the NT, during which changes are certain to have taken place in the language, among them changes that affect literary level. So we may find, for example, that a word that was in ordinary use in the time of the LXX has become obsolete and belongs to more elevated language in the NT. Similarly, when the LXX itself is quoted in the NT its level of language will tend to differ from that of the context. What has been said so far concerns distribution: by observing where a feature tends to occur we may learn something about its literary level and hence degree of formality. What other indications of level of formality are available to us? An appreciation of the long-term trends of the language, and their rate of progress, is essential. This is because it seems to be a general axiom of language that archaism and formality go together: the use of an obsolete or obsolescent feature instead of its current equivalent suggest more formal speech. (Contrast is an essential element here. An old feature is noticeable, or ‘marked,’ as regards formality only if it contrasts with some other feature which has replaced it in normal use; otherwise inherited features, which in fact make up the bulk of any language, are unmarked.)33 This principle is especially applicable to post-Classical Greek, where concern with past forms of the language was a dominant influence. We can be sure that when a writer of this period wished to be formal he or she would tend to employ obsolete features. We also depend on more elusive indications of formality. Some kinds of expression ‘sound’ formal in any language. Such reliance on impression 31 32 33

Similarly N. Turner, ‘Literary Character,’ 107–14; Wifstrand, ‘Stylistic Problems,’ 174. Kilpatrick, ‘Some Thoughts,’ 290. Cf. Frösén, Prolegomena, 134. In the case of μέν, the contrast is between use and non-use of the particle. In speech the function of μέν may well have been taken over by intonation.

56

COLLECTED ESSAYS ON THE GREEK BIBLE

is clearly hazardous, but we can define to some extent the features concerned. Elaboration and periphrasis, in particular, seem likely always to suggest formality. The preface to Luke’s Gospel, for example, clearly owes much of its highly formal tone to these, as well as to other characteristics less easy to name. That it was felt to be formal can hardly be doubted, even though detailed demonstration is not easy. In the same way, most would agree that poetry usually has a high degree of formality, despite the difficulty of proof. Finally, there are the Atticist grammarians.34 Their remarks are a valuable guide to what was felt to be ‘correct’ in their time, and must always be taken into account. But their usefulness for the NT and our present purpose is limited. The reasons for this are worth noticing. (a) Their observations date from the second century, when the Atticising tendency was at its height, and do not necessarily apply to the first century. (b) The object of their endeavours is the highest literary level; approval is reserved for the very ‘best,’ that is, Attic, Greek. Information about this does not always tell us what standards applied at less exalted levels. (c) Their statements are frequently crude and oversimplified; they need much interpretation in the light of other evidence. (d) Their material is incomplete, both because much of what they wrote has not survived and because some features were not noticed. In sum, they give us only a fragment of information about the complex picture of language varieties in Greek of the first century AD. By combining these approaches we may reach a fair understanding of the degree of formality of a given feature in the time of the NT. It must be accepted, however, that our knowledge is and will remain imperfect. Only the native speaker of a language can fully appreciate the variations in formality it employs. When the language is no longer spoken, we depend on making the best deductions we can from the available clues; certainty and exact knowledge are unattainable.

IV 4.1. εὖ It is generally accepted that in later Greek εὖ was gradually superseded by καλῶς, a near synonym from early in Classical Greek.35 Some such 34

35

A useful outline is found in Kilpatrick, ‘Atticism,’ 128. [See now Lee, ‘Atticist Grammarians’ = Essay 19.] BDF, §102.3; Chantraine, DELG, s.v. ἐύς; Mayser, Grammatik, I.3, 126; MM, s.v. εὖ; Schmid, Review of LSJ Part 4, 705.

SOME FEATURES OF THE SPEECH OF JESUS IN MARK’S GOSPEL

57

development was to be expected, given the shortness and irregularity of εὖ. It is in keeping with the tendencies of the Koine that the replacement is both ‘fuller-sounding’ and a regular formation related to a familiar word.36 Some specimen figures are enough to indicate the trend. Whereas καλῶς is hardly known to Homer (καλῶς 1, εὖ 215), and is still well outnumbered in Herodotus (καλῶς 32, εὖ 125) and Aristophanes (καλῶς 80, εὖ 130), in Plato the figures are more nearly even (καλῶς 528, εὖ 575), and by the time of Polybius the older word is definitely on the wane (καλῶς 87, εὖ 31). The papyri show the same development, as Mayser and Moulton-Milligan observe, and as may be illustrated by PTebt I–IV, containing texts of III BC to III AD: occurrences of καλῶς number 45, of εὖ 9 (not counting restorations). In Vita Aesopi G, a popular Greek text of I–II AD, καλῶς occurs 10 times, εὖ not at all.37 The LXX appears to go against the trend, with καλῶς 33 times and εὖ 56 times (also εὖγε 14 times, often doubled). But the higher figure for εὖ is largely due to the Pentateuch (καλῶς 3, εὖ 25, of which 16 are in Deut); its early date is at least part of the explanation. By the time of the NT it is clear that καλῶς has become the usual word and εὖ is obsolete. καλῶς occurs 37 times and is preferred even in the more literary books (Matthew twice, Luke 4 times, Acts 3, Epistles 18); εὖ is not only rare (6 examples, including one of εὖγε) but there is also something special about its use in every case. One example, in Eph 6:3, is in a quotation from the LXX. Another is found at the conclusion of the letter from the Jerusalem Council: Acts 15:29 ἐξ ὧν διατηροῦντες ἑαυτοὺς εὖ πράξετε. Ἔρρωσθε.

This is unequivocally labelled ‘literary language’ by BDF (§102.3). Its formulaic, and hence fossilised, character is also noticeable, whatever the sense of πράξετε.38 There remain four examples in the Gospels. Mark has one, in words spoken by Jesus: Mark 14:7 πάντοτε γὰρ τοὺς πτωχοὺς ἔχετε μεθ᾽ ἑαυτῶν καὶ ὅταν θέλητε δύνασθε αὐτοῖς εὖ ποιῆσαι.39 36 37 38 39

Cf. BDF, §126.1. Data from my own index. Cf. BAGD, s.v. εὖ; Haenchen, Acts, ad loc. There is no variant to εὖ. In place of αὐτοῖς a number of witnesses (ΑΧΨΘΠΣΦ etc.) read αὐτούς, thus making the Greek more ‘correct’ (cf. MHT, Grammar, III, 245). It is hard not to suppose the identity of the speaker helped to produce this reading (assuming it is secondary).

58

COLLECTED ESSAYS ON THE GREEK BIBLE

It is important that the more usual equivalent of this expression can be cited often from the NT: καλῶς ποιεῖν occurs 11 times, including once in Mark (7:37). There can be no doubt therefore that καλῶς could have been used in 14:7.40 The other examples in the Gospels, found in Matt 25:21, 23 (εὖ) and the parallel Luke 19:17 (εὖγε, v.l. εὖ), might be satisfactorily explained as due to the more literary tendency of these books. Moreover they are exceptional in that εὖ(γε) is here an interjection (‘well done!’).41 But it is a remarkable fact that they too occur in words spoken by Jesus (Parable of the Talents/Pounds).42 The conclusion seem inescapable. The location of εὖ in the speech of Jesus in Mark has not come about by accident. The old-fashioned word, which we may be sure had a formal, elevated tone, has been deliberately chosen for its appropriateness to the speaker. 4.2. The optative The post-Classical decline of the optative is well known and needs no demonstration here.43 The position reached by the first century AD may be summed up as follows. While not rare, the optative is no longer an integral part of the verb system; except for stereotyped expressions such as καλῶς ἂν ποιήσαις and μὴ γένοιτο, which occur at all levels, it is associated with the literary language, and even there some of its uses are greatly curtailed. Subsequently (from II AD onwards) it mcreases in frequency under the influence of Atticism but always remains an artificial feature consciously used. The NT unmistakably reflects current usage.44 There are some 67 examples, a modest total, and of these 15 are the set phrase μὴ γένοιτο and 11 the form εἴη. Some uses are barely represented, and others absent altogether. Most important of all is the distribution: all the examples except 40

41

42

43

44

A further equivalent is ἀγαθοποιῶ (10 times NT including v.l. Mark 3:4, but not in Mark otherwise. BDF (§102.3) imply that this use had a longer life. But note that καλῶς is used in a similar way: see Mark 12:32; Rom 11:20; LSJ, s.v. καλός C. 6, and Dio Cass. 67.18.1 καλῶς Στέφανε, εὖγε Στέφανε. The words are those of a speaker within the parable, not of Jesus in propria persona. It is interesting, however, that in both forms of the parable the speaker represented allegorically is Jesus. See esp. MHT, Grammar, III, 118–33; BDF, §§65.2, 384–6; Moulton, Grammar, I, 194– 9; Robertson, Grammar, 936, 1408; Debrunner–Scherer, Geschichte, 123-8; Browning, Greek, 37, 41–2; Mandilaras, Verb, §§603–58. [Evans, Verbal Syntax, 175–8.] Data from NT reference works already cited.

SOME FEATURES OF THE SPEECH OF JESUS IN MARK’S GOSPEL

59

one occur in Luke–Acts and the Epistles. Even here there are notable restrictions: the literary Greek of Hebrews, for example, has it only once (13:21 καταρτίσαι, in a prayer). Moreover, the link between this antiquated feature and its social setting is often plain to see. For example, the rare optative in a conditional clause is used by Paul in his speech of defence before governor Felix (Acts 24:19 εἴ τι ἔχοιεν); Turner aptly comments: ‘the use of slightly antique language in the presence of Felix … is understandable.’45 Similarly in Acts 26:29 we find a potential optative ‘where in the royal presence of Agrippa, Paul employs the stilted εὐξαίμην ἄν.’46 In view of all this it is not surprising that Mark, John, and Revelation make hardly any use of the optative. That there is even one example is cause for comment. When that example is found in words spoken by Jesus it is hard to doubt that the location is significant. In Mark in the Cursing of the Fig Tree, we find: Mark 11:14 μηκέτι εἰς τὸν αἰῶνα ἐκ σοῦ μηδεὶς καρπὸν φάγοι.

It is noteworthy that this example is in no way formulaic but has been created for this context, and that it is one of only two instances of the optative in imprecations in the NT.47 On the other hand, the volitive use was the most enduring of the uses of the optative and is the commonest in the NT, so there need be no suspicion that it was beyond the range of an author such as Mark.48 To complete the picture we need to notice that equivalents were definitely available; here the imperative (μηδεὶς … φαγέτω) is the probable alternative.49 There is, however, the question of the text in Mark 11:14. The majority reading φάγοι has been generally accepted without question, but Kilpatrick has argued that φάγῃ (DUWΨ fam.1 fam.13, etc., Or.) is original, on the principle that readings that are arguably stylistic improvements are to be regarded as secondary.50 Apart from the general objection to any rigid 45 46

47 48

49

50

MHT, Grammar, III, 126. MHT, Grammar, III, 123. Similarly BDF, §385.1: ‘The few examples [of the potential opt. in main clauses] are literary language befitting the occasion.’ This useful observation finds no place in BDR, §385. BDF, §384; MHT, Grammar, III, 122. The other is at Acts 8:20: τὸ ἀργύριόν σου εἴη εἰς ἀπώλειαν. As is certainly the case with Mark 12:2 ἵνα λάβοι ‫ *א‬238 579. This quite exceptional final opt. (‘the hall-mark of a pretty Attic style,’ Moulton, Grammar, I, 197) is generally agreed to be secondary. Other opts. noted from Legg’s apparatus are at Mark 12:19 (λάβοι 579); 9:34 (τίς μείζων: add. εἴη fam.13, etc.). See, e.g., 10:9; 13:15, 16, and Acts 1:20 λαβέτω for LXX λάβοι. For this and other equivalents see BDF, §384; MHT, Grammar, III, 120. Kilpatrick, ‘Atticism,’ 135; ‘What John Tells Us,’ 86.

60

COLLECTED ESSAYS ON THE GREEK BIBLE

application of this principle, there seem to me to be a number of difficulties that would need to be resolved before rejecting φάγοι. To what extent is the reading φαγη (also φαγει 472 etc.) a product of confusion of the sounds of οι, η, and ει?51 What is to indicate that the obsolescent optative has not been altered in the direction of more everyday Greek? Finally, is φάγῃ an exact equivalent of φάγοι, or does it alter the sense slightly, from a wish for evil to something like ‘no one must eat,’ ‘don’t anybody eat’?52 This seems to me a definite possibility, and if correct would be decisive against Kilpatrick’s proposal, since a clear motive for the change to φάγῃ could then be seen, that is, to tone down the curse.53 The same motive no doubt lies behind the alteration to μηκέτι ἐκ σοῦ καρπὸς γένηται in the parallel Matt 21:19.54 In my view, then, φάγοι, with its slightly archaic and elevated tone, is to be explained not as a correction of later scribes but as an original feature in keeping with the status of the speaker, and of the same character as μέν and εὖ. 4.3. ὦ The history of the incidence of ὦ is clear enough.55 In the earliest Greek it is rare, but it increases in frequency until in Classical Attic the vocative with ὦ is commonplace, while the vocative without it is less usual and in some way ‘marked.’ The details of Classical usage need not be gone into, for with Koine Greek comes a quite new picture. The bare vocative is now the norm, and the insertion of ὦ attracts notice. Very few examples of ὦ can be found in the papyri,56 and this evidence is supported, for example, 51

52

53

54

55

56

See Gignac, Grammar, 1:265–7. Mussies (Apocalypse, 246) notes Ruth 1:9 εὕροιτε B, εὕρητε A as an instance. So the other NT exx. of 3rd pers. subj. in prohibitions suggest: 1 Cor 16:11; 2 Cor. 11:16; 2 Thess 2:3 (add Mark 13:36 μὴ εὕρῃ?). Turner (MHT, Grammar, III, 94) describes these as ‘hortatory.’ Cf. Mussies, Apocalypse, 245–6, on the opposition between opt. and subj. It is evident from the commentaries that this has long been felt a difficulty: see, e.g., Lane, Mark, 402; Taylor, Mark, 460. The reading μηκέτι γένηται is that of the majority of MSS and read by NA28. Matthew’s γένηται may have helped to produce φάγῃ in Mark. Schwyzer, Grammatik, II, 60–1; Humbert, Syntaxe, §§499–502; Scott, ‘Vocative in Homer and Hesiod’; Vocative in Aeschylus and Sophocles’; ‘Additional Notes’; Moulton, Grammar, I, 71, and NT reference works cited below. Mayser, Grammatik, II.1, 55 notes one, ‘in dem feierlich gehobenen Artemisiafluch,’ UPZ 1.1 (IV BC) ὦ δέσποτ᾽ Ὀσερᾶπι. MM found BGU 2.665.III.8 (I AD) ἐρρῶσθαί σε εὔχομαι ὦ πάτερ (cf. πάτερ alone in II.11, 20). To these I can add only PFouad 25v.I.2 (II AD, document about a lawsuit) ὦ πονηρότατον πάντων τῶν ἀνθρώπων μειράκιον,

SOME FEATURES OF THE SPEECH OF JESUS IN MARK’S GOSPEL

61

by Vit. Aes. G, where vocatives are extremely frequent but ὦ occurs only three times, in fables inserted into the narrative.57 Epictetus similarly shows restricted use of ὦ.58 On the other hand, literary authors make frequent use of it, as a glance at Lucian or Plutarch and the figures for Philo (148) and Achilles Tatius (ὦ c.50, ὤ 17) are sufficient to confirm. In the same way, many of the LXX examples are found in the more literary books (4 Maccabees alone accounts for 36 out of the LXX total of 80).59 Imitation of the Classical language is obviously at work here. In sum, ὦ in late Greek shows the signs of being a prestige feature consciously used. In the NT the vocative without ὦ is usual. There is general agreement that ὦ is exceptional and calls for explanation whenever it occurs. To put it another way, ὦ is ‘marked’ and we need to try and appreciate what it conveys that the bare vocative does not. According to current opinion,60 ὦ is to be explained in three ways: (a) In Acts 1:1; 18:14, and 27:21 it is a stylistic refinement in accordance with Attic usage. (b) In Rom 11:33 it is exclamative ὦ (= ὤ), also in accordance with Attic usage. (This group is considerably expanded by BDF, following Katz, but none have followed suit.) (c) In the majority of instances it is said to express emotion, namely Matt 15:28; 17:17; Mark 9:19; Luke 9:41; 24:25; Acts 13:10; Rom 2:1, 3; 9:20; Gal 3:1, 1; Tim 6:20; Jas 2:20, to which we should presumably add 1 Tim 6:11 (this example seems to have been accidentally overlooked by all except Zerwick). We may note that Turner is compelled to observe that ‘there is no great emotion’ in some of these instances (Rom 2:1, 3; 9:20; Jas 2:20). This analysis is difficult to accept. Explanations (a) and (b) are clearly right, but (c) raises doubts. Emotion may well be present in many of these contexts, but it is puzzling that ὦ is not used in the many other cases

57 58

59

60

POxy 6.937.9 (III AD) ὦ κυρία μου ἀδελφή, and 31.2603.17 (IV AD, elegant letter of commendation) ὦ φίλτατε. Lest it be thought that vocatives (without ὦ) hardly ever occur in the papyri, I note that there are over 20 in POxy vol. 1 alone; see also exx. in Mayser, ibid. Vit. Aes. G 129.14, 131.12, 140.8. The usage of the fables is consistent with this: in Fab. Aes. 1–60 (ed. Halm) I find 16 vocs. with ὦ, 2 without. Total occurrences c.28. ἄνθρωπε illustrates the point well: it occurs 63 times, but only once with ὦ (Schenkl’s index). I have not seen Johannessohn, Präpositionen; Turner (MHT, Grammar, III, 33) reports J.’s finding that Epict. has ὦ with proper names ‘nur da, wo es sich um Grössen der Vergangenheit handelt.’ See Walters, Text, 229, 231. Walters comments (232) that the author of 4 Macc is ‘anxious to equal the best models.’ Breitenstein, Beobachtungen, 106–9, makes it clear that ὦ in 4 Macc is a stylistic device, inserted or omitted at will. BDF, §146; BDR, §146; MHT, Grammar, III, 33; Zerwick, Biblical Greek, §35; BAGD, s.v.

62

COLLECTED ESSAYS ON THE GREEK BIBLE

where emotion is equally discernible.61 The fact is, the vocative without ὦ seems to be the norm, whether emotion is expressed or not. There is also the inconsistency between (a) and (c): we are asked to accept that ὦ is a literary refinement in some places but not in others, and sometimes expresses emotion and sometimes does not. Further, it is hard to see any difference at all between some of the examples in (c) (especially Rom 2:1, 3; 1 Tim 6:11; Jas 2:20) and those in (a). I suggest that much better sense can be made of the NT evidence if, in accordance with contemporary usage, and taking up the unmistakable clue afforded by examples like Acts 1:1, we regard ὦ as primarily a feature of higher style. Its chief effect in all places is to give a formal and elevated tone.62 In emotional contexts this may well enhance the expression of emotion, but emotion is not what ὦ itself expresses. On this basis the NT distribution becomes fully intelligible. The majority of examples occur in the literary Greek of Acts and the Epistles, whose writers use it as an occasional embellishment; the rare occurrences in the Gospels share a significant characteristic: they are all found in words spoken by Jesus (Mark 9:19, Matt 17:17, Luke 9:41 ὦ γενεὰ ἄπιστος, Matt 15:28 ὦ γύναι, μεγάλη σου ἡ πίστις, Luke 24:25 ὦ ἀνόητοι καὶ βραδεῖς τῇ καρδίᾳ). Why it should occur in these particular places and nowhere else in the words of Jesus may be hard to discover, but this does not alter the point that when it does occur its appropriateness to the speaker can be felt. What we have in Mark, then, is a single example of this elevated feature, conspicuous against the background of Mark’s generally low-level Greek, but located where its effect is most telling.63 I have left aside so far the question of the differentiation between ὦ and ὤ, which has been the subject of a detailed examination by Walters.64 His results for the LXX seem clear, but are less persuasive for the NT. Here he finds exclamative ὤ in more than half the instances; in the Gospels he would read ὢ ἀνόητοι and ὢ γενεὰ ἄπιστος, leaving the vocative ὦ 61

62

63

64

E.g., Mark 8:33 || Matt 16:23 σατανᾶ. Mark 10:47, 48 || Matt 20:30, 31; Luke 18:38, 39 υἱὲ Δαυίδ (Ἰησοῦ). Matt 3:7 || Luke 3:7 γεννήματα ἐχιδνῶν. Matt 26:39 || Luke 22:42 πάτερ. Luke 22:33 κύριε. 34 Πέτρε, John 13:7, 9 κύριε, 19:26 γύναι, 21:17 κύριε. I have here the support of Jannaris (Grammar, 327): ‘its presence in postchristian compositions usually adds a certain solemnity or emphasis.’ The idea that ὦ in the NT expresses emotion seems to originate with Blass (Grammar, 1st ed. 1896, p. 85). No supporting evidence from outside the NT has yet been offered. The view that Jesus is here presented as a deity temporarily on earth complements my argument very well: see esp. Tabachovitz, Die Septuaginta, 113–5, where it is argued that several LXX reminiscences, including ὦ, aid the portrayal. Walters, Text, 228–36.

SOME FEATURES OF THE SPEECH OF JESUS IN MARK’S GOSPEL

63

only in ὦ γύναι. One may doubt the reality of the distinction here; if it is real it is a fine one and does not as far as I can see affect the conclusion I have drawn about the tone of this feature in the NT. 4.4. οὐ μή The use of οὐ μή with the subjunctive or future for a prohibition or strong denial has already been much discussed, and most of the data have been collected. I simply propose a reinterpretation, linking this feature with the phenomenon we are considering. It will be useful to summarise my argument at the outset. Οὐ μή is common in Classical Greek, but rather less so in post-Classical. In the LXX, however, it is noticeably frequent. The NT follows suit, but with the important restriction that apart from LXX quotations and reminiscences most of the examples are in sayings of Jesus. My conclusion is that the NT shows LXX influence in the use of this feature, and that it conveyed not only emphasis but more important a solemn, biblical tone, especially suited to prophetic utterances and the speech of Jesus. These points may now be taken up in more detail. The origin of the idiom has been debated, but has no importance here, nor is it necessary to enter into discussion of the differing constructions (aor., pres. subj.; fut. indic.). The wide spread of the idiom in Classical Attic is clear.65 Examples are found both in prose (Thuc., Plato, etc.) and poetry (Aesch., Soph., Eur.), in elevated language (Demosth., the dramatists) just as much as popular (Aristoph., Xen.). No particular trend is evident, except that its rarity in Aristotle and Menander suggests there was some decline in late Classical.66 The post-Classical evidence is not easy to assess, but overall there does seem to be a drop in frequency, as others have observed. Turner found no instances among 1370 negatives in his sample of six literary authors,67 and only sporadic examples are observable elsewhere in literature (e.g. one in Ach. Tat., three in Epict.).68 In the papyri, however, it is not exactly rare, as is often said. Besides the 17 Ptolemaic occurrences listed by Mayser,69 65 66

67

68

69

LSJ, s.v. οὐ μή; KG, Grammatik, II.2, 221–3; Schwyzer, Grammatik, II, 317. Aristotle only Top. 108a.28; Mir. 157b.28; Menander barely once, Fr. 893 (Körte) ap. Photius: οὐ μηκέτι· ἀντὶ τοῦ οὐκ ἔτι. Μένανδρος. Aratus, Callimachus, Lycophron, Aelian, Philostratus, and Alciphron: MHT, Grammar, III, 96 n. 2. Ach. Tat. 2.19.1; Epict. 1.25.27; 3.7.8, 22.33. Jacobitz records one ex. in Lucian: Pisc. 18. Mayser, Grammatik, II.1, 233; II.2, 564.

64

COLLECTED ESSAYS ON THE GREEK BIBLE

I have been able to find more than 20 examples.70 A further five in Vit. Aes. G should also be noticed.71 All this implies that while it was not common it was still very much alive, and more at home in popular Greek than literary.72 Turning to the LXX, one is struck by the great frequency of this feature. More than 800 occurrences are recorded by Hatch-Redpath.73 These are spread throughout, though with notable concentrations in the Pentateuch, Job, Psalms, Sirach, and the prophetical books. The exact figures for the Pentateuch and Isaiah, for example, are 82 and 103 respectively (Rahlfs’s text). Why the translators made such frequent use of it we can only guess. Hebraism can be ruled out at once: there is no equivalent Hebrew (or Aramaic) idiom. Another approach may be more fruitful. In the Pentateuch, from the first example onwards, an association with divine pronouncements is noticeable.74 Thus at its first appearence in Genesis we find: Gen 3:1 Τί ὅτι εἶπεν ὁ θεός Οὐ μὴ φάγητε ἀπὸ παντὸς ξύλου τοῦ ἐν τῷ παραδείσῳ;

and typical of later examples is: Exod 20:5 οὐ προσκυνήσεις αὐτοῖς οὐδὲ μὴ λατρεύσῃς αὐτοῖς.

The connexion between this and the generous use of the idiom in the prophetical books is hard to miss. While this is not the full story,75 it suggests that the Pentateuch translators, and then others following their lead, found in the οὐ μή construction a powerful and somewhat formal expression suited to biblical style generally and divine utterances in particular. 70

71 72

73

74 75

Already noted in MM and BAGD: POxy 1.119.4, 6, 14, 15 (II–III AD); 6.903.16 (IV AD); 12.1483.10 (II–III AD); PGM 5.279 (= PLond 1.46); 13.321 (= PLeidW); Wessely, Specimina, 26.4 (I AD). Add: PHaun 10.28 (III BC); POxy 42.3070.6 (I AD), and, with some dozen exx., SB 6.9641 recto I.17–, II.6– (II AD), a series of promises made on entering one of the Egyptian priesthoods. From the inscriptions there is SIG 2.1042.16 (II–III AD) [= IG II 1366.16 (I AD); also IAbydosMemn B.1 (n.d.); IPessinous 6.4 (II BC)]. Vit. Aes. G 28.5; 50.10; 61.11; 67.10; 80.11. Suzanne MacAlister has drawn my attention to a much later example in Malalas, Chron. 7.172.6 (VI AD) οὐ μὴ σταθῇ ἡ πόλις σου Ῥώμη. It seems significant that the context is a Pythian prophecy. [οὐ μή ‘more at home in popular Greek’: I am doubtful now of this statement.] HR, s.vv. οὐ μή, οὐδὲ (οὐ) μή, οὐκέτι (οὐ) μή, οὔτε (οὐ) μή. This figure includes instances where a v.l. is involved. In Gen 10 out of 14, in Exod 12 out of 12 exx. are in divine speech. Among other questions to be considered is the extent to which οὐ μή is used for variety (see, e.g., Exod 20:5 above). [Fuller discussion in Lee, Greek of the Pentateuch, 56.]

SOME FEATURES OF THE SPEECH OF JESUS IN MARK’S GOSPEL

65

Whatever the reason for the frequency of οὐ μή in the LXX, it must have been a very noticeable characteristic of the version for those who heard and read it. In view of this and its unexpected frequency in the NT as compared with contemporary Greek, we may reasonably conclude that it came into NT usage largely from the LXX.76 It follows that its tone is likely to have been formal and dignified, as is true generally of LXX features in the NT.77 Moreover in this case there is the association with prophecy and the like to add further to its formal character. We are now in a position to appreciate NT usage more fully. Its most striking aspect has long been observed: the numerous examples in the Gospels are almost entirely confined to sayings of Jesus.78 The conclusion I would draw is that οὐ μή is not distributed in this way by accident, but because it lends appropriate solemnity and biblical flavour to Jesus’s words, recalling in particular the LXX use of the idiom in prophecy. It is worth adding that sayings containing οὐ μή are frequently introduced by ἀμὴν λέγω ὑμῖν (and variations), which so obviously ‘adds emphasis and solemnity to that which follows.’79 The examples outside sayings need not be examined in detail here. In brief they are as follows.80 In the Gospels there are 7 (Mark 14:31 || Matt 26:35; 16:22; Luke 1:15; John 13:8; 20:25; 11:56), of which only the last (in a question) is in any way inconsistent with the view that οὐ μή is formal in tone. In Acts, the Epistles, and Revelation the great majority are in LXX quotations or reminiscences. In addition there is an example that is possibly part of an agraphon in 1 Thess 4:15.81 This last example leads us to a point of considerable interest. Moulton noted, besides 1 Thess 4:15 and Mark 16:18, the appearance of this feature twice in the Oxyrhynchus Logia (POxy 1.1.2; 4.654.1). To these can be added occurrences in Gos. Thom. A 3:2 and B 2:2 (ed. Tischendorf): 76 77

78

79

80 81

So Lagrange, Saint Marc, XCIX; Tabachovitz, Die Septuaginta, 106 n. 3; MHT, Style, 33, 69; more tentatively Moule, Idiom Book, 22. Cf. BDF, §4.2: ‘the language of the LXX appeared to be very appropriate to a solemn and dignified style.’ This aspect of Lucan style is often remarked on by Haenchen (Acts, e.g., 80, 159, 254, 351), and cf. Sparks, ‘Semitisms of Acts,’ 27. Moulton, Grammar, I, 191; BDF, §365. Robertson (Grammar, 1405) gives a useful table of occurrences, classified according to speaker. His figures are: total 66 in the Gospels, 58 in Sayings. But John 8:52 (quoted words of Jesus) should be added to the latter. Taylor, Mark, on 3:28; similarly BAGD, s.v. ἀμήν 2. In Mark, 5 out of 7 sayings with οὐ μή begin with this formula. Ἀμήν is of course another example of the phenomenon under consideration: it is used only by Jesus, and its solemn character is unquestioned. Cf. Moulton, Grammar, I, 191–2. So Moulton, ibid.; cf. Frame, Thessalonians, 171–2. But according to Jeremias (Unknown Sayings, 80 n. 3) it is ‘now generally agreed’ that the agraphon consists of 16–17a, and 15 is an ‘introductory summary’ by Paul.

66

COLLECTED ESSAYS ON THE GREEK BIBLE

both times Jesus is the speaker, and the idiom occurs only here in these texts. From this evidence it appears that οὐ μή became at some stage a conventional feature for sayings of Jesus. In this connexion it is noteworthy that Luke in his rephrasing of the saying in Mark 10:29–30 introduces οὐ μή (Mark 10:30 ἐὰν μὴ λάβῃ ἑκατονταπλασίονα || Luke 18:30 ὃς οὐχὶ μὴ [ἀπο]λάβῃ πολλαπλασίονα).82 Finally, further mention needs to be made of Moulton’s discussion, which has influenced, and I think confused, many subsequent treatments.83 Moulton reached the conclusion that the frequency of οὐ μή in OT quotations and sayings of Jesus may be put down to ‘a feeling that inspired language was fitly rendered by words of a peculiarly decisive tone.’ This conclusion is not far from the mark, but on the way to it Moulton argued that excessive use of οὐ μή is linked with translation from Semitic originals, which includes sayings of Jesus; that it has weaker emphasis in ‘translation Greek’ than elsewhere; and (less clearly) that this over-use of unemphatic οὐ μή is connected with ‘elementary Greek culture.’ Moulton here seems to me to be following a quite false trail. Without a corresponding Hebrew or Aramaic idiom, there can be no causal connexion between the use of οὐ μή and translation from Semitic originals. Translation therefore does not explain the concentration of οὐ μή in sayings any more than it explains its frequency in the LXX. In any case the sayings cannot be lumped together with the LXX as if the translation process was the same in both. As to the degree of emphasis, the question seems largely irrelevant. Examination fails to confirm a correlation between ‘translation Greek,’ that is, the LXX and the Sayings, and unemphatic οὐ μή. There is little emphasis, e.g., in Epict. 1.25.27 ἄνθρωπε, καὶ μὴ θεώρει καὶ οὐ μὴ θλιβῇς, ‘do not become a spectator and you will not be crowded’ (cf. Luke 6:37); POxy 42.3070.6 οὐκέτι οὐ μὴ δείρομέν σε ἐὰν δώσῃς ἡμεῖν τὸ ποιγίσαι, ‘we won’t hit you any more if …’; PCairZen 2.225.6; Vit. Aes. G 61.11. This is true even of some Classical examples, such as Aristoph., Pax 1226, Nub. 296. On the other hand some ‘translation Greek’ examples seem quite emphatic, e.g., Mark 10:15 ἀμὴν λέγω ὑμῖν, ὃς ἂν μὴ δέξηται τὴν βασιλείαν τοῦ θεοῦ ὡς παιδίον, οὐ μὴ εἰσέλθῃ εἰς αὐτήν, Gen 21:16 οὐ μὴ ἴδω τὸν θάνατον τοῦ παιδίου μου, Matt 5:26 ἀμὴν λέγω σοι, οὐ μὴ ἐξέλθῃς ἐκεῖθεν, ἕως ἂν ἀποδῷς. … Furthermore the same expressions can be cited from both kinds of Greek: οὐ μὴ φάγω is found in Gen 24:33; Tob 7:12S; 1 Cor 8:13; POxy 1.119.14 82

83

[Another instance: Mark 4:22 οὐδὲ ἐγένετο ἀπόκρυφον ἀλλ᾽ ἵνα ἔλθῃ εἰς φανερόν || Luke 8:17 οὐδὲ ἀπόκρυφον ὃ οὐ μὴ γνωσθῇ καὶ εἰς φανερὸν ἔλθῃ.] Moulton, Grammar, I, 39, 187–92; MM, s.v. οὐ. Cf., e.g., Lagrange, Saint Marc, XCIX; Taylor, Mark, on 9:1; Zerwick, Biblical Greek, §444. [Also Decker, ‘Markan Idiolect,’ 51 n. 31.]

SOME FEATURES OF THE SPEECH OF JESUS IN MARK’S GOSPEL

67

and SB 6.9641 recto 1.17; οὐδ᾽ οὐ μὴ γένηται is found in Matt 24:21 and ChrWilck 122.4. Clearly it would be fanciful to detect any difference of emphasis or ‘decisiveness’ between the biblical and non-biblical occurrences.

So far as it can be judged at all, the emphasis varies, both in ‘translation Greek’ and elsewhere. If οὐ μή shows loss of emphasis, it is more likely to be a case of the universal tendency for strong expressions to lose their force than a phenomenon peculiar to translators of Hebrew and Aramaic, who were no more lacking in ‘Greek culture’ than many others who used Koine Greek. 4.5. Other probable features Other features besides those examined above have been found. A full treatment is not possible here, however, and I shall simply indicate them briefly. While not all can be regarded as certain, I have examined them far enough to be confident that they are probable. (1) ἄν with indicative in the apodosis of an ‘unreal’ condition is found in Mark only in 13:20 (and very doubtfully 7:11),84 in contrast with indicative alone (Mark 9:42; 14:5, 21, and sporadically elsewhere in the NT). (2) ἔξωθεν in its original sense of ‘from outside,’ with the force of -θεν felt, occurs only in Mark 7:15, 18 in the NT; elsewhere ἔξωθεν is equivalent to ἔξω. (3) ἔσωθεν similarly has the obsolescent sense of ‘from inside’ only in Mark 7:21, 23, Luke 11:7. (4) θύραι of one door is rare in the NT; in Mark it is found only in 13:29 ἐπὶ θύραις. The appearance of equivalents using the singular (Mark πρὸς τὴν θύραν 1:33; 2:2; 11:4; Acts 3:2; ἐπὶ τὴν θύραν Rev 3:20; πρὸς τῇ θύρᾳ John 18:16) makes it likely the old set phrase was obsolescent and formal-sounding. (5) ὁράω pres. and impf. is obsolescent in the NT, being replaced by βλέπω (and θεωρῶ). Out of 11 occurrences in the Gospels (not including John 6:2 ἑώρων l. ἐθεώρουν), 9 are in words of Jesus (Mark 1:44; 8:15; Matt 8:4; 9:30; 16:6; 18:10; 24:6; Luke 12:15; 16:23.). (6) οὐαί, like οὐ μή, is a feature with OT associations and a solemn biblical flavour: all examples in the Gospels are in sayings. (7) οὖν is greatly affected by variation in the MSS, but it is hardly accidental that the five Marcan examples accepted in NA28 occur as follows: 84

It is to be understood that the noted feature occurs in words of Jesus.

68

COLLECTED ESSAYS ON THE GREEK BIBLE

in words of Jesus (10:9; 13:35), in supposed words of Jesus (11:31 [οὖν]), in words of Pilate (15:12), and in narrative only in the Longer Ending (16:19). A tendency towards localisation in sayings is also evident in Matthew (though not in Luke or John). (8) The substantival neuter participle, except for stock phrases, shows signs of being a feature of more elevated style. The majority of NT examples are in the more literary books (63 in Luke-Acts, 29 in Matt, 58 in the Epistles). In Mark there are only seven, as follows: in narrative τὸ γεγονός (5:14) and τὰ μέλλοντα (10:32); in words of Jesus τὰ κοινοῦντα (7:15), τὰ ἐκπορευόμενα (7:15), τὸ εἰσπορευόμενον (7:18), τὸ ἐκπορευόμενον (7:20), and ἐκ τοῦ περισσεύοντος (12:44). The two in narrative are set phrases whereas the others are clearly more special to their contexts. V This paper has not attempted to cover more than a part of a large subject. Attention has been concentrated on the most telling examples and a full explanation of the methods used. Nevertheless the evidence examined gives good grounds for thinking that there is some tendency in Mark’s Gospel, and probably elsewhere, for features having a formal, dignified tone to be localised in the words of Jesus. It also suggests that further investigation along the same lines would be fruitful, and lead us ultimately to a more complete appreciation of NT usage. It must be stressed that this phenomenon is no more than a tendency. The features concerned are slight and occur only sporadically. There is no suggestion that a more elevated, formal tone is consistently maintained in the sayings of Jesus or that a sharp difference between them and the surrounding contexts can be detected. The phenomenon is in fact such as to escape the notice of all but the native speaker. But this is just what might be expected: as modern sociolinguistic study has amply demonstrated, linguistic variables are often unobtrusive, and speakers show remarkable sensitivity in using and recognising them. It may be asked what proportion of the Gospels is made up of sayings. Clearly, the greater the extent of the sayings, the less meaningful any contrast between their usage and that of the remainder. The answer is reassuring. A calculation based on lines of Nestle-Aland’s text shows that sayings comprise 37% of Mark, 42% of John, 48% of Luke, and 57% of Matthew.85 85

Line-counts were: Mark 543 out of 1482; John 812 out of 1943; Luke 1228 out of 2562; Matt 1361 out of 2395. I do not know if a similar calculation has been made before.

SOME FEATURES OF THE SPEECH OF JESUS IN MARK’S GOSPEL

69

An important parallel to the suggested phenomenon is to be noticed. In the LXX Pentateuch, as I have argued in detail elsewhere,86 there is at least one case of the use of more elevated vocabulary when God is the speaker: the everyday word βρέχω, usual in narrative, is replaced by the antiquated, more dignified equivalent ὕω in the speech of God in Exod 9:18; 16:4.87 Parallels from further afield are no doubt abundant. To take only one example, it has been observed that in Virgil’s Aeneid the speech of the gods is coloured by archaisms and other features adding impressiveness.88 My argument finds further support in a related phenomenon observable in the NT, the tendency for respectful, dignified terms to be used in describing Jesus and his actions. Some examples are as follows: (a) In the description of the death of Jesus in Mark 15:37, 39; Luke 23:46, dignity and solemnity are imparted by the use of ἐκπνέω, a rare euphemism with poetic antecedents.89 (b) The obsolescent word βοῶ, largely replaced by κράζω, appears in Mark, apart from a LXX quotation, only when Jesus is being described (15:34). (c) Δακρύω, a milder and more dignified term than the usual κλαίω, is confined to a single occurrence in the NT, in John 11:35 ἐδάκρυσεν ὁ Ἰησοῦς.90 (d) A more complex example, involving also the characterization of the speaker, is seen in Luke 23:41 οὗτος οὐδὲν ἄτοπον ἔπραξεν, where ἄτοπον is a carefully chosen word, avoiding the possible offence of the more direct πονηρόν or κακόν (contrast 23:22 where Pilate says τί γὰρ κακὸν ἐποίησεν οὗτος;).91 In a similar vein are the numerous changes, especially omissions, made by both Matthew and Luke because of ‘an increasing feeling of reverence for the person of Christ’ (Allen).92 Both the phenomenon just mentioned and the one described in this paper, though extensive in themselves, are only facets of the whole subject of adjustment of language to situation and speaker in the NT. Many

86 87

88 89

90 91

92

Black (Aramaic Approach, 52) gives some estimates evidently based on impression alone. Lee, Lexical Study, 123–4. It is not clear how extensive this phenomenon is. Other probable features known to me are: ἡμέτερος (Gen 1:26), ὑμέτερος (Gen 9:5), μέσαι νύκτες (Exod 11:4), and cf. οὐ μή above. [The first two are developed in Lee, Greek of the Pentateuch, 49–53.] Palmer, Latin, 112–3. Simpson, Words, 14. Cf. Matt 27:50 ἀφῆκεν τὸ πνεῦμα, John 19:30 παρέδωκεν τὸ πνεῦμα. I owe exx. (b) and (c) to Shipp, ‘Distribution,’ 134–5. A useful discussion of the semantic history of ἄτοπος is found in Arnott, ‘The Confrontation,’ 119–22. Allen, Matthew, xxxi–iii; on Luke see Cadbury, Style of Luke, 90–6.

70

COLLECTED ESSAYS ON THE GREEK BIBLE

observations on this subject have been made, but in a piecemeal fashion; more systematic study would be likely to find much of interest, and enhance our appreciation of NT language and the impact it had in its own time. The purpose of this paper has been to observe a phenomenon, not to advance any theory about how it came into being. Nevertheless the observation, if correct, clearly has implications for the history of the sayings of Jesus. I must leave it to others with the necessary expertise to explore the subject properly; I mention only the most immediate consequences as I see them. None of the features examined has any counterpart in Aramaic (or Hebrew). This means that they cannot go back to an original Aramaic form of the sayings. It follows that, unless the sayings were originally spoken in Greek, these features do not derive from Jesus himself but were introduced into the sayings by others at some stage in their history. Who introduced them is a matter for conjecture. It does not seem likely that Mark himself was responsible, at any rate for all of them; if he was we might expect to see more use of such features in his own writing. (It would be wrong, however, to suppose that he was not fully aware of them). It is more likely that Mark simply used what came to him, and that the sayings had already undergone a certain amount of development of this kind. That the sayings were not felt to be unalterable is demonstrated in the NT itself, where Matthew and more especially Luke are seen to make changes, many of them stylistic, in sayings just as readily as elsewhere. Finally there is the question of whether anything in the nature of a convention operated to produce our phenomenon. While this could apply to some features (especially οὐ μή), it seems more likely that most of them were introduced independently and haphazardly, out of a natural desire for the dignity of the speaker to be reflected in his language.

Postscript Studies that draw on or make significant mention of this paper are: Aitken, ‘Characterization of Speech’; Bentein, ‘Particle-usage,’ 724 and n. 12; idem, ‘Dimensions of Social Meaning,’ 157 and n. 68. The subject of οὐ μή in the LXX is further developed in Lee, Greek of the Pentateuch, 54–8.

6 A NON-ARAMAISM IN LUKE 6:7 1991

Abstract The use of εὑρίσκω in Luke 6:7 with the meaning ‘find a way (to),’ ‘be able (to)’ has been suspected of being an Aramaism. While the Aramaic evidence has been established, the evidence for this use in Greek has been ignored or not followed up, both by advocates of the proposal and in the NT lexicons. Evidence is here presented to show that it is a thoroughly Greek use in the Koine period, with roots in earlier Greek. There is no reason to see it as due to the influence of Aramaic on Luke’s language.

I In Luke 6:7 εὑρίσκω is found in a context requiring the sense of ‘find a way (to),’ ‘be able (to)’: Luke 6:7 παρετηροῦντο δὲ αὐτὸν οἱ γραμματεῖς καὶ οἱ Φαρισαῖοι εἰ ἐν τῷ σαββάτῳ θεραπεύει, ἵνα εὕρωσιν κατηγορεῖν αὐτοῦ.1

This use, also found in the text of D at 11:54 and 13:24, has been the subject of remark since the time of Wellhausen, who suggested that an Aramaism was involved.2 In more recent discussion, notably by Black, Fitzmyer, and Wilcox, attention has been focused on establishing the necessary evidence on the Aramaic side, namely the existence of a verb with the two senses of ‘find’ and ‘be able.’3 It now seems beyond doubt that in Palestinian Aramaic of the NT period ‫ שׁכח‬was in use in just this way. With this settled, Fitzmyer and Wilcox are happy to conclude that the use of εὑρίσκω in the sense of ‘be able’ reflects Aramaic interference. 1

2 3

This reading continues to hold the field in NA28, with the support of P4vid ‫ *א‬B Θ f1 1241 sa. The majority reading has κατηγορίαν for κατηγορεῖν (often combined with κατ᾽ before αὐτοῦ). Wellhausen, Einleitung, 17. Black, Aramaic Approach, 133–4; Fitzmyer, ‘Aramaic Background,’ 12–3; Wilcox, ‘Semitisms,’ 1011–2. Fitzmyer’s material reappears without substantial change in his commentary, Luke I–IX, at 117, 610–1.

72

COLLECTED ESSAYS ON THE GREEK BIBLE

But what of Greek usage? Is it safe to assume that in the Koine period εὑρίσκω was used only in the sense of ‘find,’ and that ‘be able’ was not part of its semantic range? For Black and Wilcox the question simply does not arise. Fitzmyer is aware of its importance, but his investigation into Greek usage goes no further than a consideration of the one papyrus occurrence noted by BAG (1957).4 Having (rightly) rejected this as not being a true parallel to Luke 6:7, he treats the matter as closed. In fact there is good evidence for the use of εὑρίσκω in the sense of ‘be able’ in post-Classical Greek, much of it readily available in standard reference works. The examples I have been able to find, without attempting to be exhaustive, are set out below. For the first of them we need look no further than LSJ:5 (1) Epictetus (I–II AD) 2.12.2 δὸς γοῦν ᾧ θέλεις ἡμῶν ἰδιώτην τινὰ τὸν προσδιαλεγόμενον· καὶ οὐχ εὑρίσκει χρήσασθαι αὐτῷ, ἀλλὰ μικρὰ κινήσας τὸν ἄνθρωπον, ἂν παρὰ σκέλος ἀπαντᾷ ἐκεῖνος, οὐκέτι δύναται μεταχειρίσασθαι, ἀλλ᾽ ἢ λοιδορεῖ ἢ καταγελᾷ καὶ λέγει· ἰδιώτης ἐστίν· οὐκ ἔστιν αὐτῷ χρήσασθαι. At all events, give to anyone of us you please some layman with whom to carry on an argument; he will find no way of dealing with him, but after moving the man a little, in case the latter thwarts him, our man gives up trying to handle him, and thereafter either reviles him, or laughs him to scorn, and remarks, ‘He is a mere layman; it is impossible to do anything with him.’ (Oldfather, Loeb ed.)

The parallel with our Lucan example is clear: οὐχ εὑρίσκει χρήσασθαι αὐτῷ = ‘he does not find a way/is unable to deal with him.’ We notice, too, that Epictetus, varying his word, uses δύναται to express much the same thought further on in the sentence, when he says οὐκέτι δύναται μεταχειρίσασθαι = ‘he is no longer able to handle him.’ II There are further examples in literary writers of the same period: (2) Plutarch (I–II AD), Pyrrhus 6.1 ἐκ δὲ τούτου πολλὰ μὲν περινοῶν καὶ μεγάλα τῇ γνώμῃ, ταῖς δὲ ἐλπίσι μάλιστα καὶ πρῶτον ἀντιλαμβανόμενος τῶν πλησίον, εὗρεν ἐμφῦναι τοῖς Μακεδόνων πράγμασιν ἐκ τοιᾶσδέ τινος προφάσεως. 4

5

Namely, PPar 45.7 (153 BC) προσέχων μὴ εὕρῃ τι κατὰ σοῦ ἰπῖν. This was evidently taken over by BAG from MM, who unjustifiably cited it in this connexion and failed to notice any better parallel. It continues to appear in BAGD (1979), and BRAA (1988), still with the outmoded reference to PPar 45 (= UPZ 69). [Also still in BDAG (2000).] LSJ, s.v. εὑρίσκω I.3. ‘Act., also, find means, be able.’

A NON-ARAMAISM IN LUKE 6:7

73

After this, he began to revolve many large projects in his mind; but his hopes were fixed first and more especially on undertakings close at hand, and he found a way to take direct part in Macedonian affairs, on grounds something like the following. (Perrin, Loeb ed.)

Here again we find εὑρίσκω without an object and linked directly with an infinitive (ἐμφῦναι intrans. aor. 2, ‘grow in,’ ‘plant oneself in,’ etc.). (3) Achilles Tatius (II AD) 1.10.2 ἐὰν γὰρ ἡ ὠδὶς παραγένηται καὶ ἐνστῇ τῆς ἀνάγκης ἡ προθεσμία, μηδὲν πλανηθείς, κἂν πρωτοκύμων ᾖς, εὑρήσεις τεκεῖν ὑπ᾽ αὐτοῦ μαιωθεὶς τοῦ θεοῦ. For when you begin to feel the pangs and it is clear that the destined day is at hand, you cannot go wrong, even though it be your first labour, but you will find the way to bring forth and the god himself will deliver you. (Gaselee, Loeb ed.)6 (4) Dio Chrysostom (I–II AD) 33.44 καὶ τινὲς μὲν ὑμῶν ὀργίζονται καί φασί με ὑβρίζειν τὴν πόλιν, τοὺς δὲ ταῦτα ποιοῦντας οὐκ αἰτιῶνται· τινὲς δὲ ἴσως καταγελῶσιν, εἰ περὶ μηδενὸς κρείττονος εὗρον εἰπεῖν. Some of you are angry and say that I am insulting your city, but they do not find fault with those who do these things; some on the other hand are perhaps laughing at the fact that I was not able to talk about anything more important.

Next these three examples in Vita Aesopi G, a more popular text of I AD:7 (5) Vita Aesopi G 35.16–7 ὁ Ξάνθος ἀκούσας φιλοσόφου μὲν ζήτημα, μὴ εὑρίσκων αὐτὸ δὲ ταχέως ἀναλῦσαι, [τῷ Αἰσώπῳ ἑστῶτι] λέγει· πάντα τῇ θείᾳ προνοίᾳ διοικεῖται. Xanthos, hearing a problem for a philosopher, but not being able to solve it readily, said, All things are under the control of divine providence. (6) Vita Aesopi G 48.27–9 εἷς τῶν σχολαστικῶν πρὸς τοὺς ἑτέρους λέγει· διὰ τί τὸ μὲν πρόβατον ἐπὶ θυσίαν ἀγόμενον οὐ κέκραγεν, τὸ δὲ χοιρίδιον μεγάλα τονθρύζει; μηδενὸς οὖν εὑρίσκοντος λῦσαι τὴν ἐρώτησιν, Αἴσωπος λέγει. … 6

7

The meaning is defined as ‘discover how to’ in O’Sullivan, Lexicon to Achilles Tatius, s.v. 3a. Rather different, I think, is 2.35.5 ἀπῆλθεν ὁ ἐραστὴς οὐχ εὑρὼν πιεῖν, where πιεῖν is virtually a substantive and εὑρίσκω has the sense of ‘get’; cf. John 4:7, 10 δός μοι πιεῖν. O’Sullivan however defines it as ‘get a chance to’ (3b), which makes it very close to Luke’s use. The same uncertainty appears in Epict. 3.19.5 ἂν μὴ εὕρωμεν φαγεῖν ἐκ βαλανείου, where again I feel φαγεῖν is quasi-substantival, though it might be adduced as a further parallel to Luke. I wish to thank Dr Suzanne MacAlister for bringing the Achilles Tatius examples to my notice. Refs. to section and page line in Perry, Aesopica. This version of the Life, dated by Perry to I AD, is a rare example of a work of literature written (for the most part) in popular Koine Greek.

74

COLLECTED ESSAYS ON THE GREEK BIBLE

One of the students said to the others, Why is it that a sheep being led to slaughter does not cry out, while a pig makes a great deal of noise? When no one was able to answer the question, Aesop said. … (7) Vita Aesopi G 102.4–7 ἐπ᾽ ἐκείνοις δὲ τοῖς καιροῖς ἔθος εἶχον οἱ βασιλεῖς παρ᾽ ἀλλήλων φόρους λαμβάνειν διὰ τῆς ἐναρέτου μάχης· οὔτε γὰρ ἐν πολέμοις συνίσταντο οὔτε μάχαις· ἔγραφον γὰρ προβλήματα φιλοσοφίας δι᾽ ἐπιστολῶν, καὶ ὁ μὴ εὑρίσκων διαλύσασθαι φόρους ἐτέλει τῷ πέμψαντι. In those times the kings had a custom of receiving tribute from one another by an excellent form of contest: they did not engage in wars or battles, but used to send each other philosophical problems by letter, and the one who was not able to solve (the problem) paid tribute to the sender.

III In the papyri Preisigke lists some dozen examples, adding ‘usw.’ for good measure.8 Of these the clearest are: (8) BGU 3.822.28 (105 AD)9 καὶ [ἐά]ν σοι φανῇ, πέμψον μοι ἄγραφον χάρτην, ἵνα εὕρο[με]ν ἐπιστολ[ὴν] | γράψαι. And if it seems good to you, send me some clean writing paper, so that I can write a letter. (9) BGU 1.38.12 (II-III AD) ἐκομισάμην παρὰ Πτο|λεμίνου μῆλα ξγ καὶ | παρὰ τῆς γυναικὸς Σερή|νου ἄλλα ια, καὶ γινώ|σκειν σε θέλω, ὅτι πάντα | ἠφάνισται καὶ οὐκ εὕ|ρηκα δῶναι οἷς ἔγρα|ψές μοι, ἀλλὰ ἠγόρακα, | εἵνα πέν[ψ]ω αὐτοῖς. I received from Ptoleminus 63 apples and from the wife of Serenus an additional 11, and I want you to know that they have all gone bad and I have not been able to give them to the persons to whom you wrote to me (to give them), but I have bought (some) to send to them.10 (10) PCairMasp 2.III.23 (VI AD) ... ὅπως εὕρωμεν ἡσύχως βιῶναι. … so that we may be able to live quietly.

In addition to the evidence given so far, a new source of examples of this use has been brought to light in BAGD (1979). Here we find three examples 8

9 10

Preisigke, Wörterbuch, s.v. Several of these are found in LSJ, s.v. II.2. [PCairPreis 2.12 (362 AD) in the original article has been omitted here: it is mostly restored. See Postscript for new examples.] [New date from DDBDP; previously ‘II–III AD.’] ἠφάνισται ‘spoiled’: see LSJ, s.v. I.6; Shipp, MGE, 117; Lietzmann, Griechische Papyri, no. 9 (‘sind verdorben’). This fits better than ‘have disappeared/ been lost.’ The writer has received the apples, but there is something wrong with them.

A NON-ARAMAISM IN LUKE 6:7

75

cited from Astrampsychus (ed. Hercher [1863]), with the remark that ‘it is perhaps better to conclude that εὑρίσκω with inf. = be able.’ While this is a welcome addition to the evidence, BAGD simply adds it to the PPar example mentioned above, and adduces none of the other material available. BAGD’s references here can also be updated in the light of the valuable work done by Browne on this text, now known as the Sortes Astrampsychi, and edited in the Teubner series (1983). In the text now published by Browne, which he has convincingly argued was first compiled in III AD in Egypt, there are some dozen examples of the use we are looking for.11 They are all very similar, one in a question, others in responses. A selection will suffice: (11) Sort. Astr. Quaest. 83 εἰ εὑρίσκω πωλῆσαι. Sort. Astr. Resp. 3.8 οὐχ εὑρίσκεις ἄρτι δανείσασθαι. Sort. Astr. Resp. 58.1 οὐχ εὑρήσεις κερδῆσαι ἀπὸ τοῦ πράγματος. Sort. Astr. Resp. 66.3 εὑρίσκεις πωλῆσαι μετὰ κέρδους.

We may round off this collection with three examples from the Patristic literature that happen to have come to light but might be no more than the tip of the iceberg.12 (12) Apophth. Patrum PG 65.253B καὶ οὐχ εὗρον πρὸς ταῦτα ἀποκρίνασθαι αὐτῷ. (13) Apophth. Patrum PG 65.333C τίς ἐκ τῶν δύο εὑρίσκει ζῆσαι; (14) Ephraem Syrus (IV AD) 2.105C ἰδοὺ περισπῶμαι πάντοθεν, καὶ οὐδὲ τὴν μικράν μου σύναξιν εὑρίσκω βαλεῖν. I am distracted on all sides, and am not even able to recite my Small Office.13

The above evidence attests this use of εὑρίσκω from the first to the sixth centuries, in sources ranging from everyday documents to highly conscious works of literature. The conclusion seems warranted that it was a normal, though possibly infrequent, feature of Koine Greek idiom. [Moreover, the dating of many of the examples in I–II AD (including the new ones noted below) makes them contemporary with Luke.] 11

12

13

For the date see Browne, Sortes, p. v, and esp. Browne, ‘Origin and Date.’ BAGD’s note in its list of ‘Writers and Writings in Antiquity,’ xxxiii, also needs revision in the light of Browne’s researches. Remarks here and above about BAGD apply equally to BRAA (1988), which shows no change from BAGD. [Similarly BDAG.] I owe the first two to Sophocles, Greek Lexicon. There is no mention of this use in Lampe, PGL, s.v. εὑρίσκω. Quoted from Lampe, PGL, s.v. σύναξις.

76

COLLECTED ESSAYS ON THE GREEK BIBLE

IV Where did such a use come from? Can anything be said about its antecedents within Greek? In fact the origins of this use are not difficult to detect. To begin with, there is actually a Classical occurrence very like the examples we have been considering: (15) Euripides, Medea 195–8 στυγίους δὲ βροτῶν οὐδεὶς λύπας ηὕρετο μούσῃ καὶ πολυχόρδοις ᾠδαῖς παύειν, ἐξ ὧν θάνατοι δειναί τε τύχαι σφάλλουσι δόμους. But the hateful griefs of mortals, no one has found a way to stop, by music and many-stringed songs, because of which deaths and dread fortunes overthrow houses.14

Semantically this seems indistinguishable from the later use, though it differs in showing the middle and not the active.15 There is no other example exactly like it attested for the Classical period, so it might be regarded as an isolated poetic innovation. Even at that, it shows the potential for such an extension of the meaning of εὑρίσκω in ordinary language. But there are other signs that it was no very striking development. An example of εὑρίσκομαι + infinitive in Herodotus is cited by Page and Smyth in this connexion: (16) Hdt. 9.28.3 μετὰ δὲ τούτους ἵσταντο Κορινθίων πεντακισχίλιοι, παρὰ δὲ σφίσι εὕροντο παρὰ Παυσανίεω ἑστάναι Ποτειδαιητέων τῶν ἐκ Παλλήνης τοὺς παρεόντας τριηκοσίους. Next to them stood 5000 Corinthians, and they had secured from Pausanias that beside them should stand the 300 Potidaeans from Pallene who were present.

Here we have εὑρίσκομαι in the sense of ‘find/get for oneself,’ ‘obtain,’ a frequent use in Classical Greek (LSJ IV), to which is added an infinitive with subject in the accusative. While semantically the end result is somewhat different from the Euripides example (and our one in Luke), syntactically it shows the same development.16 14

15

16

LSJ, s.v. I.4: find out or discover how to, citing this example; Page, Medea, ad loc.: ‘has discovered how to stop’; Smyth, Grammar, §2135: ‘find how to.’ Cf. Humbert, Syntaxe, 105: ‘Il arrive souvent que l’interêt personnel que l’on porte à l’action donne à celle-ci une nuance « déterminée »: en face de εὑρίσκειν « trouver », εὑρίσκεσθαι signifie fréquemment «  arriver à trouver ». Further examples just like the one in Hdt. are found in Dio Cassius: 42.20.3 συγκαθέζεσθαί τε γὰρ ἐπὶ τῶν αὐτῶν βάθρων καὶ ἐς τἆλλα συνεξετάζεσθαί σφισιν … εὕρετο

A NON-ARAMAISM IN LUKE 6:7

77

In much the same way we find Thucydides amplifying εὑρίσκω with a ὅπως clause in: (17) Th. 7.67.2 … πολλοὶ δὲ καὶ ἀκοντισταὶ χερσαῖοι ὡς εἰπεῖν Ἀκαρνᾶνές τε καὶ ἄλλοι ἐπὶ ναῦς ἀναβάντες, οἳ οὐδ᾽ ὅπως καθεζομένους χρὴ τὸ βέλος ἀφεῖναι εὑρήσουσι. … and many javelin-men also, land animals so to speak, Acarnanians and others, going aboard ships, who will not even have the ability to throw their missiles while sitting down.17

Finally it may be observed that the same development was established early in the case of a verb of similar semantic force, namely ἔχω. From Homer onwards, and continuing into later Greek, this verb is occasionally found constructed with an infinitive, giving the sense of ‘have means (to),’ ‘be able (to)’ (see LSJ A.III., BAGD 1.6.a.). A NT example affords the most apt illustration: (18) [John] 8:6 τοῦτο δὲ ἔλεγον πειράζοντες αὐτόν, ἵνα ἔχωσιν κατηγορεῖν αὐτοῦ.18

V We may reasonably conclude, then, that our use of εὑρίσκω is not only amply established for Koine Greek, but also has its roots in the earlier language. There is no reason to regard it as in any way unGreek or as unavailable to a writer such as Luke.

Postscript Electronic searching provides further documentary examples of εὑρίσκω, find a way (to), be able (to), some in newly-published texts. I quote one and simply note three others:

17

18

(‘he obtained for himself that he should sit on the same benches with them and in other respects be reckoned among them’); 47.31.5 ἥ τε Κλεοπάτρα … εὕρετο τὸν υἱὸν … βασιλέα τῆς Αἰγύπτου κληθῆναι. The construction crops up again later: Dio Chrys. 11.108 οὐδὲ ὅπως ἀποκτείνῃ τὸν Ἕκτορα εὑρίσκων. [NA28 text] Interestingly, ευρωσιν appears as a variant to ἔχωσιν. Mention should also be made of εὐπορῶ (‘have plenty,’ etc.), which shows a parallel semantic development: as early as Aristotle (Top. 102a.13: LSJ I.2) it is found constructed with an infin. to give the meaning ‘have the means (to),’ and ends up as the usual Mod. Greek word for ‘be able,’ μπορῶ.

78

COLLECTED ESSAYS ON THE GREEK BIBLE

SB 26.16608.34 (c.100 AD) ἐρεῖς | δὲ Ζοιλᾶτι ὅτι | ὡς ἐδήλωσά σοι | οὐχ εὗρ[ο]ν δοῦναι | Ἑρμοδώρῳ. You are to tell Zoilas that, as I informed you, I didn’t find a way to give/ couldn’t give (it) to Hermodoros. PMich 8.477.35 (II AD); 488.15 (II AD); SB 3.6264.14 (II AD?).

7 THE UNITED BIBLE SOCIETIES’ LEXICON AND ITS ANALYSIS OF MEANINGS 1992

Abstract The paper focuses on one aspect of the Louw-Nida lexicon of the NT arranged by semantic domains, namely, its approach to the lexical analysis of individual NT words. The general characteristics of Louw and Nida’s method are first described and illustrated, then a number of problem areas are identified. These mainly concern questions of ‘lumping’ and ‘splitting,’ the lack of coverage of some meanings, problems with definitions, and the exclusion of syntactical elements. The potential of this lexicon to compete with the Bauer/ BDAG lexicon is considered as well, despite its very different character as a domain lexicon.

1. INTRODUCTION The United Bible Societies’ lexicon by Louw and Nida forms a major new contribution to the field of NT lexicography.1 Not only does it adopt a new approach to the organisation of the vocabulary, by arranging it according to areas of meaning or semantic domains; it also offers a fresh lexical analysis of each word in the NT. While taking existing dictionaries as their starting-point,2 the editors have analysed afresh the meaning or meanings of every word, re-thought the criteria of classification, and formulated new definitions for all meanings, rejecting the long-established but unsatisfactory method of defining by glosses. Although the editors have not presented their material in the conventional alphabetical form, the new lexical analysis that the dictionary offers deserves to be taken very seriously, since it forms an alternative to that 1

2

Joannes P. Louw and Eugene A. Nida, Greek-English Lexicon of the New Testament Based on Semantic Domains, Vol. 1 Introduction and Domains, Vol. 2 Indices (United Bible Societies: New York, 1988) [= LN]. Parts of this paper were presented in a seminar in the Division of Religious Studies at La Trobe University in 1990, and in a panel at the Annual Meeting of SBL in New Orleans in November 1990. See LN, 1:iv.

80

COLLECTED ESSAYS ON THE GREEK BIBLE

found in existing NT lexicons. Given the lexicographical expertise and experience of its authors, and their determination to analyse the data anew on the best principles, there is good reason to expect their treatment to be an advance on all other dictionaries, including the current leader in the field, Bauer/BAGD. That the authors intend it to be seen in this light is clear from one of their introductory remarks: ‘The principal reason for a new type of Greek NT lexicon is the inadequacy of most existing dictionaries, which for the most part are limited in indicating meanings, since they depend principally upon a series of glosses’ (1:viii). The present paper focuses on this aspect of the Louw-Nida dictionary, that is, its handling of the lexical analysis of each NT word. It is intended as a contribution towards evaluating this part of the work fully and deciding its potential as a rival to Bauer/BAGD. To be a serious competitor it would presumably need to be recast in the familiar alphabetical format, with the definitions and examples belonging to a word gathered under that word, instead of being distributed through the various semantic domains. Whether an edition in this form might be published is of course a matter for Professors Louw and Nida to decide, but it is an option that could be considered at a later stage when assessment of the work has progressed further. Most of what I have to say is directly concerned with, or arises out of, the question of ‘the fineness of the semantic grid.’ I begin, then, with some preliminary remarks on this topic. Having collected a corpus of occurrences and assigned a provisional meaning in each place, the lexicographer has the task of deciding how many senses (or ‘meanings’) to recognise, where to place the divisions, and how to define the senses recognised. In doing so, it is necessary to steer a course between setting up too many meanings and setting up too few. At the one extreme, it is possible to frame a separate definition for each occurrence; at the other, one can adopt a single meaning broad enough to embrace all the examples. How best to make the divisions cannot be reduced to rules or predicted in any way, but will be found to vary from word to word. It is also true that each lexicographer will make somewhat different decisions, and there will always be room for disagreement. All this of course is well known to Louw and Nida, and they have offered valuable comment on the process in their own writings.3 For example: 3

Besides the Introduction in the Lexicon itself, I have drawn on Nida and Louw, Lexical Semantics, which the authors have kindly shown me in a 1990 draft and given me permission to quote from in advance of publication. [The work was published in 1992.]

81

THE UNITED BIBLE SOCIETIES’ LEXICON

How many different meanings there are for a particular term depends on the fineness of the semantic grid, that is to say, the extent to which one tends to lump or to split differences. The ultimate criteria for such lumping or splitting depend upon the entire semantic system.4

Again: [O]ne must recognize the fundamental problem in all scientific descriptions, namely, the opposing tendencies in lumping and splitting. Those who favor lumping often insist that this simplifies the description and highlights the relations, but those who favor splitting insist that lumping only moves the difficulties to another level of abstraction or analysis. There is generally no final answer to the issue of lumping versus splitting. … The second principle of organizing related meanings is to adopt no more meanings than are completely necessary to account for the evidence. This is, of course, a fundamental principle of all scientific work, since the fewer the necessary categories or principles needed to explicate the data the more likely is the analysis to be correct.5

These remarks, and the lexicographical principles embodied in them, are undoubedly sound, and there can be no question that in general Louw and Nida have carried out their task well, and arrived at satisfactory decisions about the number of meanings to recognise for each word and how to define them. Nevertheless I find that there are problems. The most serious of these, and one which seems to go beyond mere difference of personal opinion, is that there are cases in which a NT use is not covered by the definitions given for the word in question. This can mostly be traced to excessive ‘lumping’: the grid is simply not fine enough to catch all the examples. There are times, however, when it seems to be due to oversight. Besides this, one can identify a number of other problems, concerning definition and the like. In what follows, these points will be set out under headings, with representative examples treated in detail. Before proceeding to this critique, however, it will be useful to describe the general features of LN’s approach to analysis of meanings. 2. GENERAL CHARACTERISTICS OF LOUW AND NIDA’S

ANALYSIS

As already mentioned, LN uses definition as the primary means of indicating meaning, an undoubtedly better method than that found in most 4 5

LN, 1:xvi. Lexical Semantics, 75 and 42; cf. also 61. These are taken from Chapter 3, a full discussion of the process of analysing different meanings of the same lexeme.

82

COLLECTED ESSAYS ON THE GREEK BIBLE

NT dictionaries.6 Glosses are not rejected altogether, however, but are added to each definition to indicate possible ways of translating. The two are distinguished by typographical means. Then the sense in question is illustrated by at least one NT example, quoted in context, with translation into English. A typical entry looks like this: 57.212 τόκος, ου m: the interest on money that has been loaned – ‘interest.’ κἀγὼ ἐλθὼν σὺν τόκῳ ἂν αὐτὸ ἔπραξα ‘then I would have received it back with interest when I returned’ Lk 19.23.

In framing definitions LN have used common sense in deciding the appropriate amount of definition required.7 It would be possible, for example, to define ἀστήρ in full scientific terms; LN sensibly opt for the equivalent English word as the definition, thus (1.30): a star or a planet – ‘star, planet.’ In the same way they frequently leave the definition out altogether and allow a gloss to stand in its place. For example: 1.28 ἥλιος, ου m – ‘the sun.’ 3.5 συκῆ, ῆς f – ‘fig tree.’ All this is unobjectionable, since the English terms are sufficiently meaningful to the English speaker and there is no ambiguity. On the other hand, some meanings are given the full treatment, as for example: 19.53 πνίγωa: to apply pressure around the neck in order to harm or kill – ‘to choke.’ In between are perhaps the majority of LN’s definitions, such as 15.227 περιπατέωa: to walk along or around – ‘to walk, to go.’, in which it is rightly assumed that a term such as walk is adequate in place of a more elaborate definition of the activity.8 It does not take long to discover that LN’s definitions are generally an improvement on the indications of meaning given in BAGD. One example will serve for the very numerous instances in which this is true. In BAGD the word δεῖπνον is divided into two senses, described thus: 1. dinner, supper, the main meal (toward) evening 2. (formal) dinner, banquet

It is not easy to grasp from these poor descriptions what the two meanings are and how they differ. LN’s analysis is not fundamentally different, but the definitions make clear what is unclear in BAGD: 23.25 δεῖπνονb, ου n: the principal meal of the day, usually in the evening – ‘supper, main meal.’ 23.22 ἄριστονa, ου n; δεῖπνονa, ου n; βρῶσιςb, εως f: a meal whether simple or elaborate (this generic meaning of ἄριστον and δεῖπνον is 6 7 8

Cf. LN, 1:vii. Cf. LN, 1:xiii. The small superscript letters attached to words mark distinct meanings; the full list of meanings for each word is given in the Greek-English Index in Vol. 2.

THE UNITED BIBLE SOCIETIES’ LEXICON

83

in contrast with the more specific meanings of ἄριστονb, 23.23, and δεῖπνονb, 23.25) – ‘a meal, a banquet, a feast.’

Another feature of LN’s approach to definition is that different parts of speech, especially verb and noun, are sometimes covered by the one definition. So for example: 25.59 ἐλπίζωa; ἐλπίςa, ίδος f: to look forward with confidence to that which is good and beneficial – ‘to hope, to hope for, hope.’

The definition given here (and in most instances, it seems) is in fact that of the verb; the user must extract the noun’s definition from it (though a noun gloss is provided). This method is considered by the authors to be one of the advantages of the semantic domain approach (see 1:x), though in practice it is not resorted to very often. One wonders whether it really is an advantage, since it means that some parts of speech are not given a definition in their own right. There is some awkwardness for example in a case like the following, where it is not really clear what the definition of εὐλογία would be: 33.470 εὐλογέωb; εὐλογίαc, ας f; κατευλογέω: to ask God to bestow divine favor on, with the implication that the verbal act itself constitutes a significant benefit – ‘to bless, blessing.’

To arrive at the meaning of the noun, it is not enough simply to turn the definition into noun form: this gives ‘asking God to bestow divine favor on. …’ In effect one needs to know the meaning already in order to get the right result. Nevertheless one can accept the principle involved, that ‘certain verbs and corresponding nouns have essentially the same meaning. … Whether a writer employs a verb or a noun depends largely upon the syntactic structure and the stylistic features of a passage’ (1:xii). LN’s treatment of the middle and passive of verbs is also rather different from what we are accustomed to see in our dictionaries. The general pattern is this: only the active of a verb is defined, and no mention is made of the middle or passive unless it is distinct semantically, when it is given its own definition and listed quite separately in the Greek–English Index. So for example ἄρχομαι ‘begin’ and ἄρχω ‘rule’ are treated as separate words, and their meanings fully defined. αἰτέω on the other hand is listed in the active with one definition, and no mention is made of the middle; likewise ἀποστέλλω is defined only in the active, and it is up to the user to know whether the passive occurs and to apply the definition in the appropriate way. This, whether or not we agree with it, is in accordance with LN’s general policy of excluding syntactical information (on which

84

COLLECTED ESSAYS ON THE GREEK BIBLE

more will be said in a moment). Even so, some inconsistencies are to be found. αὐξάνω ‘increase’ and αὐξάνομαι ‘increase (degree)’ get separate entries, but one definition has to serve for both λούω and λούομαι (47.12 λούω: to wash the body). The passive of ἀποστέλλω, as we have seen, is not mentioned, yet λυπέω ‘make sad’ and λυπέομαι ‘be sad’ are treated as semantically distinct. An important feature of LN’s analysis is that the criteria of classification are strictly semantic. This has far-reaching consequences. It means that a large amount of the material that we are used to finding in BAGD is simply excluded. In BAGD we are given a classification not only by lexical meaning but also by various other criteria, such as reference, theological content, collocation, and above all syntax – and the mixture varies from one word to another. None of this finds a place in LN unless it affects the semantic content. There is a useful discussion of this point, with illustrations, in LN’s introduction,9 but let us take our own example, ἐσθίω. In BAGD two senses of ἐσθίω are recognised, 1. eat, 2. consume, devour, but the first is broken down into five sections, with further subdivisions within these. Sections 1. a, b, c, and d have to do with the constructions found with this verb – whether it is used with a direct object, a preposition, or absolutely. Then e takes a different tack and gathers the examples of ἐσθίω in combination with πίνω, subdividing these according to the kind of eating referred to. Besides the numbered subdivisions, we find along the way further breakdown of the examples, depending on the kind of food being eaten and so on, and we are told the various expressions or phrases that crop up in the NT, such as τὸ ϕαγεῖν, διδόναι τινὶ ϕαγεῖν. All this is familiar to users of BAGD, and we have come to take it for granted. We certainly feel that it is useful information to be given, and we may also admire Bauer’s ingenuity in getting it all in. Nevertheless this type of classification, incorporating so much that has nothing to do with lexical meaning, is, as LN say, ‘not only unsystematic but misleading.’10 LN’s treatment of this word, by contrast, has a very lean appearance. The same two meanings are recognised, though of course better defined (see 23.1 and 20.44), but there is no syntactical information given, nor are any phrases or collocations noticed except for one ‘unit’ or idiom, τὸν ἑαυτοῦ ἄρτον ἐσθίω, which has its own entry and definition (57.190). 9

10

LN, 1:viii–ix, xvi–xvii. See also Louw’s discussion of πνεῦμα in Lexicography and Translation, 70–7, and Silva, Biblical Words, 172–4. Silva brings out well the inconsistency of the criteria used in BAGD. LN, 1:viii.

THE UNITED BIBLE SOCIETIES’ LEXICON

85

The absence of syntactical data is perhaps the most striking aspect here, and the one which may cause us most doubts. It is easy to recognise that omission of this material is correct according to the principles LN have set themselves; we may feel that we can live with not knowing the constructions used with ἐσθίω and a multiplicity of other verbs so carefully set out in BAGD; but we feel uneasy that ἵνα and ὅπως appear without any mention of the mood of the accompanying verb, and quite surprised that even prepositions are treated as much as possible without reference to case. More will be said on this below, where we shall consider instances in which the strain of separating syntax and semantics seems to be too great. As far as phrases or collocations are concerned, these have no place in LN unless they form a semantically distinct unit. If so, they are listed as ‘units’ under their head-words in the Greek–English Index, and then defined separately in the appropriate domain. Nearly all LN’s ‘units,’ as far as I can tell, are idioms, and are described as such in the place where they are defined.11 Thus, to continue with our example ἐσθίω, the expression διδόναι τινὶ ϕαγεῖν gets no special mention because the meaning of the whole expression is no different from the meaning of its parts; τὸν ἑαυτοῦ ἄρτον ἐσθίω, on the other hand, has a meaning all its own, and so appears in a domain far away from that of ἐσθίω, with its own definition, as follows: 57.190 τὸν ἑαυτοῦ ἄρτον ἐσθίω: (an idiom, literally ‘to eat one’s own bread’) to earn a living by one’s own efforts – ‘to earn a living, to earn a livelihood.’12

It can be seen, then, that LN’s analysis of any word will generally be characterised by fewer divisions than BAGD’s, if only because many of the criteria used by BAGD are simply not relevant: the only criterion is the semantic one. But within that, what can be said of LN’s general approach to making divisions? Do they like many fine distinctions, or do they incline towards a small number of broad definitions? It is of course impossible to assess this with precision, but the impression gained from a fairly wide range of examples is that LN prefer to lump rather than split. Let us take ἄλλος as an illustration. BAGD manage to analyse this into about a dozen divisions. Of course not all of these are really separate 11

12

The exceptions are mainly ‘titles’ (e.g., 12.15 υἱὸς τοῦ θεοῦ). I have noticed one ‘fixed phrase’ (12.38 πνεῦμα πονηρόν). It is not clear why πνεῦμα ἀκάθαρτον achieves the status of a unit: at 12.39 it is not given a description but simply defined. On the treatment of idioms see LN, 1:ix; Nida and Louw, Lexical Semantics, 106–7.

86

COLLECTED ESSAYS ON THE GREEK BIBLE

meanings, since as usual BAGD are dividing by other criteria as well, but there still appear to be about 5 or 6 different meanings – it is hard to count them exactly – including other, one … another, different, more, the other (of two). LN cover all this with two definitions of a quite broad character (also embracing ἕτερος): 58.36 ἄλλοςa, η, ο; ἄλλως; ἕτεροςa, α, ον; ἑτέρως: pertaining to that which is different in kind or class from all other entities – ‘different, differently, other than.’ 58.37 ἄλλοςb, η, ο; ἕτεροςb, α, ον: pertaining to that which is other than some other item implied or identified in a context – ‘other, another.’

Another example, briefly: παρακαλέω. This is a difficult word to handle, because the senses merge imperceptibly into one another, and it is hard to know where to draw the lines. In BAGD we find a division into five senses, with the first further subdivided into three (or is it two?), making seven or eight different meanings. LN settle for only four. As I have said, this appears to be the general trend of LN’s analysis. At the same time, instances arise in which LN split more than BAGD, usually with good reason. So for example ἥκω: only one meaning is given in BAGD, have come, be present, whereas in LN there are three, glossed as: a arrive, b be here, c happen. It should be noted that the appearance of BAGD’s entry can be deceptive in a case like this. There we find ἥκω divided into several headings, as if there is more than one sense, but inspection shows that all this classification is based on other criteria, and there is in fact only one meaning recognised (given at the beginning).13 The corpus of material analysed by LN is the NT as a whole, but no attempt is made to list occurrences exhaustively. Under each definition we are given selected examples, usually one for each word, sometimes two, rarely more. These have the virtue of being full quotations with translation into English. On the other hand, there is no listing of examples under each meaning, of the kind found in BAGD. This means that users are not told where the editors would have placed any occurrence that is not quoted, and they must make their own deductions. There is real uncertainty about this at times, as will be seen further below. It also means that there is no information about frequency or distribution. 13

For an even more deceptive case see γαμέω, in which BAGD manage to give the impression that they are dividing into some five different senses, whereas even their own definitions show that there is only one sense; the divisions are made on the basis of reference and grammar. They are thus fundamentally in agreement with LN, who naturally recognise but one sense.

THE UNITED BIBLE SOCIETIES’ LEXICON

87

Besides the restricted presentation of NT data, there is limitation of another kind: parallels outside the NT and historical data of any kind simply do not come within LN’s scope.14 On another front, LN go beyond what is usually provided in our NT lexicons. In addition to the definitions and illustrative examples, many sections have short notes which are ‘designed primarily to help translators deal with some of the frequently recurring difficulties’ (1:xiii). The lexicon thus complements the United Bible Societies’ well-known series of translators’ handbooks on the NT. In concluding this section, it is worth reiterating that here and throughout the paper I have chosen to focus on LN’s approach to the lexical analysis of individual words. The classification into semantic domains and subdomains is however an equally important part of the lexicon, and equally deserving of serious assessment. We should also be aware that the two are not really independent, but interconnect with each other, as LN themselves well explain (esp. 1:xvi). Let us turn now to the areas in which I see problems. As already mentioned, representative examples will be treated. These are not in any way exhaustive, or the outcome of a systematic search; they serve only to show that the problems exist.

3. PROBLEMS WITH LUMPING In LN’s analysis of λέγω (including its suppletives) a division is made into five meanings (plus two idioms) defined as follows: 33.69 λέγωa; ϕημίa: to speak or talk, with apparent focus upon the content of what is said – ‘to say, to talk, to tell, to speak.’ 33.129 καλέωa; λέγωb; ἐπιλέγω: to speak of a person or object by means of a proper name – ‘to call, to name.’ 33.131 καλέωb; λέγωc; ϕωνέωc; ἐπικαλέω: to use an attribution in speaking of a person – ‘to call, to name.’ 33.138 λέγωd: to mark the correspondence in the meaning of foreign expressions – ‘to mean.’ 33.140 ϕημίb; λέγωe: to say something in order to explain more fully the implications or intent of what has been said – ‘to mean, to imply.’ 14

There are occasional exceptions. At 67.183 it is mentioned that ἐπιούσιος ‘may very well have been coined by the Gospel writers.’

88

COLLECTED ESSAYS ON THE GREEK BIBLE

It does not take long to realise that the first of these senses has to cover a great deal. BAGD’s analysis, with its profusion of categories and senses, is not easy to use for comparison, but it still alerts us to various areas of meaning that LN evidently subsume under meaning a. We can probably agree that BAGD’s separate senses ask, answer, assure, maintain, s.v. II.1.a, etc.), are not really different from ‘say, speak’ and therefore do not need to be split off. But what of order? This seems to me clearly a different sense which stretches the definition in 33.69 to the limit, and certainly does not fit under any of the other four. As illustrations of this sense let us take: Matt 5:34 ἐγὼ δὲ λέγω ὑμῖν μὴ ὀμόσαι ὅλως. But I tell you not to swear at all. Matt 4:3 Εἰ υἱὸς εἶ τοῦ θεοῦ, εἰπὲ ἵνα οἱ λίθοι οὗτοι ἄρτοι γένωνται. If you are the son of God, order these stones to become loaves of bread.15

The meaning here seems to go beyond merely speaking or saying something: there is an additional notion of directing that something must be done, the semantic development being similar to that in English tell. LN’s own definition in 33.323, in the Subdomain 33F’ Command, Order, would have covered it very nicely: to state with force and/or authority what others must do. But this use of λέγω has not been recognised, and does not appear alongside κελεύω and the rest, where we might have expected to find it. This – to my mind excessive – lumping under meaning a contrasts with the careful divisions made in the case of the other four meanings. One might have thought that if λέγωd ‘mean,’ for example, deserved to be split off into a meaning of its own then this needed to be done for ‘order’ too. Consider LN’s example quoted in 33.138: Acts 9:36 μαθήτρια ὀνόματι Ταβιθά, ἣ διερμηνευομένη λέγεται Δορκάς. I think it is right to recognise a separate sense ‘mean’ here, but if ‘order’ is caught by meaning a, then this use could be too (‘a disciple named Tabitha, which interpreted says Dorcas’).16 A further difficulty is worth noticing before we leave this word. LN’s meaning a must also serve to cover a use such as Luke 9:31 οἳ ὀϕθέντες ἐν δόξῃ ἔλεγον τὴν ἔξοδον αὐτοῦ ‘who appeared in glory and spoke of 15

16

For further examples see BAGD, s.vv. λέγω II.1.c, and εἶπον 3.c, though not all those listed are sound. BAGD absurdly lump Acts 9:36 and similar exx. under II.3 call, name, at the same time as giving ‘mean’ as the translation.

THE UNITED BIBLE SOCIETIES’ LEXICON

89

his departure.’ Presumably this has not been separately recognised because of LN’s general policy of excluding syntax from consideration. Nevertheless one has doubts that the coverage is adequate. There does seem to be a semantic difference between λέγω τὴν ἔξοδον and, e.g., λέγω τὴν παραβολήν, or λέγω τὴν ἀλήθειαν. Another example is παρακαλέω. As was mentioned above, LN keep the divisions down to four, and this seems a fairly sensible analysis in general. Its weakness is that once again one meaning has to cover a great deal of ground. LN’s meanings are: 33.168 παρακαλέωa; παράκλησιςb, εως f : to ask for something earnestly and with propriety – ‘to ask for (earnestly), to request, to plead for, to appeal to, earnest request, appeal.’ 33.315 καλέωe; ϕωνέωd; παρακαλέωb: to ask a person to accept offered hospitality – ‘to invite.’ 33.310 παρακαλέωc: to call to come to where the speaker is – ‘to call together to.’ 25.150 παρακαλέωd; παράκλησιςa, εως f : to cause someone to be encouraged or consoled, either by verbal or non-verbal means – ‘to encourage, to console, encouragement.’

Meanings b and c deal with two fairly narrow senses, d takes care of ‘comfort, encourage,’ and meaning a is left to cover the rest of the field. This works well enough for most examples, but there is definitely some strain at the edges. What do we do, for example, with these two?: 1 Tim 6:2 ταῦτα δίδασκε καὶ παρακάλει. Titus 2:15 ταῦτα λάλει καὶ παρακάλει καὶ ἔλεγχε μετὰ πάσης ἐπιταγῆς.

BAGD (s.v. 2) gloss the meaning as impress upon someone, urge, exhort, which seems to be on the right track. In both places ταῦτα refers back to teaching and instructions set out in the preceding verses. The idea of recommending a course of action is what is involved, rather than simply asking for something. Yet in LN’s analysis, meaning a, ‘to ask for something earnestly and with propriety,’ is the only one of LN’s meanings that comes anywhere near, and is presumably the one intended to cover them. We also wonder if LN were right to discard ‘call upon for help’ as a separate sense, in favour of lumping into meaning a. BAGD (s.v. 1.c) find two examples of this use. One (2 Cor 12:8) cannot be relied on, but the other looks as if it ought to be taken seriously, especially in view of the long history of this sense in Greek (see BAGD and cf. LSJ, s.v. II.1):

90

COLLECTED ESSAYS ON THE GREEK BIBLE

Matt 26:53 ἢ δοκεῖς ὅτι οὐ δύναμαι παρακαλέσαι τὸν πατέρα μου, καὶ παραστήσει μοι ἄρτι πλείω δώδεκα λεγιῶνας ἀγγέλων· Cf., e.g., RSV: ‘Do you think that I cannot appeal to my Father, and he will at once send me more than twelve legions of angels?’

Again, meaning a is the only one available, and must be what LN intend here.

4. THE PLACING OF A

USE IS UNCERTAIN

In the examples looked at so far, though we may not agree with their decisions about lumping, LN still appear to be in control of the NT data. That is, their definitions can, perhaps with some stretching, be seen to cover all the occurrences, and the placing intended is not in serious doubt. We move on now to instances in which it is not clear where a use is to be placed, and we wonder whether LN have in fact addressed the question. In Luke 24:38 we find ἀναβαίνω used as follows: Τί τεταραγμένοι ἐστὲ καὶ διὰ τί διαλογισμοὶ ἀναβαίνουσιν ἐν τῇ καρδίᾳ ὑμῶν·

BAGD (s.v. 2) describe it as ‘fig(urative)’ and gloss as arise. The rendering of the RSV, for example, is: ‘Why are you troubled, and why do questionings rise in your hearts?’ LN analyse ἀναβαίνω into four meanings and one idiom, as follows: 15.101 ἀναβαίνωa; ἀνέρχομαι: to move up – ‘to come up, to go up, to ascend.’ The upward movement may be of almost any gradient, for example, in going up a road to Jerusalem … or in going up into a tree … or in ascending into heaven. … 15.99 ἀναβαίνωb: to move up onto an object, with specialization of meaning in reference to boats – ‘to go aboard, to embark.’ 23.196 ἀναβαίνωc: to grow, as of plants, from the time of sprouting to mature size – ‘to sprout and grow.’ 23.189 ἀναβαίνωd: to grow taller (restricted in the NT to the growth of plants) – ‘to grow up.’ 30.17 ἀναβαίνω ἐπὶ καρδίαν: (an idiom, literally, ‘to arise in the heart’) to begin to think about something – ‘to begin to think, to think, to have a thought occur to someone.’ ἐπὶ καρδίαν ἀνθρώπου οὐκ ἀνέβη ‘what no one ever thought could happen’ 1 Cor 2.9.

Where is our example intended to find its home? Meanings b and d are definitely out; c tempts us with ‘as of plants,’ as if our use might be

THE UNITED BIBLE SOCIETIES’ LEXICON

91

a metaphorical one based on the growth of plants, but this is not very appealing. Meaning a looks as if it is intended to describe physical movement of persons in space. That leaves the idiom in 30.17. Our example could likewise be regarded as an idiom, and the meaning given there is suitable; the difficulty is that the wording of the two expressions is not exactly the same. In the end we are left in some doubt about what LN intended to do with Luke 24:38. A second instance may be mentioned without going into full detail. In one of its uses ζητέω goes beyond simply ‘seek’ and has the idea of enquiry by discussion (BAGD s.v. 1.c, and cf. parallel development in ζήτησις). John 16:19 seems a pretty clear example of this use, which we might gloss as ‘discuss, debate’: καὶ εἶπεν αὐτοῖς, Περὶ τούτου ζητεῖτε μετ᾽ ἀλλήλων ὅτι εἶπον, Μικρὸν καὶ οὐ θεωρεῖτέ με, καὶ πάλιν μικρὸν καὶ ὄψεσθέ με·

In LN’s analysis there are 7 meanings and one idiom. None of these, as far as I can see, offers the meaning mentioned. We are thus uncertain whether LN have simply missed this use, or whether they decline to recognise it and intend John 16:19 to come under a definition such as 27.34: to attempt to learn something by careful investigation or searching – ‘to try to learn, to search, to try to find out, to seek information.’

5. A USE IS

NOT COVERED17

We come now to the area of more serious weakness in LN’s analysis. It can be shown, I believe, that there are times when LN quite fail to provide for a NT use. Whatever the reason for it, the spread of meanings recognised for some words simply does not cover all the data. Let us begin with θέλω. LN recognise four meanings and one idiom, as follows: 30.58 θέλωa: to purpose, generally based upon a preference and desire – ‘to purpose.’ … Col. 1.27. 31.4 θέλωb: to have a particular view or opinion about something – ‘to be of an opinion, to think something is so.’ … 2 Pe 3.5 … θέλω in 2 Pe 3.5 is the only NT occurrence with this meaning. … 25.1 θέλωc; θέλησις, εως f: to desire to have or experience something – ‘to desire, to want, to wish.’ … Mt 20.21 … Ga 4.9 … 1 Cor 12.18. 17

[For ‘use’ it would have been more accurate to say ‘lexical meaning.’ Cf. Lee, History, 173 n. 10.]

92

COLLECTED ESSAYS ON THE GREEK BIBLE

25.102 θέλωd: to take pleasure in something in view of its being desirable – ‘to like, to enjoy.’ … Mk 12.38. 33.136 θέλω εἶναι: (an idiom, literally ‘it wishes to be’) something which is to mean or to be understood in a particular manner – ‘it means, this is to be understood as.’ … Ac 2.12.

It is well known that from early on θέλω includes in its range of meaning a sense usually glossed as ‘be willing.’ This is distinct from ‘wish,’ ‘desire,’ which is also old, though not usual in Attic. In Koine Greek the word is used freely in both senses. In the NT a clear example of ‘be willing’ is: Matt 26:14 Τότε πορευθεὶς εἷς τῶν δώδεκα, ὁ λεγόμενος ᾽Ιούδας ᾽Ισκαριώτης, πρὸς τοὺς ἀρχιερεῖς (15) εἶπεν, Τί θέλετέ μοι δοῦναι, κἀγὼ ὑμῖν παραδώσω αὐτόν·

It is not satisfactory to see θέλω here as expressing purpose or desire; it has merely the idea of consent or acquiescence: ‘what are you willing to give me if I betray him to you?’18 As can be seen, none of LN’s definitions comes near this. Nor is θέλω mentioned in Subdomain 25E Be Willing, where ἑκούσιος and the like are gathered (25.65–67). It appears, then, that LN have simply not noticed that this sense occurs in the NT. There is another use of θέλω that has failed to achieve recognition in LN’s analysis. Of several probable NT examples, a good one is: Mark 6:48 καὶ ἰδὼν αὐτοὺς βασανιζομένους ἐν τῷ ἐλαύνειν, ἦν γὰρ ὁ ἄνεμος ἐναντίος αὐτοῖς, περὶ τετάρτην ϕυλακὴν τῆς νυκτὸς ἔρχεται πρὸς αὐτοὺς περιπατῶν ἐπὶ τῆς θαλάσσης· καὶ ἤθελεν παρελθεῖν αὐτούς.

Although this use is poorly tabulated by BAGD, and more work needs to be done on collecting the examples and parallels, it has long been clear that the sense that fits the context best is something like ‘be about to.’ To quote the note on this passage from another work co-authored by Nida: kai ethelen parelthein autous ‘and he was going to pass them by’: it is generally agreed that the verb thelo here does not mean ‘will’ or ‘wish,’ but is used in a weakened sense, being the equivalent of an auxiliary (cf. Turner … Taylor … Manson).19 18

19

BAGD treat this and similar examples under sense 2., which begins with wish, will, but goes on to embrace be ready, be prepared, be willing, and more. Other NT exx.: Matt 11:14; John 5:40; 2 Thess 3:10. [In Essay 16 I take this to be the auxiliary use of θέλω: ‘what will you give me?’] Bratcher and Nida, Translator’s Handbook, 213.

THE UNITED BIBLE SOCIETIES’ LEXICON

93

Nevertheless LN do not give this sense a separate existence. In their present analysis θέλωa is the only one available that comes near fitting the context, but this is clearly intended to be different from ‘be about to.’ Next let us take εἰ, in the treatment of which there is a glaring omission. LN give three meanings: 89.65 εἰa: a marker of a condition, real or hypothetical, actual or contrary to fact – ‘if.’ 89.30 εἰb: a marker of cause or reason on the basis that an actual case is regarded formally as a supposition – ‘since, because.’ 90.26 εἰc: a marker of an indirect question as content – ‘whether, if, that.’

In addition there are three units, εἰ τύχοι, εἰ μή, and εἰ μήν. We see that the use of εἰ in indirect questions is adequately dealt with, but where is εἰ in direct questions? It is clearly not covered by any of the three meanings, and its absence from 69.11–16, where other words introducing questions are collected, confirms that LN have failed to take account of it. It is well known that this use occurs in the NT, but lest there be any doubt, let us see an example: Acts 19:2 εἶπέν τε πρὸς αὐτούς, Εἰ πνεῦμα ἅγιον ἐλάβετε πιστεύσαντες·20

Another, but less well-known use of εἰ has also slipped through LN’s net. In the NT there are two occurrences of the Hebraistic use of εἰ in oaths, in which a negative is implied and εἰ amounts to ‘certainly not.’ So in Mark 8:12: καὶ ἀναστενάξας τῷ πνεύματι αὐτοῦ λέγει, Τί ἡ γενεὰ αὕτη ζητεῖ σημεῖον; ἀμὴν λέγω ὑμῖν, εἰ δοθήσεται τῇ γενεᾷ ταύτῃ σημεῖον.21 … Truly, I say to you, no sign shall be given to this generation. (RSV)

It would hardly be satisfactory to counter that, because this use derives (in Hebrew!) from an ordinary ‘if’ clause with aposiopesis of the main clause, it is adequately covered by εἰa, ‘if.’ No one would imagine, without being told, that a negative is to be supplied, and therefore it deserves to be described as a separate use.

20

21

Other examples: Matt 12:10; 19:3; Luke 13:23; 22:49; Acts 1:6; 7:1; 21:37; 22:25. Cf. BAGD, s.v. εἰ V.1. The other is in a repeated quotation from the LXX at Heb 3:11; 4:3, 5.

94

COLLECTED ESSAYS ON THE GREEK BIBLE

One further example. One of the senses of βλέπω, from Classical times onwards, is ‘look,’ i.e., ‘cast the gaze in a specified direction.’ We would expect this to occur in the NT at some time or other, and so it does. John 13:22 seems a convincing enough instance: ἔβλεπον εἰς ἀλλήλους οἱ μαθηταὶ ἀπορούμενοι περὶ τίνος λέγει. The disciples looked at one another, uncertain of whom he spoke. (RSV)22

LN’s coverage of βλέπω appears thorough, with 7 meanings and two units. One of these seems at first sight to provide the required sense, but inspection shows that it is dealing with something different: 82.10 βλέπωg: (a figurative extension of meaning of βλέπωa ‘to see,’ 24.7) to be oriented in a particular direction – ‘facing.’ λιμένα τῆς Κρήτης βλέποντα κατὰ λίβα καὶ κατὰ χῶρον ‘a harbor in Crete facing southwest and northwest’ Ac 27.12.

This, as the example helps to make clear, describes only orientation, and lacks the idea of using the eyes to look in a certain direction. So it appears that LN have simply failed to observe one of the standard uses of this common verb. This is all the more surprising in view of their good handling of ἐμβλέπωa, defined at 24.9 as: to direct one’s vision and attention to a particular object – ‘to look straight at, to look directly at,’ a definition which could also have embraced the missing use of βλέπω.23 There are further examples that can be added to those noticed so far in this section. A detailed treatment would be tedious, however, and I simply indicate the missing uses briefly, without reproducing LN’s analysis in full. (1) εὑρίσκω + infinitive, with the sense of ‘find a way (to), be able (to),’ found in Luke 6:7. LN’s 27.1 is the nearest, but hardly good enough. This use deserves a place in Domain 74 Able, Capable, alongside the similar use of ἔχω (74.12). (2) ζῶ. The phrase ὕδωρ ζῶν (John 4:10 etc.) has a double meaning, ‘lifegiving water’/‘fresh water.’ LN deal with the former (23.88) but fail to provide for the latter. (3) ἡγέομαι. In 1 Thess 5:13 this verb has an extra dimension not catered for by LN’s 31.1. The sense required is ‘hold a high opinion of’ (cf. BAGD’s esteem, respect, s.v. 2). 22 23

Others in BAGD, s.v. βλέπω 3. LN’s bracketed note in 82.10 above also deserves comment. It is surely clear that βλέπωg ‘face’ is not a ‘figurative extension’ of βλέπω ‘see,’ but of the unrecognised βλέπω ‘look.’

THE UNITED BIBLE SOCIETIES’ LEXICON

95

(4) καταβαίνω. LN’s one definition (15.107 to move down, irrespective of the gradient) is inapplicable to Acts 8:26, where the subject is a road. This use would have gone nicely with κατάβασις in 15.109: the slope or declivity of an object which one may descend. (5) μή and μήτι in questions. The subject deserves fuller study, but there is enough to show that these two are sometimes used to introduce neutral questions, not only those expecting a negative response. See, e.g., John 4:33; 7:35; 18:17, 25; 4:29 (μήτι).24 LN give no recognition to this at all: μή and μήτι appear in Subdomain 69E Markers for a Negative Response to Questions, and there is no Subdomain for neutral questions (not even those introduced by ἆρα). (6) ὅστις appears only in 92.18, defined as an indefinite relative, and ὅς is listed alone in 92.27 as the only definite relative. LN thus overlook the well-known fact that ὅστις in the NT (and the Koine generally) is very frequently not indefinite at all, but simply complements the declension of ὅς.25 (7) παρά of location, ‘among, at, by,’ is covered in 83.9 and 83.25, but παρά with reference to movement, ‘along, beside’ (e.g., Mark 1:16) is not catered for. The word is not included, except in the meaning ‘from’ (84.5), in Domain 84 Spacial Extensions, where we would have expected to see it alongside διά and κατά (84.29–32). (8) Words for ‘why?’ These appear in two places: πόθεν, ἱνατί, and λεμα in 89.38; τί in 92.15. Apart from the unexpected division – the definitions are almost identical – where is διὰ τί? And is the (admittedly rare) use of ὅτι totally undeserving of mention?26 I can find no sign of these two anywhere. 6. A USE IS

COVERED AFTER ALL

πολλά is worth a section of its own, for its notably misleading treatment. Suppose we are looking for the familiar adverbial use of πολλά as an intensive. BAGD handle this well enough with ‘the acc. is used as an 24

25

26

Cf. BDF, §427.2; MHT, Grammar, III, 283; Barrett, John, 439, 482. The English translations of John 18:17, 25 diverge entertainingly: NIV, NRSV dutifully render as expecting the answer ‘no’; ARV, NEB make it neutral; and KJV, RSV translate as if the question were introduced by οὐ! A good recent discussion is: Spottorno, ‘Relative Pronoun.’ LN compound the error by quoting in 92.18 an example of ὅστις which is certainly definite (Acts 5:16). Five possible examples in the NT. Mark 9:11, 28 seem secure. Cf. BAGD, s.v. ὅστις 4.b.

96

COLLECTED ESSAYS ON THE GREEK BIBLE

adv. greatly, earnestly, strictly, loudly, often, etc.’ (s.v. πολύς I.2.b.β). A typical example is Mark 5:10 καὶ παρεκάλει αὐτὸν πολλὰ ἵνα μὴ αὐτὰ ἀποστείλῃ ἔξω τῆς χώρας ‘and he begged him eagerly not to send them out of the country.’ In LN we find in the Greek–English Index ‘πολλά (adv) often,’ and a reference to 67.11. This duly provides the definition: a number of related points of time – ‘often, many times,’ with Matt 9:14 as the example. But that is all; no other entry for πολλά is to be seen. LN appear to be telling us that πολλά as an adverb is used in the NT only in reference to time. But if we happen to look under πολύς, something more emerges. Three meanings of this word are recognised, the third of which is: 78.3 πολύςc the upper range of a scale of extent, but probably somewhat less than for μέγαςb, μεγάλως, and μέγεθος (78.2) – ‘great, greatly, much, a great deal.’ ἠγάπησεν πολύ ‘the great love she has shown’ Lk 7.47; τὸ πολὺ αὐτοῦ ἔλεος ‘his great mercy’ 1 Pe 1.3; πολλὰ σπαράξας ‘caused him to convulse severely’ Mk 9.26; καὶ πολλὰ ἐπετίμα αὐτοῖς ‘and he sternly warned them’ Mk 3.12. …

So here is the rest of the story of adverbial πολλά, but subsumed under the adjective, and in a quite separate Domain (Degree). It is fair to say that this is a bad arrangement, for at least two reasons. It is misleading for the user of the dictionary, who may well think that πολλά is used only of time; and it is inconsistent, in that this use is picked out and given a standing of its own, while other uses (referring to size, number, degree) are lumped together and placed out of sight, as it were, under one use of the adjective. In fact πολλά as an adverb is hard to split into different senses, and would be better regarded simply as an intensifier. LN’s own example of the supposed sense of ‘often’ illustrates the point: Matt 9:14 διὰ τί ἡμεῖς καὶ οἱ Φαρισαῖοι νηστεύομεν πολλά, οἱ δὲ μαθηταί σου οὐ νηστεύουσιν· Here πολλά is not unambiguously ‘often,’ but might just as well be translated ‘much,’ or ‘severely.’

7. PROBLEMS WITH DEFINITIONS While it is true, as has been said above, that LN’s policy of using definitions to indicate meaning is welcome, and the results are generally an improvement on existing treatments, there is much in the way they have framed their definitions that causes uneasiness. Even after due allowance has been made for individual taste, and for the demands imposed by the domain structure, many of the definitions seem unsatisfactory.

THE UNITED BIBLE SOCIETIES’ LEXICON

97

One problem that stands out may be illustrated by the following entry: 13.3 γίνομαιb : to possess certain characteristics, with the implication of their having been acquired – ‘to be.’ γίνεσθε οὖν ϕρόνιμοι ὡς οἱ ὄϕεις ‘therefore be wise as serpents’ Mt 10.16; διότι ἀγαπητοὶ ἡμῖν ἐγενήθητε ‘because you are dear to us’ 1 Th 2.8.

What are we to make of such a definition? It may first be remarked that if we did not already know what word was being defined, it would be difficult to guess – ‘to possess certain characteristics’ sounds like something to do with ἔχω. But more important, it is only by looking at the examples (and the gloss!) that we can see that the meaning being defined is: ‘be (something specified by an adjective).’ The definition does not convey this. We cannot see how ‘to possess certain characteristics’ fits in with ϕρόνιμοι or ἀγαπητοί in the examples. Similar difficulty is encountered with γίνομαιc at 13.48, defined as: to come to acquire or experience a state (LN’s example: ὁ λόγος σὰρξ ἐγένετο Jn 1.14). How can this work in John 1:14? Is σάρξ a ‘state’? The problem here is that the definition is not a substitution equivalent. That is to say, it is not a form of words that can take the same place in the context as the word being defined and convey the same meaning. Yet the framing of definitions in this form is usually taken to be a basic procedure of lexicography. LN’s practice is inconsistent: many definitions are cast in this form, but many are not. This is clearly a deliberate choice by the editors, not an oversight, but for the user it is a difficulty running through the whole dictionary, making the definitions hard to rely on. At any time one may encounter a definition that does not ‘fit the context.’27 Further illustrations are: 71.22 χρή: that which should be or happen, with the implication of propriety – ‘should, ought.’ οὐ χρὴ . . . ταῦτα οὕτως γίνεσθαι ‘this should not happen’ Jas 3.10. 71.28 ὄϕελον: that which ought to be if one only had one’s wish – ‘would that.’ καὶ ὄϕελόν γε ἐβασιλεύσατε ‘and would that you really were kings’ … 1 Cor 4.8. … 13.68 ἐκβάλλωf: to cause a significant change of state by decisive action – ‘to cause to be, to make become.’ ἕως ἂν ἐκβάλῃ εἰς νῖκος 27

I of course exclude from this criticism the many definitions that are not intended to be any sort of equivalent, but employ a different mode of definition, namely description of the function of the word. See many examples in Domains 83–92, e.g., ‘91.3 μένa: a marker of linkage in discourse.’ Even some verbs come into this category, e.g., ‘90.45 ποιέωa: a marker of an agent relation with a numerable event.’ As this example shows, such definitions invite another kind of criticism, on the grounds of obscurity.

98

COLLECTED ESSAYS ON THE GREEK BIBLE

τὴν κρίσιν ‘until he makes justice to triumph’ or ‘until he causes justice to prevail’ Mt 12.20. 13.54 ὑπάγωe: to undergo a significant change – ‘to undergo, go to.’ καὶ εἰς ἀπώλειαν ὑπάγει ‘and he goes to destruction’ or ‘he is destroyed’ Re 17.8.

A more striking manifestation of this problem is the occasional defining of a noun as a verb. This is consistent with LN’s principle, noted earlier, that a verb and its related noun may have essentially the same meaning and one definition can cover both, but the result is distinctly odd when the definition of the verb is given for the noun alone. For example: 88.131 ὕβριςa, εως f: to be insolently mistreated – ‘maltreatment, insolence, and mistreatment.’ 88.139 ἀνομία, ας f: to behave with complete disregard for the laws or regulations of a society – ‘to live lawlessly, lawlessness, lawless living.’

Similarly, in at least one case an adjective is defined as if it were an abstract noun: 41.31 βάρβαροςc, ον: a pattern of behavior associated with a low cultural level – ‘uncivilized, barbarian.’28

Other problems with LN’s definitions vary considerably from case to case and are difficult to classify. One can only say by way of summary that at any point the reader may encounter a definition that is uninformative or inaccurate or puzzling in some way. I simply offer a few samples to illustrate. 36.1 ἡγέομαιb; προΐσταμαιa; κατευθύνω; ϕέρωd; ἄγωd: to so influence others as to cause them to follow a recommended course of action – ‘to guide, to direct, to lead.’

The definition seems very broad, even vague. This is not a consequence of having to cover five different words: all share the same meaning here. Whereas the glosses are clear, the definition, rather than helping to clarify exactly what meaning is under consideration, makes it indistinct. It sounds just as much like a definition of πείθω as of any of these verbs. How then is πείθω distinguished? At 33.301 we find πείθωa defined as: to convince someone to believe something and to act on the basis of what is recommended. This does not seem very different, and the impression is supported by the fact that the NT examples given by LN fit happily enough 28

Cf. the definition of πολύς quoted above, and other adjectives in Domain 78A, most of which will not work as substitution equivalents.

THE UNITED BIBLE SOCIETIES’ LEXICON

99

under either definition. In short, we are looking at definitions that are simply not distinctive enough to be useful. 13.85 προϕέρω: to cause to exist in an evident manner – ‘to cause to exist clearly, to bring forth, to produce.’ ὁ ἀγαθὸς ἄνθρωπος ἐκ τοῦ ἀγαθοῦ θησαυροῦ τῆς καρδίας προϕέρει τὸ ἀγαθόν ‘the good person from the good treasure of the heart brings forth that which is good’ or ‘. . . causes to exist that which is good’ Lk 6.45.

The definition on its own is quite unhelpful. What does ‘to cause to exist in an evident manner’ mean? It is only with the help of the glosses and by studying the example that we can appreciate what is going on. Part of the difficulty is LN’s decision to treat this use as an ‘established figurative meaning,’ i.e., one which has lost its figurative impact.29 This is very unexpected in a context where προϕέρω is combined with ἐκ τοῦ θησαυροῦ. And sure enough, in their treatment of θησαυρός in the parallel passage Matt 12:35, LN go the other way and take it as figurative (see 7.32). In sum, one feels a lack of sureness of touch here. 13.6 περίκειμαιd: to be in a state involving various aspects – ‘to be in various ways, to be in many ways.’ ἐπεὶ καὶ αὐτὸς περίκειται ἀσθένειαν ‘since he himself was weak in many ways’ He 5.2.

This definition suffers from all the weaknesses we have just been seeing: its import is hard to grasp, it is vague, and it cannot be fitted into the context of the quoted example. We are better off with BAGD’s ‘be subject to,’ or Newman’s ‘be subject to, be beset with.’

8. SYNTAX

IS THE HIDDEN FACTOR

As we have seen, the policy of the new lexicon is to exclude syntax from consideration. This is followed with great rigor: apart from indications of case with some prepositions, virtually no syntactical data are given. While this may seem sound as a principle, the consequences in practice give us pause. In terms of the focus of this paper, i.e., the value of LN’s lexical analysis of each word, doubts must arise whether a lexical treatment that omits syntax in this thoroughgoing way remains useful. Of course much of the syntactical material included in a dictionary such as BAGD is not essential and may be omitted without serious consequences. I am thinking rather of instances where keeping syntax out of 29

See LN, 1:xviii.

100

COLLECTED ESSAYS ON THE GREEK BIBLE

the picture seems artificial, and causes strain or is even misleading. Some specimens follow. In Domain 59 Quantity, two meanings of πολύς appear. At 59.1 πολύςa is defined as: a relatively large quantity of objects or events (example: τῷ σῷ ὀνόματι δυνάμεις πολλὰς ἐποιήσαμεν ‘in your name we did many mighty deeds’ Matt 7:22). Then at 59.11 we find πολύςb: a relatively large quantity (example: ἦν δὲ χόρτος πολὺς ἐν τῷ τόπῳ ‘there was much grass there’ John 6:10). Why two separate yet so similar meanings? The hidden factor, as we learn from the examples, is that πολύςa is plural (‘many’), and πολύςb is singular (‘much’). The difference of meaning depends on a syntactic feature which underlies LN’s analysis but is not mentioned. The user who does not already know the facts is left in the dark. At 89.49 we have ἵνα defined as: a marker of result, though in some cases implying an underlying or indirect purpose. Then at 89.52 an almost identical definition appears for ὡς and ὥστε: markers of result, often in contexts implying an intended or indirect purpose. Both entries are in the same Subdomain (89H Result). Why are they separated? Is there any real difference between them? The reason for the separation must surely be the unmentioned and unmentionable fact that their constructions are different. LN’s treatment of ἄν in ‘unreal’ conditions seems the reductio ad absurdum of the method. 71.14 ἄνa: (in combination with a past tense, as part of the apodosis of a conditional sentence) pertaining to being certain, in view of particular circumstances of a condition contrary to fact – ‘would.’4 οὗτος εἰ ἦν προϕήτης, ἐγίνωσκεν ἄν ‘if this person were a prophet, he would know’ Lk 7.39. 4

The meaning of ‘certainty’ in this type of conditional sentence cannot be attributed solely to ἄν but to the combination of ἄν with certain tense forms. This is a typical case of a combined lexical and syntactic meaning.

This word really has no meaning without the syntactic combination. LN themselves are telling us this and giving us syntactical information, but cannot bring themselves to tell us the really crucial piece, that the combination with the indicative mood is what gives this meaning. Perhaps the most difficult to understand and the most likely to mislead is LN’s handling of prepositions. Prepositions occur in a variety of Domains, usually without any reference to case, though occasional mentions occur. In Domain 84 Spacial Extensions, for example, some 18 prepositions appear. Cases are not given, except in three instances (84.5, 84.29, 84.32). The reason for the exceptions is not immediately apparent. Then we find

THE UNITED BIBLE SOCIETIES’ LEXICON

101

the strange Domain 90, called Case, in which, along with other words, many prepositions appear, now with cases uniformly indicated as far as Subdomain M Experiencer, where they suddenly stop. So in 90.41 one meaning of εἰς has the case added, but in 90.59 another does not.30 The introduction to this Domain (pp. 796–7), lengthy compared with that given to other Domains, does nothing to dispel the mystery. Case, we are told, ‘is defined in terms of the relation of participants to events or states,’ and ‘there are a number of different prepositions which are employed to mark quite diverse case relations.’ Whatever all this means, it seems clear that it is very difficult to analyse the semantics of prepositions and present them in a useful way without taking the accompanying case into account. 9. CONCLUSIONS First let us recognise that the semantic domain approach adopted by LN is of great value, offering a new way of working with the NT vocabulary. The stated objective of providing a tool for translators seems also to have been successfully achieved. The value of the work in these ways is not to be underestimated. But what of the lexical analysis as such? Again, let it be clearly stated that the greater part of it is of very high quality and an advance on previous work. Time and again LN are seen to have improved the rather muddled material often found in BAGD, by re-analysing the meanings, using definitions instead of glosses, and applying consistent criteria of classification. Nevertheless, one has to say that the work is flawed. The gaps in coverage of the NT data are the major worry. At first it appeared possible that these were few and not very serious, in that they were matters of difference of opinion about placing of NT uses. As this investigation has progressed, however, it has become clear that there are real and serious gaps, and that they are unpredictable in extent: at any time one may be looking at an analysis that fails to cover NT usage fully. The problem is made worse by the fact that it is next to impossible to tell if anything is missing without an extensive investigation. The unsatisfactory character of many of the definitions is also a serious concern. What could be done in a revised edition? I suggest that another edition in alphabetical format would be of value. In it all the senses of each word 30

σὺν χειρί (90.2) is also given a case indication, but not χράομαι, καταχράομαι (90.13) or ἐπιρίπτω ἐπί (90.18).

102

COLLECTED ESSAYS ON THE GREEK BIBLE

could be gathered under the word, and the NT examples listed more fully. But at the same time it would be necessary to correct the deficiencies in coverage and wording of definitions. To do this, nothing less than a complete check and reworking of the NT data would be required. This is no small job; yet only if it were done could the dictionary be reliable and offer the really sound lexical analysis of the NT that we still lack. Whether it is possible to carry out such a revision is a matter for the authors.

10. LN AND THE NEW MM The question of the value of LN’s lexical analysis of the NT has particular relevance to the task of producing a new edition of Moulton and Milligan’s Vocabulary of the Greek Testament, a project in which G. H. R. Horsley and this writer are currently engaged.31 The object of the work is, in essence, to present the documentary evidence available for each NT word and use. The collecting of data from the inscriptions and papyri is of course the main task, to which an essential adjunct is a clear statement of the NT meanings to be illustrated. Some such description is obviously required both as a guide to what is being sought in the documents and, at the editing stage, as a framework for the presentation of parallels. Though revisions will almost certainly need to be made in the light of the evidence found, a starting-point of some kind must be adopted. A simple but reliable lexical analysis of the NT vocabulary is therefore a necessity. Surprisingly, such a thing is not easily found. At the outset of the project it was taken for granted that BAGD’s would be the analysis on which we depended. The problems with this have become increasingly evident. The most notable of them – the inadequate handling of definition of meaning, and the classification by diverse criteria – are the same as led LN to produce a new lexicon. The effect of these and other difficulties is that a lexical analysis cannot be extracted easily from BAGD. Except in the case of the most straightforward words, anything up to a full re-analysis may be required. 31

The project was first proposed at Macquarie University in 1979 and has been pursued continuously since then, first at Macquarie, and then (since 1988) as a joint project of Sydney, La Trobe, and Macquarie Universities in Australia. Detailed testing of the proposal has appeared in G. H. R. Horsley, New Documents Illustrating Early Christianity (5 vols., North Ryde, NSW, 1981–89). [Since 2005 no further collaborative work on this project has been done. Nevertheless it is worth keeping this section as showing the issues on our minds at the time and still relevant to such a project.]

THE UNITED BIBLE SOCIETIES’ LEXICON

103

At this point, clearly, the new lexicon of Louw and Nida presents itself as a possible alternative. The investigation undertaken in this paper, however, leads to the conclusion that our problem remains unsolved. LN’s analysis, though it is certainly an advance on BAGD’s, is simply not reliable enough in its descriptions of meaning and, more particularly, in its coverage of the NT data.

Postscript Louw responded to this paper with a vigorous defence in ‘Analysis of Meaning’ (1993). A fuller assessment of LN appears in Lee, History, 155–65. In considering its place in the history of NT lexicography, I called it a ‘breakthrough’ because of its thorough application of the definition method, while also giving due appreciation to its adoption of a domain structure (155–6, 180). The issue of coverage of meanings in the NT remains, however. I returned to it in History, 162–4, presenting new examples. A more recent description and critique of LN is found in Peláez, ‘NT Lexicography,’ 37–54, and for comment on my assessment of LN, see Porter, ‘Lexical Semantics,’ 124–5. The question of an alphabetic arrangement of LN’s data has not gone away: see Peláez, ibid., 53–4; cf. Lee, History, 173 n. 6. The temptation to use the ‘Greek–English Index’ in vol. 2 with its handy glosses as a de facto small dictionary is hard to resist.

8 A LEXICAL STUDY THIRTY YEARS ON1 2003

Abstract The author’s Cambridge dissertation, A Lexical Study, was completed thirty years ago. The aims of this paper are (a) to set that work in the context of LXX studies at the time, the late 1960s; (b) to consider what investigation has been done along the same lines since then (to 2003); and (c) to present a specimen of similar work waiting to be done. Words for ‘to order’ in the LXX-Pentateuch are selected for examination. The three main words used in the Pentateuch, ἐντέλλομαι, συντάσσω, and προστάσσω are first illustrated, then the documentary evidence for the third century BC is presented, illustrated by examples; a fourth word, κελεύω, not actually used in the Pentateuch, is also noticed. Conclusions are drawn about the usage of the words and differences between them in third-century BC Greek. It is argued that any differences are found not in lexical meaning but in connotation or ‘tone.’

1. A LEXICAL STUDY My doctoral dissertation entitled A Lexical Study of the Septuagint Version of the Pentateuch was published in 1983. This was an unrevised version of a work that had been completed and submitted in 1970, from research begun in 1966. Its main objective was to demonstrate the links between the vocabulary of the LXX Pentateuch and that of contemporary Koine Greek, especially as evidenced in the Egyptian papyri. This contributed to the debate about the nature of LXX Greek, in support of the view that the Greek of the LXX is essentially the Greek of its time. It was not a new idea to use the documentary evidence for this purpose: the initial discovery had been made by Deissmann. But it had not been followed up for the 1

An earlier form of this paper was presented at a Language Colloquium at Macquarie University, 14 September, 2001. I thank Trevor Evans for the invitation to participate and for running an observant eye over this final version. [The essay found an appropriate place in Paul, et al. (eds.), Emanuel, in honour of my long-standing friend Emanuel Tov.]

106

COLLECTED ESSAYS ON THE GREEK BIBLE

LXX (unlike the NT) after the first enthusiasm. On the contrary, a reaction had set in and the phantom of ‘Jewish-Greek’ had made a comeback. Moreover, the evidence had not been thoroughly exploited by Deissmann and had continued to grow in the meantime. There was clearly a need to pursue the question further and make the case stronger. The work was well received by most reviewers and is regularly referred to in the literature. There seems to be general acceptance that the case has been proved.2 This is gratifying, but it is not the end of the story. There is much still to be done to make full use of this evidence, not only in order to strengthen the case further but, perhaps more importantly, to throw light in all sorts of ways on the meaning and usage of LXX words. In this paper I want to look back and set that study in the context of its time, consider developments since, and present a specimen of the kind of work that is still waiting to be done. My training at the University of Sydney, where I did my first degree, was primarily in Classical Greek, and as part of it I was introduced to Koine Greek in a course conducted by G. P. Shipp.3 I had also spent some years learning Hebrew, and did some work on the LXX in my honours year. Even before I left Australia to go to Cambridge, it struck me that there was a topic of interest in some combination of the LXX and the language of the papyri, though I did not have a clear idea of how it could be focused. Early in the century Cambridge had been a centre of LXX studies, but by 1966 the great names of the past, Swete, Thackeray, Brooke, and McLean, were a distant memory.4 Sebastian Brock, who had known Peter Walters, and had recently arrived in Cambridge, was, I think, the only LXX specialist there at the time. My supervisor, Barnabas Lindars, SSF, a fine scholar who was subsequently elected to the Rylands Chair of Biblical Criticism and Exegesis at Manchester, was primarily a NT expert and a little mystified by my approach. Important help came to me from John Chadwick, who taught me the basics of lexicography. 2

3

4

See, e.g., LEH, Lexicon, VIII; Muraoka, Lexicon, IX n. 14; HDM, La Bible grecque, 234; Jobes and Silva, Invitation, 263. G. P. Shipp (1900–1980), Professor of Greek 1954–1965, best known for his work on Homer, but a philologist whose range covered all of Greek and Latin. The Sydney BA provided (as it still does) the opportunity to mix subjects from different areas. I had done enough Latin as a schoolboy to satisfy me. For a helpful summary of the life and work of these scholars, see Jobes and Silva, Invitation, 247–52. On one occasion, in 1970, I dined as a guest in Christ’s College. When I mentioned my work on the LXX to one of the older Fellows, he recalled McLean, who had been a Fellow and Master of the College (he died in 1947).

A LEXICAL STUDY THIRTY YEARS ON

107

Although there was contemporary debate about the nature of LXX Greek, the issue was not a priority in current LXX research, and the idea of harvesting the rich documentary evidence did not seem to be much on anyone’s mind.5 This despite the fact that Deissmann’s discovery dated from 1895, and the subsequent work of Moulton and Milligan was focused on the NT (and covered that only partially) and was itself well out of date. I felt isolated in choosing this topic; it even seemed that I was venturing into strange and dubious territory. As to the lexicography of the LXX, though everyone said that a new lexicon was needed, there was no project in place to make it a reality. LXX studies at that time were quiet compared with the remarkable burgeoning that has occurred since. Much of the work was on the text – as it had been in the previous generation also – and the important Göttingen edition was a focus of effort. Sidney Jellicoe’s 1968 book, which marks a sort of turning-point, being the first general book since Swete’s 1902 classic, appeared during the time I was working. The IOSCS only began at the same time, holding its inaugural meeting in December, 1968.6 Contact between scholars in those days was not what it is today. Email did not exist, and flying to and fro between countries was yet to become common (my trip from Australia to England in 1966 was actually a sea voyage of four weeks). My first visit to the USA did not take place until 1985. My work in Cambridge of course predated the computer. Those who have not lived through that revolution may not fully appreciate how farreaching are the changes it has wrought. Electronic tools and resources did not exist. Not just hours or weeks, but months were spent searching for words in the indexes of documentary volumes and confirming occurrences. Every text had to be laboriously copied by hand in the library, then recopied into the manuscript when written, before finally being handed over for typing. Typing of Greek could just be achieved by the latest advance in technology, the IBM ‘golf ball’; even so, accents and other diacritics had to be inserted by hand. Photocopying was only just beginning to be possible. As to other tools, even such a basic thing as a list of editions of documentary papyri was not to hand when I began.7 5

6

7

The papyrologist Orsolina Montevecchi was a notable exception. She had already drawn attention to the links and continued to do so. See HDM, La Bible grecque, 233, 243, 248; Lee, Lexical Study, 6. For an account of the IOSCS, see Jobes and Silva, Invitation, 311–3. I still possess a copy of the original first Bulletin, of June 1968, consisting of four sides of mimeographed foolscap. This was reprinted in BIOSCS 2. I remember the kindness of E. G. Turner, Professor of Papyrology at University College, London, in showing me an advance copy of the list in his Greek Papyri (1968) and marking for me the volumes that contained Ptolemaic papyri.

108

COLLECTED ESSAYS ON THE GREEK BIBLE

These remarks indicate the practical constraints under which work of this kind proceeded. Today [in 2003] searches for words in both inscriptions and papyri can be done in seconds via the CD ROM PHI 7, texts can be inspected on screen and portions copied, and the whole LXX is available electronically. There are also two commentaries on the Pentateuch, a lexicon (or two) of the LXX, an English translation project in progress, and for good measure a CD ROM containing most of Greek literature to 1453. Obviously the book has limitations. In general it lacks the comprehensiveness that could be achieved today (even within the length limit imposed). Moreover, it was restricted to a certain kind of parallel, namely new developments in the language, so as to establish the point as persuasively as possible; and I was at all times very cautious to include only the most convincing examples. A less restrained collection of evidence (even if limited to the Pentateuch) could be twice as long, and even then would not exhaust the potential of this line of enquiry.8 What has happened since? What sort of follow-up work has there been? It is safe to say not very much. There have been minor studies in articles, but there has been no book or major work that pursues the subject thoroughly.9 There is no ‘Moulton and Milligan’ for the LXX. The two lexicons that have appeared since I completed my study (Lust, Eynikel, and Hauspie for the whole LXX and Muraoka for the Twelve Prophets)10 have performed important service in providing tools after such a long time when the only lexicon was Schleusner (1820–1821), but, as their authors admit, they have not attempted fresh searching of documentary sources: they have relied on existing collections. For myself, one discovery in Lexical Study (on ὕω and βρέχω) was the inspiration for a line of investigation on ‘formality’ in the NT; but my interests generally since 1970 have gone in other directions.11 8

9

10

11

I think the decision to publish the work as it stood was right, though the delay was regrettable. Even so, nothing occurred up to 1983 to change the picture significantly; it is not as if the book has been replaced even now. Various minor weaknesses were pointed out by reviewers, mostly rightly. But Ina Willi-Plein’s wholly negative review (in ThLZ: see bibliog.) occasioned me as much perplexity (‘Ratlosigkeit’) as she said she felt in trying to review the book. A noteworthy article is that of Hélène Cadell, ‘Vocabulaire de la législation.’ For further bibliography, see HDM, La Bible grecque, 243. Documentary evidence is used sproradically in the papers of Tov, Swinn, and Lee in Muraoka (ed.), Melbourne Symposium. That memorable occasion was one of many valued contacts with the honorand of the present volume (Emanuel Tov). LEH, Lexicon (2003); Muraoka, Lexicon (Twelve Prophets) (1993). [Now Muraoka, Lexicon for the whole LXX.] The NT study was published as ‘Some Features’ (1985). This could still be followed further in the LXX. Chapter 8 of Lexical Study on linguistic evidence for dating the

A LEXICAL STUDY THIRTY YEARS ON

109

2. ‘ORDER’ WORDS IN THE LXX-PENTATEUCH I turn now to a specimen of what might still be done to illuminate the LXX vocabulary using contemporary documentary evidence. I have selected the group of words for ‘to order.’ One naturally expects to begin with the ordinary word for ‘to order’ in ancient Greek, κελεύω. It is a surprising fact that there is not a single example in the LXX Pentateuch. Opportunities to express the idea were certainly not lacking: the usual Hebrew word (‫ )צוה‬occurs many times. The translators must have been using something other than κελεύω. What they were using is soon discovered. The following three verbs account for almost all expressions of the idea in the Pentateuch: ἐντέλλομαι (156 times), συντάσσω (82), and προστάσσω (12).12 Not so obvious is why this should be so. Why not κελεύω? What has happened to it? I think the documentary evidence helps considerably in finding an answer. Let us begin with a summary of the usage of the two commonest words in the LXX Pentateuch and a selection of examples. A look through the occurrences soon shows that ἐντέλλομαι and συντάσσω are semantically close; it is difficult to find any grounds for giving them separate lexical meanings. Both are used for orders given by an authority figure (God, Moses, Pharaoh, a parent, a priest). The manner of ordering is oral, rather than written, in both. Both are used for instructions to be immediately acted on as well as longer-term directions, and both are used to order any kind of action. ἐντέλλομαι with the law (or provisions of the law) as object is noticeably frequent. Finally, it is easy to find places in which the translators use the two interchangeably in near proximity, referring to the same event and rendering the same Hebrew word.13

12

13

LXX Pentateuch was very experimental and limited (as I knew full well). Gathering the data for this purpose could be done much more thoroughly and with less effort now; finding valid tests is the tricky part. [Ross, in ‘Lexical Possibilities (2016), reinvestigated ὁράω and βλέπω and concluded that ‘even in light of new data, Lee’s conclusions prove remarkably accurate, suggesting the potential of his methodology for further application.’ (341).] Figures from LEH, Lexicon, drawn in turn from the CATSS files. Other words used in Pent.: ἐπιτάσσω (x 1); ἐξηγοῦμαι (x 1); also λέγω/εἶπον, of which possible examples are: Exod 35:1 (cf. Le Boulluec and Sandevoir, L’Exode, ad loc.), Lev 9:6; Num 32:27 (cf. LEH). Not in Pent.: διατάσσω, παραγγέλλω. Not in this sense in Pent.: διαστέλλω. Outside the Pent. κελεύω occurs 29 times, but only in books not in the Hebrew Bible; ἐντέλλομαι is by far the favourite word. E.g., Exod 31:6 (συντ.), 11 (ἐντ.); Lev 8:4 (συντ.), 5 (ἐντ.), 9 (συντ.); Num 2:33 (ἐντ.), 34 (συντ.); Deut 5:12 (ἐντ.), 15 (συντ.), 16 (ἐντ.). All ~ ‫ צוה‬piel. Cf. Dorival, Les Nombres, 54, 56. Wevers (Numbers, 31, 494) says he can find no distinction between ἐντέλλομαι and συντάσσω. A full translation-technical study is not undertaken

110

COLLECTED ESSAYS ON THE GREEK BIBLE

A selection of typical examples first of ἐντέλλομαι (with one of προστάσσω), and then of συντάσσω: (1) Gen 3:11 καὶ εἶπεν αὐτῷ Τίς ἀνήγγειλέν σοι ὅτι γυμνὸς εἶ, εἰ μὴ ἀπὸ τοῦ ξύλου, οὗ ἐνετειλάμην σοι τούτου μόνου μὴ ϕαγεῖν ἀπ᾽ αὐτοῦ, ἔϕαγες; And he said to him: Who told you that you are naked, unless you have eaten of the tree of which I commanded you of this one alone not to eat of it? (2) Gen 12:20 καὶ ἐνετείλατο Φαραὼ ἀνδράσιν περὶ Ἀβρὰμ συμπροπέμψαι αὐτὸν καὶ τὴν γυναῖκα αὐτοῦ καὶ πάντα, ὅσα ἦν αὐτῷ, καὶ Λὼτ μετ᾽ αὐτοῦ. And Pharao gave orders to some men regarding Abram to conduct him on his way and his wife and all his property, and Lot with him. (3) Gen 28:1 προσκαλεσάμενος δὲ ᾽Ισαὰκ τὸν ᾽Ιακὼβ εὐλόγησεν αὐτόν, καὶ ἐνετείλατο αὐτῷ λέγων Οὐ λήμψῃ γυναῖκα ἐκ τῶν θυγατέρων Χανάαν· Isaak summoned Iakob and blessed him and ordered him saying, You shall not take a wife from the daughters of Khanaan. (4) Exod 7:6 ἐποίησεν δὲ Μωυσῆς καὶ Ἀαρών, καθάπερ ἐνετείλατο αὐτοῖς κύριος, οὕτως ἐποίησαν. Moyses and Aaron did as the Lord instructed them, thus they did. (5) Deut 27:1 καὶ προσέταξεν Μωυσῆς καὶ ἡ γερουσία ᾽Ισραὴλ λέγων Φυλάσσεσθε πάσας τὰς ἐντολὰς ταύτας, ὅσας ἐγὼ ἐντέλλομαι ὑμῖν σήμερον. Moyses and the council of elders of Israel ordered (them) saying, Keep all these commandments which I command you today. (6) Exod 1:17 ἐϕοβήθησαν δὲ αἱ μαῖαι τὸν θεόν, καὶ οὐκ ἐποίησαν καθότι συνέταξεν αὐταῖς ὁ βασιλεὺς Αἰγύπτου, καὶ ἐζωογόνουν τὰ ἄρσενα. The midwives feared God and did not do as the king of Egypt directed them, and they kept the male children alive. (7) Exod 12:35 οἱ δὲ υἱοὶ ᾽Ισραὴλ ἐποίησαν καθὰ συνέταξεν αὐτοῖς Μωυσῆς, καὶ ᾔτησαν παρὰ τῶν Αἰγυπτίων σκεύη ἀργυρᾶ καὶ χρυσᾶ καὶ ἱματισμόν· The sons of Israel did as Moyses ordered them and asked for silver and gold objects and clothing from the Egyptians. here, but would complete the picture. In two of my examples (Exod 1:17; 12:35) συντάσσω ~ ‫ דבר‬piel/subst.

A LEXICAL STUDY THIRTY YEARS ON

111

(8) Lev 13:54 καὶ συντάξει ὁ ἱερεύς, καὶ πλυνεῖ ἐϕ᾽ οὗ ἂν ᾖ ἐπ᾽ αὐτοῦ ἡ ἁϕή, καὶ ἀϕοριεῖ ὁ ἱερεὺς τὴν ἁϕὴν ἑπτὰ ἡμέρας τὸ δεύτερον· The priest shall give directions and shall wash [i.e., get someone to wash] the article on which the mark of infection is, and the priest shall set apart the mark of infection for seven days a second time. (9) Num 2:34 καὶ ἐποίησαν οἱ υἱοὶ ᾽Ισραὴλ πάντα, ὅσα συνέταξεν κύριος τῷ Μωυσῇ, οὕτως παρενέβαλον κατὰ τάγμα αὐτῶν καὶ οὕτως ἐξῆρον. The sons of Israel did everything that the Lord instructed Moyses, thus they encamped according to their units and thus they set out on the march.14

3. DOCUMENTARY

EVIDENCE

Now for the documentary evidence. A search (via PHI 7) in nineteen volumes of papyri produced these numbers of occurrences in third century BC texts: ἐντέλλομαι 22; κελεύω 25; προστάσσω 32; συντάσσω 285.15 Let me summarise what a perusal of this evidence indicates, and then give some representative examples. It is obvious at once that συντάσσω was a favourite word in Ptolemaic officialese in the time of the LXX Pentateuch. In meaning it is definitely in the ‘order’ field with the other verbs. This is a somewhat unexpected flourishing: the word is old, but in Classical usage it mostly means ‘arrange,’ ‘organise,’ and the like, though the beginnings of the development to ‘order’ can just be seen in some occurrences.16 ἐντέλλομαι goes back to earlier Greek, though it was not very common, it seems, outside Herodotus. It is still running along in Ptolemaic Greek with its meaning as before (‘command,’ etc.) but less favoured than συντάσσω in the kind of language found in papyrus documents. προστάσσω was well established in Classical usage as one of the ‘order’ words and maintains its ground. 14

15

16

ἐξῆρον sic Rahlfs, and Wevers, Göttingen ed.; but a 2nd aor. of αἴρω and cmpds. is unusual if not unknown. Read ἐξῆραν? PHib 1, 2; PLille 1, 2; PSI 4, 5, 6; PPetr 1–3; PMich 1; PCol 3, 4; PCorn; PCairZen 1–5. The counts cannot be absolutely precise. I have dropped grossly restored examples, but opinions would differ. Only III BC texts are counted. I have not made any attempt on the inscriptions for the usual reasons: III BC texts are scattered through numerous volumes; they come from a variety of ancient localities, not just Egypt; and their genres vary markedly. See LSJ, s.v. II.4.a and b (but Aeschin. 2.22 is quite different). More examples and comment in Anz, Subsidia, 324–5.

112

COLLECTED ESSAYS ON THE GREEK BIBLE

κελεύω is the oldest member of the group, being the standard word from Homer onwards. It is well known, however, that it may show a weaker sense translatable as ‘request,’ ‘urge,’ ‘tell.’ It loses some of its vigour as a result of long and frequent use. In the Ptolemaic documents it has clearly suffered something of an eclipse, when compared with the overwhelming popularity of its new rival συντάσσω. But this does not mean it had dropped out: it was always still available in both its weaker and its full sense. Later, from the first century AD onwards, it makes a comeback alongside other words, and συντάσσω in its turn fades away.17 All of these verbs are close semantically. The difference between them is not one of lexical meaning so much as of connotation, of ‘flavour’ or ‘tone.’ To determine this exactly is a difficult exercise; only a speaker of the time could appreciate it fully. We can choose, then, to give up the attempt altogether, or do what we can, relying on the hints gleaned from context and situation. I suggest the key is this: ἐντέλλομαι, προστάσσω, and συντάσσω are more formal and official-sounding than κελεύω. They are in use, or have come into use, to meet the need for words with just that flavour as compared to the plain, slightly worn-out feel of κελεύω. συντάσσω is the newcomer and still fresh; it is businesslike, with some hint of its origin in the meaning ‘arrange.’ προστάσσω is not markedly different in flavour from συντάσσω. How ἐντέλλομαι differs from these two is not obvious; perhaps it is a little elevated. At the same time as all this, we have to allow for the use of κελεύω in its original sense with full force, especially in literary writing. This is why most of the LXX examples are in the more literary books (2–4 Maccabees account for two-thirds), and a writer like Polybius uses it 124 times.18 A turn-over in ‘order’ words is in fact a general phenomenon: they tend to lose their punch and become weak, or to lack the authoritative sound felt desirable in certain contexts. In English we have a kaleidoscope of 17

18

A full study of the history of these words, from Class. to Mod., is a desideratum. It would include the curious later development of κελεύω to ‘please.’ Mauersberger, Polybios-Lexikon. Figures for the other words in Polybius: ἐντέλλομαι 17, προστάσσω 60, συντάσσω 84 (the last two from my own search via TLG). Other attempts to distinguish ‘order’ words have generally looked for differences in the manner or substance of the order. For Pelletier, κελεύω ‘tend à se réserver pour les cas où l’ordre est donné de vive voix,’ and ἐντέλλομαι ‘a été choisi pour exprimer le ton d’autorité bienveillante’ (‘Vocabulaire,’ 239, 240). Pelletier proceeds entirely without reference to the papyrus evidence, and uses one inscription from Teos. The idea that κελεύω refers to verbal orders is an old one, found already in Schmidt’s treatment (Synonymik, 1:199–214). Harl (La Genèse, 54, 103) follows Pelletier on ἐντέλλομαι. Cf. another treatment in Louw and Nida, Lexicon, 33.323, 325, 329.

A LEXICAL STUDY THIRTY YEARS ON

113

words and expressions of different ages and flavours at our disposal: bid, enjoin, charge, command, prescribe, direct, order, instruct, issue instructions, issue orders, issue a directive, tell.19 Next a selection of examples from third century BC papyri, the first a nice parallel to (6) Exod 1:17 above: (10) PCairZen 2.155.1–3 (Philadelpheia, 256 BC, letter) Ἀπολλώνιος Ζήνωνι χαίρειν. ὁ βασιλεὺς συνέτασσεν ἡμῖν | δισπορῆσαι τὴν γῆν. ὡς ἂν οὖν ἐχθερίσηις τὸν πρώιον σῖτον, | εὐθέως πότισον τὴν γῆν ἀπὸ χερός. Apollonios to Zenon greetings. The king has issued instructions to us to do the second sowing of the land. When therefore you have finished harvesting the early grain, at once irrigate the land by hand. (11) PCairZen 5.816.1–4 (Philadelpheia, 257 BC, letter) Ἀρτεμίδωρος Πανακέστορι χαίρειν. παραγινομένου μου ἐγ Βουβάστου εἰς Μέμ[ϕιν] | ἐνετέλλετο Ἀπολλώνιος μάλιστα μὲν αὐτὸν διελθεῖν πρὸς σέ, εἰ δὲ μή, ἀποστεῖλα[ί] | τινα παρ᾽ ἐμοῦ ὃς ἀναγγελεῖ σοι τὰ παρ᾽ αὐτοῦ. ἠκηκόει γὰρ ὅτι οὐ πᾶσα κατασπείρετα[ι] | ἡ γῆ αἱ μύριαι ἄρουραι. συνέτασσεν οὖν ἀναγγέλλειν σοι ἵνα ξυλοκοπηθῆι πᾶσα καὶ ποτισθῆι. Artemidoros to Panakestor greetings. When I arrived at Memphis from Boubastos, Apollonios instructed that preferably I myself should come to you, but failing that to send someone from me to convey to you his wishes. For he heard that not all the land, the ten thousand arourai, are sown. He gave instructions therefore to convey to you that all of it is to be cleared and irrigated.

This passage illustrates the interchangeability of the two verbs as well as their general character. Apollonios, also met in (10), was a high official in the administration. (12) PCairZen 1.57.1–4 (Philadelpheia, 257 BC, letter) Ζωίλος Ἀλεξάνδρωι χαίρειν. καλῶς ἂν ἔχοι εἰ ἔρρωσαι· ὑγιαίνομεν δὲ καὶ αὐτοί. | ἐκομισάμην τὰς ἐπιστολὰς καὶ οὐκ ἀμελῶ περὶ ὧν ἂν ἡμῖν γράϕηις. σὺ δὲ καλῶς | ποιήσεις ἐκπονήσας ἵνα ἡμῖν γίνηται ἅ σοι ἐνετειλάμεθα κατὰ τὸ ὑπόμνημα, | τούτου δὲ γενομένου ἐπί[στ]ασο ὅτι ὀϕειλήσω σοι χάριν ἱκανήν. Zoilos to Alexandros greetings. It would be good if you are well; we too are in good health. I have received your letters and I am not neglecting the matters about which you wrote to us. You will do well to ensure that what we instructed you according to the memorandum is done for us, and if this happens, know that I will owe you considerable gratitude. 19

Cf. Buck, Dictionary, 19.45. Note esp. developments from Latin to Romance.

114

COLLECTED ESSAYS ON THE GREEK BIBLE

Here in (12) we have an ordinary example of ἐντέλλομαι in a friendly though businesslike letter. (13) PCairZen 5.852.7–10 (Philadelpheia, III BC, polite letter to Zenon asking for placement) εἰ δ᾽ ἐπὶ τοῦ παρόντος ἡσυχίαν ἄγειν κελεύεις, τὸμ μὲν καιρὸν αὐτὸς | εἰδήσεις, ἐμοὶ δὲ καλῶς ἂν ποιήσαις συντάξας ὅπως ἂν παρα|δειχθῆι μεθ᾽ ὧν οἰκίαν τε ἕξω καὶ τὰ λοιπὰ δέοντα, ἵνα μὴ πρεσ|βύτερος ὢν ῥέμβωμαι. But if for the present you direct me to do nothing, you will know when it is the right time, but as for me please be kind enough to give instructions that it be indicated with whom I am to live and the other necessary matters, so that, being an old man, I may not be left unattached.

κελεύω seems to function the same way as συντάσσω here in (13). The writer is an older man, and the style shows touches of ‘better’ Greek (μὲν … δέ, τε, ὅπως ἄν). (14) PSI 4.420.1–19 (Philadelpheia, III BC, letter) Ζήνωνι χαίρειν | Σεμθεύς. συνέτα|ξάς μοι κεραμεῦ|σαι ἅπαν‹τα› τὸν κέ|ραμον ἕως τοῦ | ἰσιόντος ‹μηνὸς› ε τὴν | ἡμέραν ν. ποιῶ | οὖν κατὰ ταῦτα. | ἐκελεύοσαν δέ με | καταβαίνοντα | συγχωνεύειν. ἐγὼ | οὖν οὐκ ὠιχόμην, | ἕως ἄν μοι σὺ συντά|ξηις. νῦν οὖν ἄλλοι | πάρισιν κεραμεῖς· καὶ | γὰρ ὁ χωνεύων με|μαλάκισται, ὁ ξένος. | τί οὖν μοι συντάσσεις; | ὅπως ἂν οὕτω ποιῶ. Semtheus to Zenon greetings. You ordered me to make pots and nothing else until the 5th of the ensuing month, 50 per day. So I am doing that. But they told me to go down and join in glazing. I have not gone until you order me. Now other potters have come; and the glazer has fallen ill, the foreigner. What do you order me, then? so that I may do that.

συντάσσω is what Zenon does, κελεύω describes the others’ action. I detect a difference: συντάσσω is formal and polite in reference to the authority figure, κελεύω is informal. (15) PCol 4.66.1–9 (Philadelpheia, 256/5 BC, letter) … Ζήνωνι χαίρειν. καλῶς ποιεῖς εἰ ἔρρωσαι. ἔρρω|μαι δὲ καὶ αὐτός. ἐπίστασαι ὡς κατέλιπές με ἐν Συρίαι μετὰ | Κρότου καὶ ἐποίουν πάντα τὰ προστασσόμενα τὰ κα|τὰ ‹τὰ›ς καμήλους καὶ ἤμην σο[ι] ἀνέγκλητ[ο]ς. σοῦ δὲ προστά|ξαντός μοι ὀψώνιον διδόναι ἃ σὺ συνέταξας οὐκ ἐδίδου | μοι οὐθέν. ἐπεὶ δὴ πολλάκ[ι]ς μου δεομένου διδόναι μοι | ἃ σὺ συνέταξας οὐκ ἐδίδου μοι οὐθὲν Κρότος, ἀλλ᾽ ἐκέλευ|έμ με ἀπαλλάσσεσθαι, χρόνον μὲν οὖν πολὺν ἐκαρτέ|ρουν σε προσδεχόμενος. … to Zenon greetings. You do well if you are in good health. I also am well. You know that you left me in Syria with Krotos and I did everything

A LEXICAL STUDY THIRTY YEARS ON

115

that was ordered in regard to the camels and you found no fault with me. But though you gave orders to give me wages/provisions, what you ordered he did not give me any [of it]. When then, though I asked him often to give me what you ordered, Krotos did not give me anything but told me to leave off, I held on for a long time waiting for you.

This text (15) shows examples of προστάσσω as well as συντάσσω, all for Zenon’s action without much distinction between them. κελεύω, as in (14), is for the person cast as the bad guy.

4. CONCLUSIONS Let us draw our conclusions for the LXX. If nothing else, we have a clear explanation why the somewhat unexpected word συντάσσω makes such a strong showing in the Pentateuch: it was a standard word in Ptolemaic Egypt in the time of the translators. προστάσσω needs little comment: it is one of the staple formal words, but less frequently used than συντάσσω in the Pentateuch just as in contemporary sources.20 As to ἐντέλλομαι, while the contemporary evidence confirms that it is alive and well, the high frequency in the Pentateuch (156, as against 22 in papyri) calls for explanation. One reason can certainly be discerned. This verb is often (26 times) joined with ἐντολή, giving a neat etymological combination that matches the Hebrew (‫)צוה מצוה‬. Most of these occur in one book (Deuteronomy), which also has a very large proportion of all the examples (85). Personal taste may well come into it: this translator, like the translator of Genesis, strongly prefers ἐντέλλομαι to συντάσσω.21 Finally, κελεύω. The total absence from the Pentateuch remains a little surprising, but my conclusion is that it was just not suitable in tone. Formal-sounding words were felt more appropriate in the dignified contexts the translators were dealing with throughout the Pentateuch. They availed themselves of those in use, notably συντάσσω, the characteristic word of Ptolemaic officialdom and business. This is merely a specimen of what the documentary evidence still has to offer for illuminating the LXX. Further follow-up could certainly be done even on these words. There is a need for such work to continue in 20

21

[προστάσσω is formulaic in petitions, for the King’s action in reponse to the petition; see esp. PEnteux 2–112 (all III BC).] Distribution in Pent.: ἐντέλλομαι: Gen 20, Exod 19, Lev 16, Num 19, Deut 85: total 156. συντάσσω: Gen 2, Exod 39, Lev 12, Num 26, Deut 2: total 82.

116

COLLECTED ESSAYS ON THE GREEK BIBLE

order to place LXX lexicography on as sound a footing as possible, especially as translation and commentary work on the LXX gathers pace and calls for more refined understanding of the vocabulary. Finding the best way to collect (and store) the material systematically is the challenge for the coming generation of LXX scholars.22

22

[The discussion of ‘order’ words here formed the basis of their treatment as specimens of ‘officialese’ in Lee, Greek of the Pentateuch, 64–6.]

9 THE PRESENT STATE OF LEXICOGRAPHY OF ANCIENT GREEK 20041

Abstract The paper surveys and critiques the existing lexicographical resources for Ancient Greek, considering the questions of coverage, dependence on predecessors, and quality of the analysis of meaning. There is no single lexicon that covers everything. The lexicon that comes nearest is the renowned LSJ, but it is primarily a lexicon of Classical Greek. Others that attempt to cover the same ground are either incomplete or dependent on LSJ. Lexicons of individual authors or bodies of text, of which the NT is one, are numerous but variable in quality and dependent on predecessors. The pattern of indicating meaning by the ‘gloss’ method pervades the tradition, and has only been broken by Louw and Nida’s NT lexicon, employing definitions. In this it has been followed partially by the latest edition of Bauer’s NT lexicon, BDAG (2000). Some limited suggestions are made for improving the situation, directed mainly at the creation of a database to collect data on individual words.

I As long as Ancient Greek texts are read in the original, there will be a need for lexicons, that is to say, tools which provide, first and foremost, meanings of words, and also a certain amount of other helpful information, on occurrences, collocation, and the like. The form that they take – whether it is to be electronic or some other – is not an issue for the present purpose: we can safely leave this to common sense and market forces. But whatever form they might have, there will be a need for tools which tell the reader of a Greek text the meaning of words not already known. Lexicons will continue to be produced to meet this need. 1

This paper was originally presented at the SBL Annual Meeting in Philadelphia in November 1995, in the Biblical Lexicography Consultation. Updating and minor expansions have been incorporated; the general thrust remains the same. Some characteristics of oral presentation have been retained.

118

COLLECTED ESSAYS ON THE GREEK BIBLE

Not only will lexicography be in demand, but it will continue to carry a weighty responsibility. This is because of the special character of lexicons. Lexicons are regarded by their users as authoritative, and they put their trust in them. Lexicons are reference books presenting a compressed, seemingly final statement of fact, with an almost legal weight. The mere fact that something is printed in a book gives it authority, as far as most people are concerned. And understandably: if one does not know the meaning of a word one is predisposed to trust the only means of rescue from ignorance. Yet this trust is misplaced. The concise, seeemingly authoritative statement of meaning can, and often does, conceal many sins – indecision, compromise, imperfect knowledge, guesswork, and, above all, dependence on predecessors. Lexicographers have to make a decision and put down a definite statement, and they are fallible like everyone else. But the ordinary user has no means of knowing where the mistakes have been made, where the ignorance has been covered up, what has been lifted from somewhere else without checking, and so on. All this means that we will not only go on producing lexicons, but we have a special responsibility to ensure that the lexicons we produce get it right. The main part of my paper will be a survey and critique of the existing lexicographical resources for Ancient Greek. Ancient Greek covers Mycenaean, followed by Classical, then Koine Greek (300 BC to 600 AD), within which of course the LXX and NT find their place.2 The language continues on into Byzantine and eventually Modern Greek. All this forms a continuum, of a gradually evolving language, not a series of disparate entities. This fact is important for the lexical study of Greek words: all the evidence, from all these periods, is potentially relevant. For the purpose of this survey, however, I am going only as far as Byzantine Greek. The question I want to address is this: what lexicons or dictionaries are available to provide guidance on meanings of words in all this vast body of Greek texts? What sort of coverage is there? The ideal would be one lexicon that covered everything. This we don’t have. The coverage is in fact partial, unsystematic, and uneven in quality. To put it more bluntly: there are gaps everywhere, and even those things that seem to have been done have not been done as well as they could, and are in need of reassessment. 2

The terms are simply labels, without a commitment to any theory or judgement about the nature of these phases.

THE PRESENT STATE OF LEXICOGRAPHY OF ANCIENT GREEK

119

In saying this, my intention is not to denigrate everything that exists, but to honestly assess the present situation, so that we can go forward. I am very conscious of the centuries of work behind the resources that we now have, and that it is only because of what our predecessors have done that we can go on to build something better. Before I turn to the lexicons themselves, there is one other general point to be made, a most important one. As we all know, in the last few years Ancient Greek texts have become available in electronic form. We all know of the magnificent work done to produce the Thesaurus Linguae Graecae (currently disk ‘E’) and the databank of documentary texts (currently found on the disk ‘PHI7’). And it is already common practice to draw on these in word studies. But I don’t know if it has been generally realised that this development brings about a major shift in Greek lexicography. This resource is not just a useful extra for word studies; it means that all available evidence not only can be, but must be used in any work at an advanced level. This was not possible for earlier dictionary makers; consequently all the existing lexical entries in all our dictionaries are now obsolete and await reassessment in the light of the full evidence, or at least checking to see if there is further evidence to be added.

II Let us turn now to a survey of the present lexical resources. Given time constraints it is not possible to include illustrations of the statements I will make, but it can be taken for granted that supporting evidence is available. We begin of course with LSJ. We all know and love LSJ. It has served us well since 1843. We know it has some faults, but where would we be without it? Actually its faults are much worse than most would suppose. To sum them up briefly under three heads: it has no coherent method of defining meaning, but relies on glosses; its basic material is derived from predecessors, in some cases descending from the ancient lexicographers; and the organisation is chaotic as a result of piecemeal revisions especially in the ninth edition. All this is before taking into account new evidence afforded by electronic searching. The LSJ Supplements (1968 and 1996) could do no more than make further piecemeal changes. All this has been well known to those who have had occasion to work closely with LSJ and scrutinise its entries, but it is not generally recognised, and most scholars continue to trust LSJ. Fortunately, criticisms have now been expressed in print, particularly by John Chadwick. I refer you

120

COLLECTED ESSAYS ON THE GREEK BIBLE

especially to his 1994 paper, where abundant illustrations can be found; he says bluntly: ‘It is about time that Greek scholars recognised the need for a thorough overhaul of this indispensible tool.’3 Since then his book Lexicographica Graeca has appeared (1996), offering many word-studies that show how LSJ’s treatment needs improvement. That is how things stand with what is our only general lexicon of Ancient Greek. Moreover LSJ is basically a lexicon of Classical Greek and does not pretend to cover the post-Classical material in a thorough way. And yet this is the work on which we not only still rely heavily but which has been, for generations, the resource from which everyone, including the authors of other lexicons, has derived information. One can see its influence everywhere. The only competitor to LSJ in attempting to cover the same ground is the Diccionario Griego-Español (DGE), currently in progress.4 This shows every sign of being an improvement on LSJ and of not simply copying their material uncritically. It also makes a proper attempt to cover the postClassical period. But at the present moment it is not finished and will not be in the lifetime of many of us. And there is a further point, which raises a question that ought to be pursued at some time: given the size of the projects we are faced with, how many languages should we operate in in our major reference works? Spanish speakers and advanced scholars will be well served; how many others will use it? Another work, with the title Μέγα Λεξικὸν τῆς Ἑλληνικῆς Γλώσσης, which appeared some fifty years ago, attempts to cover Greek from Ancient to Modern, but does little to remedy the situation. It is basically LSJ’s material, with additions from various sources. III Everything else that remains deals with a special area. That is, all other lexicons are focused on a particular text or set of texts, and draw on outside material, if at all, only to illustrate the words that occur in that text. Let us take first the NT. It has a rich lexicographical tradition of its own going back to the sixteenth century. Yet there are disturbing features to be seen when one looks closely. It is not possible to go over this ground in detail here, but the salient points are these: first, there is the legacy of the long tradition of indicating meaning by glosses rather than definitions, 3 4

Chadwick, ‘Replacing Liddell and Scott,’ 10. [Vol. 7, 2009 has reached ἔξαυος.]

THE PRESENT STATE OF LEXICOGRAPHY OF ANCIENT GREEK

121

which leads to many problems (as Louw and others have shown). Secondly, there is the fact that even the latest lexicons derive their material from their predecessors, and a great deal of it has been passed on uncritically over the course of centuries. Thirdly, there is an aspect that I think is not well known: meanings given in the NT lexicons are contaminated by glosses from the standard translations, going back as far as the Vulgate. There is a fourth tendency which has become evident to me lately: NT lexicons are unsystematic in their control of other discussions, and may or may not take up useful contributions to the understanding of the meaning. All this mainly concerns the major lexicon series of our time, Bauer (1928, 6th ed. 1988) and its offshoots in English.5 Louw and Nida’s 1988 work6 presented two major innovations: it adopted a semantic domains approach, and it put into practice the method of using definitions to indicate meaning. But the understanding of the words is derived primarily from the existing tradition (mainly via Newman7), and the other tendencies that I mentioned are just as evident, even if they are under the surface. BDAG (2000), whose reviser and editor is justly honoured in this volume8 for his indefatigable lexicographical labours over nearly half a century, incorporates not only a host of minor revisions but a significant advance: for the first time in the Bauer series, definitions in the manner of Louw and Nida have been added for a majority of the words. Nevertheless, the NT lexicographical tradition, although it is in an advanced state compared with other areas, would benefit from a thorough rethinking. We come now to the LXX. Until now we have had only Schleusner dating from the 1820s,9 and we have been desperately in need of a new work for more than a century. Well, at last the time has come. We now have not one but two lexicons of the LXX. I refer of course to LEH, covering all the LXX, and Muraoka’s lexicon to the Twelve Prophets.10 I am sure we all feel grateful to these authors for having at last provided us with something. Obviously LEH, which is complete, will be the standard tool for some time to come. 5

6 7 8 9 10

These observations are argued in Lee, ‘Hebrews 5:14.’ A fuller treatment is planned in my book on the history of NT lexicography, currently in preparation. [Published 2003.] Louw and Nida, Lexicon. Newman, Concise Dictionary. [Taylor et al. (eds.), Language and Lexicography.] Schleusner, Lexicon in LXX. LEH, Lexicon (2003); Muraoka, Lexicon (Twelve Prophets) (1993). [See postscript for an update on Muraoka, Lexicon (2009).]

122

COLLECTED ESSAYS ON THE GREEK BIBLE

But we must look at it a little more closely and honestly. The most obvious strength of LEH is in making full use of previous discussions of LXX words outside lexicons. So it gathers up everything that has been done in the past by way of preparation for such a lexicon. But there are also significant weaknesses, which the authors themselves would readily admit. Two major points stand out: (a) most of the meanings are taken wholesale from LSJ; and (b) there has been no systematic gathering of non-LXX parallels that might throw new light on the meanings. In other words, it is based primarily on existing lexicons; and so we continue to move around in this circle in which the faults of one lexicon are passed on to the next. IV As regards the Greek documentary texts, i.e., the inscriptions and papyri, that vast body of data that is still being added to and assimilated, the attempt to provide any sort of lexical coverage is a struggle. There is Preisigke’s lexicon, followed by various supplements by Kiessling and others,11 all out of date or incomplete, and there is a sense of never being able to get on top of it. The only sure way to tap into this material is to work directly on the texts either in printed or in electronic form. But this is daunting for the non-specialist. The study of these documentary texts is one of the major challenges of today. Their importance for illuminating the biblical texts was of course recognised long ago, and my colleague Greg Horsley and I continue with the attempt to produce a new Moulton and Milligan, that is, an up-to-date exploitation of this evidence for the light it sheds on the NT vocabulary.12 The same task needs to be carried out for the LXX (as Lust and his colleagues well recognise). But there is more to this documentary evidence than simply quarrying it for parallels. To produce a proper lexicon of Ancient Greek, all of it needs to be brought into conjunction with the evidence from the literary texts. We are only at the beginning of doing this. The existing lexicographical tradition is based on the literary texts, which have been known all along. Documentary examples have been added piecemeal to the lexicons in recent times, but that is not the same as combining all the evidence to arrive at a complete lexical analysis of a word. 11 12

Preisigke, Wörterbuch. See recently Horsley and Lee, ‘Interim Entries, 1’; Lee and Horsley, ‘Interim Entries, 2.’

THE PRESENT STATE OF LEXICOGRAPHY OF ANCIENT GREEK

123

V So much for the documentary texts. What other lexical resources are there? We must mention, however briefly, the many dictionaries focused on individual ancient authors (Polybius, Josephus, etc.). They are all very well, and of course are very useful in relation to the particular author, but they obviously approach the language piecemeal rather than as a whole. In addition one knows that LSJ will have been the main guide to meaning. What else? There is one other major work, Lampe’s Patristic Greek Lexicon (1968). This rivals LSJ in size and is packed with mind-boggling detail related to the Patristic texts. It is obviously an indispensible and valuable tool. But its drawback is that it is intended to be used in combination with LSJ. That is, all the material in LSJ is taken for granted; Lampe aims only to add new material where necessary. In theory this might sound all right, but in practice it is a recipe for endless uncertainty. One simply cannot know the full range of meaning or usage, or even words, actually occurring in the Patristic authors. Lampe’s lexicon, then, must be recognised for what it is: a partial lexicon to a particular group of texts. Nevertheless it is the only glimmer of light in a vast territory that would otherwise be completely dark. Lampe takes us some way into the Byzantine period, but for coverage in time after that, what do we have? Not very much. A ‘lexicon of Byzantine Greek’ has just begun to appear. The first part, published in 1994, reached a point somewhere in alpha.13 Inspection immediately reveals that this is not a full dictionary at all, but takes the form of a supplement to . . . LSJ, and also Lampe. So it is rather like Lampe, a partial coverage of a certain body of authors. Actually it seems to be mainly a list of additional words.14 Finally we have the lexicon of Kriaras, covering late Byzantine popular literature. The size of this valuable work gives some idea of what we are dealing with here: there have been 14 volumes since 1968, reaching παραθήκη in 1997. But again it is a question of trying to cover a certain portion of the data, a preliminary to a full-scale lexical treatment.15

13 14

15

Lexikon zur Byzantinischen Gräzität (LBG) [completed 2017]. [The older lexicon of Sophocles, Greek Lexicon of the Roman and Byzantine Periods (1887), ought to be mentioned here. It offers useful, but again partial, coverage of its period.] Kriaras, Λεξικό.

124

COLLECTED ESSAYS ON THE GREEK BIBLE

VI This is as far as our survey needs to go, and it is time to bring things together. The general picture, as I see it, is this. First of all, in terms simply of coverage, there is only one general lexicon of the period, namely LSJ – and it is primarily a lexicon of Classical Greek. Classical Greek is covered well; post-Classical Greek has only been dealt with by focusing on patches of the total area, and there are large areas still untouched. Further, this process of working through the texts piecemeal, even if it were to be completed, would still give us a series of lexicons, each dealing with its own segment and, as it were, cut off from the rest. The essential overview of all the evidence would still be lacking. I find it hard to judge at this stage how much DGE, the Greek–Spanish lexicon, will alter this situation. Secondly, there is the problem of the quality of the coverage. It is simply a fact that what has been done so far cannot be relied on. This does not mean that it is all badly wrong; it does mean that until a thorough check has been done, both to eliminate the mistakes of the past and to use the full resources now available, we cannot know for certain that what we find in front of us when we look up a word is sound. Even without doing any investigation, one knows that all is not well from the fact that if you are reading almost any text of the post-Classical period – sometimes even one from the Classical period – you will sooner or later come across something poorly dealt with, or not covered at all, by LSJ or any other available tool. At this point one might well object that it is a question of practicalities. The resources we have, and are likely to have, are governed by what is possible, and lexicography is very slow and difficult work. I am, of course, all too aware of this. Nevertheless, it seems to me important methodologically that we recognise, and not lose sight of, the ideal, the true goal to be aimed at, if we are to get things really right. And that is a complete lexical treatment of Greek from its beginnings right through to the end of Byzantine Greek (or better still into Modern Greek), making use of all the occurrences, to ensure that all the evidence is brought to bear at once. The whole history of a word hangs together, and it is quite misleading to think in segments consisting of particular authors or groups of texts. The practical difficulties of producing an actual lexicon done in this way are of course very great, but we must recognise this as the only sound method if we are to achieve a full lexical treatment of any word.

THE PRESENT STATE OF LEXICOGRAPHY OF ANCIENT GREEK

125

Let me add that I’m well aware that even if such an ideal lexicon is achieved, there will still be a need for smaller tools that will provide concise information – using glosses if need be – on particular authors and periods of Greek. But behind the handy tools must stand the ideal complete lexical treatment that has done the work thoroughly; the smaller works should be derived from it, not stand in its stead.16

VII In conclusion, let me turn to immediate goals. My suggestions may seem modest after the far-reaching critique you have just heard. I am not about to propose that we immediately begin work on a ‘Thesaurus of Ancient Greek,’ as was suggested and fairly hastily abandoned early in the twentieth century. I make only some limited suggestions, and hope that others will be stimulated to think about these matters and come up with proposals. As regards the NT and the LXX, I think it should be recognised that the work is not over. Now that LEH is complete, a thorough revision ought to be undertaken. The NT is more difficult, because existing lexicons are generally regarded as the last word. Nevertheless all is not well with the NT lexical tradition, and long-term plans for a complete overhaul are needed. As to LSJ, we all shrink from suggesting a major revision, knowing how huge the task will be. Nevertheless, sooner or later something must be done. The availability now of LSJ in electronic form (in the Perseus digital library) is a useful step, but it is what we do next that will be important. Are there any plans for re-analysis of the entries? Will new examples be sought via the TLG and added? The idea of having an LSJ database on line for anyone to add to or comment on is worth exploring. But unless there was some ongoing sifting, not to mention reorganisation of the accumulating material under each word, we might be no better off. Everyone would enjoy adding to such a database, but they would not enjoy using it. 16

An instance of the reverse pattern is seen in Greek etymological studies. We have massive, exhaustive, and authoritative reference books on Greek etymology – Frisk, GEW and Chantraine, DELG – but for most people they are virtually unusable. The practical tool derived from them, a ‘Concise Etymological Dictionary of Ancient Greek’ is yet to be produced. [There is now also Beekes, EDG, similar to the others in size and scope. The need for a concise etymological dictionary remains: cf. concluding remarks in Essay 20, ‘Etymological Follies.’]

126

COLLECTED ESSAYS ON THE GREEK BIBLE

I have two smaller and more immediate proposals. First, I suggest that we need to establish a world database to collect the results of all published work on individual NT words. This would be not just a bibliography, but a concise summary of the result of each study. It would have a lexical focus, and be linked to words, not passages. I am suggesting that this would be initially for the NT alone, but it could eventually cast its net wider. Secondly, a matter which is in itself trivial but causes major difficulty and frustration, the lack of a consistent system of reference, in particular the abbreviations, for all the texts that we deal with. There is no one method of abbreviation even for the books of the NT. I doubt if anything can be done about them, but for the rest a plea for common sense may be worth expressing. We need to set to work to establish standard abbreviations for all authors and texts in Greek. In the case of literary texts the excellent TLG Canon provides an authoritative list of authors and their works, but does not offer any abbreviations; here I think a valuable opportunity has been missed. As far as documentary texts are concerned, I can report that Horsley and Lee have produced a list of abbreviations for volumes of Greek inscriptions, which at the moment have no agreed system. This has appeared in Epigraphica for 1994, and we hope it is the first step towards a consensus, matching the standard list of Oates (et al.) for the papyri.17 To conclude. Clearly there is much to do; perhaps it seems to be too much. On the other hand, it is good to know that there remains interesting work to be done, if scholars are willing to take it on. And I for one am optimistic that in the long run even the most daunting objectives will be achieved. For the present, if we do nothing else, we can at least recognise the true state of affairs in Ancient Greek lexicography, and be cautious. Postscript Though there have been some useful developments, the situation remains fundamentally unchanged since this paper was written in 2004. A new lexicon purporting to cover Classical and later Greek (i.e., GE, translated from GI) appeared in 2015. It represents little advance on LSJ, from which it is ultimately derived (see Essay 26). The very recent Cambridge Greek Lexicon (2021) ‘is designed primarily to meet the needs of modern students’ of Classical Greek (p. vii). While it offers a new presentation of meanings, it is confined to a mainly Classical selection of authors and cites no references beyond authors’ names. Its usefulness remains to be tested. 17

Horsley and Lee, ‘Preliminary Checklist.’ Since the presentation in 1995, BDAG has adopted this list. [Also the SBLHBS, which describes it as an ‘indispensable checklist’: see 99–100.]

THE PRESENT STATE OF LEXICOGRAPHY OF ANCIENT GREEK

127

In the NT field, the Diccionario Griego-Español del Nuevo Testamento (DGENT), employing a carefully-designed definition method, began in 2000 and has reached the end of beta (in 2012). (Cf. remarks in Lee, History, 165–6 with n. 13.) García Santos, Diccionario del Griego Bíblico (2011), covering both the LXX and NT, is useful and fairly comprehensive. It employs a mixed method of stating meaning but is discerning in its sense-divisions. The major NT lexicons (BDAG, BRAA, LN) continue as they were. In the LXX field, a significant advance has been achieved in the publication of Muraoka’s A Greek-English Lexicon of the Septuagint (2009), with complete coverage of the LXX. This lexicon applies the definition method as fully as possible, with some mixing of methods. All the LXX data have been analysed afresh. While there are some flaws, and further areas to be covered (see Essay 15), this lexicon is the best guide to the LXX at present and the sound foundation for all future development of LXX lexicography. For further commentary and update on LXX lexicons see Aitken, No Stone Unturned, 9–15. Documentary texts, both papyri and inscriptions, have become available in electronic, searchable form online (via and ) and the databases are constantly being updated. Access to the texts and fast searching have become easy, but the lexicographical tools to understand the texts have not kept pace, in fact have not changed at all. The researcher has to rely on the same printed works (LSJ; Preisigke, Wörterbuch) as before. Translations (for papyri, not inscriptions) are intermittently provided in the electronic databases, but are not entirely reliable. My suggestions for a database have been developed further in Lee, History, 182– 3 and ‘Releasing LSJ’ (Essay 17); see also ‘Lexicographic Database’ (Essay 13). There is no sign yet of action to make such a thing a reality.

10 REVIEW OF MURAOKA, A GREEK–ENGLISH LEXICON OF THE SEPTUAGINT: CHIEFLY OF THE PENTATEUCH AND THE TWELVE PROPHETS (2002) 2004

Abstract This extended review of the second instalment of Muraoka’s Lexicon first sets it in its context among other lexicons, then describes its general features, illustrated by sampling of entries. The method of indicating meaning is next examined more closely, with extensive examples. The use of the ‘definition’ method, as opposed to that of the ‘gloss,’ is an important and welcome aspect of the Lexicon, and is generally applied with success. The point is made here that a ‘one-word definition’ is an acceptable means of defining, provided the distinction between it and a gloss is maintained. Some flaws and inconsistencies in Muraoka’s treatment are noticed and suggestions made for improvement. The term ‘mixed method’ is introduced, to characterise a method by which definitions and glosses are used together to state meaning, without clear distinction of one from the other. Muraoka is not alone in falling into this pattern: it is found in some well-known major lexicons. Muraoka’s use of explanatory additions and his approach to collocation are also examined. In the final assessment this lexicon is considered to be a good lexicon, which offers the prospect of development into the standard lexicon of the LXX. (Essay 15 is a review of the complete lexicon. It is intended to be read in the light of the present one.)

I It is given to few scholars to compile a lexicon, to even fewer to compile one to the LXX. Compared with the lexicography of the NT or of Ancient Greek generally, this has been the most neglected of fields. Until the late twentieth century, lexicons devoted entirely to the LXX, from the elementary effort of Rosenbach in 1634 to Schleusner’s classic of 1820–21, could be counted on the fingers of one hand. The reasons for the long gap after Schleusner in modern times are not difficult to discern,

130

COLLECTED ESSAYS ON THE GREEK BIBLE

but need not detain us here. The LXX was certainly read, or at least consulted, and there was great interest in it as an adjunct to other disciplines. But those who ventured into it had to rely on the guidance of the general lexicons of Greek, notably Liddell and Scott, with some help from the NT lexicons and their partial coverage of LXX material. That this was the cause of much inadequate if not mistaken understanding of the Greek of the LXX cannot be doubted. By the mid-1980s the pressure to address the problem had built up greatly and more than one project was begun or mooted. A kind of scrum developed, from which two players emerged. The competition can only have been beneficial, however much one might have preferred to see, in an ideal world, a single major project combining all resources. As so often in the history of lexicography, individuals simply set to work. Around the same time as Lust, Eynikel, and Hauspie began their LXX lexicon project in Leuven, Muraoka began work in Melbourne on a lexicon to one portion of the LXX, the Twelve Prophets. This was no small objective in itself, but was achieved by 1993.1 It was, as it turned out, only the prelude to a much larger effort culminating in the present impressive lexicon. In the new lexicon the earlier material on the Twelve Prophets is expanded to cover the Pentateuch, together with, for a significant proportion of the words, coverage of the whole LXX. With the Pentateuch included, the tipping-point has been reached and the lexicon is now a major tool for work on the LXX, even though of course much still remains to be covered. It is known that Muraoka intends to continue work along the same lines and add treatment of further LXX books, perhaps all of them, until a full lexicon of the LXX is achieved.2 This step-by-step approach is both acceptable and commendable, given the size of the task of producing a major lexicon of the LXX providing full data on every word. The alltoo-familiar alternative, of a project that sets out to cover everything and makes a start with a fascicle reaching a point somewhere in alpha, then takes another forty or fifty years to complete, is not inviting. 1

2

Muraoka, Lexicon (Twelve Prophets). The present reviewer’s contribution to that lexicon, generously acknowledged in the introduction, was made at an early stage, and ended well before the lexicon reached completion; the subsequent work on the Pentateuch has been entirely Muraoka’s own. Cf. Muraoka’s entry in ‘Record of Work Published or in Progress’ in BIOSCS 35 (2002), 33: ‘Next phase started. Finished Isaiah, now working on Jeremiah.’ Professor Muraoka has kindly informed me (by email 11 Jan, 2005) that he has now covered about 70% of the whole LXX.

REVIEW OF MURAOKA, A GREEK–ENGLISH LEXICON OF THE SEPTUAGINT

131

II Muraoka’s lexicon uses the traditional method of presentation, namely a list of headwords in alphabetical order, each accompanied by an entry that gives the meaning(s) of the word along with occurrences and a selection of other information. There are 4,478 headwords, of which 1,553 are treated for the whole LXX. The layout is pleasing and easy to read, with large print and conspicuous divisions. In each entry the following information appears: headword and its morphology; the sense(s) of the word, accompanied by full (though not necessarily exhaustive) listing of occurrences (many quoted and translated); range of application and syntactic breakdown; a list of semantically related words; brief references to secondary literature; a summary of the corresponding words in the Hebrew original. A set of special symbols is used (eleven in all) to indicate various things, for instance, that data are complete for the whole LXX. It must be admitted that the symbols are not easy to remember and most users will tend to ignore them, but they are a useful system for supplying information that may be wanted at some point. The great virtue of this lexicon is its offering of a full lexical analysis of each word. All the occurrences in the books covered have been examined and classified semantically. Splits into different senses are made when required, each of the senses is defined, and under each a thorough if not complete list of occurrences is given. Along the way one gets, as already noted, an indication of the various applications of the word and the syntactic relations it enters into. The framework thus established is very valuable not only for the books covered but also, in general, as a guide to the meaning of the word in any LXX occurrence not yet treated. It will also be the basis for the expansion of the lexicon to cover the whole LXX, when that is undertaken. There are inevitably gaps at present where a word happens to be used in a different sense in some book(s) not yet dealt with, and of course the lexicon can give no assistance with words that are not yet entered at all. But as far as it goes, and that is a long way, it provides an extremely good guide to the meaning of the LXX vocabulary. How good are the lexical analyses? My impression is that they are generally sound; I can detect few faults. Muraoka has adopted, at least in principle, a definition method of indicating meaning, avoiding the problems created by glosses, and ensuring a reliable breakdown and description of the senses. (More will be said on this topic below.) That Muraoka has tackled this task in a language not his first is highly commendable; the

132

COLLECTED ESSAYS ON THE GREEK BIBLE

reader cannot fail to be impressed by the results. There are times when one might quibble over details of English expression, but this is not a serious problem: the intended meaning is always sufficiently clear. On the question of taking into account evidence from Greek outside the LXX, that essential adjunct to deciding meanings, Muraoka’s approach is as follows. First, he has relied primarily on what has so far been gathered into the major reference works or noted in special studies of the LXX vocabulary: he has not attempted any fresh searching, as he himself explains (p. ix). This is an acceptable approach, given that further searching and assessment, even of a limited kind, would have slowed progress immeasurably. It must not be forgotten, however, that the existing collection of evidence on each word is not necessarily complete, and further collecting needs to go on, especially from the documentary sources contemporary with the LXX. But this is work for the future. Secondly, none of this evidence is reported in the entries (with some exceptions). This too I think is the right way to proceed: it is simply impossible to give an adequate summary within the constraints of the printed page. It is also better for the reason that what is reported in lexicons tends to become authoritative and immutable (like the meanings). Such data will be better assembled systematically, over a longer time, in an electronic database, as I have argued elsewhere in relation to NT lexicography.3 Some mention of parallels does, however, slip in, in a very sporadic way.4 I am not sure that this is a good thing, though one can understand the temptation to mention some particularly apt parallel. The risk is that it will be taken to be definitive. In regard to secondary literature, Muraoka has made use of existing studies as much as possible, but in this case the material does gain a mention in the entries, though somewhat sparingly. Muraoka explains (p. xvi) that mention in the entry implies that the study has made a ‘substantive contribution’ to determining the semantics of the word concerned. This makes good sense, and avoids the problems created by simply listing works that may or may not have anything useful to offer. But it can be unclear whether a particular article or discussion has been omitted because it has nothing to offer, or because it has been missed. Again, a collection in electronic form that was systematic and ongoing could keep track of everything, and solve the problem of keeping up to date. But this too is for the future. 3 4

See Lee, History, 182–6. So, e.g., s.v. ἀγαπητός an example from Homer is quoted; s.v. ἐκλύω 2 one from Aristotle; s.v. ἵστημι I.3 one from Demosthenes; s.v. ἐλπίς 2 we are told that ‘this sense is also attested by Polybius’; s.v. ἐγχάσκω a parallel collocation of εἰς with χάσκω in Philostratus is noted; and s.v. ἔλεος Aristotle’s definition of the word is quoted fully (in translation).

REVIEW OF MURAOKA, A GREEK–ENGLISH LEXICON OF THE SEPTUAGINT

133

A comparison with the other current lexicon of the LXX, LEH, is inevitable.5 First it needs to be said that LEH is compact, easy to use for a quick indication of meaning, and above all complete for the whole LXX (as things stand at present). It is also, paradoxically, quite thorough in its collection of references to secondary literature. But what LEH gains in completeness it loses in brevity of analysis. Though full statistics are given, the selection of occurrences for each word is small, and most important, there is no systematic analysis and breakdown of the senses. A series of glosses is offered, mostly derived from LSJ. These cannot adequately represent the range of meanings found in the LXX. In short, LEH is ideal as a smaller lexicon, with a different aim from Muraoka’s. It was a welcome holding effort produced at a time when there was nothing but LSJ and Schleusner, and such it will remain. But the future major lexicon of the LXX will need to be built on a different and better foundation, and Muraoka’s work is that foundation. III I offer a sample of words to show the character of the work and illustrate some of the points mentioned. (1) λόγος. Muraoka makes seven divisions in his analysis, as follows: 1.[a.]6 word spoken or uttered. b. of divine message communicated to a human or humans [sic plain text]. 2. report, news, rumour. 3. esp. pl., a chain of connected events, ‘a story, an account.’ 4. what is or may be a good reason. *5. course of action, step to be taken. 6. matter under discussion.

Under each of these senses, references and brief quotations are given, for example under 6 we have ‘ἐν τῷ ~ῳ τούτῳ “in this matter” De 1.32; a legal case, 22.20.’ The number of references ranges up to 24 in the case of sense 1. It is a little disconcerting to realise, however, that all of the references under sense 1 are to the Twelve Prophets and none to the Pentateuch, though of course the meaning occurs there. The references for this sense are simply continued unchanged (with one exception, the correction of Hos 1:3 to 1:2) from the 1993 lexicon, and no Pentateuch references 5 6

LEH, Lexicon (1992, 1996); LEH, Lexicon (2003). [a.] added by me for clarity; Muraoka mostly omits this ‘a.’

134

COLLECTED ESSAYS ON THE GREEK BIBLE

have been integrated into them. There is no great harm in this; but it could mislead the user into thinking there are no occurrences in the Pentateuch. The use of the asterisk, seen here in sense 5, ‘signifies that the word is not attested earlier than the LXX Pentateuch, i.e. the third century B.C.E.’ (p. xiv). The purpose is to alert us to a possible innovation in the Greek of the Pentateuch. Muraoka is careful to add that ‘words so marked do not have to be neologisms created by Septuagint translators.’ Besides new words, Muraoka marks previously unattested meanings in the same way, as in the case of λόγος 5. It seems likely that this sense is indeed a neologism and has arisen from stereotyped rendering of the Hebrew word (‫)דבר‬, but Muraoka does not foreclose on the question: by the use of the asterisk he simply points out the absence of earlier evidence for this sense and allows the user to draw a conclusion (or not). In the analysis of λόγος, the divisions seem well taken and show subtlety in detecting differences not obvious at first sight. The division at 1.b however gives one pause. Is this a separate lexical meaning, or simply a difference of reference under the one lexical meaning word spoken or uttered? Comparison with other entries suggests that these divisions usually indicate differences of reference, syntax, or collocation, not a new lexical meaning, and that would seem to be what is intended here; but sometimes they do offer a separate lexical meaning, as, e.g., in the case of δίδωμι 2.d ‘to entrust for temporary safe-keeping.’7 This is a matter that needs clarification and a consistent policy. As to completeness of coverage of the meanings of λόγος, one might wonder where Exod 18:19 (καὶ ἀνοίσεις τοὺς λόγους αὐτῶν πρὸς τὸν θεόν) is intended to fit. The meaning required in the context seems to be ‘matter in dispute,’ something different from sense 6 and its cited occurrences (even Deut 22:20, which seems to me to belong in sense 1). This example highlights the problem of bulk that the lexicographer constantly faces. Muraoka has dealt with the 110 or so occurrences of λόγος in the Twelve Prophets and the Pentateuch as best he can, perhaps missing even some of these, but to cover the whole LXX there are still over 1000 to go! (2) κάτεργον. The meaning given, operating costs (for Exod 30:16; 35:21), has evidently benefited from the evidence of the word in documentary sources: LSJ give examples only in III BC papyri and the Pentateuch, 7

Other examples like δίδωμι s.vv. ἄρχω 3.b; αὐξάνω I.b; ἐκτός 3.b; ἐπικαλέω B.1.a, b, c, d; κινέω A.b, c. I do not find this explained in the relevant section of the Introduction (xiv–xv).

REVIEW OF MURAOKA, A GREEK–ENGLISH LEXICON OF THE SEPTUAGINT

135

offering the glosses wages … labour-costs … service (sic for Exod 30:16; 35:21). But we also see the difficulty of keeping track of sources and coverage of the evidence. Muraoka refers to Wevers on Exod 30:16 as his source (where there is no mention of papyri); the similar discussion in Le Boulluec and Sandevoir is not noted (where there is a brief mention); none of these writers refers to LSJ, which must be their ultimate source for the meaning. A full examination of the evidence is presumably still to be done: LSJ cite only five papyri; there are in fact up to 120 occurrences, 50 of them in III BC texts. Such an investigation is likely to lead to a more refined understanding of the word’s meaning in the time of the Pentateuch (e.g., does it refer to wages strictly, or more broadly to expenses including wages?)8 (3) παιδίον. Connotation, a feature difficult to handle in the traditional method of lexical description, comes into the picture for this word. Muraoka’s solution here is first to define (young child) and give the usual breakdown of occurrences, then add a separate note: ‘The feature of endearment and affection appears to be present in some cases: e.g. … [examples].’ Such notes do not appear often. In the case of τέκνον there is a different treatment. Four senses are given first with normal definitions, then comes ‘5. an endearing address to a youth: … Ge 43.29 (Joseph to Benjamin).’ Strictly speaking this implies that there is no semantic content, only connotation, which may be right. But what of the two examples in 1.b? These fall under the definition 1. immediate offspring, with only the following description: ‘b. voc. without a posses. pron.: τί ἐστιν, τέκνον; “What is the matter, child?” Ge 22.7; 22.8.’ They miss out, somewhat unexpectedly, on any special connotation. The whole subject is clearly one well worth future debate and experiment.9 (4) λαμπάδιον. Muraoka has changed his mind on the form of the headword: it was λαμπαδεῖον in the 1993 lexicon and is now λαμπάδιον, for what reason one cannot deduce. The only secondary literature mentioned in both editions is Walters, Text (1973). There is no reference to 8

9

Wevers, Exodus, 496. Le Boulluec and Sandevoir, L’Exode, 309. LSJ shows the impact of the documentary discoveries early in the twentieth century: in LS 8th ed. (1897) κάτεργον was cited only from the Pentateuch, with the meaning work. Preisigke’s somewhat baffling contribution in Wörterbuch (1925), s.v. will also need to be assessed: ‘1) Arbeitslohn … 2) Leistungsmaß, Arbeitsnachweis … 3) Lehenbeartungsdarlehen. …’ The term ‘connotation’ is used here as defined in Zgusta, Lexicography, 38–41, not as Caird uses it in ‘Towards a Lexicon. I,’ following John Stuart Mill. [Zgusta (38): ‘We can describe connotation as consisting of all components of the lexical meaning that add some contrastive value to the basic, usually designative function.’]

136

COLLECTED ESSAYS ON THE GREEK BIBLE

Boyd-Taylor’s lengthy and penetrating discussion, bearing on the meaning as well as the form of the word.10 This may have appeared too late to be taken into account, but will be relevant to future treatment. The examination of the LXX vocabulary is young and ongoing: almost every word is potentially on the brink of further elucidation by a new study. (5) διαστέλλω, διαστολή. Muraoka has a good analysis of the tricky verb διαστέλλω (but aren’t they all?), rightly recognising a sense 4. to state precisely in Gen 30:28; Lev 5:4 bis, along with five other senses. But things seem to have gone wrong with the occurrence of the noun in Num 30:7, where a parallel nominal sense such as precise statement seems required by the context (and where the Hebrew has the noun corresponding to the verb in Lev 5:4 bis). Muraoka has placed this example of διαστολή under 2. express orders or instruction, along with Num 19:2 (where it is right), but this does not make proper sense in Num 30:7. Muraoka refers to Caird’s old but decisive discussion, and to the more recent one of Dorival. The latter seems to have put him off course. Dorival (and his colleagues) saw three possible senses, but could not come to a decision between them and opted for the translation disposition, ‘qui peut avoir ces trois sens.’ This stratagem is hardly helpful to the lexicographer trying to decide the lexical meaning. Fortunately Muraoka has ignored another discussion of Num 30:7, that of Wevers, which could only have created deeper confusion.11 This instance could serve as a model of the peculiar demands of preparing an entry in a LXX lexicon. To reach a result, one must assess the entries in the major lexicons (the splendid confusion of LSJ, partially corrected in later works; the entry in the very recent DGE, which in the case of διαστολή is six or seven times the size of LSJ’s); make sense of the now generous but still somewhat raw commentary tradition; seek out other studies and try to absorb their conclusions; search for and assess, as far as one can, the evidence of the word in texts both literary and documentary outside the LXX, not trusting older collections that may be out of date; take into account the Hebrew equivalent in the source text; notice parallel formations and passages; compare the meaning of semantically related words; and check for equivalent words used in the context (in this case ὁρισμός is relevant). All this can be a confusing – and time-consuming – 10 11

Boyd-Taylor, ‘Evidentiary Value’ (2001). Caird, ‘Towards a Lexicon. I,’ 467–8; Dorival, Les Nombres, 513; Wevers, Numbers, 497–8. Yet another meaning appears in DGE, vol. 5 s.v., where Num 30:7 slips in at the end of διαστολή C II 2 glossed as compromiso (= ‘commitment’?).

REVIEW OF MURAOKA, A GREEK–ENGLISH LEXICON OF THE SEPTUAGINT

137

mixture to absorb. In the end, the only guide through the maze is one’s own sense of the likely meaning in the context. Muraoka is as aware as anyone of these demands, and has coped with them to great effect. But no one should imagine that there is no work left to be done.12

IV I turn now to a closer look at the definition method employed in Muraoka’s lexicon. My remarks are not intended to amount to a negative assessment of the work as a whole but made with a view to possible improvement in future editions, and as further reflections on the current debate over method in stating meaning in Greek lexicons. Muraoka describes at the outset (pp. xii–xiii) the method he intends to apply. Instead of giving a ‘translation equivalent,’ the problems of which are succinctly summarised, he has ‘chosen to go for definition, to describe senses of a given word in sentence form or as fragments of a sentence.’ A definition is formated in italics; occasionally a translation equivalent will be added, in plain text within single quotes. So far all is clear. The definition method will be used, and a definition is understood to have a form involving the use of more than one word. Anything else will be a translation equivalent and formated differently. But then Muraoka adds this statement: ‘Where there is no or little likelihood of misunderstanding, we have given translation equivalents: e.g., τροϕός wet-nurse instead of woman employed to suckle another’s child.’ This step causes confusion. It means that a translation equivalent (or ‘gloss’) will sometimes take the place of a definition proper, but will be formated in italics like a definition, even though it is still a gloss. It seems that the only way we will know it is a gloss and not a definition is that it will be a single word, since a definition is a ‘sentence’ or ‘fragment of a sentence’ and a gloss is by implication one word. But the user will have difficulty maintaining, or seeing any point in, the distinction. In effect what Muraoka is saying is that a single word can serve as a definition in certain cases. It would have been better to formulate it that way. To me this is a perfectly acceptable 12

I do not enter here into the issue of the theoretical basis of LXX lexicography, the focus of Boyd-Taylor’s stimulating paper (‘Evidentiary Value’). I would say only that in my opinion the ultimate criterion for deciding the meaning is the intention of the translator, who works with Greek as he knows it and as he expects readers and hearers of his own time to understand it. We cannot, of course, know the intention of the translator: we can only make the best deduction from the indications available.

138

COLLECTED ESSAYS ON THE GREEK BIBLE

method of defining in a Greek–English dictionary for English speakers, provided there is no ambiguity or lack of clarity in the chosen word, as in the case of Muraoka’s example, wet-nurse, which needs no further unpacking. When such a statement of meaning is used, it is to be regarded, not as a translation equivalent or gloss, but as a one-word definition.13 In practice however Muraoka does not keep consistently to his stated method. From what has been said in the introduction, we expect to find each word or sense defined by means of a word or phrase – one of either – formated in italics; a translation equivalent in plain text and quotation marks might sometimes be added. This would be the strict definition method. Such definitions do occur, as in: δίκαιος 2. conforming with set and agreed standards: δικαιόω *1. to pronounce innocent, ‘to acquit’: ἐϕοδιάζω to furnish with supplies for a journey: ἔτος year: θηλυκός female: πεῖρα 1. attempt: … 2. testing: … 3. experience:14

But what are we to make of the following?: καύχημα that which makes one proud, object of pride: μακρόθεν 2. at a distance, at a remove: παράϕρων wandering from reason, out of one’s mind: πορεύομαι II.3. to conduct oneself, follow a certain moral life-style:

In these we have two definitions, each one, presumably, helping to explain the other. In the same way, we often find the meaning stated by means of not one, but two or even three single words, as in: θέλημα desire, interest: λογίζομαι 4. to reason, consider: λογίζομαι 5. to deem, consider: παίγνιον plaything, toy: περιστερά dove, pigeon: ἐπιθυμέω to desire, long for, covet: καλός 1. advantageous, beneficial, desirable: καταϕρονέω to despise, belittle, treat with contempt: ὀπή opening, hole, crevice: 13

14

[Italics added, 2021.] Cf. the discussion in Lee, History, 22. Anne Thompson, in her review (2003) attributes to me (114) the term ‘definition gloss’ for such a one-word definition. I did not use it in the book, and would be reluctant to accept it: I think it is liable to be misleading. Colons are retained from the original, where they mark the end of the statement of meaning, before other material begins.

REVIEW OF MURAOKA, A GREEK–ENGLISH LEXICON OF THE SEPTUAGINT

139

πεδίον plain, open land or field: ποιέω 4. to fashion, construct, manufacture: πόνος 1. toil, suffering, hardship: πορεία 2. journey, going, marching:

Are these really one-word definitions? That is, in Muraoka’s terms, translation equivalents that have replaced a definition because there was ‘no likelihood of misunderstanding.’ It seems unlikely that they are. Rather, as the use of multiple equivalents suggests, they are in fact oldfashioned glosses that do not have the character of definitions. In addition to these, we find cases in which there is a combination of phrase and single word; that is, apparently, definition plus gloss, but with both in italics, not just the definition: δορά skin of body when taken off, hide: ἐκεῖ 1. there, in that place already mentioned: καταλαμβάνω 1. to lay hold of, seize: πλῆθος 1. large quantity, multitude: σιωπή silence, abstinence from speech:

Let us take stock at this point. What we have seen shows that glosses, the traditional method of indicating meaning in lexicons, have maintained their hold in Muraoka’s lexicon and are being used along with definitions, in a variety of combinations. This is a mixture of methods, and Muraoka is not alone in using it. The mixed method, as I think it deserves to be called, turns up often elsewhere. It appears at times in a predominantly definition lexicon like the OLD, as in: ingens 1 Of very great size or dimensions, huge, vast. insanus 1 Of unsound mind, demented, frenzied, mad.

Conversely the DMLBS, which is primarily a gloss lexicon, at times mixes the methods, as in: gratanter 1 with joy or pleasure, thankfully. improvisus 1 appearing or occurring without warning, unexpected.

The mixed method also appears occasionally even in the OED, the premier definition lexicon and pioneer of the method. For example: glad 1. Bright, shining, beautiful. subside 1. intr. To sink down, fall to the bottom, precipitate.15 15

The mixed method may be seen in Chadwick, Lexicographica Graeca, e.g., 87, 100, 207, 261. It is to be noticed that the mixed method is not used in BDAG (2000): definitions are always typographically distinguished from glosses, even though the two may occur together. The strict definition method is also followed in Louw and Nida, Lexicon. My

140

COLLECTED ESSAYS ON THE GREEK BIBLE

Can the mixed method work? Many would say yes, and one would have to concede that no great harm is done, e.g., by defining περιστερά as dove, pigeon. The user gets, as it were, two hits of information which together, presumably, are clearer than one – though one could say, even in this simple case, that we are just being given an older and a newer word for the same thing and there is no need for both. Even so, no great harm is done, nor perhaps in πόνος 1. toil, suffering, hardship, or σιωπή silence, abstinence from speech. But we should be aware of the hazards of this method when close attention to the meaning is needed in exegesis or for other purposes in reading a text. Take the case of ἐπιθυμέω to desire, long for, covet, quoted above. It matters very much what exactly this word means in Exod 20:17 and Deut 5:21 (οὐκ ἐπιθυμήσεις τὴν γυναῖκα τοῦ πλησίον σου. οὐκ ἐπιθυμήσεις τὴν οἰκίαν. …). Simply glossing it in this way (the glosses, incidentally, are the same as LSJ’s) fails to resolve the question of what it actually means, in particular whether it implies sexual desire. The English words could have more than one meaning as well as differing from one another, so all sorts of possibilities are left open by stating the meaning in the form of these three glosses. The exegete may or may not be aware of this, and may proceed to decide the meaning on the basis of the English words as he or she understands them.16 But this is not by any means the end of the variations that Muraoka allows himself in the statement of meaning. A frequent and significant element, though one on which no information is offered, are words in plain text joined with the definition or gloss in italic. Examples: ἀϕαιρέω 1. to remove and cause to disappear: βρῶμα 1. food, not cooked: γερουσία assembly of elders as a decision-making body in ancient Israel: διαβιβάζω to cause to cross from one side to the other: ἐπέχω 1. to wait without proceeding to next action: ζωή 1. life, vitality, inclusive of non-physical dimension: ζωογονέω 2. to preserve alive without killing: θῦμα slaughtered animal: κατάλοιπος left remaining untouched, unharmed or not mentioned: κατανοέω to observe closely to find out about: κῆτος huge sea-fish:

16

own thinking and practice have undergone development: the analysis of συνίστημι in Lee, ‘Συνίστημι,’ 3–4, shows the same faults I am criticising in Muraoka. The meaning of ἐπιθυμῶ is a crucial question for Loader’s discussion of the Decalogue in Sexuality, 5–25, but he seems unaware of the inadequacy of a gloss like ‘desire.’ He would get no help from Muraoka. Notice further that ‘desire’ also appears in Muraoka’s meanings for θέλω 1. to desire, wish: and βούλομαι to wish, desire: What does ‘desire’ mean there and is it the same as in ἐπιθυμῶ?

REVIEW OF MURAOKA, A GREEK–ENGLISH LEXICON OF THE SEPTUAGINT

141

κόσυμβος decorative fringe of a garment: κράσπεδον tassel on the four corners of an ancient Israelite’s outer garment: λανθάνω to escape notice of, mostly of inadvertent omission on the part of the beholder: λοιπός left and remaining out of a given number or quantity: πειρασμός putting of character to the test: πένθος grief, esp. over the dead: σαλπίζω to blow a trumpet:

The formating suggests that these are explanatory additions not intended to be part of the definition, but amplifying it in some way. Yet many of them do seem to be part of, indeed an essential part of, the definition proper. So, for example, in θῦμα slaughtered animal, where it is surely clear that animal alone is not sufficient to define the word. Similarly κῆτος is probably not just any sea-fish, but one that is of a specially large size, or huge. On the other hand, some of these examples may have to do with indicating the collocation, that is, the word(s) with which the defined word is or may be combined in context. Such may be the case in πειρασμός putting of character to the test, though it is hardly clear. Sometimes the explanatory addition is bracketed, as in: τηρέω to watch (to attack at an opportune moment). There are also instances in which the whole definition is in plain text. At first one might think these are simply the result of a formating slip, but I am not so sure: some of them look like explanatory material that has invaded the whole definition. Examples: ἁϕή 2. part of body infected by leprosy: γραμματεύς 2. a person in leadership position of one sort or another: [contrast 1. one who records] καρδία 3. seat where human thoughts, intentions and attitudes are generated and take shape: λόγιον 3. woven piece of an ephod placed on the breast worn by the high priest and used to arrive at legal decisions by means of δήλωσις καὶ ἀλήθεια, ‘oracle box’: ὄνυξ 3. a precious stone, ‘onyx’:

I note also instances in which definitions include bracketed explanatory words that are also italicised, for example: ἐπιθύω to offer (heathen, cultic) animal sacrifice: κοιμάομαι 1. to lie (down): οἰκογενής 2. belonging to the (royal) household:

Finally we come to collocation and how it is treated by Muraoka. The best and clearest method, as it seems to me, is the one seen in examples like these:

142

COLLECTED ESSAYS ON THE GREEK BIBLE

ἡγέομαι 1. to direct (organisation, group of individuals): μανθάνω 1. to learn (a skill, art, habit): ὀρθρίζω 2. to seek and turn in eager anticipation (to sbd, πρός τινα):

The collocation is in plain text enclosed in brackets, and not part of the definition itself. But we find numerous variations, with the brackets sometimes omitted, the collocation sometimes in italics, and elements of definition sometimes treated as if part of the collocation (as in μολύνω): ἀϕαιρέω 2. to shift from place X to Y: εἰσακούω 1. to give ear to sbd or sth: ἐξαποστέλλω 5. to part with, give a send-off to guest or visitor: εὐλαβέομαι 2. not to venture to do, to refrain from doing out of a sense of awe: καλύπτω 1. to place sth over an entire surface: λατομέω 1. to dig a cavity in the ground: λογίζομαι 7. to put down to sbd’s account: μολύνω 2. to violate (a woman sexually): περιαιρέω II. mid. to remove from oneself or sbd close to oneself: περιζώννυμι I. act. to put (garment) round sbd’s body:

V In the last two features discussed, I do not see any fundamental problem. In the case of collocation it is only a matter of applying consistent formating. As to the explanatory elements of definition currently formated in plain text, further thought is needed. The simplest solution would be to change all of them into italic format as being part of the definition, if they can be so regarded. If not, at least an explanation of their function needs to be formulated and a consistent policy maintained in their use. The question of the ‘mixed method’ is more difficult. It would help at least if the distinction between a one-word definition and a translation equivalent or gloss were maintained. Further revision along the lines of conversion of glosses into definitions (multi-word or one-word) and removal of superfluous glosses would be a larger enterprise. It has been truly said that it is not possible to make a good lexicon out of a bad one. In the case of Muraoka’s lexicon the possibility exists of making a good lexicon even better. These remarks are offered as a contribution to that end.

11 DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT TEXT OF THE COMPLUTENSIAN POLYGLOT 2005

Abstract This paper first notices the puzzled descriptions of the accentuation of the Greek NT text of the Complutensian Polyglot (1514), then goes on to point out the unobserved fact that the accentuation follows a monotonic system almost exactly the same as that now in use in Modern Greek. Next is considered the information on the matter in the preface to the volume. The Greek text of the preface is presented with English translation and notes. A number of current misconceptions are dealt with. The question of the identity of the inventor of the accentuation is then explored in full. The evidence in favour of Dimitrios Doukas as editor of the text and author of the preface is summarised and augmented. The paper then argues that it was he who conceived and applied the system of accentuation. Possible other sources of the idea are considered and eliminated. Finally the question of who might have been behind the initial intention to print an unaccented text is discussed.

1. DESCRIBING THE

ACCENTUATION

The fifth volume of the great Complutensian Polyglot, dated 1514, contains the Greek text of the NT, the first printed (though not the first published) edition. As such it has been of enormous importance, and the subject of scholarly scrutiny, ever since. The authorship, the nature of the text, the manuscripts on which it was based, and the typography have all received close attention, and the questions associated with them have been to a great extent unravelled. But the accentuation used in the NT text has been observed with some puzzlement, even disapproval, and not pursued further.

144

COLLECTED ESSAYS ON THE GREEK BIBLE

Metzger’s standard work on the text of the NT offers this statement: [The Polyglot New Testament] is printed without rough or smooth breathing marks and is accented according to a system never heard of before or since: monosyllables have no accent, while the tone syllable in other words is marked with a simple apex, resembling the Greek acute accent mark.1

The equivalent work of the nineteenth century, that of Scrivener, though fuller, is in similar vein (and also clearly Metzger’s main source2): The Greek type in the other volumes is of the common character, with the usual breathings and accents; in the fifth, or New Testament volume, it is quite different, being modelled after the fashion of manuscripts of about the thirteenth century, very bold and elegant … without breathings, and accentuated according to a system defended and explained in a bilingual preface πρὸς τοὺς ἐντευξομένους, but never heard of before or since: monosyllables have no accent, while in other words the tone syllable receives the acute, the grave and circumflex being discarded.

In a subsequent note, on the laudatory verses included in volume 5, Scrivener quotes approvingly those of ‘the native Greek editor, Demetrius Ducas,’ in which full accentuation and breathings are displayed, and adds this comment: ‘the fantastic mode of accentuation described above was clearly not his work.’3 Another comment, itself somewhat fantastic, is found in a 1953 paper by Tasker: The regular system of breathings and accents, the bane of all printers of Greek, is not used; but acute accents are inserted in an unparalleled manner on the tone syllable in words of more than one syllable, as though Spanish and not Greek was being printed.4 1

2

3

4

Metzger, Text of the NT, 97. An earlier, shorter version of this paper was presented at a Language Colloquium in Memory of G. P. Shipp at Macquarie University, 16 May, 2003, organised by Trevor Evans. I thank Vrasidas Karalis for his reactions and helpful information at an early stage. I am especially grateful to Michael Curran, who read a late draft of the whole article and put my arguments to the test by many acute observations. I am indebted to a number of others for assistance, as mentioned in the appropriate places. Along with Tregelles, Printed Text: cf. his statement (p. 10) that the editors say ‘that they have marked the tone-syllable of each word with a simple apex.’ The ‘simple apex’ in turn derives from simplex tantum apex in the Polyglot NT preface (Latin version). Scrivener, Plain Introduction, 2:177, 178 n. 1. Scrivener was quite wrong about the authorship of the ‘fantastic mode of accentuation,’ as we shall see. Tasker, ‘Complutensian Polyglot,’ 204. The fact that this was originally ‘a lecture given to the Spanish Department of King’s College London and others’ may go some way to explain Tasker’s facile remark.

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

145

Even Geanakoplos, a much more sensitive observer, had only this to say in 1962: Anticipating the reader’s questions as to the differences in the Greek type utilized in the printing of the Old and the New Testament volumes, the editor carefully explains why it was chosen to employ Greek characters without accents. (Actually it would be more accurate to say that a peculiar system of accentuation was employed, the acute accent alone being used and only on polysyllabic words, while aspirates were entirely omitted.)5

Basil Hall, a year later, uses the same key word ‘peculiar’ to characterise the system: The accentuation of the Greek was peculiar: it was justified by the editors on the ground that it formed no part of the genuine text and was absent from older manuscripts; and no ‘breathings’ were provided (yet the editors gave the normal accentuation to their text of the Septuagint).6

Most recently (1990) M. A. Screech offers, in passing, this puzzling remark: When we turn to the beautiful text of the Complutensian New Testament (with its absence of breathings and its accents designed to show tonic stress not modern Greek usage) we read. …7

In these discussions the writers have attempted to describe the accentuation in the NT text of the Polyglot without an appreciation of its true nature. It is in fact a monotonic system, one that is almost exactly the same as that now in use in Modern Greek, which was officially introduced in 1982.8 At this point it will be helpful to see a specimen of the Polyglot text. An illustration showing the beginning of the Gospel of John (1:1–14) is given below. This short extract is sufficient to demonstrate the main 5

6

7 8

Geanakoplos, Byzantium, 240. Whether peculiar is intended to mean ‘strange, odd’ or simply ‘all its own’ is not clear; either way it is inappropriate. CHB, 58–59. For Bentley in 1983 (Humanists, 92) ‘the Greek type is a bit odd’ in its omission of ‘all breathing marks and most accents’; he further speaks of ‘the inconvenience caused’ by the omission (to whom?). The following report the bare facts: Gregory, Textkritik, 2:925; Nestle, Introduction, 2; Darlow and Moule, Historical Catalogue, 2:4. There is no remark in Aland and Aland, Text of the NT, 3–4. Tregelles in 1854 had also used the word peculiar: ‘In their preface, the editors refer to the peculiar manner in which they had printed the Greek’ (Printed Text, 10). The views of Proctor and others regarding the type are discussed below. Reeve and Screech, Erasmus’ Annotations, xv (Screech). For a full statement of the system, see, e.g., Triantaphyllides, Γραμματική: Αναπροσαρμογή, 22–26; Holton, et al., Greek, 38–40.

146

COLLECTED ESSAYS ON THE GREEK BIBLE

characteristics of the system and to confirm that it is indeed monotonic, as the term is applied today. That is, a single accent mark, the acute, is used, and it is employed to mark the stress in words of more than one syllable. Further comment on features of the system as seen in the Polyglot will then be made.

The text of John 1:1–14 in vol. 5 of the Complutensian Polyglot (Alcalá, 1514). Image from the holdings of the Biblioteca Nacional de España.

2. FEATURES OF THE

ACCENTUATION

The mark of accentuation is clearly the traditional acute, not something resembling it but in some way different, as implied by Metzger. It is indistinguishable from the acute used, for instance, in the NT lexicon at the end of volume 5, with full accentuation.9 9

Longer and shorter varieties of the acute accent mark are detectable. In the preface it is nearly always the longer, rarely the shorter; in the NT lexicon it is the other way round; in the NT text I have not detected any of the longer form (in a very small sample).

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

147

Monosyllables are generally not accented, but sometimes they are. In the specimen text we have ο λόγος σάρξ εγένετο, with an accent on σάρξ. I have investigated this phenomenon by extensive sampling. My finding is that while the great majority of monosyllables are not accented, a substantial minority are. Those that are accented are mostly nouns and verbs, occasionally other parts of speech. There is no evident system. I interpret this phenomenon as reflecting the actual pronunciation of the phrases concerned, in which the accented word, though monosyllabic, would in fact have a stress on it and not be pronounced like an unemphatic monosyllable (mostly the old proclitics and enclitics such as εις, το, μου). The person responsible for marking the accents intuitively placed a mark on a stressed monosyllable from time to time, contrary to what the system provided, because oral realisation of the text was too strong to ignore; but it was done inconsistently and no regular system for dealing with such cases was developed. This flexibility in accentuation of monosyllables, though unsystematic, parallels to some extent the Modern Greek system, where in specified cases monosyllabic words are written with an accent.10 Some examples of accented monosyllables in the Polyglot text: όπου θέλει πνεί (John 3:8) μέλλοντες πλείν (Acts 27:2) οίνω ολίγω χρώ (1 Tim 5:23) συ τις εί (John 1:19, 8:25) είπον ουν αυτώ τις εί (John 1:22) συ εί ο διδάσκαλος (John 3:10) ηλίας ει σύ (John 1:21), but ο προϕήτης ει συ (1:21) τον βούν αυτού (Luke 13:15) ου ϕιμώσεις βούν αλοώντα (1 Cor 9:9), but βουν αλοώντα ου ϕιμώσεις (1 Tim 5:18) υμείς εστέ το ϕώς του κόσμου (Matt 5:14), but το ϕως των ανθρώπων (John 1:4, in the specimen passage) είδε ϕώς μέγα (Matt 4:16) λευκά ως το ϕώς (Matt 17:2) και χείρ κυρίου ην μετ αυτού (Luke 1:66) αυτού ϕλόξ πυρός (Rev 19:12) πρίν η συνελθείν (Matt 1:18) πού έθηκαν αυτόν (John 20:2), but που έθηκαν/ας αυτόν (20:13, 15) πού σοϕός (1 Cor 1:20) εκ της ρούθ (Matt 1:5)

10

See Holton, et al., Greek, 38–9. Modern Greek in fact accents πού the interrogative, as in the Polyglot samples below; but not σαρξ or the others.

148

COLLECTED ESSAYS ON THE GREEK BIBLE

The handling of enclitics also calls for comment. The practice of the editor(s) is to ignore the special rules of accentuation of enclitics, and treat them as ordinary words. Disyllabic enclitics, then, invariably have an acute on the second syllable, regardless of what precedes, and words before monosyllabic enclitics never receive an additional accent on the final syllable. In the former case the resulting markings, though mostly contrary to the enclitic rules, may well coincide with pronunciation in practice. So εν εισίν (1 Cor 3:8), σαρκικοί εστέ (3:3), άγιος εστίν (3:17), υμείς εστέ (1:30), οποίον εστί (3:13), των ανθρώπων εστί (1:25), εγώ μεν ειμί (1:12). But the total absence in the latter case is more unexpected. It means that in instances like απέστειλε με (1:17), το κήρυγμα μου (2:4), πνεύματι τε (4:21) the accent marks failed to help the reader place the second stress where it occurred in pronunciation, assuming it did so then as it does today, i.e., απέστειλέ με, etc. (There can be no question that the editor(s) knew the relevant rules: see, in the preface below, γνωμώνιόν τι (l. 35), ἔδοξέ τι (l. 39), and the accentuation throughout the LXX text of the Polyglot.) One must suppose that the editor(s) regarded this as a refinement that was not worth attempting in a reduced system of accent marking.11 Besides the acute accent, one other diacritic is used, the diaeresis. This is exactly as in the modern system. A few examples: ησαΐας (John 1:23), βηθσαϊδά (1:44), νοΐ (1 Cor 1:10), γάϊον (1:14). The iota subscript is omitted altogether, as can be readily seen from the opening words of the sample, εν αρχή. This is not a new step, but was already standard practice at the time, even in texts written or printed with full polytonic accentuation.12 What we have in the NT text of the Polyglot is a fully thought-out and effective system of monotonic accentuation, consistently applied apart from occasional variations in the treatment of monosyllables. It is not some unheard of, as if to say outlandish, idea without a place in the history of the language and unworthy of serious notice.

11

12

For modern survival and writing of the accent pattern in απέστειλέ με, etc., see Holton, et al., Greek, 30–1. As, e.g., the preface to the NT and the LXX text in vols. 1–4. The letter υ often appears in the NT text with a small dot between the prongs. This is clearly not a breathing. I take it to be an accidental feature of one of the sorts of upsilon in the Polyglot font. It occurs not only initially but in all positions; it is even found with an accent added above (see, e.g., Παύλος in 1 Cor 1:13). I notice the same sporadic dot in eta and omega. This then will account for Tregelles’ assertion that breathings are omitted ‘except in the case of Υ’ (Printed Text, 10).

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

3. THE ACCENTUATION

149

IN HISTORICAL PERSPECTIVE

Let us now try to put this remarkable feature of the Polyglot into historical perspective. The official introduction of the monotonic system of accentuation in 1982 was preceded by a long period of experimentation and debate in the nineteenth and twentieth centuries. Its competitor was the traditional polytonic system inherited from late antiquity, which had been preserved all through into the modern period and was the prevailing system, with some modification, until that time; it can still be seen today in formal writing. Debate over the accentuation (and use of breathings, iota subscript) formed part of the larger language question, that is, what the modern language actually was – or ought to be – and what its orthography ought to be. The first attempt at reform of the traditional method of writing the accents appeared in 1814 with the work of Y. Vilaras, who published texts without any accents and breathings. By 1900 a monotonic system essentially the same as that in use today had been proposed by D. Melandinos, which gathered supporters as the century progressed, culminating in the result we have indicated. Such a system is practical, economical, and adequate, removing superfluous distinctions while not abandoning accent marking entirely.13 The movement towards a revised system of accent marking begins in modern times no earlier than the nineteenth century. The Polyglot NT thus anticipates by 300 years the earliest modern proposal for change, and by over 450 years the establishment of a full monotonic system. This is not to suggest a connexion, but to observe the surprising but apparently unnoticed fact that someone in the early sixteenth century arrived at exactly the same solution as modern thinkers on the subject. In its own time the accentuation of the Polyglot NT was, as far as I can discover, quite without precedent. Other early printed Greek texts show the traditional system, or, less frequently, complete omission of markings.14 13

14

For a historical summary of the question, on which I rely here, see Argyriades, Νεοελληνική Γλώσσα, 154–9. A specimen of Vilaras’s orthography may be seen in Triantaphyllides, Γραμματική 1: Ιστορική Εισαγωγή, 443–4. I am grateful to Elizabeth Kefallinos of Macquarie University for drawing my attention to the former work. Numerous illustrations may be seen in: Proctor, Printing of Greek; Scholderer, Greek Printing Types; Morison, Politics and Script; and Barker, Aldus Manutius. Barker’s collection of 46 illustrations includes script as well as type of the period: none show anything but the polytonic system. See also, e.g., the facsimile edition of Lascaris’s grammar (Amsterdam, 1966); Geanakoplos, Byzantium, 206; Staikos and Sklavenites, Πεντακόσια Χρόνια, 11–17. For texts without any accent or breathing marks, see esp. Proctor, figures 2–4, 6, 9, 10, 13, 20, 26, 28, 31, 35, 38. Some of these are the earliest attempts at printing Greek, with a few words of Greek embedded in Latin text. On fig. 31, which is

150

COLLECTED ESSAYS ON THE GREEK BIBLE

The specimen page of a polyglot projected by Aldus in 1504 uses the traditional system, and is therefore no forerunner of the Polyglot in this respect.15 Besides this evidence, the fact that the accentuation needed to be explained and defended confirms that it was a novelty. In the Complutensian Polyglot itself, the full polytonic system is used in the Greek text of the Septuagint in volumes 1–4, in the NT lexicon at the end of volume 5, and in all the other places where Greek is printed, including some where the NT font is re-used.16 As for MSS of the NT, the general picture is well known: older MSS, such as the great uncials, are unaccented (unless markings are added in a later hand, as in Codex Vaticanus), but by the ninth century both uncials and minuscules are regularly adorned with full polytonic accentuation and two breathings. A similar situation applies in the case of MSS of other ancient literary texts.17 The Polyglot accentuation, then, was an innovation in its time. We are, however, not without information on the thinking behind it. As mentioned in the quotations above, there is a preface on the subject in the work itself. I propose now to reproduce this preface in full, with a translation, since not all of it has been adequately or correctly appreciated – in fact a number of gross misinterpretations persist – and no one has yet, to my knowledge, made a translation of it in its entirety. I shall examine its contents closely to see what it reveals about the accentuation, and then go on to consider the question of the identity of its inventor.

15

16

17

of special interest, see below. Some may also be seen in Scholderer, facsimiles 2, 3, 6, 19 (= Proctor, fig. 31). A facsimile can be seen in Renouard, Annales, 390–1 (I am grateful to Erik Hamer for checking this for me). Cf. Geanakoplos, Byzantium, 246 n. 89. See below for a list of the last-mentioned. Doukas’s editions of the Erotemata, etc. and of Musaeus, Hero and Leander (both Alcalá, 1514), published in the same year but after vol. 5 of the Polyglot, use the Polyglot NT font but with standard accentuation. Illustrations of the Erotemeta, etc. may be seen in: Geanakoplos, Byzantium, 241; Legrand, Bibliographie, 1:118; Proctor, Printing of Greek, pl. 24; of the Musaeus in: Thomas, Sixteenth-Century Printing, pl. 12; Scholderer, Greek Printing Types, fig. 24; Morison, Politics and Script, pl. 171. (Scholderer, followed by Morison, gives the date of the Musaeus as 1510: this seems to be a mistake.) For illustrations of NT MSS, see, e.g., Metzger, Manuscripts. On accentuation, p. 12. A great variety of texts and MSS from III AD to XVI AD can be seen in Wilson, Mediaeval Greek Bookhands, showing accentuation at various stages of development, from none through incomplete to full polytonic. Instances of partial accentuation do not show anything approaching a monotonic system. Further illustrations in: Cavallo and Maehler, Greek Bookhands, and Turner, Greek Manuscripts. Cf. Turner’s summary (11) of accentuation in his sample of MSS (late IV BC to VI/VII AD): ‘Accents are rarely written in prose literature, still more rarely in private letters. … But they are likely to be used fairly frequently in texts of lyric verse, especially in verse in difficult dialects. …’ The earliest occurrence of accents so far known is from II BC. The establishment of the polytonic system was a lengthy and somewhat complicated process reaching finality in IX/X AD; for a summary, see Schwyzer, Grammatik, I, 373–5.

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

151

4. THE PREFACE 4.1 Introduction The NT volume of the Polyglot, that is, volume 5 (1514), opens with a lengthy preface in Greek, entitled Πρὸς τοὺς ἐντευξομένους, filling one folio page and four lines of a second. It is followed at once by a translation into Latin. The heading of the latter makes clear that the Latin is a translation, not the original: Precedens Greca prefatiuncula in latinum versa. Ad Lectorem. We are thus spared the need to establish this point. The Latin version (but not the Greek) is printed also in volume 1 (1517), with the heading Prologus in nouum testamentum: et de causis quare in eo apices graeci sunt praetermissi. Ad lectorem. This text matches the one in volume 5 very closely, with some minor differences in orthography and very slight differences in substance.18 Differences between the original Greek and the Latin translation are another matter, to be noticed in passing as we proceed. 4.2 Text of the Preface Πρὸς τοὺς ἐντευξομένους. ῞Ινα μὴ θαυμάσῃς ὦ σπουδαῖε ϕιλολόγε μηδὲ σχῇς πρὸς ἡμᾶς ἀηδῶς εἰ ἐν παρούσῃ τῆς νέας διαθήκης γραικοτυπώσει ἑτεροτρόπως ἢ ἐν τῆς παλαιᾶς τὰ στοιχεῖα μόνα ἄνευ τῶν πνευμάτων 5 καὶ τῶν τόνων ἐντετυπωμένα ἐξεδόθη, προύργου νενομίκαμεν τὴν τοῦ πράγματος τούτου αἰτίαν πᾶσιν κατὰ τὰ πρῶτα ἐμπεϕανίσθαι. ἔστι δὲ καὶ τοιαύτη. τοὺς ἀρχαιοτάτους τῶν ἑλλήνων χωρὶς τούτων ἐν τοῖς χαρακτῆρσι κορυϕῶν γράϕειν εἰθισμένους, σαϕέστερόν ἐστιν ἢ πολ10 λῶν δεῖσθαι μαρτυριῶν. καὶ γὰρ δηλοῖ τοῦτο ϕανερῶς παλαιά τινα κοὐκ ὀλίγα τῶν ἀντιγράϕων, οἷον καλλιμάχου ποιήματα καὶ τὰ σιβύλλης ἔπη, καὶ πεπαλαιωμέναι ἐν τῇ πόλει λίθων γλυϕαὶ μόνοις ἁπλῶς γράμμασι ἐγκεχαραγμέναι. ὥστε πάντως εἶναι πρόδηλον, ῥαβδίσκων τούτων καὶ κεραιῶν ὑπερθέσεις ἐν τῇ πρώτῃ 15 ἐκείνῃ τῆς γλώττης ἑλληνικῆς γενεσιουργίᾳ μὴ ἐπινενοημένας, μήτε πρὸς ὁλόκληρον τῆς αὐτῆς ϕωνῆς τελειότητα πανταχόθεν συντετακυίας. 18

Differences in substance: heading (above); the word hac is omitted from in hac noui testamenti greca editione to suit relocation to vol. 1; rursus is replaced by rursum (once). Differences of orthography: mainly more frequent use of ae for e; th in authoritas is changed to t (twice). There are differences of punctuation, esp. more frequent use of the colon (:) in vol. 1.

152 20

25

30

35

40

45

50

55

COLLECTED ESSAYS ON THE GREEK BIBLE

ἐπειδὴ δὲ καὶ πᾶσαν τὴν νέαν διαθήκην τοῦ κατὰ ματθαῖον εὐαγγελίου καὶ τῆς πρὸς ἑβραίους ἐπιστολῆς δεόντων, ἑλληνικῇ ἀπ᾽ ἀρχῆς διαλέκτῳ ὥσπερ καὶ ὑπὸ τοῦ ἁγίου πνεύματος ἐχρηματίσθη, συγγεγράϕθαι πάντες ὁμολογοῦσιν, ἔδοξε καὶ ἡμῖν ἀρχαίαν ἐν αὐτῇ τῆς αὐτῆς γλώττης παλαιότητα καὶ μεγαλοπρέπειαν ὁσίως διατηρεῖν, καὶ τὴν βίβλον ἄνευ ὁπωσοῦν ἐλαχίστων προσθηκῶν, ἀρχαίων γραϕῶν δίκην ἐκδιδόναι, ὡς μὴ ϕανῶμεν πρᾶγμα οὕτως δή τοι ἱερὸν καὶ αἰδεσίμης μεγαλοϕροσύνης πλῆρες κατ᾽ ἀλλοτρίας αὐτῷ προσεπιβεβλημένας καὶ νέας ἐργασίας ἀνακαινοτομῆσαι. πρὸς δὲ τούτοις εἰ τἀληθὲς λεκτέον ἐστί, ἡ τῶν πνευμάτων καὶ τόνων ἔνδεια τοῖς ὁσονδήποτε ἐν τοῖς ἑλληνικοῖς λόγοις γεγυμνασμένοις οὐδὲν πρόσκομμα ἐπιϕέροιεν ἄν. λέγω δὲ πρὸς τὴν εἰλικρινῆ τῶν λεγομένων ἔννοιαν ἀναϕέρων. οὐ μὴν ἀλλὰ ἵνα μή τις ἀπορήσῃ ἐν ποίᾳ τῶν συλλαβῶν προσήκει τὸν τόνον ἐϕαρμόζειν, ἁπλῆ μόνον ἐν ταῖς πολυσυλλάβοις λέξεσι κεραία προσηρτήθη. οὐ μέντοι τόνος αὐτὴ ἑλληνικὸς ὑπολαβέσθω, ἀλλὰ γνωμώνιόν τι καὶ σημεῖον ὑϕ᾽ οὗ ἀπευθύνοιτ᾽ ἂν ὁ ϕιλομαθής, ὥστε μὴ διαμαρτάνειν ποτὲ ἐν τῇ ἐκϕορᾷ καὶ εὐρυθμίᾳ τῶν λέξεων. ἐν δὲ τῆς ἀρχαίας διαθήκης ἑλληνικῇ ἐκδόσει ὅτι αὐτὴ μετάϕρασις παντῇ που ἀλλ᾽ οὐ πρωτοποίητος σύνταξις οὐκ ἔδοξέ τι ἐκ τῆς κοινῆς ἐν τῷ γράϕειν συνηθείας ἢ ἀϕαιρεῖν ἢ ἐναλλάξαι. ἐπειδὴ δὲ οὐ τοῖς πολυμαθέσι μόνον ἀλλ᾽ ἅπασι καθόλου τοῖς περὶ τὴν ἁγίαν γραϕὴν σχολάζουσι συντελεῖν ἡ βίβλος αὕτη πεϕιλοτίμηται, ἐϕ᾽ ἑκάστῃ τῶν λέξεων γραμματίδια ῥωμαϊκὰ κατὰ στοιχεῖον ἐπετέθη, ἃ ἐπιδεικνύει τὴν ἀμϕοτέρων πρὸς τὰς ἑτέρας καταντικρὺ κειμένας ἐπάλληλον σύγκρισιν, ὅπως οὐδὲν ᾖ πρὸς τὸ σϕάλλεσθαι ἐνδόσιμον τοῖς μαθητιῶσι καὶ μήπω εἰς ἄκρον τῶν ἑλληνικῶν ἀϕιγμένοις. ἔτι δὲ ἐπείπερ ἐγχωροῦσιν ἐνίοτε λέξεις ἑλληνικαὶ πολυσχιδεῖς δοκοῦσαι ἔχειν καὶ ἀμϕιβόλους καὶ ἄλλας τινὰς διχογνώμους σημασίας, διὰ σπουδῆς ἡμῖν ἐγένετο σημειοῦν καὶ τοῦτο, ὑποκειμένῃ στιγμῇ ῥωμαϊκῷ γραμματιδίῳ ὑπὲρ τὴν ἑλληνικὴν καθεστηκότι λέξιν. καὶ ἵνα παύσωμεν προοιμιάζοντες, κακεῖνο τὸν ϕιλομαθῆ μὴ λανθανέτω, οὐ ϕαῦλα ἡμᾶς οὐδὲ τυχόντα ἐπὶ τῇ ἡμετέρᾳ ἐντυπώσει ἐσχηκέναι ἀντίγραϕα, ἀλλ᾽ ἀρχαιότατα καὶ καθόσον οἷόν τε ἦν ἐπηνορθωμένα, καὶ δὴ καὶ κατὰ τὴν παλαιότητα οὕτως ἀξιόπιστα, ὥστε μὴ πείθεσθαι αὐτοῖς, πρὸς δυσκόλου εἶναι τὸ παράπαν καὶ βεβήλου. ἃ δὴ καὶ αὐτὰ ὁ ἁγιώτατος ἐν χριστῷ

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

153

πατὴρ καὶ κύριος ἡμῶν ὁ μέγιστος ἀρχιερεὺς λέων δέκατος τῇ ὁρμῇ ταύτῃ συλλαμβάνειν προθυμούμενος, ἐκ τῆς ἀποστολικῆς βιβλιοθήκης ἀγόμενα, ἔπεμψε πρὸς αἰδεσιμώτατον κύριον τῆς ἱσπανίας καρδηνάλιον, οὗ χορηγοῦντος καὶ κελεύσαντος τὸ παρὸν βιβλίον ἐτυπώσαμεν. 65 ὑμεῖς δὲ οἱ τῆς παιδείας ἐρῶντες θεῖον τουτὶ καὶ ἱεροπρεπὲς τοὖργον νεωστὶ ἐντετυπωμένον μετὰ πάσης δέχεσθε προθυμίας. κἂν τοῦ χριστοῦ τοῦ κυρίου καὶ θεοῦ ἡμῶν μιμηταὶ ϕανῆναι καὶ τοῖς ἔργοις γενέσθαι σπουδάζετε, ἴστε ὅτι οὐδεμία ὑμῖν ἔτι πρόϕασις λοιπὴ τὸ μὴ τῇ θείᾳ γραϕῇ συνομιλεῖν. οὐκέτι 70 ἀντίγραϕα διεϕθαρμένα, οὐ μεταϕράσεις ὕποπται, οὐκ ἀπορίαν τῆς ἀπογράϕου ϕράσεως ἀπορεῖτε. τὸ θέλημα μόνον τὸ ὑμέτερον καὶ προθυμίαν προσδεχόμεθα. ἧς μὴ ὑπολειπούσης, ἀναντιρρήτως ἔσται οὕτως. τῆς γὰρ γλυκύτητος τῶν θείων λόγων γευσάμενοι, τὰς λοιπὰς τῶν ἐπιστημῶν μακρὰν χαίρειν ἐάσετε. 75 εὐτυχεῖτε, καὶ τὸν καινουργισμὸν τῆς ἡμετέρας ἐργολαβίας μὴ ἐκϕαυλίσητε. 60

4.3 Notes on the Greek text The original is in general very accurate in regard to spelling, accentuation, punctuation, and word-spacing. Some errors in these are clearly the typesetter’s, not the author’s. I have attempted to keep changes to a minimum consistent with present-day practice and comprehensibility. The punctuation is that of the original, with a change in two places only, as noted below (lines 26, 47; Legrand made the same changes in his printing of the text in Bibliographie, 1:115–7). There are no capital letters except pi, used once: I have retained this feature. The original is printed in a solid block without paragraphs: I have introduced them. There appears to be only one place where the spelling is in error: line 37 γνωμώνιον, of which the attested form is γνωμόνιον. This is likely to be a morphological as much as phonological slip: the writer could have been misled by the association with γνώμων as well as the phonetic identity of ο and ω. Cf. εὐγνομόνως in the extract from Doukas’s preface to Plutarch, Moralia (1509; see below), but εὐγνωμόνως correctly in his preface to the Erotemata, etc. (1514; Legrand, Bibliographie, 1:119). Throughout the text I have made two general changes, not individually noted. Iota subscript, not used at all in the original, is here introduced where required. Second, in initial diphthongs the original regularly has the breathing, or combination of breathing and accent, placed over the first vowel, in accordance with a common (though not universal) practice at that time and still sometimes found long afterwards: I have moved these to the second vowel. Similarly, in the (rare) instances where an accent is placed over the first vowel of an internal diphthong I have moved it to the second. All other changes are noted.

154

COLLECTED ESSAYS ON THE GREEK BIBLE

As far as I know, the text has been reproduced only once before, by Legrand in 1885 (Bibliographie, 1:115–7). Legrand’s text differs from mine in a few places, the most important as follows: he inserts an article twice (lines 2, 62), alters to τ[οι]ούτων in line 8 and corrects the spelling of ἐπηνορθωμένα in line 57. These I regard as improvements beyond necessity. Legrand also omits the words τοῦ χριστοῦ in line 67, presumably by accident. In the following, the text I have printed above is given first, the original (or Legrand’s emendation) second. 2 ῞Ινα] Ινα, first letter drop cap 2 μηδὲ σχῇς] μὴ δὲσχῆς 3 ἐν παρούσῃ] ἐν [τῇ] παρούσῃ Legrand 8 τούτων] τ[οι]ούτων Legrand 13 ἁπλῶς] ἀπλῶς 22 ἀπ᾽ ἀρχῆς] and 28 κατ᾽ ἀλλοτρίας] elision mark and breathing printed together over α 26 ἐκδιδόναι, ὡς] ἐκδιδόναι. ὡς 27 δή τοι] δήτοι 30 λεκτέον ἐστί,] λεκτέον ἐστὶ, 31 οὐδὲν πρόσκομμα] οὐδέν πρόσκομμα 32 ἐπιϕέροιεν sic 32 ἄν.] ἂν. 34 μή τις] μήτις 35 ἁπλῆ] ἀπλὴ 35 ἐν ταῖς] ἔν ταῖς 37 γνωμώνιόν τι sic. γνωμόνιόν τι Legrand 37 καὶ σημεῖον] καί σημεῖον 37 ὑϕ᾽ οὗ] ὑϕοῦ 37–8 ἀπευθύνοιτ᾽ ἂν] ἀπευθύνοιτἂν 40 παντῇ που] παντῆπου 40 ἀλλ᾽ οὐ] ἀλλ οὐ 43 ἀλλ᾽ ἅπασι] ἀλλ ἅπασι 45 ἐϕ᾽ ἑκάστῃ] ἐϕ ἑκάστη 47 σύγκρισιν, ὅπως] σύγκρισιν. ὅπως 50 ἐνίοτε] ἑνίοτε 54 κακεῖνο sic. κἀκεῖνο Legrand 56 ἀλλ᾽ ἀρχαιότατα] ἀλλ ἀρχαιότατα 56 καὶ καθόσον] και καθόσον 57 ἐπηνορθωμένα] ἐπηνωρθωμένα Legrand 59 τὸ παράπαν] τοπαράπαν 61 ἀποστολικῆς] similarly original. ἀποιολικῆς Legrand 62 πρὸς αἰδεσιμώτατον] πρὸς [τὸν] αἰδεσιμώτατον Legrand 65 ἐρῶντες] ἑρῶντες 65 καὶ ἱεροπρεπὲς] και ἱεροπρεπὲς 66 τοὖργον sic. Cf. Smyth, Greek Grammar, §173.a. 67 κἂν τοῦ χριστοῦ τοῦ κυρίου] κἂν τοῦ κυρίου Legrand 69 λοιπὴ τὸ] λοιπή τὸ

4.4 Translation To future readers. So that you may not be surprised, diligent scholar, or be displeased with us that in the present Greek printing of the NT, in a way different from that of the Old, only the letters, without the breathings and accents, have been set in print and published, we have thought it important that the reason for this be made clear to all at the outset. It is as follows. That the most ancient of the Greeks were accustomed to write without these points (κορυϕαί) on the letters is too clear to need many testimonies. For certain old copies (ἀντίγραϕα), not a few in number, clearly show this, such as poems of Kallimakhos and the verses of the Sibyll, and carvings of great age on stone in the city, engraved simply with letters alone. So it is quite evident that, in that first bringing into being of the Greek language, the placing on of these small strokes and marks was not devised, nor contributed to the full completeness of the said language in any way. Since also all acknowledge that the whole NT, apart from the Gospel according to Matthew and the Epistle to the Hebrews, was written down

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

155

in the Greek language from the beginning just as it was imparted by the Holy Spirit, we too decided piously to preserve the archaic antiquity and majesty therein of the same language, and to publish the book without the least addition whatever, in the manner of the ancient writings, so that we may not seem to have introduced novelty into something so holy and full of revered lofty thought, by means of alien and new operations imposed on it. Moreover, if the truth be told, the lack of breathings and accents could cause no obstacle to those with any training at all in Greek letters. I mean this with reference to the pure thought of what is said. Nevertheless, lest anyone be in doubt on which syllable it is proper to apply the accent, a simple mark (κεραία) has been attached only to words of more than one syllable. Let it not be supposed, however, that this is the Greek accent [mark], but a small pointer and sign by which the studious person might be guided, so as never to go astray in the pronunciation (ἐκϕορά) and modulation (εὐρυθμία) of the words. But in the Greek edition of the OT, because it is more or less entirely a translation and not an original composition, it was decided not to remove or change anything from what is common practice in writing. Since this book earnestly aspires to be of service not only to those advanced in learning but to all universally who study the holy scriptures, small roman letters in alphabetical order have been placed on each of the words, which indicate in succession the match between each word and the other lying opposite, so that there may be no cause for those who are learners and have not yet reached the peak of Greek studies to go wrong. Further, since Greek words are capable at times of seeming to have multiple and ambiguous meanings and others that give rise to difference of opinion, we have been diligent to indicate this also, by a dot underneath the small roman letter standing above the Greek word. And so that we may make an end of our preface, this also should be brought to the notice of anyone eager for learning, that we did not have inferior or just any copies for our printing, but the most ancient and correct as possible, and in particular so reliable in regard to (κατὰ) their age that not to trust them is altogether the act of an obstinate and impious (βέβηλος) person. Which very copies our most holy father and lord in Christ, the most high chief priest Leo the tenth, being eager to assist this enterprise, sent, when they were brought out of the apostolic library, to the most reverend lord cardinal of Spain, at whose provision of funds (χορηγοῦντος) and command we have printed the present book. You who love learning, receive with all zeal this holy and reverent work newly printed. And if you are eager to be seen as imitators of Christ our

156

COLLECTED ESSAYS ON THE GREEK BIBLE

Lord and God, and in your deeds to be such, know that there is no excuse remaining to you for not engaging with (συνομιλεῖν) holy scripture. No more corrupted copies, no suspect translations, you are not at a loss from the copied manner of expression (τῆς ἀπογράϕου ϕράσεως). We await only your wish and readiness. If this is not lacking, it will undoubtedly be so. For when you have tasted the sweetness of the sacred words, you will put far from your mind the remaining branches of learning. Farewell, and do not disparage the novelty of our undertaking. 4.5 Notes on the Greek and the translation The language is a learned, almost timeless Byzantine Greek, unalloyed by vernacularisms and ultimately based on Classical models, yet idiomatic within its own parameters. The style is dense, verbose, and showy. There are no false forms, even in the case of such difficulties as adjectives in -ης, and no mistakes in concord. The syntax is sometimes idiosyncratic, but hard to fault. Confident handling of particles, the relative pronoun, and most noticeably the participle, is on display. Hyperbaton is used, but not to excess. Vocabulary from all eras is deployed as necessary, including neologisms called for by the recent advance in technology, the invention of printing. Besides LSJ (with Supplement) and Lampe, PGL, I have consulted seven lexicons of Ancient, Byzantine, and Modern Greek: Contopoulos, Neon Lexikon; BDAG; DGE; Lex. Proia; Mega Lexikon; Sophocles, Lexicon; and OGELD. Even these do not completely cover the vocabulary available to the author. TLG CD ROM E (1999) has been checked for the hapax legomena ῥαβδίσκος, ἀνακαινοτομῶ, πρωτοποίητος. There are some unexpected omissions of the article, notably: 3 ἐν παρούσῃ τῆς νέας διαθήκης γραικοτυπώσει, 4 ἐν τῆς παλαιᾶς [γραικοτυπώσει], 14 ῥαβδίσκων τούτων καὶ κεραιῶν ὑπερθέσεις (for, e.g., τὰς τῶν ῥαβδίσκων...), 24 ἀρχαίαν... παλαιότητα καὶ μεγαλοπρέπειαν, 39 ἐν δὲ τῆς ἀρχαίας διαθήκης ἑλληνικῇ ἐκδόσει. Rather than signs of inadequate command of the language, these are to be seen as a deliberate mannerism of the style: the writer boldly plays against expectation for effect. There is a noticeable preponderance of perfect participles. I take this to be a sign of conscious display. 2 σπουδαῖε. Perhaps as I have translated, in the original sense and as the Latin version has it (studiose); but it might be no more than an empty cliché, ‘excellent.’ 12 πεπαλαιωμέναι. Not just equivalent to ‘old,’ but with a certain flavour, like ‘aged,’ ‘of long standing.’ 14 ῥαβδίσκων. ῥαβδίσκος is not found in LSJ; Lampe, PGL; Mega Lexikon; Contopoulos, Neon Lexikon; Lex. Proia; OGELD; Sophocles, Lexicon. The neologism (if such it is) provides a neat descriptor of the accent mark (literally ‘little rod/stick’). 14 κεραιῶν. κεραία is old as a term for any mark or serif above or forming part of a letter (see LSJ, s.v. II.3, BDAG, s.v.). Here it evidently refers to the breathing marks, perhaps alluding to the horn shape. But in 36 it refers to the acute accent mark used to indicate the stress.

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

157

15, 24 γλῶττα. The Atticistic form signals literary language. There are no other occurrences of σσ/ττ here. 16 πανταχόθεν. It makes the best sense if taken closely with the negative, i.e., ‘not in every way’ = ‘not in any way,’ ‘not at all.’ I do not know of a parallel with πανταχόθεν, but cf. οὐ᾿ πανταχοῦ, οὐ πάντῃ, οὐ πάντως (LSJ, s.vv.). Otherwise it is very tautological if taken with ὁλόκληρον τελειότητα (‘the full completeness throughout’). 17 συντετακυίας. For συντείνω intr., ‘contribute,’ ‘conduce’ (to, πρός), see LSJ, s.v. συντείνω II.2, Lex. Proia, s.v. The thought is that the marks were not necessary to make the writing of the language complete. One might translate ‘nor was important for …,’ ‘nor had any bearing on. …’ 26 δίκην + gen., ‘as, like.’ An ancient usage (LSJ, s.v. I.2; DGE, s.v. A.II) still sufficiently alive in modern times to be recorded in Lex. Proia. 29 ἐργασίας. An exactly suitable sense is not evident in the lexicons. A translation such as ‘operations’ or ‘procedures’ fits the context. 29 ἀνακαινοτομῆσαι. Not attested, only καινοτομῶ (with καινοτομία, etc.) from Classical onwards. A model for the compound in ἀνα- can be seen in ἀνακαινοποιῶ (recorded in Lampe, DGE). For the meaning here, cf. LSJ and Lampe, PGL, s.v. καινοτομῶ. Kαινοτομῶ is used in Doukas’s preface to Plutarch’s Moralia (1509; Legrand, Bibliographie, 1:93) and Mousouros’s preface to Epistolae philosophorum, etc. (1499; Legrand 1:54). 32 ἐπιϕέροιεν. Strict grammar would require ἐπιϕέροι, with singular subject ἡ ἔνδεια. But it is plural ad sensum, the subject being in effect ‘the lack of breathings and the lack of accents.’ 36 αὐτή. I take this to be the post-Classical (and Modern) demonstrative use (subject of ὑπολαβέσθω, complement τόνος ἑλληνικός). For earlier examples of the development, see Horrocks, Greek (1st ed.), 74, 93 (papyri), 181–3 (Malalas), 226 (in general); BDAG, s.v. 2.a (NT). 37 γνωμώνιον. See above on the spelling. LSJ’s meaning for γνωμόνιον is ‘pointer or dial-hand.’ Cf. γνώμων, ‘pointer’ of a sundial. Here we have a new application, to describe the accent mark. The Latin version renders notula (‘a little mark’). 39 εὐρυθμία, ‘modulation.’ I have used a rather general equivalent; I am not sure if the word is meant to have a more precise reference, such as ‘correct accentuation.’ 40 πρωτοποίητος. Not attested. There are 37 other compounds in -ποίητος (-τός) in Buck-Petersen, Reverse Index, 482. The meaning is evidently ‘made for the first time,’ hence ‘original,’ ‘first-hand.’ 44 συντελεῖν. For the meaning ‘help,’ be of service’ (to, dat.), see LSJ, s.v. II.2; Lex. Proia. 45 ἐϕ᾽ ἑκάστῃ τῶν λέξεων, ‘on each of the words.’ 53 ὑπὲρ τὴν ἑλληνικὴν καθεστηκότι λέξιν, ‘standing above the Greek word.’ The small roman letters are superscripts standing at the beginning of the word to which they relate. 46–7 ἃ ἐπιδεικνύει … σύγκρισιν. The intended meaning is clear, but the expression appears somewhat awkward. σύγκρισις = ‘comparison,’ ‘parallel,’ according to the lexicons. τὴν ἀμϕοτέρων πρὸς τὰς ἑτέρας καταντικρὺ κειμένας ἐπάλληλον σύγκρισιν literally = ‘the of both [words] in relation to the others lying opposite successive comparison.’ I have paraphrased slightly, in the interests of natural English.

158

COLLECTED ESSAYS ON THE GREEK BIBLE

48 ἐνδόσιμον. A post-Classical meaning, ‘incentive, cause, occasion,’ for which see Mega Lexikon, s.v. 3; Contopoulos, Neon Lexikon; Lex. Proia. The earliest attestation is in Arrian, Anabasis 1.7.8 (Mega Lexikon). 50–1 πολυσχιδεῖς … καὶ ἀμϕιβόλους καὶ ἄλλας τινὰς διχογνώμους. All three adjectives qualify σημασίας. 51 διχογνώμους. A rare word (one citation in LSJ; DGE), but there is sufficient evidence to determine its meaning: see it and its group in LSJ; Mega Lexikon; Lex. Proia. 63 ἱσπανίας (hispanie in the Latin version). Likewise (e.g.) in Doukas’s prefaces to the Erotemata, etc. and Musaeus, Hero et Leander (both 1514). The use of the rough breathing implies that its original value was known and it was still potentially realisable as [h] in pronunciation at this time. 63–4 καρδηνάλιον, οὗ χορηγοῦντος καὶ κελεύσαντος τὸ παρὸν βιβλίον ἐτυπώσαμεν. The device of the relative clause with participle and hanging finite verb, in the best Classical manner. Cf., e.g., Demosthenes 19.1 τὸν ὅρκον ὃν εἰσελήλυθεν ὑμῶν ἕκαστος ο᾿μωμοκώς. For a liturgical example, see Αἶνοι at Ὄρθρος, ἦχος πλ. ά, 1: τὸ μυστήριον, ὃ ἀνυμνοῦσι ἀπόδος ἡμῖν ἀγαλλίασιν καὶ τὸ μέγα ἔλεος. The form with gen. abs., as in the preface, is a clever variation. For a use of χορηγῶ like that here, cf. Doukas’s preface to Plutarch, Moralia (1509): τοιαύτης μὲν οὖν ἠξιώθητε δωρεᾶς, Ἄλδου χορηγοῦντος, ἡμῶν τε διορθούντων. 67–8 κἂν … σπουδάζετε. Is this a slip for subj. σπουδάζητε (after κἄν = καὶ ἐάν)? Probably not. Rather, it reflects the advance of the merging of indic. and subj. endings, and the loss of distinction esp. between εἰ and ἐάν/ἄν, ὅτε and ὅταν, which had begun much earlier. Cf. Mandilaras, Verb, 241–8; Horrocks, Greek (1st ed.), 181, 246–7; BDAG, s.vv. ἐάν, ὅταν for NT examples and references to the standard treatments. LSJ’s example of κἄν with indic. (s.v. κἄν II.1) in Apollonius Dyscolus, Synt. 70.22 (II AD) reads: κἂν γὰρ οὕτω ϕαμέν. … 69 συνομιλεῖν. An interesting choice of word. It appears to be a metaphorical use of the standard meaning ‘converse with.’ 71 τῆς ἀπογράϕου ϕράσεως, ‘the copied manner of expression.’ ἀπόγραϕος is often neut. as subst., ‘copy,’ but here the adj. ‘copied’ (exx. in Sophocles, Lexicon; none in LSJ; DGE), hence ‘at one remove,’ etc. (A modern meaning ‘crossed out, cancelled’ leads nowhere.) ϕράσις can also mean ‘an (individual) expression, idiom, phrase’ (cf. LSJ), but this does not fit as well. I take this to be a succinct description of one of the problems of reading a translation, i.e., trying to understand translated idiomatic language without access to the original. The Latin version gives up and paraphrases all of οὐκ ἀπορίαν τῆς ἀπογράϕου ϕράσεως ἀπορεῖτε as: non inopia textus originalis, ‘no lack of the original text.’ 75 ἐργολαβία. A more general meaning than the older ones in LSJ (‘contract for the execution of work’; ‘profit-making’). See Contopoulos, Neon Lexikon. The new vocabulary of printing is seen in 5, 66 ἐντυπῶ, 55–6 ἐντύπωσις, 64 τυπῶ, all old words now put to work in a new way. Other prefaces of the time use the same terminology. 3 γραικοτύπωσις seems to be a novelty, but γραικότυπος/ον occurs in Mousouros’s preface to Etymologicum Magnum (1499: Legrand, Bibliographie, 1:61).

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

159

5. MISTAKEN VIEWS It will be as well to begin by clearing away some of the misconceptions that have gathered about this preface. First, it is sometimes said that it was written by the printer, Brocar. There is no reason whatever to suppose this or evidence to support it. On the contrary, its content and the character of the Greek clearly suggest, to put it no stronger, that it was composed by the editor(s) of the volume, as would be one’s first expectation in any case. This error originates with Robert Proctor, in 1900, who simply refers to ‘the preface by the printer, Arnaldo Guillen de Brocar,’ without explanation or argument.19 I surmise that Proctor was so focused on the type that he just assumed that anything said about it came from the printer. He was certainly not well acquainted with the content of the preface: he misunderstood even the Latin version at a crucial point, as we are about to see. A second, more serious mistake is the assertion that the preface states that the font was cut on the model of a single MS sent from the Vatican library. This has spread through a number of writers, not all of whom acknowledge a source, but it can be clearly traced to Proctor. Proctor says that ‘it can be affirmed with certainty that [the Greek type] is based on the writing of a particular manuscript,’ and goes on to say ‘it seems … to be beyond dispute that the type was cut on the model of the writing in the ‘archetypa tantae uetustatis, ut fidem eis abrogare nefas uideatur,’ sent to Cardinal Ximenez by Leo X from the Vatican Library.’20 Even without returning to the original Latin text (let alone the Greek), and relying just on Proctor’s elliptical quotation, one can see that something is wrong: eis is plural; it follows that archetypa probably is also. In fact the context runs as follows (Proctor’s quoted words underlined): illud lectore[m] no[n] lateat: no[n] queuis exemplaria i[m]pressioni huic archetypa fuisse: s[ed] antiquissima eme[n]datissimaq[ue]: ac tante preterea vetustatis: vt fidem eis abrogare nefas videatur. Que 19

20

Proctor, Printing of Greek, 144. Repeated in Lyell, Cardinal Ximenes, 48; Tasker, ‘Complutensian Polyglot,’ 201; Hall, ‘Trilingual College,’ 5:143; cf. Scholderer, Greek Printing Types, 9. Proctor, Printing of Greek, 144. Similarly Scholderer, Greek Printing Types, 9–10; Turner, Printed Editions, 12; Tasker, ‘Complutensian Polyglot,’ 204; Hall, ‘Trilingual College,’ 143; Morison, Politics and Script, 298–9; Barker, Aldus Manutius, 6. In Bentley, Humanists, 93 there seems to be a remnant of Proctor in: ‘The preface to the New Testament claimed great antiquity and accuracy for them [the manuscripts used], but troubled to mention only a single one, lent to Ximénez by Pope Leo X from the papal library.’

160

COLLECTED ESSAYS ON THE GREEK BIBLE

s[an]ctissimus in [Christo] p[ate]r [et] d[omi]n[u]s n[oste]r Leo decimus pontifex maximus huic i[n]stituto fauere cupie[n]s ex ap[osto]lica bibliotheca educta misit ad Reuerendissimu[m] d[omi]n[u]m Cardinale[m] hispanie: Let the reader be aware, that not just any copies were the originals for this printing, but very ancient and correct ones, and moreover of such antiquity that to refuse to trust them is wrong. These our most holy father and lord in Christ Leo the tenth, supreme pontiff, desiring to favour this plan, sent, after they were brought out of the apostolic library, to the most reverend lord cardinal of Spain.21

It is quite clear that the preface here (and in the Greek original, lines 55– 62) speaks of manuscripts, plural: no[n] queuis exemplaria i[m]pressioni huic archetypa fuisse … (‘not just any copies were the originals for this printing …’). It is clear moreover, as pointed out by Woody in 1971 in a valuable but neglected article, that the remarks are not about the font, but about the text and the authorities lying behind it. The ‘manuscript origin’ of the font ‘belongs to a well-populated world of historical myth,’ as Woody puts it.22 In the perpetuation of this myth, what is most striking is that writers not only take it on trust from a previous source but often repeat Proctor’s Latin quotation word for word, oblivious of what it actually says. Proctor was responsible for another false assertion, which bears on the question of the accentuation. He thought the reason given in the preface for the lack of accents and breathings in the NT text was an ‘ingenious excuse,’ and the true reason was the ‘defective state of the type’ at this time. This is instantly disproved by the fact that the same font appears with full accentuation elsewhere in the volume, namely: in the preface itself; in the letter of Eusebios to Karpianos; in the miscellaneous introductory matter to the Epistles (12 pages); in the verses of Doukas and Fausto; and in the lexicon at the end of the volume (75 pages). All of this amounts to a substantial body of text.23 The fact that some accentuation, albeit in the 21

22

23

I quote the Latin text in the form found in vol. 5, not that in vol. 1. Square brackets indicate resolution of abbreviations. This passage had been quoted and (correctly) translated in Tregelles, Printed Text, 4. Woody, ‘Greek Fonts,’ 143. Geanakoplos already in 1962 ignored the myth and stated the facts correctly (Byzantium, 240–2). Woody (145) makes much out of the preface to the OT as a means of elucidating the meaning of the NT preface: the OT preface, in a similar discussion, is clearly talking of the text not the font. This is true and helps to clinch the case, but the NT preface is already clear. Even Woody manages to misquote the Latin, printing archetype for archetypa (144). The Cardinal’s prologue in vol. 1 also clearly speaks of codices of both the Greek OT and the NT sent from the Vatican library. The same point is made by Woody (‘Greek Fonts,’ 146 and n. 7), but he instances only the poem by Doukas. (The OT volumes, though fully accented, are not relevant here,

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

161

reduced form of the acute alone (and the diaeresis), was in fact employed in the NT text further undermines the theory of a defective font: a full set of letters with the acute (and the diaeresis) was available. We may doubt, too, that the editors would have bothered to concoct such an elaborate and plausible justification just to conceal the true reason. A final argument against Proctor’s assertion is seen in Nicolaus de Lyra’s Liber differentiarum veteris testamenti, which was printed before the Polyglot, in about 1512. In it a smaller version of Brocar’s Polyglot NT font is used for occasional patches of Greek: it has full polytonic accentuation.24

6. INFORMATION

IN THE PREFACE

Let us return now to the monotonic accentuation and what can be learnt about its origin from the preface. The question of the accentuation of the text was clearly of deep concern to the editors. It is the first topic on their minds, and occupies more than half the preface. The tone is defensive, expecting adverse criticism and attempting to forestall it. The question of the text and the manuscripts is deferred till well on in the preface, and on this subject, to the regret of later scholarship, it is tantalisingly brief. The argument presented is as follows. Ancient Greek writing did not originally use accents and breathing marks. They had not yet been devised, and were in fact unnecessary. Two kinds of evidence are available as proof: old ‘copies’ of some authors, such as Kallimakhos and the Sibylline oracles, and ancient inscriptions still visible in ‘the city.’ In both cases texts without any accent or breathing marks are observable. The NT was compiled and written in the same ancient period, and may be assumed to have been originally written in the same manner. It is therefore historically accurate to print it without them; and indeed they are not essential. Moreover, it is an act of piety to present the sacred text imparted by the Holy

24

because a different, Aldine font is used.) On the technical difficulties of printing Greek with accents and breathings, cf. Proctor, Printing of Greek, 17–21. I do not pretend to expertise in this area, and am willing to be corrected if I have missed something that supports Proctor’s view. Proctor also put a false slant on the preface’s statement about the Septuagint. Where the editors are clearly speaking of the accentuation and giving their reason for not departing from the usual practice (lines 39–42), Proctor sees a statement about the choice of font: ‘According to the printer … the Greek of the Old Testament is merely a translation, and therefore not worthy of his fine special type’ (144). I first learnt of this work from Norton, Printing in Spain, 40. I rely on Norton for the attribution to Brocar and the dating (‘not later than 1512’): there is no statement of either in the volume. Cf. Norton, Descriptive Catalogue, 9: ‘c. 1512?’, and 10: ‘dating based on ornamental material.’ Norton makes no remark on the accents.

162

COLLECTED ESSAYS ON THE GREEK BIBLE

Spirit in its original form without any innovations. It is clear that authenticity is the aim. The NT will be presented in its ancient, original form, that is, the Greek text without the accentual markings that historically came later. To the same end, the text is based, as we are told further on in the preface, on the ‘most ancient and correct’ manuscripts.25 The two pieces of evidence mentioned by the editors invite further attention at this point. First Kallimakhos and the Sibylline oracles. Woody astutely perceived that the singling out of these authors for mention ought not to be accidental, and he found out the reason for it. In the Miscellanea of Angelo Poliziano, published in Florence in 1489, portions of these very authors are printed without accentuation or breathing marks of any kind. As Woody pointed out, Poliziano states or implies that he was following his manuscripts in this respect. Clearly, the Polyglot editors selected Kallimakhos and the Sibylline oracles for mention because they knew this book and it provided visible testimony to ancient practice. By ‘copies’ (ἀντίγραϕα) they mean manuscripts (as they do elsewhere), i.e., the manuscripts lying behind Poliziano’s text.26 There is something more. In the same work, the Kallimakhos verses are introduced by Poliziano in these words: Sed aures ad Callimachi iam versiculos subscriptos veteri more sine ullis accentiunculis arrigamus (‘Let us now pay attention to the verses of Callimachus written below in the ancient manner without any accent marks’). The words veteri more, ‘in the ancient manner’ are significant: Poliziano already uses the same justification for omission as the Polyglot editor(s). They could hardly have been unaware of his remark. It does not follow that they knew of ancient practice only from this source, but Poliziano’s work was a precedent for printing Greek in this way.27 25

26

27

Woody (‘Greek Fonts,’ 145) makes an interesting observation about the fonts used in printing the Latin texts. In the OT volumes the font is the more modern roman, in the NT the archaic gothic. The Latin fonts thus parallel the Greek, i.e., the more modern Aldine cursive in the OT, the traditional ‘Greco–Latin’ in the NT. Woody, ‘Greek Fonts,’ 146. Illustrations of Poliziano’s Kallimakhos text may be seen in Proctor, Printing of Greek, fig. 31 and Scholderer, Greek Printing Types, facsim. 19. The Greek in the rest of the volume has normal accents and breathings (Woody, 146 n. 8). I have had access to the Poliziano volume only in a later edition (Antverpiae, 1567), which seems to be a fairly faithful copy. At any rate, it prints the Kallimakhos and Sibylline oracles without accents or breathings, even though the font used is a later Aldine cursive. Poliziano’s reliance on MSS is stated in the case of the Sibylline oracles (576) but only implied in that of Kallimakhos (604–5), it seems to me. Scholderer (Greek Printing Types, 5) quotes the words veteri more sine ullis accentiunculis in his discussion of the font, obviously aware of their significance; Woody missed them. Proctor (Printing of Greek, 133) again saw only typographical motives: ‘this

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

163

The other evidence mentioned is ancient inscriptions that can be seen in ‘the city.’ To anyone reading or writing in Greek, this could mean only one thing: Constantinople. Personal acquaintance with that city, either by the author or by others who had visited it and could report what they had seen, is implied. The Latin version, somewhat surprisingly, translates this as ‘Rome.’ Two possible explanations present themselves. Either the translator simply misunderstood ἡ πόλις, or, as seems more likely, it was a deliberate shift by someone who could not allow Rome to appear less well endowed with ancient monuments than the old capital of the Eastern Empire. At any rate the change throws into relief the Greek orientation of the preface, a matter that is important for the question of authorship, as we shall see. The argument of the preface now takes a new turn. The text of the NT, it has just told us, will be printed without accents or breathings, in fact ‘without the least addition whatever.’ Nevertheless, there will be some accentuation after all, a simple mark placed on words of more than one syllable. In practice this means that the inessentials, the three varying accent marks and the breathings, are dispensed with, and a single accent, the acute, is used to indicate where the stress is to be placed in all cases where there might be any doubt. It is a system that supplies the minimum necessary for practical purposes. Though the diaeresis is not mentioned, the author of the system knows that it too is necessary, and uses it where required. We are told that this system is to provide guidance in the correct ‘pronunciation and modulation’ of the words. The purpose, then, is practical. It is also pedagogical: the learner needs this assistance. We note that reading aloud is taken for granted. The pedagogical concern is visible, and explicitly stated to be the motive, in another feature of the text, the keying of each Greek word to the corresponding Latin word. Elsewhere in the volume, too, the needs of beginners in Greek are served, namely in the Introductio quam brevissima ad grecas litteras and the NT lexicon that appear at the end of the work. It might be said that volume 5 of the Polyglot is both an authoritative presentation of the Greek and Latin texts and a tool for learning Greek in order to compare them.28

28

rejection of accents [in the Poliziano vol.] must be held to be due to a deliberate decision in favour of greater simplicity. This is certainly attained; for the entire fount consists of twenty-seven Greek sorts.’ The keying of the Greek and Latin texts is at the expense of the general appearance of the page. Cf. Scholderer, Greek Printing Types, 9. Besides the small roman letters, the text also indicates, by a small forward slash, every Greek word that has no equivalent

164

COLLECTED ESSAYS ON THE GREEK BIBLE

There is more than a hint of compromise in the fact that some accentuation is provided after all. It is a step back from the original intention. These, it seems, are the accents you have when you’re not having any accents. Why the compromise? Was it simply to answer the stated need? Or was it from fear that the break with tradition was too radical, despite the earlier example of Poliziano’s book? We can only speculate: it could have been either or both. But the latter reason is not very satisfying: critics would be just as offended by the novelty of a reduced accentuation as by none at all. There is a third possibility. The compromise could be the outcome of conflicting views among the editors. While one side wanted to go all the way and print nothing but the bare letters, the other insisted that there must be some accentuation, and if it was not to be the full polytonic system, at least some minimal indication must be given as a practical guide. We shall return to this later and see if anything can be deduced about the parties involved. But even if we can find out nothing about them, a conflict of opinion would provide an explanation for something rather puzzling otherwise. The accent system that resulted was clearly not proposed for general use. The editors are careful to warn the reader that it is not the standard system of Greek accentuation (‘let it not be supposed …’). The standard remains the full polytonic system, as used in the printing of the Septuagint and elsewhere. The novel one has been created only for use here in the special circumstances of the NT text. Again we see a concern for the student, in this case not to allow any misconception to arise. We also see a conservative outlook on Greek accentuation.

7. DOUKAS THE EDITOR OF THE TEXT AND AUTHOR OF THE

PREFACE

I turn now to the question of authorship. Who was the inventor of the monotonic accentuation? It is not possible to say with certainty, but there must be a strong suspicion that it was Dimitrios Doukas the Cretan. Before we approach that conclusion, let us set out what can be said about his editorship of the volume and authorship of the preface. in the Latin. Most are occasioned by the Greek definite article. One might have thought they could be dispensed with, but they are inserted with great diligence, further detracting from the general appearance. The OT volumes likewise provide the tools for learning Hebrew and studying the original text. On the dotted letters (ll. 49–53) see below. On the significance of the NT lexicon, see Lee, History, 47–51.

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

165

Bataillon, in 1937, conjectured that Doukas played a leading role in the preparation of the NT text of the Polyglot. Then Geanakoplos, in his important work Byzantium and the Renaissance (1962), attempted to bring this role into clearer focus. Geanakoplos’s cautious conclusion from the evidence he could find was that ‘it would seem justifiable to affirm that Ducas was not only the author of [the] preface but primarily responsible for editing the Greek text of the NT volume.’ This view was accepted and developed further by Bentley in 1983. The case depends on a series of data, some no more than clues but others good evidence, which it will be useful to summarise.29 (1) Doukas is known to have held the chair of Greek at Alcalá. The date of his arrival cannot be determined, but there is definite evidence of his presence by 1513. He came to Spain at the invitation of Cardinal Ximénes, and was already experienced in editing Greek texts: he had been the principal editor of two major Greek works of the Aldine press in Venice. In Spain, there is evidence of his work as a teacher and publisher of textbooks on the rudiments of Greek for students. (2) In the series of dedicatory verses addressed to the Cardinal in the NT volume (five sets in all, two in Greek, three in Latin, with their authors’ names attached), Doukas’s head the list, followed by those of Nikitas Faustos, Juan de Vergara, Hernán Núñez de Guzmán, and Bartolomeo de Castro. The placing of Doukas’s name and verses appears significant. None of the others named, or of the other Spanish scholars known to have been involved, Antonio de Nebrija and Diego López de Zúñiga, can be shown to have a better claim as leading editor. 29

Bataillon, Érasme et l’Espagne, 21–2, 42; Geanakoplos, Byzantium, 223–55, esp. 239– 43; Bentley, Humanists, 76–9. Cf. also Sáenz-Badillos, La filología bíblica, 399: he sees Doukas as the only person in the Polyglot team with the capacity to edit the Greek columns in the NT and the OT, and concurs with Geanakoplos that he was the author of the Greek preface. Legrand (Bibliographie, 1:117; similarly p. cv) already asserted that the Greek text of the NT was due ‘aux soins réunis de Démétrius Ducas et de Nicétas Faustos.’ See also De Jonge, Opera Omnia Erasmi, IX.2, 14–7 for an analysis of Zúñiga’s share in the Polyglot: there is no sufficient ground to speak of him as ‘the principal editor,’ as has often been done (17 with n. 72). Lowry (Aldus Manutius, 286 with n. 121) dissents from Geanakoplos, finding the attribution of the preface to Doukas ‘on stylistic grounds’ to be ‘ingenious rather than convincing,’ and doubting that Doukas could have been the editor of ‘a text which appeared in January 1514 if he had been in Alcala [sic] only since the previous year.’ Lowry’s first argument has no substance; and as far as I can see there is no certainty that Doukas did not arrive in Spain before 1513. His last known location before Spain is Venice with Aldus in March 1509. Aldus shut down his business and left Venice later the same year, when a French invasion threatened (see Lowry, 159–60); Doukas is likely to have been out of work then.

166

COLLECTED ESSAYS ON THE GREEK BIBLE

(3) A strong link to Doukas is provided by the textual note attached to Matt 6:13 in the NT, concerning the conclusion to the Lord’s Prayer. ‘Its author exhibits a precise knowledge of the Greek Church’s liturgy’ (Bentley, 79). Doukas, who was both a Greek and subsequently the editor of the editio princeps of the Greek liturgies (Rome, 1526), certainly had this knowledge. The only other person in the team who might conceivably have been as well informed was Fausto, concerning whom it will be useful to say a little more at this point. Vittore/Vittorio/Vettor Fausto (c.1480–c.1540) was certainly an Italian, a Venetian, not a Greek, as has sometimes been said: Νικήτας Φαῦστος is simply the Greek equivalent of his name. Apart from noting that he was elected to the chair of Greek at Venice in 1518, and was the editor of a Greek liturgical book, writers on the Polyglot have offered little information on him. He seemed a shadowy figure who might or might not have played an important role in the Polyglot. But a lot more is known about him, and, most surprisingly, that his ‘chief claim to fame is his work as a naval architect’ (Wilson, 92). This began in Venice in 1525–1526 when he proposed the building of a quinquereme, supposedly based on ancient models. The vessel was in fact built, with some success, and Fausto went on to further experiments in ship-building and to become ‘a celebrity in Venice and elsewhere’ (95). Wilson seems sceptical that Fausto even took part in the Polyglot project, and says (89) that his collaboration ‘does not appear to have been the original purpose of his travels [in Spain and elsewhere], and he cannot have stayed very long with Ducas.’ Fausto’s experience of editing up to this point was in Latin texts (Cicero and Terence); later, in Paris, he edited a Latin translation of Aristotle’s Mechanica. The liturgical book, the Paraklitiki, was published in Venice in 1522 by the de Sabio press at the expense of Andreas Kounades. There is a Greek preface by Fausto, from which it appears that he was called in at a late stage as a metrical expert to correct the text from that point of view. He clearly writes as a non-Greek, addressing his preface τοῖς καθ᾽ Ἑλλάδα χριστιανοῖς, and employing ‘you’ throughout. All this seems to justify regarding Fausto as a junior player in the editing of the Polyglot NT text and not specially expert in liturgical matters.30 30

On the subject of this paragraph, see esp. N. G. Wilson, ‘Vettor Fausto,’ and Dizionario Biografico, 45:398–401. Cf. Legrand, Bibliographie, 1:cv; Bataillon, Érasme et l’Espagne, 42; Geanakoplos, Byzantium, 166 n. 179, 240. For details of the Paraklitiki, see Legrand, 1:173 and, with much additional material including Fausto’s preface, Ladas and Chatzidimos, Προσθῆκες, 7–11. Fausto’s explanation of his part in editing this book begins: ἔπειτα δὲ [ὁ Κουνάδης] πολλῶν ἀντιγράϕων συνειλεγμένων τὸ ἐλλεῖπον ἐϕ᾽ ἑκάστου ἐξ ἀλλήλων ἀναπληρώσας τὸ σύνολον, ἅτε ποιητικοῦ τινὸς [sic] ἀνδρὸς δεόμενον διὰ τὸ ἔμμετρον αὐτοῦ, ἐμοὶ παραδέδωκεν, ἵνα δὴ τὸ ϕαῦλον ἐκκρίνας τὸ ὀρθὸν δοκιμάσαιμι. Incidentally, the Paraklitiki (also called the Oktoïkhos) is not, as Geanakoplos has it (Byzantium, 240 n. 72), a book of ‘hymns to the Virgin,’ but the book containing the hymns of the eight-tone cycle. The Dizionario Biografico (‘una raccolta di preghiere alla Vergine’) betrays its source. Geanakoplos (240) says Fausto was ‘probably a pupil of Ducas.’ This is a guess, but a plausible one. Bataillon’s guess (Érasme et l’Espagne, 42) that he was ‘peut-être le typographe qui a composé le texte hellénique sous la direction de Doucas’ seems improbable: surely this was Brocar’s task.

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

167

(4) While the older Spanish sources all play down the role of Doukas, Nebrija once ‘let slip,’ as Bentley puts it (78), ‘the information that Hebrews (i.e., conversos) and Greeks were charged with editing, respectively, the Hebrew and Greek scriptures.’ Doukas was the only native Greek working on the project. (5) Geanakoplos observed (243) that the language of the preface ‘seems to be that of one intimate with the nuances of Greek style.’ This is right, but can be put a good deal more strongly. In my opinion the character of the Greek is such that it could not have been written by anyone but an advanced Hellenist. It is a tour de force in the Byzantine style. While there were other Hellenists among the Polyglot scholars, they were all nonGreeks, and all Spaniards apart from Fausto. Doukas stands out among them all as the person with the necessary expertise to write it. I offer an additional observation that reinforces this. The Latin translation is inexpertly done, with simplifications, gaps at difficult points and some outright mistakes in translation. This suggests that none of the Polyglot scholars who could write Latin – and that would have been all of them apart from Doukas – had sufficient skill in Greek to translate such a text well. Still less could they have composed it.31 (6) The reference to ancient inscriptions in ‘the city,’ i.e., Constantinople, implies acquaintance with the place, something unlikely to be available to anyone but Doukas. Geanakoplos was inclined to see this as evidence that Doukas himself had visited Constantinople. This of course cannot be ruled out, but it is not certain: word of mouth could have conveyed this information to the Greeks in the West; in fact it may have been common knowledge among them. To me the strength of this point lies more in the Greek perspective evidenced here and highlighted by the change to ‘Rome’ in the Latin version. The writer writes as a Greek. Doukas seems the best, if not the only, candidate.32 (7) The preface uses the first person plural frequently, but the singular ‘I’ occurs at one point (l. 30 λέγω …, ‘I mean this with reference to the pure thought of what is said’). Geanakoplos was rightly cautious about 31

32

Geanakoplos (Byzantium, 242 n. 78) pointed out that the Latin version is ‘simpler in its wording.’ But he went on to say, ‘which may mean it was translated from the Greek,’ showing he had missed the statement in the heading. Geanakoplos, Byzantium, 225 with n. 7, 242–3. There seems to be good evidence that a journey to Constantinople was possible after 1453. But the ‘vividness’ of the phrasing in the preface, appealed to by Geanakoplos as a sign that Doukas had seen the inscriptions in situ, does not strike me. In fact the Latin version suggests autopsy in a way the Greek does not: monumenta vetustissima que rome adhuc visuntur, ‘monuments of great age which are still on view in Rome.’ The Greek simply says (l. 12) πεπαλαιωμέναι ἐν τῇ πόλει λίθων γλυϕαί, ‘carvings of great age on stone in the city.’

168

COLLECTED ESSAYS ON THE GREEK BIBLE

using this, but thought that it ‘may well indicate that one person was mainly responsible for editing the Greek text.’ I am not sure that it does that; but it certainly suggests that a single person was responsible for writing the preface. The ‘we’ elsewhere, though it could mean ‘I,’ is more probably inclusive, reflecting the collaborative character of the project. The ‘I’ emerges in explanatory λέγω, ‘I mean,’ because the singular is usual in this expression unless one really means ‘we.’ The Latin translation does not, as Geanakoplos asserted, change ‘I’ to ‘we’ here, but drops the whole sentence! In fact the Latin version eliminates all but two of the thirteen occurrences of ‘we/us/our.’ I am not sure what significance is to be seen in this. Perhaps it is simply stylistic. It has a depersonalising effect, at any rate.33

8. FURTHER INDICATIONS To the above can be added further evidence that I have not seen adduced before. The first is in lines 67–68, where we read κἂν τοῦ χριστοῦ τοῦ κυρίου καὶ θεοῦ ἡμῶν μιμηταὶ ϕανῆναι καὶ τοῖς ἔργοις γενέσθαι σπουδάζετε, ‘and if you are eager to be seen as imitators of Christ our Lord and God, and in your deeds to be such.’ The phrase τοῦ χριστοῦ τοῦ κυρίου καὶ θεοῦ ἡμῶν has the sound of Greek liturgical language. Though I have not found this exact phrase, combinations of a similar kind are commonplace throughout liturgical texts: τοῦ Κυρίου καὶ Θεοῦ καὶ Σωτῆρος ἡμῶν Ἰησοῦ Χριστοῦ Κύριε Ἰησοῦ Χριστὲ ὁ Θεὸς ἡμῶν Χριστὸς / Χριστὲ ὁ Θεὸς ἡμῶν Χριστῷ τῷ Θεῷ Χριστῷ τῷ Βασιλεῖ καὶ Θεῷ ἡμῶν τοῦ ἀγαπητοῦ Υἱοῦ αὐτοῦ, Κυρίου δὲ ἡμῶν Ἰησοῦ Χριστοῦ 33

The two that survive are 60 ὁ … κύριος ἡμῶν … λέων ~ d[omi]n[u]s n[oste]r Leo; 64 ἐτυπώσαμεν ~ imprimi fecimus. The others are at lines 2, 5, 23, 27, 53, 54, 55 bis, 65, 70, 73. That first person plurals do not necessarily mean a plurality of persons is demonstrated with clarity by another preface of Doukas’s, that found in the Erotemata, etc. (Alcalá, April 1514). Here we have χαρακτῆρσιν οἷς ἐνετύχοδεν [sic] in a series of first person singulars where Doukas is definitely speaking, or rather complaining, of his own labours in preparing the book. The text of this is reproduced in Legrand, Bibliographie, 1:119. Geanakoplos (Byzantium, 235) translated: ‘using letters [characters in Greek type] which I found here.’ His translation implies a reading ἐνέτυχον ὧδε. But I suggest the true reading is ἐνετύχομεν, the δ being simply a typesetter’s error. Cf. in Doukas’s prefaces to Rhetores (1508; Legrand, 1:88) and to Plutarch (1509; Legrand, 1:93): τοῖς ἀντιγράϕοις οἷς ἐνετύχομεν. Bataillon translated χαρακτῆρσιν οἷς ἐνετύχοδεν as ‘avec les caractères que j’avais sous la main’ (Érasme et l’Espagne, 22).

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

169

The similarity is evident and the point reinforced by the strikingly different formulation in the Latin version: et si christi Opt[imi] Max[imi] sectatores videri vultis [et] esse. Just as the textual note on Matt 6:13 links Doukas to the editing of the NT text, this – possibly unconscious – reminiscence of Orthodox liturgical language links Doukas to the writing of the preface.34 The application of the title Optimus Maximus to Christ in the Latin text is noteworthy. It was of course originally a title of Jupiter in Roman antiquity. It was then transferred at some point to the Christian God, as seen in the well known D.O.M. (Deo Optimo Maximo) inscribed on churches and elsewhere, and modelled on I.O.M. (Iovi Optimo Maximo). The application to Christ rather than God the Father is unexpected, possibly even of questionable orthodoxy. Who at Alcalá could have written such a thing? And is it an isolated instance? I have no answer to the first, but in answer to the second can point to some parallels, though far from Spain of 1514. Optimus Maximus Jesus occurs in a Scottish source of the mid sixteenth century.35 The application of the title to Christ is also found in Rabelais: in a passage in the Quart Livre (1552), in which he interprets the death of Pan in Plutarch as the Crucifixion, he adds ‘… car cestuy tresbon, tresgrand Pan, nostre unicque Servateur, mourut lez Hierusalem, regnant en Rome Tibere Cesar.’36 The whole topic is clearly one worth further study. In the same sentence there is another difference of interest. In μιμηταὶ … γενέσθαι we have a distinct NT allusion (cf. 1 Cor 4:16; 11:1 μιμηταί μου γίνεσθε, Eph 5:1 γίνεσθε οὖν μιμηταὶ τοῦ θεοῦ ὡς τέκνα ἀγαπητά, 1 Thess 1:6; 2:14). The Latin translator misses the allusion by translating μιμηταί as sectatores (‘followers,’ ‘adherents,’ ‘enthusiastic supporters’), whereas the Vulgate renders μιμηταί as imitatores in all places; the Complutensian NT lexicon likewise gives imitator as its only gloss on μιμητής. The reason for the rendering is not quite clear, but it may have arisen, as Harm W. Hollander points out to me, from the association of μίμησις with ζῆλος/ζήλωσις, and of μιμητής with ζηλωτής. Cf. Moeris p. 168 (ed. Pierson, 1759) ζήλωσις, Ἀττικῶς. μίμησις, Ἑλληνικῶς 34

35

36

In the NT the only comparable passages are 1 Thess 1:12 κατὰ τὴν χάριν τοῦ θεοῦ ἡμῶν καὶ κυρίου Ἰησοῦ Χριστοῦ, which however is usually taken as ‘according to the grace of our God and the Lord Jesus Christ’ (RSV); and John 20:28 ἀπεκρίθη Θωμᾶς καὶ εἶπεν αὐτῷ· ὁ κύριός μου καὶ ὁ θεός μου. There is a parallel use of liturgical language in Doukas’s preface to Αἱ Θεῖαι Λειτουργίαι (Rome, 1526; Legrand, Bibliographie, 1:193–4), where the memorable phrase τὸ ἀρχαῖον κάλλος, ‘[their] ancient beauty,’ occurs twice. This comes from the Orthodox funeral service: … εἰς τὸ καθ᾽ ὁμοίωσιν ἐπανάγαγε, τὸ ἀρχαῖον κάλλος ἀναμορϕώσασθαι (Νεκρώσιμα Εὐλογητάρια 2). Ferr[erius], Kinloss 81, recorded in DMLBS, s.v. optimus. I thank Jim Adams, David Howlett and Theodor Christchev for their assistance, and especially for making the dictionary entry available to me in advance of publication. I also thank Edwin Judge for valuable help along the way. Marichal, Quart Livre, 138. See also Screech, Rabelais, 328. I am grateful to Michael Screech for supplying me with this information. He also kindly informs me, in answer to my query, that he has never consciously come across such a use of the title in Erasmus’s works and would expect to recall it if he had.

170

COLLECTED ESSAYS ON THE GREEK BIBLE

[ζ 4 Hansen]; Ammonius 209 (ed. Nickau, 1966) ζῆλος καὶ ζηλοτυπία διαϕέρει … ζῆλος δὲ μίμησις καλοῦ …, cf. 210–211; Hesychius p. 677 (ed. Schmidt, 1867) ζηλωτής· ἐρεθιστής. μιμητής. In the NT text, 1 Peter 3:13 ζηλωταί has a v.l. μιμηταί, the majority reading. Finally, the rendering of ζηλωτής as sectator occurs in the Vulgate at Titus 2:14. At any rate it seems that the writer of the Greek preface was familiar with the Greek NT in a way the Latin translator was not: the latter neither recognises the Greek allusion nor recalls the Vulgate equivalent.

The preface also contains a not so veiled criticism of the Vulgate and its manuscript witnesses. When it says (ll. 69–71) ‘no more corrupted copies, no suspect translations, you are not at a loss from the copied manner of expression,’ the target can only be the Vulgate. What this means, put bluntly, is: the Vulgate text is corrupt, its translations faulty, and its meaning difficult to grasp. Instead of grappling with these problems, the scholar can now engage directly with τῇ θείᾳ γραϕῇ, the original Greek NT. The sentiments expressed earlier about ‘piously’ preserving the text in its ancient form and avoiding the introduction of ‘novelty into something so holy’ are further evidence of the same high regard for the original. While this may represent the views of some of the Polyglot scholars, they were in general deeply conservative in their approach to the Vulgate; at the extreme end of the spectrum was one like Zúñiga, who later vehemently defended the traditional Latin text against the perceived attacks of Erasmus. The Cardinal himself, though he could assert the need for recourse to the originals when the Latin manuscripts differed or appeared corrupt, revered Jerome’s translation and took the conservative side in the dispute between Nebrija and the other editors over the Latin version, restraining Nebrija from presenting a new translation based on the Greek in place of the Vulgate. The conclusion all this tends to is that the views espoused in the preface are those of one who takes for granted the superiority of the original Greek over the Latin translation, that is, someone like Doukas or Nebrija.37 37

Cf. Bentley, Humanists ch. 3, esp. 89–91 (dispute with Nebrija), 97–111 (respect for the Vulgate; evidence of unpublished annotations); also Homza, Religious Authority, 87–8. For a summary of Zúñiga’s outlook, see De Jonge, Opera Omnia Erasmi, 18–20. The Cardinal is said to have exclaimed to Nebrija, ‘God forbid that I should alter a word of the Blessed Jerome’s!’: so Fernández-Armesto, ‘Cardinal Cisneros,’ 157. While the Cardinal’s prologue expresses much in common with the views of Nebrija, it also displays what Fernández-Armesto characterises as ‘pre-humanist techniques,’ ‘echoes of cabbalistic lore,’ and ‘the world of mystical devotion’ (157–8). For the well known passage Accedit … examinetur, in the Cardinal’s prologue, see, e.g., Darlow and Moule, Historical Catalogue, 2:4; Homza, Religious Authority, 248–9. An English translation of the whole prologue can be found in Olin, Catholic Reform, 61–4. The Latin version of

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

171

The fact that there is a Greek preface at all is something to reflect on. Only in this volume of the Polyglot is there a Greek preface; all the other prefaces are in Latin. A connexion is thereby made to all the other publications of Greek texts up to this time, each usually equipped with a learned preface in Greek, written by a Greek scholar involved in the editing of the volume. These range from the 1476 epitome of Laskaris’s grammar, with a preface by Dimitrios the Cretan (a different Dimitrios), to the 1513 edition of Plato’s works, introduced in Latin by Aldus and in Greek by a 200-line poem (soon to be famous) by Markos Mousouros, the joint editor. And the practice continued long after the time of the Polyglot. The Greek preface in the NT volume of the Polyglot signals the presence of a Greek scholar involved in the work, and the quality of his preface is a sign of his competence in Greek.38 In other Greek prefaces both before and after the Polyglot the indication of the author’s name at the top is standard practice. There is no name at the head of our preface. This absence, however, seems to me readily explicable. The fact that the volume was a team effort would have played some part, but a more compelling reason can be suggested. The great project was the Cardinal’s, and all honour for it belonged to him. As Doukas’s own verses said, ἔργα αὐτοῦ ἥδε βίβλος, ‘this book is his work.’ Of course this was exaggerated, but it was the right tone to adopt in dealing with a powerful hierarch and patron. It would have been unseemly for someone else’s name to appear at the top of the first page. The collaborators on the volume could put their names to their verses later on in the book, but the opening preface remained tactfully anonymous. In due course the Cardinal’s own prologue, addressed to Pope Leo and setting out his titles in their full glory, appeared on the opening page of volume 1.

38

the NT preface does not tone down the above passage much, No[n] mendosa exemplaria: non suspecte translationes: non inopia textus originalis, though mendosa (‘faulty’) is less harsh than διεϕθαρμένα, and the last clause is a weak paraphrase (cf. notes above on l. 69): perhaps the translator missed what was going on. A further point that seems to have drawn no comment before: the preface to vol. 5 has nothing to say about the Latin text printed in the volume. It is only from the title page that one learns that it is the Vulgate (cum latina beati hieronymi translatione); there is no other information. A large body of Greek prefaces from 1476 to 1563 can be seen in Legrand, Bibliographie, vol. 1. Five other Greek prefaces or notes by Doukas, three of them quite lengthy, are found in: (1) Rhetores Graeci (vol. 1. Venice, 1508; Legrand, 1:85–88); (2) Plutarch, Moralia (Venice, 1509; Legrand, 1:92–93); (3) Erotemata, etc. (Alcalá, April 1514; Legrand, 1:119); (4) Musaeus, Hero et Leander (Alcalá, [1514]; Legrand, 1:120); (5) Αἱ Θεῖαι Λειτουργίαι (Rome, 1526; Legrand, 1:193–4). There does not appear to be one in Doukas’s final publication, Alexandri Aphrodisiei Commentaria (Rome, 1527; Legrand, 3:316–7).

172

COLLECTED ESSAYS ON THE GREEK BIBLE

There is then a strong case for concluding that Doukas was the editor of the NT Greek text and the writer of the preface to the volume. This of course is not to rule out or diminish the contribution of others: it is certain that the book as a whole was a collaborative effort, and that is likely to have included some of the labour on the Greek text. The preface also, though written by one person, could have received input from others in one way or another. It certainly had to reflect the general policy decided on by the team under the direction of its head, Cardinal Ximénes de Cisneros. 9. DOUKAS THE INVENTOR OF THE ACCENTUATION Was Doukas the inventor of the monotonic accentuation? There is no way to prove decisively that he was; we can only gather the available clues and draw the most probable conclusion. The clues seem to me to be as follows. Doukas had general responsibility for editing the Greek text, and wrote the preface. He must at least have been closely involved with the decisions about the accentuation of the Greek and the execution of them. We know that he was a native Greek, a Greek scholar and teacher, and an experienced editor of Greek texts. He was the only person in the team with all those qualifications. Concern over the accentuation of the Greek would seem to belong to such a person, rather than the Latinists and experts in the Vulgate; they are more likely to have felt it was a matter of small consequence. That Doukas himself did indeed worry about accents is shown by another preface of his, in the edition of Plutarch’s Moralia (Venice, 1509). At one point the following memorable passage occurs: Ὑμέτερον δ᾽ ἂν εἴη τὸν ἡμέτερον κάματον εὐγνομόνως ἀποδέξασθαι, μὴ δ᾽ ἐκϕαυλίζειν, εἴ τι που γράμμα διεστραμμένον ἢ ἄλλο τι τοιοῦτον ἡμᾶς παρέδραμεν, οὐκ ὂν ἐμποδὼν οὐδὲ τοῖς μήπω ϕωνῆς γευσαμένοις ἑλληνικῆς. Πολλοὶ μὲν ἀπείρως ἔχοντες τῆς περὶ τὸ διορθοῦν ϕιλοπονίας ἤ, βέλτιον εἰπεῖν, ταλαιπωρίας, καὶ τοῦ γνῶθι σαυτὸν ἀμνημονοῦντες, ὅλως τε σοϕοὶ δοκεῖν ἐϕιέμενοι, μεγάλῃ τῇ ϕωνῇ λαρυγγίζουσιν· «ὀξεῖα ἀντὶ βαρείας ἐτέθη· παῖε, παῖε τὸν κατάρατον, διέϕθαρται γὰρ ἡ βίβλος ἣν ἐκεῖνος ἐπετέτραπτο κατασκευάσαι τοῖς ϕιλολόγοις ὁλοσχερῆ τε καὶ ἀνενδεᾶ.» Your role would be to receive the product of our toil kindly and not disparage it if some misshapen letter or other such thing has got past us, when it is no hindrance even to those who have not yet tasted the Greek

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

173

language. Many who have not experienced the labour, better to say pain (ταλαιπωρία), of editing, and forget the maxim ‘Know yourself,’ and are anxious to seem very clever, shout out with a great voice: ‘An acute has been put instead of a grave! Hit him, hit him, the cursed fellow! For the book that he was entrusted with preparing for scholars complete and without deficiency is ruined.’39

We can see that Doukas had suffered over Greek accents. This outburst is a cry from the heart. He is unlikely to have given the matter anything less than his best attention when it came to the printing of the NT text. The extensive treatment of the topic in the preface is another sign of that preoccupation.40 That Doukas was responsible for the execution of the system in the Polyglot text seems extremely likely, not only from the points made so far, but from the nature of the task. Let us consider what was involved. The manuscripts exhibited the polytonic accentuation. To make the changeover to monotonic would have been too complicated for a typesetter to carry out alone, even with detailed instructions. Someone needed to go through the entire text and mark the new accents and diaereses on the copy that went into the print-room. That person would have been able to follow the polytonic accents of the manuscripts as a guide to the position of the stress, but expertise in NT Greek would be required to ensure that the appropriate changes into the new system were made; some acquaintance with the pronunciation tradition of the Greek NT, if not with spoken Greek, would have been an advantage. There are some indications that the latter was indeed brought to bear. The occasional accentuation of monosyllables, discussed above (§2), 39

40

Greek text from Legrand, Bibliographie, 1:92–3; my translation. See Lowry, World of Aldus, 240 for a lively picture of the methods – or lack of them – in Aldus’s workshop when this edition was produced. In Doukas’s ed. of the Erotemata, etc. (April 1514), there is another heartfelt preface on the difficulties of the editor’s lot, including almost the same phrase as above: ἐν ταῖς ταλαιπωρίαις τῆς διορθώσεως. Sáenz-Badillos (La filología bíblica, 392), after noting what the preface says about omission of accents and breathings and the use of ‘pequeñas señales’ as an aid to pronunciation, attributes the decisive voice in both to Doukas: ‘Es muy probable, que sea Demetrio Ducas, – el más familiarizado con los antiguos textos, y el que mayor experiencia editorial tiene – quien haya hecho prevalecer su opinión en este punto concreto.’ Cf. also 399: ‘… puede ser suya la decisión de editarlo sin acentos ni espíritus.’ Sáenz-Badillos does not discuss the accentuation further. It may be no coincidence that a work published long afterwards by one of Doukas’s pupils was on accentuation: Diego Sigeo, De ratione accentuum (Lisbon, 1560). Sigeo refers explicitly to his old teacher (… praeceptorem meum in schola Complutensi Demetrium Ducam. …). I owe the information here to Legrand (Bibliographie, 1:194 n. 2); cf. Geanakoplos, Byzantium, 233. I have not (yet) been able to see this ‘livre rarissime,’ or to check whether it contains any remnant of Doukas’s views on accentuation.

174

COLLECTED ESSAYS ON THE GREEK BIBLE

indicates more than a mechanical approach to placing the stresses. Though it created inconsistency, this phenomenon arose from familiarity with the actual pronunciation of Greek. The placing of the diaereses was also not a simple matter. The manuscripts often have superfluous diaereses, and the retention of only those that are necessary requires judgement. Again, actual pronunciation is the guide. Someone who knew what he was doing carried out these tasks. Whether it was Doukas himself, or whether he delegated some or all of the work, he at least supervised it.41 It is harder to say for certain whether Doukas himself conceived the basic idea of the monotonic accentuation, that is, the idea of a system using an acute accent mark to indicate the stress to the extent necessary and no more. It could have come from some other person or source, either in Spain during the work on the Polyglot, or at an earlier time when he was in Venice or even Crete. So far everything points to Doukas in response to a certain need at Alcalá, but we need to consider other possibilities.

10. PARALLELS IN THE OLD TESTAMENT

TEXT

First there is something of direct relevance to be seen in the other part of the Polyglot, the OT. The OT volumes (1–4, 1517) are introduced by their own preface, which runs in part:42 Illud est etia[m] co[n]sidera[n]du[m]: q[uod] in hebraicis characterib[us] scie[n]ter omisim[us] apices illos: quib[us] nu[n]c vtu[n]tur Hebraei p[ro] acce[n]tibus. Na[m] hi cu[m] ad nulla[m] vel significati vel p[ro]nu[n]ciationis differentia[m] pertinea[n]t: sed ad sola[m] ca[n]tus ipsoru[m] modulatione[m]: merito a veteribus Hebraeis reiecti sunt: quos in hoc imitari maluimus. Veru[m] ne locus acce[n]tus cuiusq[ue] dictionis ignoraretur: hoc modo p[ro]uidimus: vt quonia[m] dictiones hebraicae vt plurimu[m] in vltima habe[n]t accentu[m]: o[mn]es huiusmodi dictiones nullo prorsus apice notare[n]tur: reliquae vero no[n] habentes accentu[m] 41

42

One might speculate that Fausto was Doukas’s main assistant (cf. above). An illustration of what might be involved in the diaeresis: in John 1:19 the Polyglot text prints λευίτας, where λευΐτας is found in some MSS, as in a gospel lectionary of XII AD (Metzger, Manuscripts, no. 38). The lack of a second accent in απέστειλε με, etc. (see above) remains puzzling, however. Why did they leave these out, if pronunciation was the guide? This preface is headed Prologus. Ad lectorem. De his que ad lectionem veteris testamenti diuersis linguis nunc primum impressi sunt praenotanda and comes next after the Cardinal’s prologue addressed to Pope Leo X. Both are printed in all four OT volumes.

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

175

in vltima (quae rarissime occurru[n]t) Super syllaba[m] vbi praedominatur acce[n]tus: apice signare[n]tur: hoc modo ‫א ֵר ץ‬. ֶ֨ [sic for ‫]א ֶר ץ‬ ֶ This also is to be observed, that in the Hebrew characters we have knowingly omitted those marks which the Jews now use for accents. For since these do not relate to any difference of meaning or pronunciation, but only to the melody in chanting them, they were with good cause left out by the Jews of old, whom we have chosen to imitate in this matter. Nevertheless, so that the place of the accent of any word may not be unknown, we have made provision in this way, namely, since most Hebrew words have their accent on the final syllable, all the words of this kind are distinguished by no mark at all, but the remainder that do not have the accent on the final syllable (which occur very rarely) are indicated by a mark over the syllable where the accent predominates, thus ‫א ֵר ץ‬. ֶ֨

We learn that the cantillation marks, which also function as indicators of the accent, will be omitted from the Hebrew text as not being part of the ancient way of writing. An indication of the position of the stress will however be given by a simple system which marks the stress when necessary, namely, when there is an exception to the rule that the final syllable receives the stress. The mark (apex) illustrated is not unlike a Greek grave accent and is placed above the syllable. Thus the student or anyone else not completely familiar with Hebrew receives the necessary guidance in reading aloud. Inspection readily confirms that this system has been applied, and with great consistency, throughout the Hebrew text. Moreover, although nothing is said about it, the system has already been put into practice in volume 6 containing the Hebrew lexicon, printed in 1515. The whole procedure here runs parallel to that seen in the editing of the Greek text of the NT. In both, an unnecessary part of the accentuation is abandoned for the sake of a return to ancient practice, and instead a simplified accent is introduced as a practical guide to pronunciation. There is a further notable parallel between the OT Hebrew and the NT Greek texts. Not only do they both show the same method of keying each word to the corresponding word in the Vulgate by a small roman letter, but both use the method of marking selected words by a dot under the letter. This phenomenon is easy to miss, and I am not aware that it has been commented on before. It is introduced in the NT preface in the sentence ἔτι δὲ ἐπείπερ … (see above, lines 49–53) and in the Old by an explanation beginning Ex his praeterea litterulis nonnullas reperies: quibus tenue punctu[m] subscribitur hoc modo a Quod quidem punctum designat: … (‘Moreover among these small letters you will find some under which is written a slight

176

COLLECTED ESSAYS ON THE GREEK BIBLE

point, thus a, which point indicates. …’). Examples can readily be seen on inspection of the texts.43 This phenomenon deserves a study of its own, which will not be undertaken here. For the moment we note it only as a further significant parallel between the texts of the two Testaments.44 It is evident that the editors of the two parts conferred, and decided on similar procedures in printing the accents and other features of the texts. But can we deduce which came first? The order of printing of the volumes is not much of a guide. Decisions about how the Hebrew text was to be printed would have been made long before 1517, given the huge size of the task. We know the system of accentuation was already in place in 1515 in the Hebrew lexicon, and work on it must have begun well before then. Further, the writer of the NT preface already knows by 1514 how the LXX will be printed in the OT volumes, that is, with full accentuation (see lines 2–5 above). So priority is difficult to determine by the dates.45 Perhaps it is futile to attempt to determine priority, but the decision about the Hebrew accentuation appears to me, at least, to be based on what was done in the Greek rather than independently evolved for the Hebrew. If authenticity was the aim, we might have expected the editors of the Hebrew to decide to omit all vowel signs and print just the bare letters. This would indeed have been a return to the most ancient practice. Instead they decide to omit the so-called ‘accents,’ which they know are primarily musical notation marks. There is no mention of precedents, as there is in the case of the Greek, no allusion to old manuscripts without accents. Further, the accent system introduced is, as far as I am aware, an extreme novelty for Hebrew, unlike the Greek one, which arises out of the existing 43

44

45

[An instance can be seen in the image of John 1:1–14 above: εγεννήθησαν in the second-last line (1:13) is preceded by a dotted superscript ‘c.’ The dot indicates that there is a discrepancy between the Greek meaning and the Vulgate Latin nati sunt.] Woody (‘Greek Fonts,’ 145) already noticed the statement about omission of accents, but not the rest. Homza (Religious Authority, 86) mistakenly speaks of the Polyglot as printing Hebrew without vowel markings. The OT preface echoes the NT (Latin) preface in the discussion of the Greek MSS employed, even using the same words at some points: see in the NT preface illud lectore[m] no[n] lateat … hispanie (quoted above) and cf. in the OT preface: Quod aut[em] ad Graeca[m] scriptura[m] attinet: illud te no[n] latere volumus: no[n] vulgaria seu temere oblata exemplaria fuisse huic nostrae impressioni archetypa: sed vetustissima simul et eme[n]datissima: quae sanctissimus D[omi]n[u]s noster Leo Decimus Po[n]tifex maximus caeptis nostris aspira[n]s ex ipsa apostolica Bibliotheca ad nos misit: ta[n] tae integritatis: ut nisi eis plena fides adhibeatur: nulli reliqui esse videa[n]tur: quibus merito sit adhibe[n]da. It is natural to suppose that the OT preface of 1517 was written later and drew on the earlier one.

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

177

system. And the accent symbol chosen is like the Greek one but turns the opposite way. Why so? Because it mirrors the Greek. Moreover, in the Hebrew lexicon we find that an accented Hebrew word and its transliteration into roman letters show accents facing opposite ways, thus: áleph. ‫ ָ֨א ֶלף‬and: ‫ ָ֨מיִ ם‬Máim. The suggestion of a mirror image is further supported by the way the small roman letters are placed. In both the Greek and the Hebrew texts they come before the word: in the Greek that means to the left, since one reads from left to right, but in the Hebrew it means the opposite side, that is, to the right, because one reads from right to left. I conclude from this that the idea of using an acute accent mark to indicate the stress where necessary originated in the editing of the NT Greek text and was adapted for use in the Hebrew text. The Hebrew editors’ primary reason for giving up the cantillation marks may well have been convenience: it was challenging enough to print the intricate vocalisation in full, without attempting those as well. The justification that they were not ancient could then, with some truth, be brought in.46

11. LATIN, SPANISH,

AND ITALIAN

The basic notion of the monotonic accentuation, then, does not seem to have come from the Hebrew of the Polyglot. But there are other possible sources to be considered and eliminated. A system of marking Latin accentuation was long used, and is still found, in Latin service books, as also in various related works. In its basic form it indicates the stress by an acute in words of three or more syllables, this being the minimum necessary. Its purpose was obviously to guide those whose grasp of Latin accentuation needed some assistance when reading aloud. How old is this system? Might it already have been established in Spain and Italy in the time of the Polyglot? I see no evidence that it was. I have checked a number of mapping points, as follows. There are no marks in (1) the editio princeps of the Missale Romanum (Milan, 1474), in (2) Missale ad Usum Sarum (Basle, c.1489), and in (3) Missale Romanum (Venice, 1549). There is frequent but still incomplete marking of the stress (the circumflex is also 46

On the accents/ cantillation marks, see Bauer and Leander, Historische Grammatik, 136– 54; Joüon-Muraoka, Grammar, 61–9. There is a cantillation mark ‫’( ֨א‬azla), that is similar to the grave used. But the choice of it from among the thirty or so marks available can only have been motivated by its similarity to the Greek accent.

178

COLLECTED ESSAYS ON THE GREEK BIBLE

used) in (4) Missale Romanum (Salamanca, 1587), and very limited marking, sometimes none at all in (5) Missale Romanum (Munich, 1613). Finally, full application of acute marks where necessary, and also occasional use of grave and circumflex (for a different purpose), can be seen in (6) Missale Parisiense (Paris, 1738), and (7) Uffizio della B. Vergine Maria (Rome, 1782).47

Latin of the early sixteenth century in general is written without marks, as in the prefaces and Latin texts of the Polyglot. At the same time, however, there begins the occasional use of the grave, not to mark stress but to indicate vowel length, as, e.g., in adverbs (longè, verò), a phenomenon still seen long afterwards. But there is no sign of a system of marking the stress.48 Nevertheless the occasional use of an acute to indicate stress can be found, when it is needed for a special purpose. Thus Melancthon’s Greek grammar (1527), in the discussion of accents, illustrates some points by writing Latin words with an acute; Erasmus does the same in his work on Greek and Latin pronunciation (1528); and Nebrija marks the stress of some Latin words in his Latin-Spanish lexicon (1492).49 Similarly, as we have just seen, volume 6 of the Polyglot itself uses an acute to mark the stress in transliterations of Hebrew words. The idea as such was certainly about at the time. But there is no sign of the application to Latin of a fully developed system; moreover the inspiration for the sporadic use of the acute mark seems to come from Greek. The same question needs to be asked regarding Spanish and Italian. Both employ accentual markings today. How far back do these systems go? Could Doukas or someone else associated with the Polyglot have seen in the contemporary writing of the vernacular languages a model for Greek? The Spanish system was clearly not in place at this early date. In Nebrija’s grammar of Spanish (Castilian) (1492) and his Latin/ Spanish lexicons (1492, 1495), accent marks are not used.50 Spanish 47

48

49

50

A facsimile of (2) can be seen in Johnson, First Century of Printing, pl. 4, and of (4) in Thomas, Sixteenth-Century Printing, pl. 46. John Sheldon kindly made available his copy of (7). Other copies or facsimiles were seen in the Veech Library, Catholic Institute of Sydney (I thank Charles Hill, Margaret Watt and Michael Cullen), and Fisher Library, University of Sydney. I have been unable to discover a discussion of this phenomenon. See the plates in Johnson, First Century of Printing; Johnson, Italian Sixteenth Century; Thomas, Sixteenth-Century Printing. Melancthon [sic], Grammatica graeca. I have not seen the earlier ed. (1518), entitled Institutiones graecae grammaticae. Erasmus, De recta latini graecisque sermonis pronuntiatione, e.g., 200–1. For Nebrija, see next note. Antonio de Nebrija, Gramatica castellana (1492); Diccionario Latino-Español (1492); Vocabulario español-latino (1495?). Nebrija does use the acute as a means

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

179

documents up to and well after the time of the Polyglot are also seen to be unaccented.51 In Italian there is still ‘chaos’ in the writing of accents today (or at least up to 1966), in the words of a leading authority, but the beginnings of accent marking are seen further back than in Spanish. The situation in the early sixteenth century is summed up by Migliorini (trans. Griffith) as follows: Accents, too, [like the apostrophe] were modelled on Greek, as can be seen from the preference for an acute accent internally in a word (in the rare cases when an accent was used) and the grave accent on the final syllable. After sporadic appearances in the Quattrocento, they were used by Bembo and Manuzio in the Asolani (1505), which sometimes had the grave on the final vowel (menò, altresì; but also amista, castita, etc.) and very rarely an accent within the word (restío).52

It is clear that at this time there was no regular system based on the acute alone as far as necessary, as in the Greek monotonic. Moreover, Greek itself was the source of the sporadic and unsystematic accentuation that there was. None of these, then, provides a model for what was done in the accentuation of the NT Greek text of the Polyglot. Nor do we have any indication that any other person had thought of such a system for Greek before. It seems most probable that the idea was conceived by Doukas, who developed it into a workable system and saw to its application in the printing of the text. He had the knowledge and experience – one might add confidence – to invent a new system in response to the need that had arisen, and to come up with one that was simple, practical, and elegant.

51

52

of indicating the stress in Spanish when discussing points of accentuation, but only then: see Gramatica castellana, lib. 2, cap. ii–iiij. In his Latin-Spanish lexicon, some Latin headwords have the stress marked (e.g., Dídymus, Tolétum), but the Spanish is unaccented. I have checked the 129 plates in Arribas Arranz, Paleografía documental hispánica, showing documents ranging from 850 to 1641, including several in the period 1490– 1516: there are no marks of accentuation (unlike ç and ñ, which appear sometimes). In Thomas, Sixteenth-Century Printing, I see no marks in the Spanish texts until, and only in, pl. 48, Vida del P. Ignacio de Loyola (Madrid, 1594), which has very occasional use of final grave or acute. The description as ‘chaos’ is in Migliorini, The Italian Language, 474; quotation 237 (= Migliorini, Storia della lingua italiana, 383–4). See further Migliorini, The Italian Language, 181, 231, 278, 322–3, and the plates in Migliorini, Storia, esp. the Aldine Petrarch of 1501. A description of the present-day system may be seen in Dardano and Trifone, La lingua italiana, 395–6. It is clearly not comparable to the monotonic accentuation of Greek: both acute and grave (also, rarely, circumflex) are used only in certain words and conditions, and not solely to indicate stress. (I am grateful to Gianluca Alimeni of Macquarie University for assistance here.)

180

COLLECTED ESSAYS ON THE GREEK BIBLE

12. THE PROPONENT OF AN UNACCENTED TEXT Who was responsible for the initial intention to print the Greek text without accents? This is an interesting question that does not have to be answered, but is too tantalising to leave untouched. One could of course argue that it was Doukas himself. Persuaded by the ancient model of the inscriptions in Constantinople and aware of the precedent in Poliziano’s Miscellanea, he as editor of the Greek text could have initiated the idea and led the rest of the team to agree with him. On the other hand, the final outcome smacks so much of compromise that it is tempting to see behind it a difference of opinion among the editors, as I have argued above. If there was such a conflict, on which side is Doukas likely to have been? In my view Doukas is likely to have been conservative in this matter. As a native Greek educated from an early age in the writing of Greek, and later a teacher of the language, he would have been totally at home with the polytonic system and, in the general nature of teachers, committed to its retention despite the difficulty of learning it. The suggestion to print without accents on the ancient model I think came from someone else, and it was Doukas who clawed back some accentuation to serve as a practical guide. Who was that someone? We need to look for a radical. Woody came up with a name behind the proposal to omit accents and breathings, though he makes no mention of the other ingredient in the situation, the decision to insert a simplified accentuation: The fact that the omission of accents was borrowed from Poliziano may enable us to guess who among the Complutensian editors was responsible for the idea. It was probably Elio Antonio de Nebrija, who had studied for ten years in Italy and whom we find, on at least one occasion, placing Poliziano at the head of a list of Italian humanists (even before Pico della Mirandola).53

This is an inviting suggestion, but there is a problem about dates. Nebrija did not actually join the Polyglot team until the autumn of 1513, when preparation of the NT text must have been well advanced (printing was complete in January 1514). Nevertheless the idea is not to be abandoned entirely. The link to Poliziano is stronger than Woody supposed. Sáenz-Badillos has shown that some of Nebrija’s material in his Tertia Quinquagena (Alcalá, 53

Woody, ‘Greek Fonts,’ 147; cf. 144, 146. Woody voiced his reservations: ‘[t]here is, however, no proof that the idea was Nebrija’s, and the point is not one I will insist upon.’

DIMITRIOS DOUKAS AND THE ACCENTUATION OF THE NEW TESTAMENT

181

1516) depends directly on Poliziano’s Miscellanea, the very source of the notion of printing Greek without accents, as we saw.54 Nebrija (1441 or 1444–1522) was among the most famous and distinguished scholars of his day. He was a renowned linguist, the author of the first grammar of Spanish and a compiler of lexicons. His approach to the NT was along humanist lines and his work was of the calibre of Valla’s. In Spain ‘il devance Erasme lui-même’ (Bataillon, 27). He held the chair of grammar at Salamanca until mid 1513; it was not until then that Cardinal Cisneros was able to attract him to Alcalá, ‘at the cost of a handsome salary’ (Bentley, 88). We also know that Nebrija held radical views on some other matters bearing on the content of the Polyglot, so much so that he withdrew rather than go on associating himself with it. From a surviving letter we know that he felt thwarted on two counts, both issues in which he was on the side of innovation and the rest of the team were conservative, namely his desire to revise the Vulgate, and his strong objection to the planned inclusion of a ninth-century list of interpretations of proper names by Remigius. Nebrija was radical enough to envisage a NT text without accents, and senior enough to be listened to. If he was not in Alcalá at the relevant time, his views could still have had an impact at a distance, especially through those whom he had taught and who were members of the project team, Núñez de Guzmán and López de Zúñiga. In the end there is of course no necessity for Nebrija alone to have been the source: the contents of Poliziano’s book could have been widely known among Spanish Hellenists, including those at work on the Polyglot.55 54

55

Sáenz-Badillos, La filología bíblica, 71–3. Bentley (Humanists, 84–5) argues that Nebrija began work on the Tertia Quinquagena well before 1505. On Nebrija’s NT work, date of arrival, contribution to the Polyglot, and letter to Cardinal Cisneros: Bataillon, Érasme et l’Espagne, 26–41; Sáenz-Badillos, La filología bíblica, 162–7; Bentley, Humanists, 88–91; Homza, Religious Authority, 248 n. 34 (letter). Sáenz-Badillos (166–7) finds some reflection of Nebrija’s views in the interpretations of proper names provided in vol. 5. If this is right, it is an important clue that Nebrija did have some input even into vol. 5, whether in person or through his writings. On Núñez and Zúñiga as pupils of Nebrija: Sáenz-Badillos, 175, 199. But Zúñiga’s teacher of Greek was Barbosa: see De Jonge, Opera Omnia Erasmi, 14. Barbosa, too, had studied in Italy under Poliziano (De Jonge, ibid.; Sáenz-Badillos, 199). Juan de Vergara (born 1492) was still a youngster, though full of promise. According to Sáenz-Badillos (322), he studied Greek under Doukas and Núñez. Homza (4) says an inventory of his books after his death included a work or works of Poliziano (his later brush with the Inquisition is described in entertaining detail in Homza’s ch. 1, pp. 1–48). Full polytonic accentuation was used in the LXX, of which Doukas was probably editor also (Bataillon, Érasme et l’Espagne, 42; Geanakoplos, Byzantium, 239 n. 70; Sáenz-Badillos, 399). If my reconstruction is correct, Doukas got what he wanted there, and it had nothing to do with the

182

COLLECTED ESSAYS ON THE GREEK BIBLE

13. CONCLUSION Whatever the truth may be on the question of who proposed the printing of an unaccented text, the outcome was the invention of the monotonic system, and it occurred in Spain nearly four centuries before the system emerged again in Greece in modern times. Its inventor was almost certainly Dimitrios Doukas of Crete, principal editor of the NT volume of the Polyglot.

reason given in the NT preface. This would explain its specious logic. The justification given is that the LXX was a translation and not the original. But it was just as ancient as the NT, as they well knew: if printing without accents was historically authentic for the NT, it was also for the LXX.

12 ΕΞΑΠΟΣΤΕΛΛΩ 2007

Abstract This paper was contributed to a collection of essays with the theme ‘Septuagint Greek and its significance for the New Testament.’ Each contributor chose a word or words found in both texts and explored the relationship between them, considering also the possible input of other sources. In this study of ἐξαποστέλλω it is shown that the word was adopted by the Pentateuch translators when it was in vogue in the third century BC, but that over the next two centuries it gradually dropped out of use in ordinary Greek. It reappears in the NT almost exclusively in Luke-Acts. I argue that Luke used it as a ‘Septuagintalism,’ rather than taking it from contemporary Greek. Later reappearances of the word suggest that it had entered the ‘biblical’ style from the NT (and LXX). The study illustrates the necessity of taking into account contemporary Greek in tracing the influence of the LXX on NT Greek. The other ‘send’ words, πέμπω and ἀποστέλλω, are included in the discussion, as being in the same semantic field as ἐξαποστέλλω and part of its history.

INTRODUCTION The word ἐξαποστέλλω is used in both the LXX and the NT. It occurs commonly in the former and occasionally in the latter, almost exclusively in Luke–Acts. It would be possible to proceed at once to assess the LXX and NT data and begin to draw conclusions about the links between the two. But there is a third ingredient to be taken into account, namely the usage in Greek outside the biblical texts and contemporary with them, that is, in the period from the third century BC to the first century AD. As this paper will show, a proper appreciation of the LXX and NT cannot be achieved without knowing what was going on in the background. A further point is that ἐξαποστέλλω was part of a semantic field, which it did not occupy alone but shared with other words, notably πέμπω and ἀποστέλλω. These also will need to be taken into acccount in our investigation.

184

COLLECTED ESSAYS ON THE GREEK BIBLE

ΕΞΑΠΟΣΤΕΛΛΩ OUTSIDE THE BIBLICAL TEXTS The reference books tell the story up to a point. LSJ offer some 19 instances, nearly all in post-Classical Greek, including some from documents and from the LXX and NT. A few more documentary references are to be found in MM. BDAG provides the usual brief summary and occasional citations, including some not in LSJ (e.g., Josephus), without figures. Lampe, PGL does include the word (not always the case) and lists a few occurrences in patristic authors. The Mega Lexikon covers much the same ground as LSJ.1 Lee in 1983 drew attention to some further examples in III BC papyri and its frequency in Polybius, and noted that the one apparently Classical example (in a purported letter of Philip in Demosthenes 18.77) is likely to be a later insertion.2 In TDNT there is a brief survey, which sums up with ‘this word became common in Gk. from the time of Polybius with essentially the same meaning as ἀποστέλλειν.’3 All this is more or less what we expect from our present reference works, which have necessarily relied on the old methods of gathering data. The picture so far is of a word that came on the scene in III BC as ‘a more vigorous form of the older word [ἀποστέλλω]’ (Lee), and continued in use in Koine authors right through to the patristic period, being equally at home, as far as one could tell, in both vernacular and literary Greek. But let us now consider what can be learnt from more comprehensive data obtainable by electronic searching. We begin with documentary texts, accessible via the disk PHI7, which contains most of the published inscriptions and papyri.4 The number of occurrences of ἐξαποστέλλω in these texts turns out to be around 270. The majority of these by far are in inscriptions (over 200). The figures for ἀποστέλλω and πέμπω in the same corpus, however, are somewhat over 3,000 each.5 We can immediately deduce that ἐξαποστέλλω was never a strong contender against these two older, better established words. The documentary evidence is also interesting for another, more significant reason. Inspection of the search results reveals that nearly all the 270 or so occurrences of ἐξαποστέλλω are dated to the first three centuries BC; texts from I AD and later number fewer than ten. Within the latter group, 1

2

3

4 5

[DGE has since reached this word and has an extensive selection of examples. I do not see anything that contradicts my findings. See n. 9 below for one inscr.] Lee, Lexical Study, 93–4. References to the useful older studies of Anz (1894) and Glaser (1894) are found there. Cf. also Cadbury, Style of Luke, 32. TDNT, 1:406; cf. 398-400 on ἀποστέλλω and πέμπω. Tagliaferro, in ‘Nota linguistica filoniana,’ explores Philo’s attempted distinction between ἐξαποστέλλω and ἐκβάλλω in Gen 3:23, 24, together with a review of their attestation in non-biblical Greek. Packard Humanities Institute CD ROM 7. It is important to note that figures given are approximate; it is very difficult to make them exact.

ΕΞΑΠΟΣΤΕΛΛΩ

185

the papyrus texts show something interesting too. There are six, and only two date from I AD, the rest from IV and V AD. The two I AD examples are in petitions, PRyl 2.127.22 (29 AD) and SB 6.9150.42 (5 AD).6 The relevant wording in PRyl 2.127 runs … ἀναζητῆσαι ὑπὲρ τοῦ μέρους καὶ τοὺς αἰτίους ἐξαποστεῖλαι ἐπὶ σὲ πρὸς τὴν ἐσομένην ἐπέξοδ(ον), ‘… to enquire into the matter and despatch the culprits to you for the punishment they will receive.’ This is a formulaic section of the petition, the conclusion requesting action of the official, and something of the kind occurs with variations in countless documents of this type. The SB 6.9150 example is similarly located: … ὅπως ἀσϕαλισάμεν[ος] τὸν ἐνκαλούμενον ἐξαπο[σ]τείλῃς ἐϕ᾽ οὓς καθήκει. From earlier centuries the same use of ἐξαποστέλλω in petitions can be readily exemplified, as in PDion 10.13 (109 BC) ὅπως οἱ αἴτιοι ἀναζητηθέντες ἐξαποσταλῶσι ἐπὶ τὸν στρατηγόν.7 It is clear that these two texts of 29 and 5 AD are employing traditional language continued from petitions of previous centuries. This not only explains the two occurrences in documents of I AD but offers a clue to the ‘tone’ that ἐξαποστέλλω might now have, given that it was becoming rare. Its tone is likely to have been formal and official. The later occurrences in papyri, i.e., in IV and V AD as mentioned, are rather isolated and not inconsistent with what has been suggested. They are mainly in official letters in formal style; only one is in a private letter.8 I interpret these as coming from a time when ἐξαποστέλλω had undergone a revival – but not in popular speech: it remained a word of formality. In any case these much later examples are an insignificant number in the overall picture. As to the occurrences in inscriptions, there are, as we have already noted, over 200. Among these I can find only one of certainly AD date, SEG 18.143.45 (Corinth, c.43 AD, copy of decrees and letter of the confederation of the Lycians).9 The inscriptional texts, like this one, tend to employ formulaic, traditional, and formal language. 6

7

8

9

SB 6.9150 is entered in PHI7 with the date ‘Vspc’ but in the edition it is ‘Jahr 5 n. Chr.’ (not ‘5. Jahrh. n. Chr.’). Others are: SB 14.11274.13 (4 BC), PRainCent 51.17 (I BC), PGrenf 1.38.18 (170 BC). The editor of PRainCent 51, Bastianini, notes that ἐξαποστέλλω is frequent at this point in petitions; similarly Tagliaferro, ‘Nota linguistica,’ 417 n. 11, with further references. The official letters are: PPanopBeatty 2.115, 154 (300 AD); POxy 50.3577.6 (342 AD); PPrinc 3.183.5, 11, 19, 21 (345 AD); The private letter is SB 6.9158.[11], 21 (V AD) θέλησον δὲ ἐν τάχι ἐξαποστῖλαι τὸ σημῖον μὴ ὀκνουμένη, ἀμμή. Notice also PapGraecMag 15.16 (III AD) Βαρβαρε Λαιλαμψ …, ἐν τῷ οὐρανῷ θεὸς ὁ μονογενής, ὁ ἐκσαλεύων τὸν βυθόν, ἐξαποστέλλων ὕδατα καὶ ἀνέμους. In BDAG there is a reference to ‘PFouad 203, 2ff (I-IIAD)’ s.v. ἐξαποστέλλω 1.b, but no such text can be located. Editor’s statement of date: ‘paullo post a. 43p.’ Another text of possibly AD date is IG VI 1.14.7 (Sparta), given as ‘primo a. Chr. n. saeculo vel paulo post.’ There remains

186

COLLECTED ESSAYS ON THE GREEK BIBLE

The documentary evidence, then, gives a clear indication that ἐξαποστέλλω was a word that enjoyed a vogue in III–I BC, but thereafter became obsolete in ordinary language. Its use continued then sporadically, but with some noticeable ‘affect’ or additional connotation. Let us now turn to the evidence of Greek literature.10 The overall figure for occurrences of ἐξαποστέλλω is high, around 2,400, but dwarfed by those for ἀποστέλλω (13,000+) and πέμπω (17,000+). The data are more diverse and complicated than in the case of the documents, involving, as they do, a great range of authors of all dates. I offer the following representative sample:

Polybius (III–II BC) Diodorus Siculus (I BC) Dionysius Halic. (I BC) Philodemus (I BC) Philo (I BC–I AD) Josephus (I AD) Vita Aesopi G (I AD) Arrian (I–II AD) Dio Chrysostom (I–II AD) Epictetus (I–II AD) Plutarch (I–II AD) Achilles Tatius (II AD) Lucian (II AD) Pausanias (II AD) Cassius Dio (II–III AD) Xenophon Ephes. (II/III AD) Basilius (IV AD) Eusebius (IV AD) John Chrysostom (IV–V AD)

ἐξαποστέλλω

ἀποστέλλω

πέμπω

208 102 4 0 15 12 0 0 0 0 6 1 0 0 1 0 33 106 103

66 331 104 0 7 55 3 9 3 5 126 2 5 43 61 1 108 228 497

89 93 60 0 14 297 13 62 12 9 332 12 34 18 376 1 23 28 500

We see the predominance of occurrences in the centuries BC, followed by a decline and then a revival in IV AD. It turns out that only Polybius, often thought of as a writer of ‘literary Koine,’ strongly preferred ἐξαποστέλλω

10

a small residue of texts undatable for me, thanks mainly to the reluctance of epigraphers to indicate any kind of date, even within wide parameters (the refusal of some to add ‘AD’ and ‘BC’ (or ‘CE/BCE’) to dates is a further source of uncertainty and frustration). But even if all the uncertainly dated inscriptions turned out to have an AD date they would be too few to affect the general picture. [DGE record one additional AD inscr., IAdramytteion 16.25 (‘II d.C.’), but this turns out to be a mistake: in PHI Greek Inscrr. the date is ‘106 v.Chr.’] Accessed via TLG CD ROM E (1999). Again, figures are not exact.

ΕΞΑΠΟΣΤΕΛΛΩ

187

to its rivals in all its history.11 Other pointers are the following. Philo’s examples are all either in LXX quotations or in re-uses of LXX wording following a quotation. In Josephus, ἐξαποστέλλω, though not in quotations as such, often occurs in contexts with obvious biblical associations. The single example in Achilles Tatius is a curious one: it looks as if it was deliberately used to secure a succession of ἐξ- compounds.12 When we come to the sample patristic authors, Basil, Eusebius, and John Chrysostom, the majority of their examples by far are in biblical quotations or allusions, or in discussions using the words from a previous quotation. In the rather popular Greek of the Apophthegmata Patrum (V–VI AD), ἐξαποστέλλω is not one of the standard ‘send’ words: there is but one occurrence, in a LXX quotation.13 All the data support the interpretation that the heyday of ἐξαποστέλλω was in the third to first centuries BC and it declined thereafter until it entered the vocabulary of ecclesiastical writers under the influence of the biblical texts. It was not an established vernacular word in the first and later centuries AD. I have said nothing so far about the semantics of ἐξαποστέλλω and its competitors ἀποστέλλω and πέμπω, i.e., about their lexical meanings and range of application; I have simply compared the figures. It is of course unlikely that the three words were totally interchangeable. Nevertheless existing analyses show that the overlap was great, especially in the case of ἐξαποστέλλω and ἀποστέλλω, and little can be done to separate them without an exhaustive examination of the examples (together with the minor players ἐκπέμπω and ἀποπέμπω).14 For the present study this would yield little of interest; what is of interest here is the frequency of one word visà-vis its main competitors over time. The result is that we can track its rise and fall, and gain an insight into its currrency and its probable ‘feel,’ or connotation, at any given time. The first of those times is when the LXX came into being. 11 12

13

14

[For useful remarks on ‘literary Koine’ and on Polybius see Horrocks, Greek, 96–8.] Ach. Tat. 6.9.1 ἐβόα πάλιν, Τὸν μοιχὸν ἐξέκλεψας σύ, τῶν δεσμῶν ἐξέλυσας, καὶ τῆς οἰκίας ἐξαπέστειλας· Abba Petros Pionites 2. Figures for ἀποστέλλω and πέμπω in the Apophthegmata: 47 and 41. I have left many loose ends because they do not affect the picture. For example, the TLG search produces an occurrence of ἐξαποστέλλω in Aeschylus (Fragmenta 314c.11), but it turns out not to be in any play of Aeschylus but in a later summary by Diod. Siculus. Dates of authors are taken from the TLG Canon. See Muraoka, Lexicon, s.vv.; Lee, Lexical Study, 94; Dorival, Les Nombres, 308; BDAG, s.vv.; Louw and Nida, Lexicon, 15.66, 67, 68. An analysis of πέμπω, based mainly on Classical data, is found in Chadwick, Lexicographica Graeca, 233–40.

188

COLLECTED ESSAYS ON THE GREEK BIBLE

ΕΞΑΠΟΣΤΕΛΛΩ IN THE SEPTUAGINT The overall figures are: ἐξαποστέλλω 287, ἀποστέλλω 691, πέμπω 22.15 When we look within these at the Pentateuch, we find that ἐξαποστέλλω and ἀποστέλλω are used almost equally, the former 80 times, the latter 94. The Pentateuch translators clearly had a liking for ἐξαποστέλλω, despite the availability and semantic equivalence of the older word ἀποστέλλω. The big loser is πέμπω, which is used only once (Gen 27:42): for some reason it was not thought to be up to the job. Outside the Pentateuch, ἐξαποστέλλω is spread through the different books without any obvious concentrations, but ἀποστέλλω, as the massive figure suggests, overtakes it in frequency in most books after the Pentateuch. πέμπω, however, is noticeably concentrated in books not in the Hebrew Bible (17 times), and among these in books known for their literary Greek (there are 3 in Wis, 12 in 2–3 Macc). I interpret this as due to a feeling on the part of those writers that πέμπω, as the old-fashioned, unadorned word, was appropriate to the classicising style.16 It was with the Pentateuch translators, then, that the fondness for ἐξαποστέλλω began, and we naturally assume this to be a reflection of the currency of the word in their time. This is true up to a point. But let us dig a little deeper. A search of third-century BC Egyptian papyri reveals 556 examples of ἀποστέλλω, 21 of πέμπω, but only 5 of ἐξαποστέλλω.17 These figures are quite surprising. Such a marked disparity between ἐξαποστέλλω and ἀποστέλλω in particular is at odds with the Pentateuch frequencies. It begins to look as if the Pentateuch translators favoured ἐξαποστέλλω even beyond what would have been usual at the time. Why would that be? We have no means of knowing for certain, but the explanation could be that ἐξαποστέλλω was marked as the slightly more formal and impressive word in comparison with ἀποστέλλω, and might therefore have been regarded as especially suited to the biblical style. It was not a 15

16

17

Figures from LEH, Lexicon, in turn derived from the CATSS files. The Hebrew equivalent in nearly all instances is ‫ שׁלח‬qal or piel. The figures for πέμπω in authors of literary Greek (above) tell a similar story. There is a parallel in the distribution of κελεύω in the LXX: see Lee, ‘Thirty Years On,’ 520 [= Essay 8]. In twenty vols. containing the majority of III BC texts, viz. PHib 1, 2; PLille 1, 2; PSI 4–6; PPetr 1–3; PMich 1; PCol 3, 4; PCorn; PCairZen 1–5; PYale 1 (part). The larger figures are not precise. The five III BC examples are: PSI 4.384.4; PPetr 2.37.3; PCairZen 1.93.7; 4.578.2; PYale 1.39.12. [A fresh search via Papyri.info locates five more: PHal 7.2; PKoeln 6.258.2; SB 18.14042.9; 28.16923.1; PSorb 1.50.4. The previously noted PYale 1.39.12 is now dated c.190 BC, not quite in III BC, but near enough. The tally, then, is 10.]

ΕΞΑΠΟΣΤΕΛΛΩ

189

total equivalent of ἀποστέλλω, but because of the doubled-up prefix carried some additional connotation, however slight. Of course it is obvious that often the translators are simply ringing the changes on the two words, but this need not invalidate the point.18 The result we have for the LXX is this. The LXX translators adopted a word that was current in the early Koine of their time and made it a characteristic LXX word by frequent use; there is, in addition, a possibility that it was slightly more formal in tone than its main equivalent. With the passage of time, however, changes would occur in these words. Writers of the next period would make different choices, in accordance with the vocabulary of their own day; their choices might also reflect acquaintance with an older text that had significance for them. ΕΞΑΠΟΣΤΕΛΛΩ IN THE NEW TESTAMENT In the first century AD ἐξαποστέλλω was no longer a standard part of the Greek vocabulary. That it might still appear sporadically is not to be denied – no word ever disappears completely from Greek! – but the evidence has shown that both vernacular texts and literary authors made little use of it now. We would not expect to see it in the Greek of the NT writers, unless there were special factors at work. The statistics for the three main ‘send’ words in the NT are:

NT as a whole Matt Mark Luke–Acts John Gal All other Epistles Rev

18

ἐξαποστέλλω

ἀποστέλλω

πέμπω

12 0 0 10 0 2 0 0

132 22 20 49 28 0 9 3

79 4 1 21 32 0 16 5

E.g., Gen 8:7 ἀπέστειλεν, 8 ἀπέστειλεν, 10 ἐξαπέστειλεν, 12 ἐξαπέστειλεν. Exod 3:10 ἀποστείλω, 12 ἐξαποστέλλω, 13 ἀπέσταλκεν, 14 ἀπέσταλκεν, 15 ἀπέσταλκεν, 20 ἐξαποστελεῖ. 1 Kgdms 5:8 ἀποστέλλουσιν, 10 ἐξαποστέλλουσιν, 11 ἐξαποστέλλουσιν, ἐξαποστείλατε. In these passages ἐξαποστέλλω has the role of ‘varier’ and ἀποστέλλω that of ‘varied.’ I thank Michael Curran for this observation and his comments on this paper generally.

190

COLLECTED ESSAYS ON THE GREEK BIBLE

We see that there are some examples of ἐξαποστέλλω after all, but they are quite few in number, and concentrated in the writings of one author, Luke. The predominant word throughout is ἀποστέλλω, even in Luke. The old plain word πέμπω makes a strong showing in John, who has a special fondness for it in ὁ πέμψας με (22 times) and similar expressions, but otherwise it shows up most in the more literary Greek of Luke and the Epistles.19 It is a commonplace of NT commentary that Luke was a ‘Septuagintaliser.’ That is, he not only incorporated many direct quotations from the LXX into his work, but consciously imitated its language and style by frequent use of LXX expressions and vocabulary. Moreover, some degree of adaptation to context is observable: that is, LXX features may be deliberately placed to suit the character of the context. Though details remain debatable, this seems to be the consensus view; it is certainly the best explanation of the phenomena.20 The conclusion is easily reached. Luke displays a fondness for LXX features in his writing; our word was a characteristic LXX word; none of the occurrences in Luke are simply a matter of direct quotation; and we know that ἐξαποστέλλω was obsolete in contemporary Greek. We conclude that Luke drew it from the vocabulary of the LXX.21 We can also conclude that the two occurrences in Galatians must be explained as either older Greek or Greek in the Septuagintal style, not contemporary usage. Let us take a closer look at the NT occurrences in the light of this conclusion and see what else can be observed. The first example is in the Infancy Narrative, in Mary’s Song of Praise: 19

20

21

Mark’s one instance of πέμπω is odd: 5:12 πέμψον ἡμᾶς εἰς τοὺς χοίρους, ἵνα εἰς αὐτοὺς εἰσέλθωμεν. Other NT words: ἐκπέμπω twice (Acts 13:4; 17:10), ἀποπέμπω not at all; perhaps ἐκβάλλω, included with ἐκπέμπω and ἐξαποστέλλω by Louw and Nida, Lexicon in 15.68, instanced by Matt 9:38. From the extensive literature: Fitzmyer, Luke I–IX, 113–25; Ó Fearghail, ‘Imitation of the LXX’; Haenchen, Acts, 74–5, 80; Barrett, Acts, 2:xlv–xlviii; Witherington, Acts, 123–5; Arnold, ‘Luke’s Characterizing Use.’ Sparks’s articles ‘Semitisms of Luke’ and ‘Semitisms of Acts’ retain their value (ἐξαποστέλλω is not mentioned). ‘Context’ embraces situation, speaker, subject-matter. Some Marcan instances of adaptation to social context are presented in Lee, ‘Some Features.’ Commentators often point out that ἐξαποστέλλω is restricted to Luke–Acts (and Galatians) in the NT, and may add that it occurs in the LXX, but the conclusion does not seem to be drawn. Fitzmyer lists ἐξαποστέλλω among characteristic Lucan vocabulary (Luke I–IX, 110), but not as a ‘Septuagintism.’ Tagliaferro, ‘Nota linguistica,’ 417 n. 13 drew the opposite conclusion to mine, namely that ἐξαποστέλλω in the NT ‘appare conforme al contemporaneo greco profano piuttosto che a quello dei LXX.’ But this was based on an incomplete survey of the evidence.

ΕΞΑΠΟΣΤΕΛΛΩ

191

Luke 1:53 πεινῶντας ἐνέπλησεν ἀγαθῶν καὶ πλουτοῦντας ἐξαπέστειλεν κενούς.

It is well known that the Magnificat (vss. 46–55) is a canticle composed in OT style, with similarities to the Song of Hannah in 1 Kgdms 2:1–10. It is, in Fitzmyer’s words, ‘a mosaic of OT expressions drawn from the LXX.’ The actual combination seen here, ἐξαποστέλλω + κενόν, is attested in the LXX (Gen 31:42; Deut 15:13; 1 Kgdms 6:3; Job 22:9). Many commentators argue moreover that Luke has re-used a pre-existing hymn, with his own adaptation only in vs. 48, if that. However that may be, the Septuagintal character of the whole context and of this instance of ἐξαποστέλλω is very distinct, and our investigation does no more than confirm what is already clear.22 Next are the two examples in Luke 20:10, 11. Here it is useful to see the context as a whole and to take into account the Synoptic parallels. Luke 20:10 καὶ καιρῷ ἀπέστειλεν πρὸς τοὺς γεωργοὺς δοῦλον ἵνα ἀπὸ τοῦ καρποῦ τοῦ ἀμπελῶνος δώσουσιν αὐτῷ· οἱ δὲ γεωργοὶ ἐξαπέστειλαν αὐτὸν δείραντες κενόν. 11 καὶ προσέθετο ἕτερον πέμψαι δοῦλον· οἱ δὲ κἀκεῖνον δείραντες καὶ ἀτιμάσαντες ἐξαπέστειλαν κενόν. 12 καὶ προσέθετο τρίτον πέμψαι· … 13 εἶπεν δὲ ὁ κύριος τοῦ ἀμπελῶνος· τί ποιήσω; πέμψω τὸν υἱόν μου τὸν ἀγαπητόν· || Mark 12:2 καὶ ἀπέστειλεν πρὸς τοὺς γεωργοὺς τῷ καιρῷ δοῦλον … 3 καὶ λαβόντες αὐτὸν ἔδειραν καὶ ἀπέστειλαν κενόν. 4 καὶ πάλιν ἀπέστειλεν πρὸς αὐτοὺς ἄλλον δοῦλον· κἀκεῖνον ἐκεϕαλίωσαν καὶ ἠτίμασαν. 5 καὶ ἄλλον ἀπέστειλεν· … 6 ἔτι ἕνα εἶχεν υἱὸν ἀγαπητόν· ἀπέστειλεν αὐτὸν ἔσχατον … || Matt 21:34 ὅτε δὲ ἤγγισεν ὁ καιρὸς τῶν καρπῶν, ἀπέστειλεν τοὺς δούλους αὐτοῦ … 35 καὶ λαβόντες οἱ γεωργοὶ τοὺς δούλους αὐτοῦ ὃν μὲν ἔδειραν, ὃν δὲ ἀπέκτειναν, ὃν δὲ ἐλιθοβόλησαν. 36 πάλιν ἀπέστειλεν ἄλλους δούλους … 37 ὕστερον δὲ ἀπέστειλεν πρὸς αὐτοὺς τὸν υἱὸν αὐτοῦ …

It can be seen that the other Synoptics use only one word and repeat it as often as necessary, but Luke introduces variation. His resources include all three ‘send’ words: their word ἀποστέλλω, the old standby πέμπω, and his LXX-flavoured ἐξαποστέλλω. Where Mark had simply ἀπέστειλαν κενόν, he changes the verb to ἐξαποστέλλω, thus securing a match 22

Fitzmyer, Luke I–IX, esp. 358–62; quotation 359. Cf. Marshall, Gospel of Luke, 77–9, 82–5.

192

COLLECTED ESSAYS ON THE GREEK BIBLE

with the LXX (and an echo of Luke 1:53), and then repeats the expression in the next verse in place of what Mark had.23 It appears that Mark’s ἀπέστειλαν κενόν provided the trigger for Luke to introduce a Septuagintalism. We notice also Luke’s introduction into the context of the obvious Septuagintalism προσέθετο + inf., ‘do x again’ (11, 12), which is not present in the parallels.24 In Acts there are seven examples. In some, though not all, there is discernible adaptation to context in the choice of this word. Three of the examples are in speeches, which are usually significant contexts in Acts. For one thing, they may contain concentrations of OT quotations, which are part of the characterisation of the speaker as a figure in ‘biblical history’ speaking in a biblical style. This is especially noticeable in the speeches in 1–4, 7 and 13.25 Besides quotations proper, OT allusions and linguistic features may be introduced for a similar purpose. All this is one aspect of the generally higher register of the speeches in Acts. Thus we have in the speech of Stephen before the Sanhedrin: Acts 7:12 ἀκούσας δὲ Ἰακὼβ ὄντα σιτία εἰς Αἴγυπτον ἐξαπέστειλεν τοὺς πατέρας ἡμῶν πρῶτον.

Here the content of Gen 42:2 is summarised. But ἐξαποστέλλω is not actually used in that place in the LXX. Luke introduces it, so producing a Septuagintal turn of phrase. We can almost watch Luke at work. Similarly in Paul’s speech in the synagogue of Antioch, ἐξαποστέλλω plays a part in the ‘biblical’ (and elevated) effect: Acts 13:26 Ἄνδρες ἀδελϕοί, υἱοὶ γένους Ἀβραὰμ καὶ οἱ ἐν ὑμῖν ϕοβούμενοι τὸν θεόν, ἡμῖν ὁ λόγος τῆς σωτηρίας ταύτης ἐξαπεστάλη.

Another example occurs in the speech of Paul from the steps of the temple in Jerusalem: Acts 22:21 καὶ εἶπεν πρός με· πορεύου, ὅτι ἐγὼ εἰς ἔθνη μακρὰν ἐξαποστελῶ σε.

The Septuagintal style can be seen as appropriate to God as the speaker of the words Paul is reporting, as well as the right style for the audience. 23

24

25

The distinctiveness of ἐξαποστέλλω κενόν is hightlighted by the fact that ἀποστέλλω is never used in this combination in the LXX. Cadbury (Style of Luke, 167) sees the change to ἐξαποστέλλω in vs. 10 as due to Luke’s preference for compound verbs; this is an undeniable feature of Luke’s style, but not the main factor here. Marshall (Gospel of Luke, 729) aptly comments on vs. 11: ‘πέμπω is a literary variant for ἀποστέλλω, diff. Mk.’ Cf. esp. Arnold, ‘Luke’s Characterizing Use,’ with references there.

ΕΞΑΠΟΣΤΕΛΛΩ

193

A similar use of ἐξαποστέλλω, in that God is again the one doing the sending, is in: Acts 12:11 καὶ ὁ Πέτρος ἐν ἑαυτῷ γενόμενος εἶπεν· νῦν οἶδα ἀληθῶς ὅτι ἐξαπέστειλεν [ὁ] κύριος τὸν ἄγγελον αὐτοῦ καὶ ἐξείλατό με ἐκ χειρὸς Ἡρῴδου.

The other examples in Acts are in narrative, as follows: Acts 9:30 ἐξαπέστειλαν αὐτὸν εἰς Ταρσόν. Acts 11:22 ἐξαπέστειλαν Βαρναβᾶν [διελθεῖν] ἕως Ἀντιοχείας. Acts 17:14 εὐθέως δὲ τότε τὸν Παῦλον ἐξαπέστειλαν οἱ ἀδελϕοὶ πορεύεσθαι ἕως ἐπὶ τὴν θάλασσαν.

The lexical meaning is not quite the same in all of these. In 11:22 we have simply ‘despatch’ a person to perform some action or mission, as in 7:12; 12:11; 22:21 above; but in 9:30 and 17:14 it is something like ‘see (a person) off on his/her way,’ a LXX usage seen, e.g., in Gen 26:31; 31:27; 1 Kgdms 9:19, 26. Whether or not there is anything significant here in terms of location, the use of the word in this sense confirms Luke’s intimacy with LXX usage.26 Outside Luke–Acts there are only the two examples, in Galatians, which occur in the same context and are close in thought and wording: Gal 4:4 ὅτε δὲ ἦλθεν τὸ πλήρωμα τοῦ χρόνου, ἐξαπέστειλεν ὁ θεὸς τὸν υἱὸν αὐτοῦ, γενόμενον ἐκ γυναικός. … 6 ῞Οτι δέ ἐστε υἱοί, ἐξαπέστειλεν ὁ θεὸς τὸ πνεῦμα τοῦ υἱοῦ αὐτοῦ εἰς τὰς καρδίας ἡμῶν κρᾶζον·

There is no reason for Paul to introduce this word instead of his usual ἀποστέλλω and πέμπω (note Rom 8:3 ὁ θεὸς τὸν ἑαυτοῦ υἱὸν πέμψας), unless one wishes to argue that he chose it specially to suit this weighty pronouncement. But it would remain a little odd that Paul used the word only here. The explanation probably lies elsewhere. Other lines of enquiry by NT scholars have led to the conclusion that Paul incorporates a prePauline statement into vs. 4b, and possibly another in vs. 6, or his own adaptation of the preceding one.27 ἐξαποστέλλω may not be Paul’s usage, then, but someone else’s. On the basis of what we have established about the character of the word, it would seem to be a case of deliberate choice 26

27

In 17:14 there is an element of variatio, since ἐκπέμπω is used in the same way shortly before (vs. 10 οἱ δὲ ἀδελϕοὶ εὐθέως διὰ νυκτὸς ἐξέπεμψαν τόν τε Παῦλον καὶ τὸν Σιλᾶν εἰς Βέροιαν). See Betz, Galatians, 205–10; Bruce, Galatians, 194–5; Longenecker, Galatians, 166– 74; Lambrecht in EDNT, 2:2–3; Tarazi, Galatians, 201–8.

194

COLLECTED ESSAYS ON THE GREEK BIBLE

of ‘biblical,’ i.e., Septuagintal, vocabulary by the composer of the statement (a parallel with Wis 9:10 has been noted). Finally, two other occurrences not counted above may be briefly noticed: Luke 24:49 καὶ [ἰδοὺ] ἐγὼ ἀποστέλλω τὴν ἐπαγγελίαν τοῦ πατρός μου ἐϕ᾽ ὑμᾶς·28 εξαποστελλω ‫ א‬2 Β Δ 33. … ‫ آ‬εξαποστελω L 892 ‫ آ‬txt ो75 ‫ *א‬Α C DΚΝWΓΘΨ…ै

The reference is to the sending of the Spirit; hence there is a connexion to Gal 4:6 just discussed. But given the uncertainty of the text, it is hard to see what might follow. As to the choice of reading, one could argue that ἐξαποστέλλω was introduced because it was felt to be appropriate to Lucan style in such a context. Mark 16:8 [[Shorter Ending: μετὰ δὲ ταῦτα καὶ αὐτὸς ὁ Ἰησοῦς ἀπὸ ἀνατολῆς καὶ ἄχρι δύσεως ἐξαπέστειλεν δι᾽ αὐτῶν τὸ ἱερὸν καὶ ἄϕθαρτον κήρυγμα τῆς αἰωνίου σωτηρίας. ἀμήν.]]

This occurrence, though in a passage unlikely to be a genuine part of Mark’s Gospel, leads us into the next stage in the history of our word in Christian circles. I have suggested above that ἐξαποστέλλω ‘entered the vocabulary of ecclesiastical writers under the influence of the biblical texts.’ The Shorter Ending, whatever the date of writing, clearly shows how ἐξαποστέλλω might be used in compositions in the biblical style in the early Church.29 Another sign of this is to be seen in the occurrences in the ‘Apostolic Fathers.’ A couple of samples will give an idea of the flavour of the passages. The model now is not only the LXX but the NT. Thus 2 Clement 20:5 (II AD): Τῷ μόνῳ θεῷ ἀοράτῳ, πατρὶ τῆς ἀληθείας, τῷ ἐξαποστείλαντι ἡμῖν τὸν σωτῆρα καὶ ἀρχηγὸν τῆς ἀϕθαρσίας, . . . αὐτῷ ἡ δόξα εἰς τοὺς αἰῶνας τῶν αἰώνων. ἀμήν.30

Hermas, Vis. 4.2.6 (II AD): πιστεύσατε τῷ κυρίῳ, οἱ δίψυχοι, ὅτι πάντα δύναται καὶ ἀποστρέϕει τὴν ὀργὴν αὐτοῦ ἀϕ᾽ ὑμῶν καὶ ἐξαποστέλλει μάστιγας ὑμῖν τοῖς διψύχοις.31 28 29

30 31

[Text and app. crit. NA28.] Note this characterisation of the Shorter Ending in Metzger, Textual Commentary, 125– 6: ‘Besides containing a high percentage of non-Markan words, its rhetorical tone differs totally from the simple style of Mark’s Gospel.’ Ed. K. Bihlmeyer (Tübingen, 1956). Ed. M. Whittaker (Berlin, 1967). Others at Sim. 8.6.3; 9.14.3. It is notable that God is the sender in all of these.

ΕΞΑΠΟΣΤΕΛΛΩ

195

Subsequently, ἐξαποστέλλω enters into liturgical texts, and not just in quotations, so clearly it had become established in the vocabulary of the Church. So in the Divine Liturgy of St John Chrysostom: καὶ ἐπὶ πάντας ἡμᾶς τὰ ἐλέη σου ἐξαπόστειλον.

And in this poetic composition of VII AD: Λελυμένους σοὺς δεσμίους, Σιών, ἐξαπόστειλον καὶ ἐκ λάκκου ἀγνωσίας ἀνύδρου ἐξάγαγε·32

CONCLUSION The word ἐξαποστέλλω was a new formation that entered Koine Greek in the third century BC. Unlike numerous other newcomers of the period, it was not destined to be the vernacular replacement for some older obsolescent word, but remained less common than its near synonyms ἀποστέλλω and πέμπω. While it was close to both in lexical meaning, there are signs that its connotation differed, even from the beginning, in the direction of greater formality. The Pentateuch translators of the LXX took it up and made extensive use of it, followed by other translators, thereby making it a characteristic LXX word. It continued in use until the end of the first century BC, particularly in the formal language of inscriptions and in the ‘literary Koine’ of writers like Polybius, but thereafter it declined, so that by the first century AD its occurrence in ordinary Koine Greek was unusual. The NT author Luke took it up not from the language of his day but from the LXX, whose language he extensively imitated. It imparted a ‘biblical’ flavour, whether by being part of a LXX expression or by being in itself an old-fashioned word associated with the LXX. In Luke’s writings it is to some extent significantly located, in that it appears in contexts where its flavour plays a role in the general effect. Two examples outside Luke, in Galatians, are also likely to derive from Septuagintal language. After the NT the word quickly entered the language of Christian writers as a NT as well as LXX word, and continued into ecclesiastical Greek of the 32

The first is in the Holy Anaphora in the section beginning Μνήσθητι, Κύριε, τῆς πόλεως ταύτης, the second in the Kanon (Ode 6) for Orthros of Palm Sunday, attributed to Kosmas the Monk (or, according to some, St Andrew of Crete). ἐξαποστέλλω continued into medieval Greek and survives today in modern Greek (demotic form ξαποστέλνω). It is unclear what role may have been played by ecclesiastical language or puristic Greek. It seems significant that in the modern Greek versions of the NT by Kolitsaras (Athens, 1964) and the UBS (Athens, 1989), ἐξαποστέλλω is always replaced, mostly by στέλνω. For medieval Greek, see Kriaras, Λεξικό, s.v. εξαποστέλλω. For modern, see, e.g., Lex. Proia; Mega Lexikon; Babiniotes, Λεξικό; all s.v. ξαποστέλνω.

196

COLLECTED ESSAYS ON THE GREEK BIBLE

patristic period. It also survived to be used sporadically in late Koine, especially in official contexts. Some reflections on method may be offered. In NT language study it is usual to regard the usage of the LXX as having a major, sometimes crucial, role to play in understanding the NT, and rightly so. But it is never a simple matter of comparing the one with the other. The ‘third ingredient,’ the usage of the language contemporary with both, must be taken into account. Before any feature of NT language can be categorised as a Septuagintalism, or as anything else for that matter, it is necessary to know what non-biblical Greek usage was at the time of the NT. A moment’s reflection will show that this knowledge is a prerequisite in all such attempts at categorisation, otherwise we might simply be looking at a feature common to the NT and Koine Greek. The frequent NT expression ἀποκριθεὶς εἶπεν, for example, is not recognised as a Septuagintalism simply because it is common in the LXX, but also because it is unlike anything that normally occurs in Greek at any time; conversely no one would suggest that λέγει, ‘he says’ is a Septuagintalism, even though it is common in the LXX as well as the NT, because we know it is ordinary Greek. This methodological underpinning, though obvious once pointed out, is easy to neglect and has often been inadequately followed in practice, partly because of the difficulty of gathering the data. Today the situation has changed. Electronic resources make it possible to assemble fuller data than ever before. A truer picture of the usage of any word can be seen, as ἐξαποστέλλω has demonstrated. This potential for improved information on the state of the language cannot be neglected in future study. A further important point is evident. Koine Greek was not a static entity. Over the long span of 900 years change was inevitable. This too is obvious, but it is easy to forget to look for change when studying some LXX or NT feature. In the past there has been a tendency to rest content with establishing that a feature was found in Koine Greek, as though it would have been spread evenly through all periods. Attention to the progress of the language is also needed. There are many signs that in the 300 years between the LXX and the NT there were important shifts in usage. In fact there are grounds for dividing Koine Greek at the point where I BC ends and I AD begins, as Thumb long ago suggested.33 Our word ἐξαποστέλλω has proved a good instance of the changes that could occur from one period to the next. 33

Thumb, Hellenismus, 9–10. Cf. Moulton, Grammar, I, 41. It is time to go further, I think, and recognise a division into Early (III–I BC), Middle (I–III AD) and Late Koine (IV– VI AD).

ΕΞΑΠΟΣΤΕΛΛΩ

197

Postscript The proposed division of Koine Greek into three periods, put forward in passing in footnote 33, has been taken up by several scholars in the field. It is proving a useful adjunct to the study of post-Classical Greek.34

34

See Aitken, ‘Phonological Phenomena,’ 262, 277; Bentein, ‘Perfect Periphrases,’ 207; La Roi, ‘Diachronic Analysis,’ 230; ‘Insubordination,’ 21; Stolk, ‘Dative and Genitive,’ 188 n. 10.

13 A LEXICOGRAPHICAL DATABASE FOR GREEK: CAN IT BE FAR OFF? THE CASE OF AMPHODON 2008

Abstract This paper was presented in Wuppertal in 2006, in a session on the possible contribution of databases to LXX studies. I chose the case of ἄμϕοδον as an illustration. Information on this word in lexicons and scholarly discussions and the data on its occurrences have never been systematically collected or analysed. The meaning of ἄμϕοδον consequently remains unsettled and false meanings have been perpetuated. There are two occurrences in the LXX in Jeremiah, which I discuss. An interim lexical analysis is offered. The overall conclusion is that the only solution to the problem of scattered information is to store data on each word in an ongoing electronic database.

I A partial database of the Septuagint vocabulary already exists. It was created by James Aitken, initially as part of the AHRB Greek Bible Project. From that starting-point it has continued to be developed and will shortly be available on line as the ‘Demetrios database.’1 The benefits of the database approach are already evident from the samples available. But this paper is designed to take matters a step further. A database for the LXX, valuable though it is, inevitably constitutes the first step towards a database that covers all of Greek, which is also an important desideratum. Once electronic data are collected on any word in the LXX, data have also been collected for the study of that word wherever else it may occur in Greek. Conversely, any collection of data related to the occurrence of a word in some other text outside the LXX offers data of value to the Septuagintalist if the word also occurs in the LXX. Instead, therefore, of 1

[Archived at: https://ota.bodleian.ox.ac.uk/repository/xmlui/handle/20.500.12024/2528 (Jim Aitken).]

200

COLLECTED ESSAYS ON THE GREEK BIBLE

creating different databases for different Greek texts and keeping things in separate compartments, there is everything to be gained by aggregating the data and serving the needs of everyone at once. Given the nature of the electronic medium, there is nothing to stop us doing so. As to the importance of such a tool, let me briefly describe the situation in Greek lexicography generally and mention previous contributions I have made on this topic. In a short paper presented in 1995 (but not published until 2004), I surveyed the state of lexicography of ancient Greek and pointed out that ‘the coverage is partial, unsystematic, and uneven in quality.’ Ideally, what is needed is not a series of updated lexicons but ‘a complete lexical treatment of Greek from its beginnings right through to the end of Byzantine Greek … making use of all the occurrences, to ensure that all the evidence is brought to bear at once.’ I made only tentative suggestions about an electronic database. One of them was that we could establish ‘a world database to collect the results of all published work on individual New Testament words,’ this being not just a bibliography but a summary of results. I also considered the possibility of using the recently created electronic LSJ as the basis of ‘an LSJ database on line for anyone to add to or comment on.’2 I was, however, wary of this because of the problem of sifting and organising the contributions.3 Much more developed ideas are to be found in my History, where I put forward proposals for the direction that NT lexicography might take in the future. Fundamental to my view is the creation of an electronic resource for gathering the material. The goal would remain the same, that is, ‘the practical one of delivering an accurate description of the meaning of each Greek word in the New Testament,’ but the database would be an ongoing, long-term means of assembling and assessing all the data relevant to the task.4 Similar ideas, applied to the lexicography of ancient Greek as a whole, were developed in my paper ‘Releasing LSJ from its Past,’ designed to address the question: what is the future of LSJ?5 This represents my fullest views so far. There are serious flaws in LSJ, despite its great value. 2

3

4 5

Lee, ‘Present State,’ 67, 72–3, 74, 73 [= Essay 9]. That paper was presented at the SBL Annual Meeting 1995 in Philadelphia. My remarks misled at least one person into thinking I was advocating ‘a kind of lexicographical chat-room’ – which is far from what I intended: see Büchner, Review of LEH, Lexicon, 147. Lee, History, ch. 11; quotation 182. Delivered in 2002 at Cambridge Colloquium on Ancient Greek Lexicography; MS for publication 27.10.03; a volume of colloquium papers is projected. [Published 2010 = Essay 17.]

A LEXICOGRAPHICAL DATABASE FOR GREEK

201

They are the result of the constraints, both practical and methodological, within which earlier lexicographers worked. An old-style revision would not solve them but perpetuate them. The way forward is by the creation of an electronic database to assemble and reassess all the data on every word in ancient Greek. II I turn now to an illustration, taking a word in the LXX as my startingpoint. From it we see how inadequate is the information currently contained in our lexicons, how extensive the data available if all sources are tapped, and how useful a full collection would be for deciding the meaning of the word, in the LXX or anywhere else. The word is ἄμϕοδον, which occurs twice in the LXX, translating Hebrew ’armôn: Jer 17:27 καὶ ἀνάψω πῦρ ἐν ταῖς πύλαις αὐτῆς, καὶ καταϕάγεται ἄμϕοδα Ιερουσαλημ καὶ οὐ σβεσθήσεται. Jer 30:16 (MT 49:27) καὶ καύσω πῦρ ἐν τείχει Δαμασκοῦ, καὶ καταϕάγεται ἄμϕοδα υἱοῦ Αδερ.

It is not entirely clear what the word ἄμϕοδον means. Let us suppose we want to find out. The first step would be to consult a lexicon, let us say LEH, which gives the meaning as: block of houses surrounded by streets.6 You would most probably turn then to LSJ, where you would find: ‘I. street … II. block of houses surrounded by streets …: hence, ward, quarter of a town, LXX Je. 17.27. …’7 You might think this was LSJ’s final word on the subject and not think to check the 1996 supplement, which has further adjustments, no small ones.8 We are now told to delete all of ‘hence … town’ and after the Jeremiah reference to add ‘as administrative unit.’ The upshot is rather puzzling: we lose ward, quarter, generally understood to be the commonest meaning, and gain a clarification of Jer 17:27 that seems irrelevant to it. But nothing has been done about block of houses surrounded by streets, which is a questionable meaning, though the ordinary user would have no reason to suspect this, not knowing that it goes back to the 5th ed. of Liddell and Scott (1861) and takes no account of new evidence or discussion since that time. Its perpetuation in LEH is an unfortunate effect of relying on LSJ for meanings. 6 7 8

LEH, Lexicon, s.v. Part 1 of LSJ containing ἄμϕοδον appeared in 1925. LSJ Suppl. (1996).

202

COLLECTED ESSAYS ON THE GREEK BIBLE

Let us suppose we investigate further in the lexicons. There are several to sample. In BDAG the focus is naturally on the one NT occurrence (Mark 11:4), given as ‘street,’ but we are also told that this comes from the meaning ‘a city quarter.’ Sophocles’s old lexicon offers a few references, in all of which it is taken to mean ‘street, in a city,’ including Jer 17:27.9 DGE is also to be consulted. Its meanings are 1 calle … 2 plazuela … 3 barrio … with Jer 17:27 and some documentary examples placed under sense 3, i.e., ‘quarter.’ Montanari’s lexicon (GI) similarly assigns Jer 17:27 to isolato, quartiere, rione.10 There is also the Mega Lexikon, which has its own take: (1) ‘road running around a neighbourhood’; (2) ‘group of houses separated by roads from other habitation, separate neighbourhood, quarter.’11

III Vistas are opening up, and we seem to be finding more questions than answers. Yet we have hardly begun. Next must be considered what is in the literature, that is, scholarly discussions of the word in articles and books. The lexicons give almost no clue to this: LEH (2003) lists Milligan12 and one article by Tov;13 BDAG (2000) refers to MM and Preisigke, Wörterbuch, in all a total of four items. LSJ, DGE and others attempt no coverage of this kind. How would one know that there are two very recent articles, a short but useful discussion by Lolos (2003)14 and a much fuller treatment by Du Bouchet (2004)?15 Or that there is a long history of discussion of this interesting word? Among the older items is one by Buck (1922), a master philologist whose opinion ought not to be neglected.16 Buck declared that the meaning commonly adopted in the lexicons is simply wrong: the primary notion is likely to be ‘that which is on both sides of the street,’ so the entity is ‘the complex of houses facing a street and forming a court which may readily be closed at each end.’ 9 10 11

12 13 14 15 16

Sophocles, Lexicon. [GE translates GI as block, quarter, district.] My translation. Orlandos and Travlos (Λεξικὸν Ἀρχαίων Ὅρων) work with two similar senses, adding a long list of references including Mark 11:4 and Jer 17:27, but without indicating which meaning is to be assigned to them. Lampe, PGL does not cover this word. [Muraoka, Lexicon (2009): quarter of a town.] Milligan, Selections, 81. Tov, ‘Three Dimensions,’ 530–1. Lolos, ‘Greek Roads,’ 165–6. Du Bouchet, ‘Quartiers.’ Buck, ‘Greek ἄμϕοδον,’ 114.

A LEXICOGRAPHICAL DATABASE FOR GREEK

203

By following other trails and sources I have brought to light over two dozen items.17 Even so, the list does not include translations and commentaries, or anything published before 1900. Though not all these discussions may be useful (and some are indeed useless), they all have some relevance and need to be assessed for their contribution. And it is not enough simply to read the last of them: what is noticeable is that there has been no systematic accumulation of data, such as to allow one to find out where the discussion is up to by consulting one item. The lexicons, whose task one supposes is to do just that, are the worst of all. For reasons beyond their control, lexicons simply cannot cover the data in a thorough and up-to-date way. The authoritative and reliable statement that everyone supposes they give may be anything but. But this is the end of the story, isn’t it? By no means. We have not yet dealt with evidence of the word in use, i.e., occurrences in ancient Greek texts. The collection of citations in the lexicons may seem extensive, but electronic searching reveals that there is much more available. In Greek literature, on the basis of TLG CD-ROM E, there are over 200 occurrences; in documents, if we use PHI CD-ROM 7, there are over 1,700 examples. The majority of the latter are in papyri, where it is a word of common occurrence, mostly in connexion with named ἄμϕοδα.18 This evidence has not yet been systematically studied. Ideally, all of it needs to be gathered and assessed to ensure that the usage of our word is correctly appreciated. Du Bouchet’s study, the most recent in the field, offers a good summary, but it is not a systematic presentation with lexical meanings, nor is it clear how complete his collection is. The nature of the problem in front of us is clear, and its solution ready to hand. An ever-growing body of data has never been systematically collected and analysed; it could all be gathered electronically in one place, where it could be organised, discussed, and kept up to date. Furthermore, whatever is collected in this way will be relevant to the study of ἄμϕοδον wherever it occurs.

17

18

Additional to those noted already or below are the following (11 items): Garlan, ‘Cités, armées,’ 21–2; id., Recherches, 382; Hennig, ‘Staatliche Ansprüche,’ 250–1; Hohlwein, L’Égypte Romaine, 95–6, 334; Jouguet, La vie municipale, 136–7, 282–91; Klamp, ‘Testament des Taharpaesis,’ 113; Llewelyn, New Documents, 7:41–2; Preisigke, Fachwörter, s.v.; Rink, Strassen- und Viertelnamen, 7–17; Robert, À travers l’Asie Mineure, 152–4; Thackeray, Septuagint and Jewish Worship, 34. Daris (‘I Quartieri’) has recently compiled a list of the known ἄμϕοδα of Oxyrhynchus, which number over 40.

204

COLLECTED ESSAYS ON THE GREEK BIBLE

IV Let us return to the two examples of ἄμϕοδον in the LXX. What can be said about their meaning? Obviously it would be advisable to approach them with a prior knowledge of the meaning in Greek generally, including any progression over time. A final statement of this cannot be offered here; it must await the full study of the evidence that is still required. Nevertheless I put forward the following as an interim analysis and a corrective to the misleading information in the lexicons: 1 a complex of houses facing a street and forming a court closable at either end 2 a street in such a complex 3 one such complex or a group of them forming an administrative district of a town, often bearing a name; a ‘quarter’19

The Hebrew original in LXX Jer 17:27 and 30:16 (MT 49:27) must be considered next, before any attempt to decide what meaning was intended by the translator. I note the following meanings of ’armôn from a few standard sources: HALOT: (fortified) palace BDB: citadel, then: citadel, castle, palace KB, Lexicon: Wohnturm … dwelling-tower (fortified building of small square base a[nd] several stories) DCH: fortress … fortress, fortification Zorell: domus elegantior et magis munita divitum, palatium [‘a more elegant and fortified house of the rich, a palace’]20

I do not myself see clearly what the entity is, beyond a certain vague impression. Each of the equivalents given by the Hebrew lexicons means something slightly different. We should note that the Hebrew word in Jeremiah is plural (like the Greek), so that there are multiples of them in one city – which hardly suits ‘citadel’ or ‘palace’ (if the royal palace is meant). If the translator was trying to represent one of these senses, ἄμϕοδον is nothing like what we would expect. More likely equivalents would be βασίλειον, βᾶρις, πυργόβαρις, or πύργος, the first three of which do occur elsewhere, along with other more far-fetched renderings.21 For Tov, 19

20

21

Cf. Du Bouchet, ‘Quartiers,’ 50: ‘Quelles que soient les incertitudes de détail, il faut donc distinguer pour ἄμϕοδον trois référents distincts, … la parcelle urbaine, le quartier et la rue.’ Zorell, Lexicon Hebraicum. Caird (‘Towards a Lexicon. I,’ 460–1, discussing ἄντρον) confidently states the meaning of ’armôn as ‘the inner keep or redoubt of a fortified enclosure or palace.’ He mentions no source or authority. βασίλειον Prov 18:19; βᾶρις Ps 47(48):4 + 4; πυργόβαρις Ps 121(122):7. Other renderings: ἄντρον, γῆ, ἐναντίον, θεμέλιον, ναός, οἶκος, πόλις, χώρα.

A LEXICOGRAPHICAL DATABASE FOR GREEK

205

the variety of renderings indicates that the Jeremiah translator and others did not know the meaning of ’armôn and guessed from context. Tov nevertheless regards ἄμϕοδον (taken to be a ‘block of houses surrounded by streets’) as close to the meaning of the Hebrew word (understood as ‘palace’), and intended to be so by the Jeremiah translator. I do not see this: there is no similarity between an ἄμϕοδον and a palace; and if the translator does not know the meaning of the Hebrew word, it cannot be used as an indicator of the intended meaning of ἄμϕοδον.22 All in all, it seems to me fruitless, in the present case, to attempt to draw out the meaning of ἄμϕοδον from the Hebrew. The best guide is what the word would normally mean in Greek, combined with the indications of the (Greek) context. On that basis, sense 1 or 3 above would suit well, and sense 1, the less technical, would be preferable. The picture is of fire consuming the dwelling complexes that make up the large towns of Jerusalem and Damascus. The translator chooses a term that would be an everyday one and immediately meaningful to his readers, many of whom would have lived in an ἄμϕοδον (unlike us, who have to work to imagine one).

V I come now to a conclusion. One could say that all scholarly work on Greek words up till now has been a preparation for the day when it could be systematically collected and organised. We publish articles and books on the assumption that everything will somehow be read and absorbed into the scholarly consciousness, or at least taken into the lexicons at the next updating. But we know it doesn’t happen. The process is sluggish, haphazard, and subject to all kinds of omission and duplication, simply because of the practical constraints. Who has time to read everything, let alone remember it? But the day has come when a systematic collection can become a reality. An electronic gathering-point is now possible; it is also, as I have tried to show, necessary. Good decisions about the meaning 22

Tov has discussed this word four times, each with slight variations, but his understanding of ἄμϕοδον, dependent on LSJ, does not advance (Jeremiah and Baruch, 33; ‘Three Dimensions,’ 530–1; ‘Did the Septuagint Translators?,’ 56–8; ‘The Septuagint,’ 164). Du Bouchet’s discussion of the LXX examples (‘Quartiers,’ 47–8) is hampered by the attempt to get the meaning to match the Hebrew, taken to be ‘palais.’ Case’s discussion (‘The Use of ἄμϕοδον’) achieves little. Like Tov, Caird (‘Towards a Lexicon. I,’ 461) noted the variety of renderings as an indication that the word ’armôn ‘gave the LXX translators a great deal of trouble.’ The variety of definitions in the Hebrew lexicons suggests its meaning is still uncertain today.

206

COLLECTED ESSAYS ON THE GREEK BIBLE

of words cannot be made without taking full account of the available discussions and evidence, and this is not happening with current methods. Even if one were simply to take the bibliography noted in this paper, enter it into a database and keep it up to date, it would be a big step forward.23 At least the next person to work on this word would not have to do it all again. If a list of the occurrences of ἄμϕοδον, merely that, were added as well, we would be well on the way to a major resource for lexical study. My paper title is cast in the form of a question. Let me end with another. If, as seems certain, the creation of a database for Greek lexicography would be of great value, is there any reason why it can’t be done?

Postscript There has been no answer to my final question, and (as of 2021) no news of the creation of a database along the lines suggested.

23

An existing print resource attempting this is the RBLG (1998). The problem is that it cannot be readily updated or augmented. In the case of ἄμϕοδον it records only three items. [RBLG Suppl. 3 (2008) (online) adds two more, Lolos and Du Bouchet.]

14 THE COMPLUTENSIAN POLYGLOT, THE TEXT OF SIRACH, AND A LOST GREEK WORD 2009

Abstract An otherwise unknown word, πάλαθος, recorded in the Complutensian lexicon of the NT, Wisdom, and Sirach (1514), is tracked down and explained in this paper. It turns out to occur in Sirach at 24:15 in one part of the MS tradition known as ‘Gk II,’ found in MS 248, which the Polyglot editors used for their text of Sirach. Though πάλαθος is probably a corruption of a similar word in Sir 24:15, its status as a ‘real word’ is worth considering. Comparison of the Old Latin leads to a proposal for an improved reading of another word accompanying πάλαθος in the text of Sir 24:15 in the Göttingen edition based on ‘Gk I.’ Some other instances of weakly attested words occurring in the Gk II variants are noted as useful additions to the lexicographical record.

This paper tells a tale of detection. It all started with an entry for an unknown Greek word in a forgotten lexicon. The quest for an explanation led to the differing recensions of the book of Sirach, to a textual problem in the Greek text, a suggestion for amendment in the light of the OL, the discovery of a nest of unrecorded words, and some lessons for Greek lexicography.1 The ‘forgotten lexicon’ is not really forgotten, just old and not well known. It is found in the great Complutensian Polyglot, printed at Alcalá (Latin Complutum) in Spain in 1514–1517. This six-volume work presents the biblical texts in their original languages, together with the ancient versions. In volume 5 (1514) the Greek NT is printed for the first time; the volume also contains a lexicon of the NT, another first. This lexicon, 1

This is an expanded version of a paper delivered at the SBL International Meeting in Vienna, in July 2007. I am grateful to the participants for their comments, especially Anneli Aejmelaeus for drawing my attention to O’Connell’s book (see n. 8), and to Michael Curran for reading the final version. The BIOSCS reviewers’ comments have also contributed to its improvement.

208

COLLECTED ESSAYS ON THE GREEK BIBLE

somewhat surprisingly, sets out to cover not only the NT but also the two wisdom books in the Greek Apocrypha, namely, Wisdom and Sirach (Ecclesiasticus). Each entry in the lexicon gives the Greek word, its forms, and a Latin equivalent, but not references. The number of entries totals over 9,000. My interest in this lexicon originated in the investigation undertaken for my history of NT lexicons.2 I am now engaged in preparing a new edition of the lexicon with a full study of its content.3

A

MYSTERY WORD

In the Complutensian lexicon the following entry appears: πάλαθος. ου. ὁ. Massa. et πάλαθος ἀρωμάτων. vulgo poma.

We have first the headword, πάλαθος, with an indication of the genitive (-ου) and gender (masc.), then the meaning, ‘lump/cake,’ followed by a phrase in which it occurs, namely, πάλαθος ἀρωμάτων, ‘cake of spices,’ and finally a comment, ‘commonly [applied to] fruits.’ The quotation of a context is unique in the lexicon, and additional comments are almost as rare. Apart from that, there is nothing unusual about the entry. But the problem is, the word πάλαθος does not exist. If we look in LSJ at the point where we would expect it, we find this: παλάθ-η... ἡ, cake of preserved fruit, Hdt. 4.23, Thphr. HP 4.2,10, LXX 1 Ki. 25.18, al., Amynt. ap. Ath. 11.500d, Luc. Pisc. 41, Vit. Auct. 19. -ιον, τό, Dim. of foreg., Polem. Hist. 88; cf. παλάσιον. -ίς, ίδος, ἡ, = foreg., Ph. Bel. 89.28, Str. 2.3.4. -ώδης, ες, like a παλάθη, Dsc. 1.67.

This is a nice little word-group, and clearly the one to which πάλαθος belongs – or would belong if it existed – but πάλαθος is not there and is apparently unknown. A search of other lexicons, old and new, failed to find any entry for it.4 So why is πάλαθος entered in the Complutensian lexicon? 2 3

4

Lee, History, esp. 45–51; for Polyglot title details see 329–30. [Collecting the data was in progress at the time of this paper but was diverted by work for the Grinfield Lectures on the LXX and the ensuing book, The Greek of the Pentateuch (2018). Some fruits of that research may still be hoped for.] Such as Hesychius (ed. Hansen); Etymologicum Magnum (ed. Gaisford, 1848); Suda (ed. Adler); Συναγωγὴ Λέξεων (ed. Cunningham); Kriaras, Λεξικό; Sophocles, Lexicon. A TLG search also did not produce any examples (though many of παλάθη). It was only at the final stage of preparing this paper that I checked Schleusner, Lexicon in LXX (1822) and found that he includes it, from the Complutensian text: ‘πάλαθος, pila vel massa,

THE COMPLUTENSIAN POLYGLOT, THE TEXT OF SIRACH

209

The NT can quickly be ruled out as the source. A glance in a concordance, if such were needed, establishes its absence. This leaves Wisdom and Sirach. But the usual tools fail to help: πάλαθος does not appear in those books or any others, as far as the concordances know.5 It was at this point that I began to learn more about the Greek text of Sirach, in particular the form of it printed in the Complutensian Polyglot (in vol. 3, 1516), and this led to the solution. The text of Sirach in the Polyglot is based on a MS that represents a markedly different version from that in the standard editions of Rahlfs and Ziegler, namely, MS 248, on which more will be said shortly. It is in that other version that πάλαθος occurs, at Sir 24:15, and accordingly in the Polyglot printed text, which reads as follows: Ὡς κιννάμωμον καὶ ὡς πάλαθος ἀρωμάτων, καὶ ὡς σμύρνα ἐκλεκτὴ ἔδωκα εὐωδίαν. Ὡς χαλβάνη καὶ ὄνυξ καὶ στακτή, καὶ ὡς λιβάνου ἀτμὸς ἐν σκηνῇ.

The text in Ziegler, on the other hand, based on a majority of MSS including the major uncials, is rather different:6 ὡς κιννάμωμον καὶ ἀσπάλαθος ἀρωμάτων καὶ ὡς σμύρνα ἐκλεκτὴ διέδωκα εὐωδίαν, ὡς χαλβάνη καὶ ὄνυξ καὶ στακτὴ καὶ ὡς λιβάνου ἀτμὶς ἐν σκηνῇ.

The older editions of Rahlfs (1935) and Swete (1891) likewise read ἀσπάλαθος ἀρωμάτων (+ δέδωκα ὀσμήν). Since most concordances and lexicons depend on this form of the text, the reading of the Polyglot text is not covered; hence the absence of πάλαθος.7 THE LEXICON ENTRY The lexicon editor included πάλαθος in the lexicon because it was in the Polyglot text, and he quoted the phrase in which he found it. He also

5

6 7

i.q. παλάθη. Sir. XXIV.15 sec. Compl. ὡς πάλαθος.’ He goes on to suggest that ἀσπάλαθος should be read. That is, HR and HR Suppl.; Auwers, Concordance. There is no mention of πάλαθος in: Ziegler, ‘Zum Wortschatz’; idem, ‘Ursprüngliche Lesarten’; Smend, Griechisch-SyrischHebräischer Index; Wagner, Septuaginta-Hapaxlegomena. For further discussion of the indexing problem in Sirach, see Ziegler, ‘Vokabel-Varianten,’ 188–90. Ziegler (ed.), Sapientia Iesu Filii Sirach (Göttingen ed.). Even Auwers’s Concordance to Gk II does not fully cover the Polyglot text of Sirach: it covers Gk II as printed in Ziegler’s text, with some (most?) of the variants of 248, but not all of them. I eventually found πάλαθος when I thought to look for ἄρωμα in Sirach.

210

COLLECTED ESSAYS ON THE GREEK BIBLE

gave it a meaning. How did he know what it meant? Before answering that, we must go back a step and ask where he found the text that he worked from. The lexicon was printed in the fifth volume dated January 1514; the text of Sirach is in the third volume, printed later, certainly after May 1515 and most likely at the end of 1516.8 It is improbable that the sheets of the third volume were already printed and accessible three or more years earlier. The next possibility is that the editor worked from the fair copy, that is, a final handwritten copy prepared by the text editor for the printer to work from. This too seems unlikely: a fair copy was probably made, but not as early as 1513.9 In fact the answer is that the lexicon editor compiled his lexicon direct from the MS of Sirach (and Wisdom). This MS is 248, which has long been known to be the basis of the text of Sirach and Wisdom printed in the Polyglot.10 A comparison of entries in the lexicon with 248 and the printed text proves that the lexicon editor worked from the MS: in certain instances the editor who prepared the text for printing in the Polyglot made changes to what was in 248, or a typesetting error occurred, but these deviations are unknown to the lexicon editor and he enters only the original reading of 248.11 The lexicon editor, then, compiled his word-list for Sirach from MS 248. Deciding the meanings of the words came next. For help with this he had very limited resources. In the Introductio to the lexicon some are mentioned, among them the lexicon of Cyril, Suidas (or the Suda), and the Etymologicum Magnum; in addition it can be shown that the editor made extensive use of another current work, the Greek–Latin lexicon of Crastonus.12 Apart from these he had Jerome’s Vulgate (= OL in Sirach), that was all. For a word resembling πάλαθος the information at his disposal was: 8 9

10

11

12

Cf. O’Connell, Ancient Sources, 6–7. See O’Connell, Ancient Sources, 144, on the likelihood of an editor’s fair copy of Sirach. The task of type-setting the Greek text and Latin interlinear matching word-for-word would, I think, make a fair copy essential. MS 248 (Holmes and Parsons) is a minuscule of the thirteenth century in the Vatican Library (Vat. gr. 346), lent to Cardinal Ximénes for the editing of the Polyglot. It was the primary source for all the LXX books in vol. 3. See Ziegler, Sirach, 42; O’Connell, Ancient Sources, 127–8. An edition of 248 in Sirach is available in Hart, Ecclesiasticus (but not without errors: Ziegler, Sirach, 53). One example will suffice from many: Sir 37:11 μεταβουλίας 248; μεταβολίας Compl., Ziegler; μεταβουλία Lexicon (μεταβολία not in). Deviations of the Complutensian text from 248 are quite numerous: see Ziegler, Sirach, 42, with a select list. O’Connell (Ancient Sources, 144–5) argues that all are explicable as editorial changes to 248. First published in 1478, followed by many editions; the one available in Alcalá was probably the Aldine, i.e., [Crastonus,] Dictionarium graecum (1497).

THE COMPLUTENSIAN POLYGLOT, THE TEXT OF SIRACH

211

Crastonus, Dictionarium (1497): παλάθη. ης. ἡ. massa. Suda (ed. Adler): παλάθαι· μᾶζαι σύκων. Etymologicum Magnum (ed. Gaisford): not in.13 Lexicon Cyrilli (MS): ?14 Vulgate (as in Polyglot, vol. 3): Sicut cinamomu[m] [et] balsamu[m] aromatizans odore[m] dedi.

The real help obviously came from Crastonus. Though πάλαθος was not showing up, the lexicon editor made an intelligent guess on the basis of παλάθη in Crastonus. He had no reason to doubt it was a real word, and took it as a variation on παλάθη. If παλάθη meant massa, πάλαθος would probably mean the same; and it made good sense in his Greek text: ‘Like cinnamon and like a cake of spices, and like choice myrrh I gave a pleasant odour.’ His text of the Vulgate offered balsamum (‘basalm’) as the meaning of πάλαθος, but he evidently didn’t trust it: he preferred to rely on Crastonus. He had no reason to think of the word ἀσπάλαθος, or to suspect that πάλαθος in his text might be a corruption. He added the remark that πάλαθος is commonly applied to fruits (vulgo poma) on the basis of the statement in the Suda, that παλάθαι (pl. of παλάθη) is used with reference to ‘cakes of figs.’15 He quoted the phrase in which he found πάλαθος because it showed a rather different, though similar use. So we arrive at the entry with which we began. The volume containing the text of Sirach was printed subsequently, and it presented not only the Greek text and the Latin Vulgate but an interlinear Latin rendering of the Greek, as in all the Old Testament volumes of the Polyglot. It is known that in Sirach this translation was the work of Juan de Vergara.16 It retains the Vulgate where possible, but changes the wording to match the Greek, which is often very different. It is a fair guess that this rendering was prepared later than the lexicon, and that Vergara was able to make use of the lexicon, already printed in 1514. From it he 13

14

15

16

The first editions of both the Suda and the Et. Mag. had appeared in 1499. Hesychius (1514) was not yet to hand. The copy of Lexicon Cyrilli now in Madrid, a MS of X/XI AD, and almost certainly the one used by the Complutensian editors, does not contain the relevant page, as far as I can ascertain from the scanned images online. If it did, the entry was probably very similar to that in the Suda (above): the same lemma, thought to derive from Lexicon Cyrilli, appears in the Συναγωγὴ Λέξεων and Hesychius. [A return visit to the MS online in the Biblioteca Histórica, Universidad Complutense de Madrid revealed the relevant entry in the lexicon of Cyril. It is indeed the same as that in the Suda, except for the accents: πάλαθαι· μάζαι συκῶν. (22.9.21)] This itself is based on LXX examples, cf. 4 Kgdms 20:7 παλάθην σύκων, Isa 38:21 παλάθην (ἐκ) σύκων. See Sáenz-Badillos, La filología bíblica, 327. He was also responsible for the translation of Wisdom and several other books.

212

COLLECTED ESSAYS ON THE GREEK BIBLE

took massa as the meaning of πάλαθος.17 The Polyglot text of Sir 24:15, with Vergara’s interlinear rendering, is as follows: Sicut ci[n]namomu[m] [et] sicut massa aromatum: Ὡς κιννάμωμον καὶ ὡς πάλαθος ἀρωμάτων, [et] q[ua]si myrrha electa dedi suaue[m] odore[m]. καὶ ὡς σμύρνα ἐκλεκτὴ ἔδωκα εὐωδίαν. Quasi galbanu[m] [et] onyx [et] stacte: Ὡς χαλβάνη καὶ ὄνυξ καὶ στακτή, [et] q[ua]si thuris vapor in tabernac[u]lo. καὶ ὡς λιβάνου ἀτμὸς ἐν σκηνῇ.

THE TEXT OF SIR 24:15 Having found πάλαθος, we could proceed at once to the question of what status it has as a word. But the text of Sir 24:15 invites attention, and will prove to be an interesting trail to follow. The text history of Sirach is one of the most difficult and complex in the Greek Bible.18 It is not the aim of this paper to make a contribution to this subject; I simply report the current consensus, as background to a closer look at the text. The original Hebrew version of Sirach (Hb I) was the basis of a Greek translation (Gk I); then came an expanded version of the Hebrew (Hb II), which was in turn the basis of a revised and expanded Greek version (Gk II). Gk I is transmitted in the major uncials and dependent minuscules; Gk II is represented by a number of other witnesses, including 248, but neither it nor any other MS preserves a pure text of Gk II.19 Ziegler’s text and app. crit. present the data on the MS readings in Sir 24:15, as follows: ὡς κιννάμωμον καὶ ἀσπάλαθος ἀρωμάτων B S A V O L l a b c min. La verss. ἀσπάλαθος] παλ. 248-672 46 336 534ʹ; απαλ. V 705; σπαλ. 543; fort. κάλαμος Sm.; pr. ως O-Sc-V 248-672 46 336 534ʹ 543 Aeth ArmII 17

18

19

It is of course theoretically possible that Vergara’s translation was made first and the lexicon editor used it, but practically very unlikely, both because of the time-frame and the fact that the lexicon editor worked direct from MS 248. Ziegler, ‘Ursprüngliche Lesarten,’ 461; Skehan and Di Lella, Ben Sira, 59. Cf. Jellicoe, Septuagint, 306–10. Ziegler, Sirach, 74; on witnesses to Gk II, 58–69.

THE COMPLUTENSIAN POLYGLOT, THE TEXT OF SIRACH

213

The reading ἀσπάλαθος ἀρωμάτων adopted by Ziegler is that of BSAV and various minuscules. The Polyglot text matches 248 and others that share ὡς πάλαθος ἀρωμάτων. Clearly Ziegler regarded ἀσπάλαθος as original and variants such as ὡς πάλαθος as secondary, and there is no reason to argue with him. At some point in the tradition ὡς was introduced before ἀσπάλαθος, leading to various corruptions including ὡς πάλαθος. But there is still a problem in Ziegler’s text: ἀρωμάτων does not make proper sense. The word ἄρωμα itself (aromatic herb or spice, LSJ) is not the problem: it is well attested in Greek from early on, and occurs in the LXX and NT. Likewise ἀσπάλαθος is a well known word, even if its meaning is somewhat hazy: it is the term for some sort of thorny aromatic plant, and, though found only here in the LXX, is attested from the fifth century BC to Modern Greek.20 But how does the gen. pl. ἀρωμάτων fit syntactically? The best one can think of is ‘among (the) spices,’ but this is awkward. At this point it may be of interest to see how the translations have dealt with it. The older ones reflect a text that is different from Ziegler’s and similar to Rahlfs’s, i.e., with the additional words δέδωκα ὀσμήν, ‘I gave a sweet smell,’ after ἀρωμάτων, but this does not affect their rendering of ἀρωμάτων, except in the case of the NEB (see further below): KJV (1611): I gave a sweet smell like cinnamon and aspalathus, and I yielded a pleasant odour like the best myrrh, as galbanum, and onyx, and sweet storax, and as the fume of frankincense in the tabernacle. [London: Bagster, n.d.]: I gave a sweet smell like cinnamon and aspalathus, … = KJV Smend (1906): Wie Zimmt und wohlriechender [Kalmus und Kassia], und wie Myrrhenfluss duftete ich süss. Wie Galbanum und Räucherklaue und Stakte, und wie Weihrauch war mein Duft in der Hütte. Giannakopoulos (1955–68): Ὡς ἡ εὐώδης κανέλλα καὶ ὁ ἀρωματικὸς σπάλαθος δίδω καὶ ἐγὼ τὴν εὐωδίαν μου. … NEB (1970): Like cassia or camel-thorn I was redolent of spices; I spread my fragrance like choice myrrh, like galban, aromatic shell, and gum resin; I was like the smoke of incense in the sacred tent. Kolitsaras (1981): Ὅπως ἡ εὔοσμος κανέλλα καὶ ὁ ἀρωματικὸς ἀσπάλαθος ἔδωσα καὶ δίδω ἐγὼ τὴν εὐωδίαν. … NJB (1985): Like cinnamon and acanthus, I have yielded a perfume, like choice myrrh, have breathed out a scent, like galbanum, onycha, labdanum, like the smoke of incense in the tent. 20

See LSJ, s.v.; Chantraine, DELG, s.v.; Kriaras, Λεξικό, s.v.; Lex. Acad., s.v.

214

COLLECTED ESSAYS ON THE GREEK BIBLE

Skehan-Di Lella (1987): Like cinnamon, or fragrant cane, or precious myrrh, I give forth perfume; Like galbanum and onycha and mastic, like the odor of incense in the holy Tent. NRSV (1989): Like cassia and camel’s thorn I gave forth perfume, and like choice myrrh I spread my fragrance, like galbanum, onycha, and stacte, and like the odor of incense in the tent. NETS (2007): Like cinnamon and camel’s thorn for spices,* and like choice myrrh I gave forth a fragrance, like galbanum and onycha and stacte and like the vapor of frankincense in a tent. [Fn. *Possibly of spices; + I gave off a fragrant smell = Ra.] LXX.D (2009): Wie Zimt und Gewürzstrauch und wie ausgewählte Myrrhe habe ich den Wohlgeruch verbreitet; wie Galbanum und Onyx und Myrrhen-Öl und wie Duft von Weihrauch im (Heiligen) Zelt.

Some of these, namely, KJV, NJB, NRSV, LXX.D, do not appear to render ἀρωμάτων at all: they simply pass over it.21 NETS makes the attempt, but ‘for spices’ is not a possible meaning of the genitive and does not make much sense. NEB, working with Swete’s text, joins ἀρωμάτων with ὀσμήν, which is possible with that text but not Ziegler’s. Skehan and Di Lella’s ‘fragrant cane’ is not a rendering of ἀσπάλαθος (+ ἀρωμάτων?) but, as far as I can make out, a rendering of κάλαμος εὐώδης, the phrase found in Exod 30:23. They take it for granted that Sir 24:15 is based on Exod 30:23, 34, in the passage describing the perfumes and incense used in the service of the Tent, and allow that passage to influence the interpretation here, where Wisdom likens herself to a similar list of perfumes and incense.22 Smend works from the same premise, but at least his alterations are overt.23 Giannakopoulos and Kolitsaras render ἀρωμάτων as equivalent to ἀρωματικός, ‘aromatic,’ which makes good sense and is what we would like it to say, but is not an accurate rendering of what we actually have in the text.24 21

22

23

24

I do not know if LXX.D’s Gewürzstrauch (‘spice-bush’) could include representation of ἀρωμάτων. Skehan and Di Lella, Ben Sira, 328, 334–5. Cf. Wright, No Small Difference, 248, with n. 26, expressing some caution. ἀσπάλαθος is not in the Exodus passage (or anywhere in the LXX outside Sir 24:15). Translation in Smend (ed.), Weisheit des Jesus Sirach, 41; discussion 219–20. Smend emends to match Exodus: ‘Vielleicht ist (ἀσ)πάλαθος Fehler für (ὡς) κάλαμος’ (219); hence ‘fort. κάλαμος Sm.’ in Ziegler’s app. crit. (above). Smend’s wohlriechender (‘fragrant’) implies ἀρωματίζων not ἀρωμάτων. He notes the OL reading without comment. Giannakopoulos, Ἡ Παλαιὰ Διαθήκη, vol. 26; Kolitsaras, Ἡ Παλαιὰ Διαθήκη, vol. 4.

THE COMPLUTENSIAN POLYGLOT, THE TEXT OF SIRACH

215

All this demonstrates the difficulty of ἀρωμάτων: none of the translations has been able to make sense of it.25 There are no MS variants to the word, so we get no help from that direction. But there is another avenue to follow.

THE OLD LATIN The oldest witness to Sirach, apart from the original Hebrew (not extant in Sir 24:15), is the OL version. This predates the earliest Greek MSS, the IV AD uncials, and is generally regarded as a witness of high value.26 What does it have as the rendering of ἀρωμάτων? It reads as follows in a modern critical edition:27 sicut cinnamomum et aspaltum aromatizans odorem dedi quasi myrra electa dedi suavitatem odoris et quasi storax et galbanus et ungula et gutta et quasi libanus non incisus vaporavi habitationem meam

The OL corresponding to ἀρωμάτων is aromatizans, a pres. part., agreeing with aspaltum.28 This clearly implies a Greek original ἀρωματίζων, masc. pres. part. agreeing with ἀσπάλαθος, with the straightforward meaning ‘aspalathus/camel’s thorn giving off an aroma.’29 It is hard to see how the Latin rendering could have arisen from any other form of the Greek, when Latin aromatizo is a rarity that appears to have been created for this place, on the model of ἀρωματίζω (which is a normal Greek word). Lewis and Short cite aromatizo only here; TLL adds two examples, one in a glossary and one in Oribasius (VI AD). The interpretation offered in Lewis and Short, aromatizans odorem dedi (‘giving off an aroma I gave a sweet smell’), 25

26

27

28

29

There is of course no such difficulty with the Complutensian text πάλαθος ἀρωμάτων, as understood by the lexicon editor and Vergara, i.e., ‘cake of spices.’ Interestingly, the Geneva Bible (1560) reflects the same text: ‘I smelled as the cinnamom, and as a bagge of spices.’ See Ziegler, Sirach, 14, 75; Skehan and Di Lella, Ben Sira, 56–7; Wright, No Small Difference, 5–6; Reiterer, ‘Review of Recent Research,’ 26. The displacement of chapters seen in the Greek MSS is not present in the OL, which therefore precedes all the extant Greek MSS. Vetus Latina: Sirach, at 24:20. The text of the Vulgate in Weber’s edition (Biblia Sacra) is identical, except in having murra for myrra (at Sir 24:20). For the form aspaltum (v.l. balsamum), see TLL, s.v. aspalathus. It is obviously a (neuter) variation on aspalathus, itself derived from ἀσπάλαθος. The gender of ἀσπάλαθος is commonly fem., but masc. is also found (LSJ); by Mod. Greek the masc. is standard, as already earlier: see Lex. Acad., s.v.; Kriaras, Λεξικό, s.v.

216

COLLECTED ESSAYS ON THE GREEK BIBLE

though a possible reading of the Latin, would not be possible in the Greek original, which would require ἀρωματίζουσα (Wisdom/Σοφία is subject). The attestation of ἀρωματίζω includes occurrences in Aquila (see HR, s.v.). The additional words δέδωκα ὀσμήν in most MSS (~ OL odorem dedi) are regarded by Ziegler as secondary (‘ex 15b’); the implications for the OL reading are not clear. In Herkenne, De Veteris Latinae Ecclesiastici capitibus I–XLIII, 191–2 the OL of Sir 24:15 is noted without remark on aromatizans or ἀρωμάτων. Thiele (Vetus Latina: Sirach, 24:20, app. crit. to aromatizans) had the solution almost within his grasp: ‘habuitne interpres latinus (αρωμα)των pro participio?; non neglegenda est lectio sequens (murra) [sic] electa (substantivum + adiectivum).’

My proposal, then, is that the original form of the text was ἀσπάλαθος ἀρωματίζων, which was corrupted early to ἀσπάλαθος ἀρωμάτων, by miscopying of ἀρωματίζων. The original reading of the Greek was the basis of the OL translation, but the corruption occurred soon after, early enough to enter our oldest Greek witnesses and to be transmitted in all subsequent extant MSS. For another instance where the OL preserves a better reading than all the Greek witnesses and has been used to restore the Greek original, there is Sir 35(32):9 ὅπου γέροντες μὴ πολλὰ ἀδολέσχει (Ziegler’s text). Here γέροντες is not found in any of the Greek MSS, which all read λέγοντος/ ες, but it matches the OL senes. The correction was first made by the Complutensian editor, probably by retranslation from the Latin, as Ziegler says, and is accepted by Ziegler, with support now from the Hebrew and the Syriac (and the Sahidic?).30 Before leaving the question of the text, it will be useful to consider what the original Hebrew of ἀσπάλαθος ἀρωματίζων might have been. It seems likely to have been the same phrase as in Exod 30:23, ‫( קנה־בשׂם‬lit. ‘reed/cane of perfume’: LXX κάλαμος εὐώδης, ‘sweet-smelling cane’), but translated differently. A match of ἀρωματίζων with ‫ בשׂם‬is very probable: ἄρωμα often renders ‫בשׂם‬ elsewhere, in fact more frequently than εὐώδης. The rendering of ‫ קנה‬by ἀσπάλαθος, however, would be a one-off: the common renderings are κάλαμος and καλαμίσκος. Even if this means that κάλαμος was the original rendering (in Gk I?), later changed to ἀσπάλαθος (in Gk II?), the change is too far back to justify restoring κάλαμος to the surviving Greek text. 30

Ziegler, Sirach, 42–3. Ziegler records the support inconsistently: app. crit. at 35:9 (p. 275) has ‘= La (ubi sunt senes) Sa: cf. H,’ but at p. 43 he says ‘… diese Lesart, die auch dem hebr. und syr. Text entspricht. …’ An anonymous reviewer of this paper has confirmed that the Syriac does have ‘elders.’ Further comment on this example is found in O’Connell, Ancient Sources, 144. At Sir 3:17 there is what seems to me a missed opportunity for improvement of the text on the same basis: the reading of the MSS is ὑπὸ ἀνθρώπου δεκτοῦ, for which Ziegler adopts Smend’s ὑπὲρ ἄνθρωπον δότην, ‘cf. super hominum gloriam La.’ But the OL implies ὑπὲρ ἀνθρώπων δόξαν, which gives good sense: ‘you will be loved beyond the glory of men.’ Smend’s δότην agrees with the Hebrew ‫ )מ(נותן‬not the OL: see Beentjes, Book of Ben Sira, 23.

THE COMPLUTENSIAN POLYGLOT, THE TEXT OF SIRACH

THE STATUS OF

217

ΠΑΛΑΘΟΣ

Πάλαθος, ‘lump/cake’ is found in six MSS of the eleventh to fourteenth centuries AD, then in the Complutensian lexicon dependent on MS 248, in Vergara’s translation dependent on the lexicon, in the Complutensian text of Sirach based on MS 248, in Schleusner’s lexicon dependent on the Complutensian text, and nowhere else in any lexicon or text. Moreover it appears to be a corruption of another word. Nevertheless it makes sense as a word in context, as read by the lexicon editor and the text editor. As far as formation goes, it could be a word: πάλαθος belongs to a type of formation in -θος (-θον), -θη.31 Mostly only one or the other suffix appears, as, e.g., in the case of κάλαθος and σπάθη, but if one exists there is potential for the other to be formed, as in: κολοκύνθη and κολόκυνθος; λαπάθη, λάπαθος and λάπαθον; ὄχθη and ὄχθος; Πλαθάνη and πλάθανον. The potential for a masc. form alongside παλάθη is therefore clear, and πάλαθος, though not attested, could have existed. Is πάλαθος then a real word, deserving of a place in our lexicons? This of course raises the question of what a ‘real word’ is. Obviously, not every misspelling and corruption in the surviving Greek MSS can be regarded as a real word. Something else is needed. We need to establish in some way that the word existed at some time in the language of Greek speakers. We might appeal to the knowledge of the copyists, who copied πάλαθος as if it was a word they knew and understood. But that is not very reliable; they might equally well not have understood it and thought nothing as they copied it. The case is different with another variant in Sir 24:15, namely, σπάλαθος, the reading of MS 543, dated 1186 AD (see Ziegler’s app. crit. above). The form σπάλαθος is one of the large number of variant forms of ἀσπάλαθος recorded in medieval and dialectal Greek and is still alive today, as evidenced by Giannakopoulos’s translation (above).32 Thus the variant spelling has the support of evidence of the living language outside the MS, and is a ‘real word.’ We do not have such evidence for πάλαθος – yet. For these reasons I think πάλαθος cannot count as a real word; but at the same time I think it ought to be recorded in some way in the lexicons – with a suitable indication of uncertainty – because if one day another 31

32

See Chantraine, Formation des noms, 366–8; Buck-Petersen, Reverse Index, 444–9; Schwyzer, Grammatik, I, 510–1. See esp. Lex. Acad., s.v.; cf. Kriaras, Λεξικό, s.v. There is no sign of πάλαθος among the variant forms of ἀσπάλαθος.

218

COLLECTED ESSAYS ON THE GREEK BIBLE

example is found, the link can be made and its status upgraded.33 If one were inclined to think that all the evidence of Greek is in, here is a fact to ponder: the Oxyrhynchus papyri that have been deciphered and published to date are only 1% of the total held in Oxford.34 From this source alone, new data on the Greek language will certainly be brought to light.

OTHER

OVERLOOKED DATA

In πάλαθος we have discovered a gap in the lexicographical record, but it is not the only one of its kind. As pointed out above, the lexicons do not fully cover the text of Sirach preserved in the MS tradition. Data from Gk II are not noted in LSJ (and others) because they depend on editions based on Gk I, covering only part of the tradition. The problem affects not only the additional verses of Gk II (printed by Ziegler in smaller type), but Gk II variants within verses common to Gk I and Gk II. The latter type may make their appearance only in the app. crit. of Ziegler. Sometimes we may discover an unknown word like πάλαθος, but this is rare; more often we gain a useful attestation of a word that is weakly attested so far. Here are some examples. (1) ἔκπρακτος in Sir 10:8: φιλαργύρου μὲν γὰρ οὐδὲν ἀνομώτερον· οὗτος γὰρ καὶ τὴν ἑαυτοῦ ψυχὴν ἔκπρακτον ποιεῖ.

This is a verse found in the representatives of Gk II and printed by Ziegler in smaller type. ἔκπρακτος is unknown to LSJ and Suppl. and other lexicons.35 But it is a plausible member of the group ἐκπράκτης, ἔκπραξις, ἐκπράσσω, ‘exact payment,’ etc., i.e., an adjective meaning ‘payment-exacting.’ So we might translate ‘there is nothing more lawless than a person who loves money, for he makes his own soul a debtcollector.’36 33

34

35 36

The place for this is of course not in a printed book but an electronic database. See Lee, ‘Lexicographical Database,’ 214–20. Michael Theophilos, at SSEC conference, Macquarie University, Sydney, 5 May, 2007. The POxy series has reached vol. 72 (2008). LEH, Lexicon; Lampe, PGL; Mega Lexikon; Sophocles, Lexicon, s.vv. The NETS rendering ‘… makes his own soul a commodity’ is evidently influenced by the OL (10:10) … animam suam venalem habet (venalis = ‘for sale, open to bribes’),

THE COMPLUTENSIAN POLYGLOT, THE TEXT OF SIRACH

219

(2) λαλητός in Sir 18:33: ἔσῃ γὰρ ἐπίβουλος τῆς ἰδίας ζωῆς. + λαλητός 248-672

The additional word found in two MSS is recorded by LSJ only in Job 38:14 (‘endowed with speech’) and the Etymologicum Magnum (‘talked of’); two later examples in the Fathers are noted in Lampe (‘endowed with speech’; ‘argumentative’). An additional occurrence is not without value. Whether or not the text is better with λαλητός added at the end, it yields the meaning ‘for you will be talked of as a schemer against your own life.’37 (3) πολύλαλος in Sir 21:25: χείλη ἀλλοτρίων ἐν τούτοις διηγήσονται, ἀλλοτρίων] πολυλάλων 248 Anton. p. 993 Mal.

This example of πολύλαλος can be added to five out-of-the-way occurrences in LSJ, one of them in Symmachus Job 11:2. The variant makes good sense in the text of 248, which as a whole reads: χείλη πολυλάλων τὰ οὐκ αὐτῶν διηγήσεται, ‘the lips of the talkative will narrate things not their own.’38 (4) περιψήχω in Sir 30:7: περιψύχων υἱὸν καταδεσμεύσει τραύματα αὐτοῦ, καὶ ἐπὶ πάσῃ βοῇ ταραχθήσεται σπλάγχνα αὐτοῦ. περιψύχων] -ψήχων 248

LSJ’s entry for περιψήχω reads (in full): ‘περιψήχω, sine interpr., Gloss.’ Further data of any kind would obviously be welcome. LSJ’s ‘Gloss.’ refers to a seventh century glossary in Cod. Harl. 5792.39 Sir 30:7 could be the

37

38

39

which was probably arrived at by (wrongly) connecting ἔκπρακτον with πρᾶσις, ‘sale,’ πιπράσκω, ‘sell,’ and related forms. Et. Mag. 588.54 (ed. Gaisford) notes the word without meaning; ‘talked of’ is LSJ’s. The KJV rendering, ‘For thou shalt lie in wait for thine own life, and be talked on,’ shows that they, or rather a predecessor, worked from a text with the additional word. NRSV and NETS: ‘For you will be plotting against your own life.’ KJV again reflects this text: ‘The lips of talkers will be telling such things as pertain not unto them.’ Similarly NRSV: ‘The lips of babblers speak of what is not their concern.’ It appears that the Complutensian text was the basis of the KJV or an earlier English version; cf. n. 25 on the Geneva Bible. Goetz and Gundermann, Glossae, 406.

220

COLLECTED ESSAYS ON THE GREEK BIBLE

source of the lemma in that glossary. The reading περιψήχων is not to be dismissed as a mere misspelling. Though this is the only occurrence we know of, it is not an improbable compound and its meaning is readily deducible as ‘wipe clean’: cf. περιψῶ (άω), ‘wipe all around, wipe clean,’ and the simplex ψήχω, ‘rub down,’ etc.40 The resulting meaning is at least satisfactory: ‘Wiping clean his son he will tie up his wounds, and at every cry his insides will be agitated.’ The majority reading περιψύχων, however, remains preferable, even though the meaning ‘cherish’ etc. rests on a slender foundation.41 Another variant, περὶ ψυχῶν, found in several MSS and reflected in the OL, is difficult to make sense of. But whatever the merits of the reading, the variant text of MS 248 yields an occurrence of περιψήχω worth noting in the lexicons. (5) συγγηρῶ (άω) in Sir 11:16: τοῖς δὲ γαυριῶσιν ἐπὶ κακίᾳ συγγηρᾷ κακία.

Another verse included by Ziegler from Gk II. The point of interest is the attestation, for only the second time, of a present stem συγγηρῶ, as opposed to συγγηράσκω: the latter is attested in Classical texts since Herodotus, the former known from one occurrence in Aretaeus (II AD).42 Clearly there are data here that would be valuable in the lexicographical record but have been missed. These five words are only a sample: the text of Sirach, with MS traditions that vary markedly, is certain to have more to offer of the same kind; but it is likely that variant texts in other parts of the LXX will have similar useful material. The lesson for Greek lexicography is that standard critical texts are not the only potential source of vocabulary items. The variant readings of those texts, the ones that end up on the editor’s cutting-room floor, are equally worthy of attention. 40

41

42

Mega Lexikon, rightly: περιψήχω γλωσσ. ἄνευ ἑρμ. δ[ιάφορος] τ[ύπος] τοῦ περιψῶ. The Complutensian lexicon editor made a good fist of it: circunfrico. consumo. rado. emundo. abstergo (‘I rub around; I consume; I scrape; I clean, I wipe off’). Vergara’s interlinear follows the lead of the lexicon with abstergens. LSJ, s.v. II: ‘metaph. refresh, revive, cherish,’ citing Sir 30:7 and ‘D.H. 7.46 (cj. Reiske), Alciphr. 1.39.’ The conjecture carries no weight; the two examples in Alciphron (Rhet. et Soph. II/III AD), Epistulae 4.14.3, 8 (ed. Schepers, 1905) are much later and describe a courtesan’s behaviour towards her lover. Cf. NETS: ‘When one cherishes a son, one will bind up his wounds.’ Other versions add their own spin: KJV: ‘He that maketh too much of his son shall bind up his wounds.’ NRSV: ‘Whoever spoils his son will bind up his wounds.’ Skehan and Di Lella, Ben Sira: ‘Whoever spoils his son will have wounds to bandage.’ All these renderings surely require ὁ before περιψύχων, as in vss. 1, 2, 3. See LSJ. The NETS rendering ‘and evil things grow old along with those who take pride in evil’ appears to render the v.l. κακά (to κακία).

THE COMPLUTENSIAN POLYGLOT, THE TEXT OF SIRACH

221

CONCLUSION A previously unknown word, πάλαθος, recorded in the Complutensian lexicon of the NT, Wisdom, and Sirach turns out to exist in the MS tradition of Sirach, in one of its text-types, Gk II, of which a leading representative, MS 248, was the basis of the text of Sirach in the Complutensian edition. The lexicon editor worked direct from MS 248, where he found the word at Sir 24:15, and entered it in his lexicon. He assigned a plausible meaning with the aid of the tools of the time and some guesswork. The word arose from a corruption and is not yet secure as a real word; nevertheless it should be placed in the lexicographical record. The same verse contains an undetected corruption of a different word, ἀρωμάτων, which should be restored to ἀρωματίζων on the basis of the OL. There are other instances of unrecorded words that occur in the MS tradition of Sirach, of which five examples have been given: ἔκπρακτος (Sir 10:8), λαλητός (Sir 18:33), πολύλαλος (Sir 21:25), περιψήχω (Sir 30:7), συγγηρῶ (Sir 11:16). Greek lexicography could record these and be more aware generally of variant texts as a potential source of new attestations of words.

Postscript The paper is admittedly something of a showpiece. The results are less significant than the demonstration of what comes to light when a simple question is asked about a word, and what a tour of manuscripts, texts, and obscure vocabulary is required to find the answer. For the history of Greek lexicography, the paper illustrates the challenges the Renaissance pioneers faced and the limited resources they had at their disposal. The most useful result is not so much the explanation of the unknown word πάλαθος (‘lump,’ ‘cake’), as the discovery of a case for emendation of the problematic ἀσπάλαθος ἀρωμάτων in Sir 24:15 to ἀσπάλαθος ἀρωματίζων on the basis of the OL.

15 REVIEW OF MURAOKA, A GREEK-ENGLISH LEXICON OF THE SEPTUAGINT (2009) 2010

Abstract This review assesses the final instalment of Muraoka’s lexicon, completing the coverage of the whole LXX. It builds on the review (2007) of the preceding lexicon that covered not all, but a substantial part of the LXX. The features criticised in the previous review are briefly revisited and changes or reactions noticed. The review then turns to the question of Muraoka’s approach to determining meaning in the LXX and how it compares with that of NETS, the new translation of the LXX into English. NETS was preceded by lengthy discussion of method, and this evolved over time. The aim is to reach a clear understanding of how the two approaches differ and the issues involved, in particular the widely misunderstood ‘interlinear model’ introduced by NETS. The discussion is illustrated by examples. The review concludes with suggestions for future development of LXX lexicography on the basis of Muraoka’s lexicon.

I In a mere twenty-five years, a short span in lexicographical time, T. Muraoka has singlehandedly brought his lexicon of the Septuagint to completion. After two earlier instalments, the full work now appears, covering both the translated books of the Hebrew Bible and the original Greek works. It is not dependent on any other Septuagint lexicon but is based on a fresh lexical analysis of the material; this too is unusual in the world of lexicon production. Thus the desideratum of Septuagint scholars for over a century has been fulfilled. Muraoka is to be congratulated and his achievement celebrated. In an earlier review in this journal I examined the previous instalment of the lexicon in some depth, describing its general characteristics as well as offering a critique of Muraoka’s method of stating meaning.1 That 1

Muraoka, Lexicon (Pentateuch and Twelve Prophets), reviewed in BIOSCS 37 (2004), 127–39 [= Essay 10].

224

COLLECTED ESSAYS ON THE GREEK BIBLE

material does not need to be repeated here. Readers are encouraged to read the present review in the light of the earlier one. Here I consider (a) what may have changed in the present lexicon, especially in response to my earlier comments on definition; (b) the general issue of how lexical meaning is determined in the LXX, comparing Muraoka’s approach with that of the NETS project; and (c) future work in LXX lexicography. The general contours of the lexicon remain unchanged from the previous instalments. That is, it presents a list of headwords for the LXX vocabulary in alphabetical order, with information on morphology, a breakdown into senses with full or selective listing of occurrences, information on collocation, and finally a list of semantically related words, and references to selected literature. One change from its predecessors is the omission of the summary of corresponding Hebrew words, which had appeared at the end of each entry. This is an appropriate step, and Muraoka reports that he intends to publish the data in a separate work.2 The Introduction, which is short and to the point, will repay reading by all who use the lexicon. It not only explains the layout of the entries but sets out – rather too briefly perhaps – Muraoka’s approach to LXX lexicography and his method of indicating meaning. The choice of the ‘definition’ method as the basis of the latter is the most important feature of this lexicon. Whatever faults may be found with the execution, the application of this method offers the best prospect of stating meaning unambiguously and separating different lexical meanings, as opposed to the ‘gloss’ method with its great propensity for imprecision. In the earlier review I looked in some detail at Muraoka’s practice in framing definitions, with lists of examples. I pointed out that although the definition method was generally applied, there were many meanings that had the appearance of glosses, either in combination with a definition, or standing alone, or in groups of two or three. I called Muraoka’s approach the ‘mixed method’ and noted that this method is also encountered in some leading ‘definition’ lexicons (OED, OLD). It seemed to me that it would be helpful to clarify the difference between a gloss and a one-word definition, and to recognise the latter as an acceptable form of definition. Further, I questioned Muraoka’s highly developed practice of adding explanatory additions (in plain text) to the definition or gloss (in italic), and occasional inconsistency in the formating of collocations. In the present edition, Muraoka has taken the point about one-word definitions and in the Introduction has spelt out his position (p. xii); on this 2

See ‘Introduction,’ xv. The data are retained in a few instances ‘for which HR provide no information.’

REVIEW OF MURAOKA, A GREEK-ENGLISH LEXICON OF THE SEPTUAGINT

225

we are now in agreement, at least in principle. As regards the examples that I thought required adjustment, changes have been made, but only to about half. Moreover, a great many other instances of the same features remain throughout the lexicon. Further, there is no new information on the purpose or role of the explanatory additions, nor improved consistency in their use. It is still not quite clear if these are part of the definition proper, or an explanation of the definition, or an indicator of collocation (or a combination of these). Collocations also remain inconsistently formated. It will be useful to note a few examples to illustrate these points. One or two are from my earlier selection, but the rest are new; all are taken from the entries as they now stand in the 2009 edition. Two or three glosses, not definitions: καλός 1. advantageous, beneficial, desirable: νεανίσκος young man, lad: νεῖκος quarrel, contention: πόνος 1. toil, suffering, hardship: πρᾶγμα 1. deed, action: ῥαντός spotted, speckled: τάχος swiftness, speed: φοβερός awe-inspiring, formidable, frightful: φόβος 1. fear, dread: φόβος 2. religious fear, awe, reverence:

Explanatory additions: ἀκούω 1. of sense perception: to hear, + acc. δεῖπνον meal, dinner, usually sumptuous or formal: ἐπακούω 5. to react to oral message: ἐπανέρχομαι to return to the point of origin: ἐπανήκω to move back to the point of origin: καταλήγω to leave off speaking: κῦμα wave of the sea: λουτήρ washing-tub for ritual use: παιδεία 1. education, instruction (in religion and morality): ῥάβδος 2. shepherd’s staff, crook: σθένος 1. bodily strength: τέρας portentous, extraordinary event with some symbolic meaning performed by God, or by man (though ultimately by God):

Collocation mostly without brackets, sometimes with: γυμνόω 3. to remove one’s upper garment: εἰσδύω to move oneself into a space: κύριος 1. one who owns and controls (possessions): τίθημι I. 1. to place, lay (+ acc.): τίθημι I. 5. to direct sbd to do sth: τίθημι II. 2. to institute (law etc.):

226

COLLECTED ESSAYS ON THE GREEK BIBLE

It can be seen that Muraoka is inclined to variation. These instances are of course only a selection: many more could be added. At the same time, there are just as many other entries that maintain a consistent approach. The reason for some of the variations can be surmised. For instance, it is certainly difficult to compose a definition of καλός sense 1, though presumably possible (cf. sense 2. morally good and acceptable). Hence three glosses, each helping to focus the intended meaning: the time-honoured gloss method comes to the rescue. But I confess to finding the system behind the explanatory additions (if there is one) elusive. I admit that they are not necessarily unhelpful, if one suspends one’s wish to know exactly what they are doing; some may even like this idiosyncratic feature of the lexicon.3 A general tidying up of these matters could be hoped for at a later stage in the life of the lexicon. In the meantime, users are not likely to be seriously misled by the inconsistencies in definition method discussed here. There are of course other aspects that invite comment; some are taken up in the fourth part of this review, which considers future steps in LXX lexicography.

II The production of Muraoka’s lexicon coincided with another major initiative in LXX studies, a new translation into English (NETS), which was set in train under the leadership of Al Pietersma in the mid-1990s. Translating involves deciding what the text means, and very soon the editors of NETS found themselves facing questions very like those faced by a lexicographer. The difficulty of the questions is greatly magnified in dealing with a work that is itself a translation. To their credit, Pietersma and his colleagues began addressing theoretical issues early on, and the work has been accompanied by a wealth of discussion and explanation. It has also been accompanied by controversy and at times incomprehension.4 My aim is not to wade into this difficult and sometimes bewildering debate, but 3

4

Danker in his review of Muraoka, Lexicon (2009) draws attention to the ‘different formulae and formulaic combinations used in entries,’ concluding in characteristically mild fashion that ‘some of them invite the reader to be on special alert.’ See, e.g., the polemics between Muraoka and Pietersma in Muraoka, ‘Recent Discussions,’ and Pietersma, ‘A Response’; and the corrective to Dines, Septuagint in Wright, ‘The Septuagint,’ esp. 107–10. It is evident from Muraoka’s ‘Introduction’ (p. ix) that he misunderstands the approach of NETS (and LEH); likewise Rajak in Translation and Survival, 143–5.

REVIEW OF MURAOKA, A GREEK-ENGLISH LEXICON OF THE SEPTUAGINT

227

to try to reach a clear understanding – for myself as much as the interested reader – of where the difference lies between the approaches of Muraoka and NETS. My own view necessarily emerges. The NETS approach. It was easy to get the impression in the initial stages of the project that the basic principle of NETS was to treat the Hebrew original as determining the meaning of the Greek translation. The NRSV, i.e., a translation of the Hebrew text, was to be the base text, with revisions as required to match the Greek. Further, the ‘interlinear’ model was introduced as a means of describing the translators’ approach to their original. This seemed to imply that the translators’ method was a mechanical one in which each Greek word would have the meaning of its Hebrew counterpart. Statements of the kind still found in the introduction to NETS added to the impression, e.g., ‘the Greek had a dependent and subservient linguistic relationship to its Semitic parent.’5 Extended discussion of translation theory, though valuable in itself, tended not to clarify the issue for the non-expert.6 In all this I speak of impression. The outcome. I think it fair to say that the NETS approach has evolved over time, gaining focus and consistency. The relationship to the NRSV can be seen to have moved from greater dependence to less, in the face of dealing with Greek that departs in its own peculiar ways from the Hebrew, even in the ‘literal’ translations.7 ‘Interlinearity’ is clearly stated to be a metaphor, a way of conceptualising the translators’ approach, not a description of an actual written form.8 At any rate, the true nature of the NETS approach is now the opposite of what is commonly supposed: NETS attempts to translate the Greek according to the meaning it has as Greek, not to transfer the Hebrew meaning to the Greek and thence to the English translation. One might think a lexicographer’s aim to be the same, but this is not necessarily so. To understand how Muraoka’s approach differs, we need to look deeper. The focal point. The focus of NETS is on the meaning at the point of translation, that is, ‘what the original translator thought his text to mean.’9 Muraoka on the other hand looks to the subsequent meaning, that is, ‘what sense a reader in a period roughly 250 B.C. – 100 A.D. who was ignorant 5 6

7

8 9

‘To the Reader of NETS,’ xiv. Similarly Pietersma, The Psalms, ix. See, e.g., papers from the Panel Discussion ‘LXX and Descriptive Translation Studies’ in BIOSCS 39 (2006). See, e.g., Silva, ‘Esaias,’ 824 on the shift in dependence on the NRSV in Isaiah. In van der Meer, Review of NETS (2007), ‘interference’ from the NRSV is assessed as slight (118, 119). ‘To the Reader,’ xiv. ‘To the Reader,’ xv.

228

COLLECTED ESSAYS ON THE GREEK BIBLE

of Hebrew or Aramaic might have made of the translation.’10 This is a significant difference, and it can lead to fundamentally different results between the translation in NETS and the lexical meaning in the lexicon. (This is not always the case, of course; in the majority of instances NETS and Muraoka would agree.) A simple example: in Gen 13:14 θάλασσα, in a context where it is one of the four points of the compass, is translated ‘sea’ in NETS, but given the meaning *2. west in Muraoka. Thus NETS treats θάλασσα as having its normal Greek meaning ‘sea,’ while Muraoka takes it as having the sense of the word it translates, Hebrew ‫‘( ים‬west’ as well as ‘sea’).11 The difference of focal point is the cause, but how exactly did they arrive at these different meanings? Context. As θάλασσα shows, the role given to context is crucial. In a non-translated Greek text, context is the standard tool of the lexicographer for determining the meaning: the meaning required by the context is what the text means. Muraoka uses it in that way. This is natural if one’s aim is to discover what the LXX would mean to a later reader. The context in Gen 13:14, as it stands in the Greek, requires θάλασσα to be taken as meaning ‘west,’ with confirmation from the meaning of the original. For NETS, on the other hand, the Greek context does not have this determinative role.12 When context requires a Greek word to have a meaning outside its normal range, NETS does not accept it, but gives the word its normal Greek meaning. This is because of its view of how the translator operates within the ‘interlinear’ paradigm: he may simply write down a standard equivalent (θάλασσα = ‫)ים‬, disregarding how it fits in the Greek context as a whole. So NETS translates ‘sea’ because the translator himself disregards context in his choice of rendering. The Hebrew of the source text is obviously relevant to the debate here and we must consider it next. The Hebrew original. Muraoka does take into account the original, even though his target is the meaning as understood by a later reader who knows no Hebrew. How it is done is not clearly spelt out.13 Examination of entries suggests that Muraoka’s practice is to allow the meaning of the Hebrew to be transferred to the Greek where the Greek context seems to require it. The meaning is then usually marked with an asterisk, indicating nonstandard Greek. Our example θάλασσα is one such case. The application 10

11

12 13

‘Introduction,’ viii. The approach of La Bible d’Alexandrie is in essence the same as Muraoka’s: see Harl, La Genèse, 10. Discussions in, e.g., Boyd-Taylor, ‘Lexicography and Interlanguage,’ 58–60; Muraoka, ‘Recent Discussions,’ 235; and Pietersma, ‘A Response,’ 334–6. See esp. Pietersma, ‘Context is King.’ ‘Introduction,’ viii is rather vague.

REVIEW OF MURAOKA, A GREEK-ENGLISH LEXICON OF THE SEPTUAGINT

229

is largely ad hoc, each instance being decided as it arises; this is not unexpected, given the lexicographer’s experience that every case is different. The NETS project also attaches importance to the original Hebrew: ‘what this Septuagint says, and how it says it, can only be understood in its entirety with the help of the Hebrew.’ But apart from ‘arbitrating between competing meanings of the Greek,’14 the Hebrew meaning is not permitted to override the normal Greek meaning. If the Hebrew and the normal Greek meaning match well enough, we notice nothing; but if the Hebrew meaning is outside the normal range of the Greek word, NETS keeps the Greek meaning and does not allow a transfer from the Hebrew. So the NETS introduction can truthfully say, ‘perhaps paradoxically, the interlinear paradigm safeguards the Greekness of the Septuagint.’15 Intention. Another issue lurks in the discussion so far, namely translator’s intention. Though nothing is stated in the introduction, NETS is wary of resort to the ‘intention’ of the translator: we simply cannot read the mind of an ancient translator.16 This is true scientifically, but in practice unworkable. A translator has some intention when he translates, however difficult it may be to determine it. The postulated ‘interlinear’ model of the translator’s method itself involves an attempt to capture to some degree the intention of the translator. The editors of NETS say as much in using the words ‘what the original translator thought this text to mean.’17 Muraoka alludes to the subject, but it does not loom large because his focus is on the subsequent meaning.18 In my view, if we seek to understand the text ‘as produced’ (NETS’s words), we have no choice but to attempt to deduce the translator’s intention, and I believe it can be done, using certain clues. These are: (a) context (of the Greek as created by the translator); (b) meaning of the Hebrew original (as understood by the translator); (c) Greek usage (in the translator’s time). From these we deduce, however imperfectly, the meaning intended by the translator, and this in turn is the meaning, at the point of production. In this I adhere to the insight of Emanuel Tov, enunciated in 1976.19 14 15 16

17

18

19

‘To the Reader,’ xv. ‘To the Reader,’ xv. Compare Boyd-Taylor, ‘Evidentiary Value,’ 51–55; idem, Reading Between the Lines, 4–5. ‘To the Reader,’ xv. Boyd-Taylor’s older paper, ‘Place in the Sun,’ includes subtle (and convincing) attempts to reconstruct the translator’s thought processes. Muraoka, ‘Introduction,’ viii: ‘The translator’s intention is something rather elusive and not easy to comprehend with confidence.’ See Tov, ‘Three Dimensions,’ 529–30, 532. Compare Lee, ‘Equivocal Renderings,’ 104 [= Essay 3].

230

COLLECTED ESSAYS ON THE GREEK BIBLE

I am, then, in basic agreement with the NETS approach, but with translator’s intention as the path to establishing the meaning of the text as produced.20 Bilingual interference. This is a large topic, extensively discussed in the context of language generally. In a translated text, all sorts of potential for interference arises; there are also many problems of definition and understanding. It is not possible to deal with the topic adequately here. For the present purpose I focus on one issue, raised by our example of θάλασσα. Is it possible that the translator himself intended θάλασσα to mean ‘west’ in the context of Gen 13:14 (and elsewhere)? That would be a ‘loan-shift’ or ‘semantic loan.’21 It is very difficult, if not impossible, to know. We have no means of showing that the shift had occurred in the mind of the translator. The alternative, that θάλασσα was simply the default rendering of the Hebrew word, not intended to mean something different from its usual Greek sense, remains the safer assumption. That is the position of NETS. Are there, then, any cases of interference of this type? That is, are there any genuine loan-shifts? I used to think there were, but now I am not so sure.22 Pietersma has argued persuasively that their existence remains to be demonstrated.23 The interesting consequence is that renderings like θάλασσα, ‘sea,’ instead of a normal Greek word for ‘west,’ reflect a choice to translate in that way and thus to retain some of the un-Greek character of the original in the translation.24 To what extent the choice was free and conscious is a topic for further thought. Conclusion. The approaches of NETS and Muraoka to determining the meaning of a word in the LXX are fundamentally different, and the main reason is the difference of focus, between the meaning ‘as produced’ on 20

21

22

23

24

Aejmelaeus’s thoughtful exploration of the topic in ‘Translation Technique’ (1989) led her to a position like Muraoka’s and to rule out translator’s intention as a means of interpreting the text. For her, intention could even be absent, as in the case of standard renderings. I see intention as involved in these choices like all others; it is simply very variable or flexible. See Adams, Bilingualism, 461–3; Silva, Biblical Words, 87. Older Septuagintalists spoke simply of ‘Hebraisms,’ without much analysis: see, e.g., Thackeray, Grammar, 39– 55. See Lee, ‘Note on Septuagint Material,’ 237–8; ‘Equivocal Renderings,’ 116–7. [I have changed my mind about κῶλον and now agree it is not a ‘loan-shift’: see Essay 1, n. 8. But in Greek of the Pentateuch, 195, 198 I identify what I think are definite cases of ‘loan-shifts’ (certain exx. of ἁμαρτία, πλημμέλεια).] Pietersma, ‘Context is King,’ 172–5. We have come a long way since the days when Gehman could claim that ‘the translator had in mind a certain kind of tree’ when he used ἄρουρα (‘field’) to translate ‫אשׁל‬, ‘tamarisk,’ and even more startlingly that ἀπό meant ‘toward’ in ἀπὸ λιβός (‫)נגבה‬: see Gehman, ‘Adventures,’ 130, 126. The point is well made by Satterthwaite, at ‘Judges,’ 199.

REVIEW OF MURAOKA, A GREEK-ENGLISH LEXICON OF THE SEPTUAGINT

231

the one hand, and what a later reader ‘might have made of the translation’ on the other. The two approaches are not reconcilable and are bound to lead to different results when the conditions require them. Yet they are not unable to co-exist, and the LXX student or scholar can accept and make use of each on its own terms. As well as a lexicon on Muraoka’s principles, it is valuable for us to have a rendering of what the LXX appears to have meant as Greek at the point of production. In practice, of course, things are not always so simple. There are difficulties for both in attaining their intended targets. Muraoka faces the prospect of not one but multiple later readers spread over centuries, and uncertainty about how any of them would have understood the LXX, especially where it presents extremes of un-Greek usage.25 The NETS approach at least has only one person to focus on (at a time) and has a clear theoretical basis for assigning a meaning to the Greek, but it is not easy to do, and NETS is seen to compromise its principles at times.26 Part of the difficulty is the ever-present problem of determining what Greek usage actually was in the translator’s time. A change in the evidence can require a change in the rendering, from a Greek meaning differing from the Hebrew to one matching it, and vice versa. The NETS project and the debates generated by it have been beneficial to LXX studies. The insistence of the NETS editors on establishing a theoretical basis for their work has brought out and clarified issues that were unclear to most of us. It seems to me, indeed, that we have seen the most significant advance in a century in our understanding of what used to be called ‘the Semitic element’ in the LXX. III I offer now just a few examples to illustrate the issues discussed in the preceding section. Num 5:12–13 ἀνδρὸς ἀνδρὸς ἐὰν παραβῇ ἡ γυνὴ αὐτοῦ καὶ παρίδῃ αὐτὸν ὑπεριδοῦσα, (13) καὶ κοιμηθῇ τις μετ᾽ αὐτῆς. Of a man, of a man, if his wife transgresses and disregards him disdaining (him) and someone sleeps with her. ‫ ושׁכב אישׁ אתה‬:‫אישׁ אישׁ כי־תשׂטה אשׁתו ומעלה בו מעל‬ 25

26

For example, ἐν at Hos 12:12, where Muraoka has it under *4. in return for, for the price of. Would a reader without Hebrew guess this? See examples below. One understands, of course, the difficulty of maintaining consistency across the work of over thirty contributors.

232

COLLECTED ESSAYS ON THE GREEK BIBLE

The rendering is isomorphic (‘interlinear’), i.e., the Greek and Hebrew words match in order and meaning (more or less). But the interesting thing is the way the translator, using the syntactic markers natural to Greek, creates a context that makes the sentence hang together and convey a meaning, despite its strangeness as Greek. Especially significant is his choice of genitive case in ἀνδρὸς ἀνδρός: he looks ahead and links them to αὐτοῦ, instead of using the totally ‘literal’ rendering ἀνὴρ ἀνήρ. In this I see clear proof of ‘intention.’ But there is nothing to allow us to take the step of transferring the meaning of the Hebrew idiom to the Greek (‘if any man’s wife ...’). The translator certainly knows what the Hebrew idiom means, as renderings elsewhere show,27 but chooses not to represent it in normal Greek but to keep the un-Greek flavour of the original. The right translation into English, then, is in accordance with NETS principles, i.e., as above, ‘of a man, of a man.’28 The rendering of ‫ אישׁ‬by τις in the next clause shows the translator choosing to go the other way and use natural Greek instead of the stereotypical equivalent ἀνήρ.29 The same phenomena could be illustrated at length. For example, at Gen 8:10–12, the translator first renders the Hebrew idiom ‫ יסף‬hiph + infin. by a natural Greek equivalent (πάλιν ἐξαπέστειλεν) and then two verses later by a ‘literal’ equivalent, unnatural for Greek (οὐ προσέθετο τοῦ ἐπιστρέψαι). To represent what the translator intended when he chose this rendering we must translate ‘she (the dove) did not add to return.’30 Determining a later reader’s understanding of it is harder.31 Deut 19:5 καὶ ἐκκρουσθῇ ἡ χεὶρ αὐτοῦ τῇ ἀξίνῃ κόπτοντος τὸ ξύλον, καὶ ἐκπεσὸν τὸ σιδήριον ἀπὸ τοῦ ξύλου τύχῃ τοῦ πλησίον. and (if) his hand is knocked aside as he is cutting the tree with the axe, and the iron axe-head falling out of the wooden handle hits his neighbour. ‫… ונשׁל הברזל מן־העץ ומצא את־רעהו‬

To render ‫ מצא‬the translator declines the stereotypical equivalent εὑρίσκω and draws on the deep resources of the Greek language for the appropriate word. His choice is τυγχάνω + gen. in its ancient, indeed original, 27 28 29

30 31

For example, Num 1:4 ἕκαστος (κατὰ …). NETS itself compromises: ‘Man by man – if his wife goes astray …’ with note: ‘I.e. Any man.’ Both are concessions to natural English and not what the Greek says. The Leviticus translator adopts the same strategy for ‫ אישׁ אישׁ‬in Lev 15:2 ἀνδρὶ ἀνδρὶ ᾧ ἐὰν γένηται ῥύσις. He too can render idiomatically if he chooses, as in Lev 20:2 ἐάν τις. … NETS again compromises, with ‘did not continue to turn back’: LSJ, s.v. προστίθημι B.III is the culprit. Muraoka, s.v. προστίθημι *2., takes it to be that of the Hebrew idiom: ‘still to do sth as formerly, do sth again.’

REVIEW OF MURAOKA, A GREEK-ENGLISH LEXICON OF THE SEPTUAGINT

233

meaning ‘hit’ (the mark/target), attested since Homer. The total context he creates tells us exactly what he intends. This use of τυγχάνω is not an everyday one and possibly came to the translator from an education in the Classics.32 Isa 48:22 (simil. 57:21) οὐκ ἔστι χαίρειν τοῖς ἀσεβέσι, λέγει κύριος. There is no ‘greetings’ for the ungodly, says the Lord. :‫אין שׁלום אמר יהוה לרשׁעים‬

This example shows how the original Hebrew can be decisive in discerning the translator’s intention; acquaintance with Greek usage is also essential. If the Greek is read on its own it appears to mean ‘it is not possible for the ungodly to rejoice’ (compare Muraoka, s.v. εἰμί 1.d) or ‘there is no rejoicing, says the Lord, for the impious’ (NETS). But χαίρειν translates ‫ שׁלום‬in the original. How does that work? Why not εἰρήνη? The answer strikes us when we bring together the twin facts that ‫ שׁלום‬is a standard greeting in Hebrew, and χαίρειν has the same function in the greeting formula in Greek letters (‘x to y greetings’). The Isaiah translator skilfully captures in the Greek of his day the meaning of the Hebrew as he understands it. 1 Kgdms 1:26 (+ 4) ἐν ἐμοί, κύριε. ‫בי אדני‬

This classic has been discussed so often there is not much left to say. But I include it as a final test of the approaches of Muraoka and NETS. The basic facts are: Hebrew ‫ בי‬is an idiom meaning ‘I pray, excuse me’ (BDB), or ‘by your leave’ (HALOT). The translation ἐν ἐμοί simply replicates the (perceived) components of the Hebrew (‘in/on me’) and conveys nothing more than ‘in me’: there is no comparable idiom in Greek. We do not know if the translator understood the Hebrew idiom; we cannot say that he intended ἐν ἐμοί to have its meaning; we only know that he chose to represent the Hebrew ‘literally.’ The translation of the Greek as produced is therefore ‘in me.’33 In the introduction to NETS, the editors offer ἐν ἐμοί as a specimen of what they call an ‘isolate,’ giving the meaning as ‘in/with me.’34 But 32

33 34

NETS is not quite on the mark: ‘happens to strike his neighbour.’ Dogniez and Harl, Le Deutéronome have got it: ‘atteint son prochain.’ τυγχάνω is only here in the Pentateuch. Homer, Iliad 23.857, with τυγχάνω and its opposite, is too good to miss: ὃς δέ κε μηρίνθοιο τύχηι, ὄρνιθος ἁμαρτών (‘he who hits the string [with his arrow], missing the bird. …’). Compare Barr’s unerring analysis in Typology, 393. ‘To the Reader,’ xvii.

234

COLLECTED ESSAYS ON THE GREEK BIBLE

the actual translations are: ‘by me …?’ (Judg 6:13AB, 15AB), ‘by me’ (Judg 13:8AB, 1 Kgdms 1:26) and ‘with regard to me’ (3 Kgdms 3:17, 26). There have been some compromises, apparently in the direction of more intelligible English. I do not think ‘with regard to’ is possible, and ‘with’ and ‘by’ (instrumental?) would be hard to pick up without a context to suggest them; the treatment as introducing a question is simply puzzling. So the NETS outcome is only approximately in line with what one expected. Muraoka (s.v. ἐν *18) rightly labels ἐν ἐμοί a calque on the Hebrew.35 But he gives no meaning. He thus does not commit himself to a decision on what it would mean to a Hebrewless later reader, no doubt because it is difficult to say, just as that reader would be hard put to make sense of it. IV Muraoka’s lexicon is a major step forward in LXX lexicography. It will now be both the standard tool and the foundation for future development. As to the latter, the lexicography of the LXX is young and will certainly not stand still in coming years. Let us consider what might happen next. It goes without saying that refinements to the lexicon will need to be made. I am not suggesting a major overhaul, but minor improvements that will ensue from further observation and discussion. These are of the kind already pointed out by other reviewers; my own list can wait.36 They can easily be done in future editions – which are certain to come, if the history of NT lexicography is any guide. At the same time there are other needs in LXX lexicography that go beyond small adjustments to a printed work, though the results will end up there. First and most important is the further pursuit and evaluation of evidence of Greek outside the LXX. This evidence obviously plays a vital role in determining meaning in the LXX. The easy assumption that the evidence can all be found in works like LSJ is mistaken. Muraoka, as he explains, drew on whatever was available in the reference books and special studies, but did not undertake fresh searching of his own; he also 35

36

NETS has its own use of the term calque that seems at variance with the usual understanding. For a recent definition see Adams, Bilingualism, 459. NETS’s stock example, διαθήκη ~ ‫ברית‬, cannot be a calque in that sense. See, e.g., the reviews by Danker; Shaw; and Silva. The long review of Muraoka, Lexicon (Pentateuch and Twelve Prophets) in Andersen, ‘Advances’ raises many questions but is short on answers.

REVIEW OF MURAOKA, A GREEK-ENGLISH LEXICON OF THE SEPTUAGINT

235

does not attempt to report the evidence in his entries.37 This approach was the right one. But the task of completing the evidence for all the vocabulary of the LXX now needs to be undertaken in earnest. This will be ongoing and involve many contributors, not one lone lexicographer. I also do not believe that the lexicon should attempt to assemble this evidence; it should do no more than draw on the results. The place for the assembly of the evidence is in an electronic database. A database is also the place for the material that meets the second need, which we will come to in a moment. I give just one example of the importance of non-LXX evidence and how it can change. The word προσήλυτος has been supposed, on the basis of lack of attestation, to be a creation of the LXX translators and even, despite the inherent improbability, to have had the sense ‘proselyte’ when it was coined.38 But at the 2009 SBL meeting David M. Moffitt presented a papyrus document containing an occurrence of the word: the date is III BC, there is no suggestion of a Jewish connection, and the meaning can hardly be other than the expected one, i.e., ‘newcomer, temporary resident.’39 Of course, proper searching of the evidence is not just for the purpose of finding parallels outside the LXX, but also to establish when a word or use is unattested and so might be a creation of the translators. The other need is for an ongoing, up-to-date, complete record of existing discussions of words in the LXX, and indeed in all of Greek.40 Muraoka includes a selection in each entry, following his sound principle that he names only works that have made a contribution to his decision.41 LEH include much fuller lists. But we have no way of knowing if the lists are complete, and they are impossible to keep up to date; further, mere lists tell us nothing about what the authors have to say. The only way this can be remedied efficiently is in an electronic database. Again just one example, the ‘famished bear’ of Hos 13:8 (ἄρκος ἀπορουμένη). The discussions of this disputable case seem to be proliferating, almost before one has had time to take it in. Muraoka mentions only Bons 2001. Besides that there is his own discussion in the Introduction 37

38

39

40 41

‘Introduction,’ ix. Compare comments in my review of Muraoka, Lexicon (Pentateuch and Twelve Prophets), 129 [= Essay 10]. Labelled ‘neol.’ by LEH, marked * by Muraoka. See both for literature as far back as Allen in 1894. The papyrus is P.Duk.inv.727R. Publication is expected in the near future (email from David Moffitt, 6.5.10). [See Lee, Greek of the Pentateuch, 15 for an update.] See Lee, ‘Lexicographical Database.’ Muraoka, ‘Introduction,’ xv. Compare remarks in Lee, Review of Lexicon (Pentateuch and Twelve Prophets), 129–30.

236

COLLECTED ESSAYS ON THE GREEK BIBLE

(p. viii), where the views of Harl and Joosten are mentioned; LEH add no more; but I have come across one by Boyd-Taylor, with a further reference to Jobes and Silva.42 Are there more? Who can say? And what do they each contribute? To see where the discussion is up to, it would be helpful, one might say essential, to have this information. The same applies to countless other cases. The place for all this information is in an electronic database, and I believe this to be the new desideratum of LXX studies. Reviewers tend to focus on faults, often [self-]indulgently, showing off their skill at spotting mistakes and raising questions. In the final impression the flaws become magnified out of proportion. The criticisms voiced in this review cannot detract from Muraoka’s outstanding achievement. The work as a whole is reliable, helpful, and thorough. In it students of the LXX have a tool that we can use with confidence and respect long into the future.

42

Boyd-Taylor, ‘Evidentiary Value,’ 54–5.

16 AUXILIARY THELO 20101

Abstract Over the history of Greek, the verb θέλω, ‘wish,’ ‘be willing,’ evolves into an auxiliary expressing futurity. Instances as early as Classical Greek are known; partial collections of examples have been made. My aim in this paper is to gather as many instances as possible up to 600 AD. Existing sources as well as my own observations are used, and 70 + instances are found. Even so, this is not an exhaustive collection. The conclusion is reached that θέλω was available as a future auxiliary over the course of 1000 years, but was never more than sporadic until after the end of Koine Greek. All the examples are of θέλω + infinitive, none of θέλω ἵνα + subjunctive, a detail of significance and contrary to what is sometimes assumed. In the course of the investigation another, previously unnoticed use of θέλω, namely, imperative of θέλω + infinitive with the meaning ‘(please) do/don’t,’ was discovered; instances of this use are presented.

1. THE GRAMMATICALISATION

OF ΘΕΛΩ

The future tense in Modern Greek is formed with θα + subjunctive, as for example, θα γράϕω, θα γράψω. This form of expression has its origin in a periphrasis with θέλω. The ultimate base is θέλω + infinitive, with θέλω in its original meaning ‘wish to,’ which evolves into an expression of simple futurity. The development is parallel to that in many languages, among them of course English, in which futurity is expressed by an auxiliary that originally meant ‘wish/want’; or to put it in terms of grammaticalisation, the lexical item ‘wish/want’ has evolved along the cline of grammaticality to a grammatical function, namely, to express futurity.2 The detailed history of the development in the Byzantine period is not the concern of this paper and will be touched on only briefly. It is more 1

2

This paper was originally presented at the conference ‘Buried Linguistic Treasure: The Potential of Papyri and Related Sources for the Study of Greek and Latin,’ at Christ Church, Oxford in 2006, convened by Trevor Evans and Dirk Obbink. Cf. Hopper and Traugott, Grammaticalization, 6–7.

238

COLLECTED ESSAYS ON THE GREEK BIBLE

complicated than one might have expected, and debate continues on the details. It is not simply a matter of a single line of development θέλω + infinitive > θέλω ἵνα + subjunctive > θέλω να + subjunctive > θα + subjunctive; there are more steps and variants involved, as shown especially by Brian Joseph’s study, which is a warning against over-simplification.3 For our purposes, let us simply note that when the ancestor of the Modern Greek particle first appears in the twelfth century, in the form θε να, it does so alongside θέλω + infinitive as an expression of futurity: the latter was still in use and continued to be for some time before its final displacement by θα + subjunctive. By the Byzantine period the periphrasis with θέλω had clearly prevailed over the other, earlier contenders as the means of expressing the future. The other main contenders, at the end of the Koine period, were: the old monolectic form; the present with future sense; μέλλω + infinitive; ἔχω + infinitive; the aorist subjunctive.4 We know with hindsight that θέλω was to prevail, but the issue had not yet been decided.

2. A NEW ASSESSMENT OF THE EVIDENCE BEFORE 600 AD Where are the beginnings of this development of θέλω? When, in the period before 600 AD, does θέλω start to show signs of being a future auxiliary? Where are the examples, and how many are there? That is the question that I want to (and will) address in this paper.5 It is commonly said that θέλω + infinitive appears as a future-equivalent in the Koine period. Browning is the most authoritative voice on the subject. He simply states it as a fact, adding the rider that it is not common till after 600, but gives no examples.6 Browning’s book was intended as a general survey, without detailed references; but if we look elsewhere it 3

4 5

6

Joseph, Morphology and Universals, 114–59. Cf. Horrocks, Greek, 130, 228–9. See also Pappas, ‘Microcosm’; Joseph and Pappas, ‘On Some Recent Views’; Holton, ‘Formation of the Future,’ 119–20, 127–8; Hock and Joseph, Language History, 402–5 (Balkan developments); earlier Jannaris, Grammar, 552–9; Thumb, Handbook, §226; Hatzidakis, Μεσαιωνικά, 1:197. See Browning, Greek, 33–5. Other less frequent future-equivalents are also noted there. I am well aware of the older usage of shall and will maintained by some, but as this is not my own practice and is artificial for me, it is not followed in this paper. Browning, Greek, 34. He goes on (35) to list the numerous ways of expressing futurity in John Moschos, again without citing examples except one (not of θέλω). The unnamed source from which these data are derived, Mihevc-Gabrovec, Études, 64–5, noted only one instance of θέλω as a future auxiliary in Moschos (see under no. 1 in my list of examples below).

AUXILIARY THELO

239

is much the same. Gignac, in his generally thorough grammar of the papyri of the Roman and Byzantine periods, speaks of the ‘increasingly frequent replacement of the future tense by periphrastic constructions in the later Koine, mainly by θέλω ἵνα and the subjunctive,’ but offers only two examples – in the same text, and in fact of θέλω + infinitive.7 Mandilaras likewise asserts it, but gives no examples.8 Joseph simply refers to Browning.9 Horrocks takes it for granted and does not amplify.10 Back in 1898, Karl Dieterich did much better: he noted some instances in late funerary inscriptions, a source which proves to be a rich one when modern searching techniques are applied; but his observations slipped out of sight.11 Besides these there are a number of specialised studies, notably those of Joüon, Riesenfeld, Rödiger, Schrenk, and Wifstrand, that offer useful collections of examples.12 But they all focus on their own area of interest; they do not connect with one another nor study the phenomenon across time. In the lexicons there is a certain amount of material, very partial, but useful as far as it goes. LSJ offer a sense II.1 ‘to express a future event, like our will or shall,’ with eight instances cited (plus an ‘etc.’). The examples are all Classical, and only half seem to me to be right, but even so, this is a beginning.13 Lampe also recognises this sense, but has only two examples, the same two that had been noted (with three others) by Sophocles 7

8 9 10 11

12

13

Gignac, Grammar, 2:290, with n. 3. He adds a reference to Burguière, Histoire de l’infinitif, but this work yields no Koine Greek examples of auxiliary θέλω. Gignac’s examples are at no. 6 below. Mandilaras, Verb, 180. Joseph, Morphology and Universals, 114, 116, with nn. p. 150. Horrocks, Greek, 76, cf. 229–32. Dieterich, Untersuchungen, 245–6. See nos. 7 to 10 in the list of examples. The lengthy discussion in Mirambel, ‘Essai,’ 179–88, yields one example, the same one as noted by Mihevc-Gabrovec in Moschos (see n. 6 above). Joüon, ‘Θέλειν’; idem, ‘Les verbes’; Riesenfeld, Zum Gebrauch von θέλω; Rödiger, ‘βούλομαι und ἐθέλω’; Schrenk, ‘θέλω, θέλημα, θέλησις’; Wifstrand, ‘Verba für wollen.’ Other works consulted: Adrados, ‘Sobre los orígenes’; Braun, ‘Nota sui verbi’; Fleischman, The Future; Fox, ‘βούλεσθαι und (ἐ)θέλειν’; Pinkster, ‘Development of Future Tense’; Preisigke, Wörterbuch. I have not been able to see Psichari, Quelques travaux. LSJ, s.v. ἐθέλω. The whole section II is headed ‘of inanimate things’ and examples of that kind are cited first under II.1; then LSJ add ‘very rarely of living things’ and proceed to cite an equal number. The distinction has no effect on the lexical meaning, but, as Willy Clarysse pointed out to me after my paper at the ‘Buried Linguistic Treasure’ conference, examples applied to inanimates are strong proof of the development. Andreas Willi made the somewhat similar point that the clearest examples will be those where the verb is in the third person, and not in an if-clause (as no. 35 below). The ‘etc.’ in LSJ covers some good Plato examples that had been in the 7th and 8th editions but were dropped in the 9th, leaving only Rep. 370b, an unconvincing case.

240

COLLECTED ESSAYS ON THE GREEK BIBLE

back in 1887.14 The NT lexicons, on the other hand, are not aware of the question at all; even the probable NT examples escape notice, let alone others.15 Most surprising is DGE, which has no instances of this sense and apparently does not recognise its existence.16 My purpose has been to gather as many examples of θέλω as a future auxiliary as I can from all previous sources, as well as those I have found myself. It must be said at once that the collection is not exhaustive. While most of the documentary evidence has been checked (via PHI7), I have not done the full examination of Greek literature that would be possible – though forbidding – by means of the TLG and would be likely to yield further material. But what I have goes some way towards answering the question. My list of examples is presented below, in reverse chronological order. Needless to say, all the items in the list have been thoroughly vetted; I have rejected any proposals that are doubtful. The list is therefore not a list of all the proposals but only of those that have a good chance of being what we are looking for. What are we looking for? This needs to be clear at the outset. We are looking for cases where the usual or established senses of θέλω do not seem to work, where any such sense has faded away to the point where there is not much left but futurity. By the usual senses I mean ‘wish/want’ and ‘be willing’; I do not include among them a meaning ‘intend,’ as I am not sure that it is clearly established for this word (as it is for μέλλω). This fading does not rule out the possibility, even likelihood, that θέλω retained some nuance that distinguished it from the monolectic future expressing simple futurity and from other future expressions. But it is difficult if not impossible for us at this distance to appreciate such a nuance; even to define the usual senses of θέλω is notoriously difficult. In a quest to find any new semantic development, one needs to be able to produce examples that are better than just possible, but highly probable (or as John Chadwick would have put it, ‘incontrovertible’). It is a severe test in this case, because it is in the nature of the phenomenon that there is gradual shading from one meaning into another, and it is hard to know in a particular instance whether the meaning really has shifted from the lexical area into the grammatical.17 I cannot claim that my examples all pass this test, but there are certainly some. 14 15

16 17

Lampe, PGL, s.v. θέλω IV; Sophocles, Lexicon, s.v. θέλω 5. See BDAG, s.v. θέλω; Louw and Nida, Lexicon, Subdomains 25.1,102; 30.58; 31.4. Louw (‘Analysis of Meaning,’ 142) specifically rejects Mark 6:48 (no. 23 below) in reply to me (Lee, ‘UBS Lexicon,’ 179). DGE, vol. 6, s.v. ἐθέλω tb. [también] θέλω. Cf. Hopper and Traugott, Grammaticalization, 6–7, 9.

AUXILIARY THELO

241

3. SAMPLES Let us take some samples from the list to illustrate these points. In the case of no. 38 Hdt. 1.109.4 εἰ δ᾽ ἐθέλει … ἀναβῆναι ἡ τυραννίς, it is hard to see how, with the inanimate subject ‘sovereignty,’ the verb can continue to have its sense of ‘wish’ or any other distinct semantic content; we are left with futurity. The same can be said of no. 35 Pl., Rep. 423b μέχρι οὗ ἂν ἐθέλῃ … εἶναι μία, where the subject is ‘the city.’ These are just two items from the surprisingly extensive evidence in the Classical period, notably in Herodotus and Plato. For good examples from much later, consider no. 21 Aesopi Fab. 142 πῶς πάλιν ἐξ ὄνου ἵππον θέλεις ἔχειν; and no. 12 POxy 14.1763.10 λέγουσι δὲ ὅτι ǀ μέχρι ιε θέλομεν ǀ ἐξελθεῖν. The latter in particular seems to be a periphrasis for the future. The context, with its reported speech and time expression, makes it clear that the volition of the parties described as ‘we’ is not in the picture. In no. 7 MAMA 1.160.4 ἄν τις θεǀλήσι ἀνῦξεν, it is not a question of someone merely wanting to open the tomb; the text envisages someone actually doing so in the future and suffering the penalty. All these funerary texts (nos. 7–10, with list) are similar and provide a strong bloc of examples. We note that the dates are not from the end of Koine Greek but from the fourth, third and even second centuries AD (many are of course not precisely datable). Among them no. 10 IPrusaOlymp 1.83.9 is an interesting variation. The words ἐὰν δέ τις θ[ε]ǀλήσει stand without an infinitive expressed: it is to be supplied by extrapolation from ἀνεξοδίαστον, ‘inalienable,’ ‘not to be taken over,’ to give the meaning ‘if anyone shall/does (alienate it, take it over).’ The simple future-auxiliary function of θέλω seems inescapable. Included in the list are one or two examples where there is some doubt. Example no. 32 PHib 1.65.25 was proposed by Mayser.18 At first sight one would be inclined to take θέλομεν as ‘I want to,’ but a reading of the whole letter suggests that Mayser was right. The writer is explaining his plans and simply states what he will do to make up the deficit if he can get some help with the rest. It is not, then, an expression of a wish but a description of future action. If accepted, this would be contemporary with no. 31 LXX Exod 2:14 μὴ ἀνελεῖν με σὺ θέλεις, which I think is sound.19 18 19

Mayser, Grammatik, II.1, 226. Cf. Evans, Verbal Syntax, 229, where θέλω is taken to be ‘mean/intend’: ‘surely you don’t mean to kill me?’ My understanding of μή here as introducing a neutral question makes a slight difference.

242

COLLECTED ESSAYS ON THE GREEK BIBLE

In an example like no. 16 Hermas, Vis. 3.1.9 θέλοντος οὖν μου καθίσαι εἰς τὰ δεξιὰ μέρη, we seem to catch θέλω at the point of transition. Is it ‘wishing to’ or ‘being about to’? Either is possible. But the two clearer examples of ‘be about to’ in the same author (no. 17, and 3.3.1) tip the balance in favour of the latter. Two general observations may be made at this point. Quite a number of the examples are of a past tense (ἤθελον, ἠθέλησα), where I have translated ‘was going to,’ ‘was about to.’ Some overlap or competition with μέλλω, the standard word for this from early Greek onwards, seems obvious. A full study is needed before any firm conclusions can be reached about μέλλω in Koine Greek and its relation to θέλω, but one suggestion may be put forward here. If we take the NT, where μέλλω is common (109 occurrences), as a sample, we see that while μέλλω is sometimes used like θέλω (as Acts 16:27 σπασάμενος τὴν μάχαιραν ἤμελλεν ἑαυτὸν ἀναιρεῖν), it most often refers to the more distant future (as John 6:71 οὗτος γὰρ ἔμελλεν παραδιδόναι αὐτόν). This loss of immediacy could be the reason for another contender to appear, to supply the meaning ‘be on the point of.’ Secondly, I draw attention to a noteworthy fact: none of the examples in my collection shows θέλω ἵνα + subjunctive; all are of θέλω + infinitive. This is significant in the light of what came later. It is consistent with the evidence of the continuing use of θέλω + infinitive as a future expression in Byzantine Greek. A shift from infinitive to ἵνα + subjunctive in this expression appears not to have been a feature of the Koine period at all.

4. AN

UNOBSERVED AUXILIARY FUNCTION OF ΘΕΛΩ

Before I come to my conclusion on θέλω as a future auxiliary, there is another to be noticed. In the course of this investigation I came across what amounts to another auxiliary function of θέλω, one that as far as I know has not been observed before. It was from the search of the papyri for θέλω that this discovery emerged; a connection could then be made to some literary examples not considered in this light before. A selection of examples is given below (§§6.2, 3). What we see is θέλω in positive and negative commands + infinitive (rarely imperative: POxy 14.1776.6 below) in which the full semantic content of θέλω has faded and the verb is simply a means of introducing or

AUXILIARY THELO

243

in some way nuancing the instruction contained in the infinitive. The combination appears to form a polite request, but the semantic value of θέλω itself is hard to pin down. ‘Please’ or ‘be so kind as to’ are makeshifts and not true equivalents, though they give the general effect. The development would seem to come from θέλω in its sense of ‘be willing.’ The upshot is that we have an auxiliary reminiscent of the English auxiliary do in ‘do say,’ ‘do not say,’ etc.

5. CONCLUSION Finally, a conclusion on θέλω as a future auxiliary. We have a large number of examples, over seventy, spread across a time span of more than 1000 years. How do we assess the significance of this evidence? What does it mean? On the one hand, it seems clear that θέλω as a future auxiliary was more common than has been supposed. If we apply the argument that people write more carefully than they speak, and assuming that θέλω as future auxiliary was a vernacular feature, it could be concluded that it was in fact in frequent use in speech throughout the whole time, and has simply not surfaced much in our evidence. On that basis it might already have been the front-runner well before the end of Koine Greek. On the other hand, we find as late as the end of the Koine period, in reasonably vernacular texts, various other future expressions still competing with θέλω. Moreover, some of these are not old but new contenders that had appeared in the middle and later Koine (present with future sense; ἔχω + infinitive; aorist subjunctive), thus showing that no one form had yet established itself. So one might conclude that θέλω as future auxiliary was never very common, and even by the end of the Koine period was still some way from establishing the dominance that it was to gain later. I incline to the latter view, that θέλω was available as a future auxiliary for 1000 years, but was never more than sporadic until after Koine Greek. Though the time span of θέλω as future auxiliary seems remarkably long, and one might be surprised that different, nearly synonymous ways of expressing the future could co-exist for centuries, grammaticalisation studies have shown that just such characteristics are part of the phenomenon.20 20

Cf. Hopper and Traugott, Grammaticalization, 97 on the persistence of alternative future markers in English since the time of Beowulf.

244

COLLECTED ESSAYS ON THE GREEK BIBLE

Obviously what would be helpful next is a statistical study, in as large a corpus as possible, of all the ways of expressing futurity up to the end of Koine Greek. But that is a task for the future.21

6. EXAMPLES 6.1 θέλω as future auxiliary 1. John Moschos 19 (M 87:2865C–D) (c.600 AD) [Sophocles; Jannaris; Dieterich; Lampe]22 διὰ τοιαύτην οὖν ἡδονὴν βλέπε πόσους κόπους θέλεις ἀπολέσαι, ἴδε διὰ ποίαν ἁμαρτίαν θέλετε ἑαυτοὺς ἀποστερῆσαι τῆς βασιλείας τῶν οὐρανῶν. ἀβάλε τῇ ἀνθρωπότητι. διὰ μίαν ὥραν ὅλον ἐκεῖνον τὸν κάματον θέλεις ζημιωθῆναι; For the sake of such pleasure look how much work you are going to destroy, see for what kind of sin you are going to deprive yourselves of the kingdom of heaven. Oh human nature! Are you willing to/going to forfeit all that toil for one hour [of pleasure]? See also 184 (M 87:3057A) [Mihevc-Gabrovec] 2–4. Apophthegmata Patrum (M 65:76–440) (c.500 AD) 2. Arsenios 29 ἦλθέ ποτε μαγιστριανός, ϕέρων αὐτῷ διαθήκην τινὸς συγκλητικοῦ συγγενοῦς αὐτοῦ, ὃς κατέλιπεν αὐτῷ κληρονομίαν πολλὴν σϕόδρα· καὶ λαβὼν αὐτήν, ἤθελε σχίσαι. καὶ ἔπεσεν ὁ μαγιστριανὸς εἰς τοὺς πόδας αὐτοῦ, λέγων, δέομαί σου, μὴ σχίσῃς αὐτήν· Once a magistrianus came to him bringing the will of a certain senator his kinsman, who had left him a very large inheritance. He took it and was going to tear it up. The magistrianus fell at his feet saying, ‘I beg you, don’t tear it up.’ 3. Paphnoutios 1 [ὁ ἀρχιλῃστὴς] … ἐγέμισε ποτήριον οἴνου, καὶ τὸ ξίϕος ἐν τῇ χειρὶ αὐτοῦ, καὶ λέγει τῷ γέροντι· ἐὰν μὴ πίῃς, ϕονεύω σε. γνοὺς δὲ ὁ γέρων ὅτι ἐντολὴν Θεοῦ θέλει ποιῆσαι, βουλόμενος αὐτὸν κερδῆσαι, ἔλαβε καὶ ἔπιεν. 21

22

A recent Cambridge PhD dissertation by Theodore Markopoulos, ‘The Category “Future” in Greek: A Diachronic Investigation of Three Future-referring Periphrastic Forms,’ deals with μέλλω, ἔχω and θέλω from the Hellenistic to the late Medieval period. I thank the author and others for bringing this to my notice after my paper. My examples have been collected independently. [Published in 2008; see Postscript.] A name in square brackets after a reference indicates the scholar who proposed this example; if there is no name, it is my own proposal. See bibliography for works referred to.

AUXILIARY THELO

245

[The robber chief] … filled a cup of wine and with his sword in his hand said to the elder, ‘If you don’t drink, I’ll kill you.’ The elder, knowing that he was about to perform a command from God, and wishing to win over the robber, took it and drank. 4. Silouanos 1 καὶ ἐξελθόντων αὐτῶν, εὗρεν ὁ μαθητὴς αὐτοῦ ὕδωρ ἐν τῇ ὁδῷ καὶ ἤθελε πιεῖν· καὶ λέγει αὐτῷ ὁ γέρων· Ζαχαρία, νηστεία σήμερον· After they had set out, his disciple found water on the way and was going to take a drink. The elder said to him, ‘Zacharias, fast day today!’ See also Makarios 1; 11. 5. Acta Conc. Ephes. 1.1.2, p. 40 l. 19 (ed. Schwartz) (431 AD) ἢ πῶς Χριστιανοὶ θέλουσιν ὀνομάζεσθαι οἱ λέγοντες εἰς ἄνθρωπον ἅγιον ὡς ἐπὶ ἕνα τῶν προϕητῶν ἐληλυθέναι τὸν λόγον καὶ μὴ αὐτὸν ἄνθρωπον γεγονέναι λαβόντα ἐκ Μαρίας τὸ σῶμα, ἀλλ᾽ ἕτερον εἶναι τὸν Χριστὸν καὶ ἕτερον τὸν τοῦ θεοῦ λόγον τὸν πρὸ Μαρίας καὶ πρὸ αἰώνων υἱὸν ὄντα τοῦ πατρός; Or how will they be called Christians who say that the Word came into a holy man as upon a prophet and that he did not become a man by taking his body from Maria, but that Christ is one thing and the Word of God, who was the Son of the Father before Maria and before the ages, is another? 6. PMichael 39.10, 14 (V AD?) [Gignac] ει (= ἡ) μητέρα σου ǀ ἀσθενῖ, ἀποθανῖν θέλι. ǀ ἐὰν δυνασθῇς σοὺ βάλλε ǀἠ ινοψεσουσωπου. ει μιǀτέρας σου εἶπι ἀποθανῖν θέλιν. [Punctuation and some accents added.] Your mother is sick, she is going to die. If you can, put. … Your mother said she is going to die. 7. MAMA 1.160.4 (IV AD?) ας ᾽Ωαλεντίνη ǀ ζῶν ϕρονῶν ἀ[ν]ǀήστησα μνήμης ǀ χάριν. ἄν τις θεǀλήσι ἀνῦξεν (= ἀνοῖξαι) ἔǀξ[ωθε]ν τοῦ γένǀο[υς, ἕξ]η πρὸς τὴǀν [Τριάδ]αν. I …-as Valentini while alive and in my right mind set up [this tomb] as a memorial. If anyone outside my family shall open it, he will have to face the Trinity. 8. IG XIV 1563.2 (Rome) [Dieterich] Εἰρῆνα ἔζησεν ἔτη ζ. ἐὰν οὖǀν τις αὐτὴν θελήσῃ ἀνορύξǀαι, τὸν μέλλοντα αἰῶνα μὴ κλǀηρονομήσῃ. Eirena lived seven years. If anyone shall dig her up, may he not inherit the age to come.

246

COLLECTED ESSAYS ON THE GREEK BIBLE

9. TAM 5.1.213.7 (261/2 AD) εἴ τις θελήσει ǀ σκυβαλλίσαι τὸ μνῆμα τοῦτο, ἕǀξει τὸν Ἀπόλλωνα κεχολωμέǀνον καὶ τὴν κυρίαν Ἀναεῖτιν διὰ ǀ τέκνα τέκνων, ἔγονα (= ἔκγ-) ἐγόνων. If anyone shall foul this tomb, he will incur the wrath of Apollo and the lady Anais for his children’s children, his descendants’ descendants. 10. IPrusaOlymp 1.83.9 (II AD) … τὸ μνημ[εῖ]ǀον κατασκεουασάντων ǀ Μαρκέλλας τῆς γυναικὸς ǀ αὐτοῦ καὶ τῶν υἱῶν αὐτοῦ ǀ [- - -] ἀνεξοδίαστον σ[ὺν] τ[ῷ] ǀ περικήπῳ. ἐὰν δέ τις θ[ε]ǀλήσει, δώσει τῇ πόλει προσǀ[τ]είμου (δην.) ͵ βϕʹ. χαίρετε. … the tomb, prepared by Marcella his wife and his sons, … not to be taken over by someone else, together with the garden around it. If anyone does [take it], he will give to the city a penalty worth 2500 dinaria. Farewell. Similarly (total 12): IG XIV 238.10 [Dieterich]; 625.6 [Dieterich]; MAMA 6.234a.6 (c.275 AD); IGBulg 1.218.7; 3.1.996.2; IKios 100.8; TAM 2.3.1086.5; TAM 5.1.741.8 (244/5 AD); 5.1.776.11 (305/6 AD); 5.2.1077.10; 5.2.1083.7; 5.2.1107.6. 11. Acta Xanthippae et Polyxenae 7.25 (ed. James) (III AD) [Jannaris] εἰ ἦν μοι δυνατόν, ἤθελον ἅψασθαι τοῦ κρασπέδου τῶν ἱματίων αὐτοῦ, ἵνα ἴδω [1. εἰδῶ?] τὴν εὐμένειαν καὶ τὴν πρόσδεξιν αὐτοῦ καὶ εὐωδίαν. [Xanthippe has caught sight of Paul walking in the street outside her house and says:] If it were possible for me, I would take hold of the hem of his clothes, so that I might see/know his goodwill and acceptance and fragrance. 12. POxy 14.1763.10 (III AD, after 222) [Costas; Joüon, ‘Les verbes’] οὔπω μέχρι σήμεǀρον τὰ πλοῖα τῆς ἀνǀνώνας ἐξῆλθεν ǀ ἵνα δυνηθῶμεν ἐξελǀθεῖν, καίτοι ἐμοῦ μηǀδὲν ἔχοντος πρᾶξαι ǀ ἐνθάδε. λέγουσι δὲ ὅτι ǀ μέχρι ιε θέλομεν ǀ ἐξελθεῖν σὺν θεῷ. Up till today the grain-supply ships have not left so that we could leave, yet I have nothing to do here. They say that we will leave by the 15th with god’s help. 13. PMichael 17.3 (II–III AD) εἰ ǀ οἶδας ὅτι θέλεις μετενέγκαι τὸν ǀ σῖτον εἰς Πέψα, γράψον πῶς ǀ μέλλομεν αὐτὸν μετενέγκαι. If you know that you will transfer the grain to Pepsa, write [and tell me] how we are going to transfer it. 14. X. Eph. 2.13.3 (II–III AD) ἔδει δὲ τὴν Ἀνθίαν οὕτως ἱερουργηθῆναι. ὡς δὲ πάντα ἕτοιμα ἦν καὶ κρεμνᾶν τὴν κόρην ἤθελον, ψόϕος τῆς ὕλης ἠκούετο καὶ ἀνθρώπων κτύπος.

AUXILIARY THELO

247

It was necessary for Anthia to be sacrificed in that manner. When all was ready and they were about to hang the girl up, a rustling was heard in the bushes and the sound of men moving. 15. POxy 10.1293.18 (117–138 AD) τοῦτο οὐχ ἕνεκα ἡμῶν ǀ ποιῶ ἀλλὰ ἕνεκα τῶν καμηλειτῶν, ǀ μὴ θελήσῃ τις ἀϕεῖναι μέρος ǀ μὴ ἐνένκας. I am not doing this [i.e., asking for confirmation of receipt of a previous load of oil] on our account, but because of the camel drivers, in case any of them leaves part [of the load] and does not bring it. 16–17. Hermas, Vis. (II AD)

16. 3.1.9 λέγει μοι· κάθισον ὧδε. λέγω αὐτῇ· Κυρία, ἄϕες τοὺς πρεσβυτέρους πρῶτον καθίσαι. ὅ σοι λέγω, ϕησίν, κάθισον. θέλοντος οὖν μου καθίσαι εἰς τὰ δεξιὰ μέρη οὐκ εἴασέ με, ἀλλ᾽ ἐννεύει μοι τῇ χειρὶ ἵνα εἰς τὰ ἀριστερὰ μέρη καθίσω. She says to me, ‘Sit here.’ I say to her, ‘Lady, let the elders sit first.’ ‘Do as I say,’ she says, ‘Sit.’ Then when I was about to sit on her right she stopped me and signalled to me with her hand to sit on her left. 17. 3.2.3 ταῦτα εἴπασα ἤθελεν ἀπελθεῖν· πεσὼν δὲ αὐτῆς πρὸς τοὺς πόδας ἠρώτησα αὐτὴν κατὰ τοῦ κυρίου ἵνα μοι ἐπιδείξῃ ὃ ἐπηγγείλατο ὅραμα. After she said this she was going to leave; but I fell at her feet and asked her by the Lord to show me the vision which she had promised. See also 3.3.1 18. Hypothesis of Euripides, Alexandros. POxy 52.3650.29 (early II AD)23 … οἵτινες ἡττῆσθαι διαλαβ[ό]ντες ὑπὸ δούλου κατηξίωσαν τὴν Ἑκάβην ὅπως ἂν αὐτὸν ἀποκτείνῃ. παραγενηθέντα δὲ τὸν Ἀλέξανδρον Κασ[σάν]δρ[α] μὲν ἐμμανὴς ἐπέγνω καὶ π[ερὶ τῶ]ν μελλόντων ἐθέσπισεν, Ἑκάβη [δὲ ἀπο]κτεῖναι θέλουσα διεκωλύθη. π[α]ρα[γενό]μενος δ᾽ ὁ θρέψας αὐτὸν διὰ τὸν κίνδυνον ἠναγκάσθη λέγειν τὴν ἀλήθειαν. … who, supposing they had been defeated by a slave, urged Hekabe to kill him. When Alexandros arrived, Kassandra in a raving state recognised him and prophesied what would happen; but Hekabe as she was about to kill him was prevented: the man who reared him arrived and because of the danger was compelled to tell the truth. 23

I first noticed this example some years ago at a seminar on this text by the late Kevin Lee in Sydney. [My suggestion that θέλουσα might mean ‘being about to’ was met with little enthusiasm at the time.] Coles’s translation in the ed. princ. in POxy was ‘Hecabe who wished to kill him’; in Collard et al., Euripides, 2:50–1, it is ‘Hecuba, who was ready to kill him.’ There is no comment on θέλουσα in either edition.

248

COLLECTED ESSAYS ON THE GREEK BIBLE

19–20. Vita Aesopi G (ed. Perry; Ferrari) (I AD) 19. 99.6 καθ᾽ ὃν καιρὸν ἦν ὁμόϕωνα τὰ ζῷα τοῖς ἀνθρώποις, πένητα ἀπορούμενον τροϕῆς ἐπιλαβέσθαι [δὲ] ἀκρίδας τὰς λεγομένας ‹τερετιστρίας› καὶ ταύτας ταριχεύειν καὶ πωλεῖν ϕανερῆς τιμῆς. πιάσας δέ τινα ἀκρίδα ἠθέλησεν αὐτὴν ἀποκτεῖναι. ἡ δὲ ἰδοῦσα τὸ μέλλον πρὸς τὸν ἄνθρωπον εἶπεν. … At the time when animals had the same speech as human beings, a poor man lacking sustenance caught grasshoppers called hummers and pickled them and offered them for sale at a certain price. He caught a certain grasshopper and was about to kill her, but she, seeing what was going to happen, said to the man. … 20. 91.9 ἡ μὲν οὖν τοῦ σημείου λύσις ἐστὶν αὕτη· πάντως τις τῶν βασιλευόντων θελήσει ὑμῶν τὴν ἐλευθερίαν καταδουλῶσαι καὶ τοὺς νόμους ἀκυρῶσαι καὶ ἐπισϕραγίσαι τῇ ἰδίᾳ δυνάμει. The interpretation of the sign is this: one of the reigning kings will for certain reduce you from freedom to slavery, nullify your laws and put the stamp of his power on you. 21. Aesopi Fab. (ed. Chambry) 142 ῞Ιππος καὶ στρατιώτης. ὅτε δὲ ὁ πόλεμος κατέπαυσεν, εἰς δουλείας τινὰς καὶ ϕόρτους βαρεῖς ὁ ἵππος ὑπούργει. … ὡς δὲ πάλιν πόλεμος ἠκούσθη καὶ ἡ σάλπιγξ ἐϕώνει, τὸν ἵππον χαλινώσας ὁ δεσπότης καὶ αὐτὸς καθοπλισθεὶς ἐπέβη. ὁ δὲ συνεχῶς κατέπιπτε μηδὲν ἰσχύων· ἔϕη δὲ τῷ δεσπότῃ· ἄπελθε μετὰ τῶν πεζῶν [τῶν] ὁπλιτῶν ἄρτι· σὺ γὰρ ἀϕ᾽ ἵππου εἰς ὄνον με μετεποίησας, καὶ πῶς πάλιν ἐξ ὄνου ἵππον θέλεις ἔχειν; When the war ended, the horse served at various tasks and carried heavy loads. … When war was declared again and the trumpet called, the master put the bridle on the horse, put his armour on and mounted. But the horse continually fell down because he had no strength. He said to his master: ‘Go with the foot soldiers now. You turned me from a horse into a donkey; how will you get a horse again from a donkey? 22–26. New Testament (I AD) 22. Matt 26:15 Τότε πορευθεὶς εἷς τῶν δώδεκα ὁ λεγόμενος Ἰούδας Ἰσκαριώτης πρὸς τοὺς ἀρχιερεῖς εἶπεν, Τί θέλετέ μοι δοῦναι, κἀγὼ ὑμῖν παραδώσω αὐτόν; Then one of the Twelve called Ioudas Iskariotes went to the chief priests and said, ‘What will you give me and I will hand him over to you?’ 23. Mark 6:48 [Turner; Schrenk; Joüon, ‘Θέλειν’; Taylor; Bratcher and Nida; al.] καὶ ἰδὼν αὐτοὺς βασανιζομένους ἐν τῷ ἐλαύνειν, ἦν γὰρ ὁ ἄνεμος ἐναντίος αὐτοῖς, περὶ τετάρτην ϕυλακὴν τῆς νυκτὸς ἔρχεται πρὸς αὐτοὺς περιπατῶν

AUXILIARY THELO

249

ἐπὶ τῆς θαλάσσης· καὶ ἤθελεν παρελθεῖν αὐτούς. οἱ δὲ ἰδόντες αὐτὸν ἐπὶ τῆς θαλάσσης περιπατοῦντα ἔδοξαν ὅτι ϕάντασμά ἐστιν, καὶ ἀνέκραξαν· Seeing them struggling in their rowing, since the wind was against them, about the fourth watch of the night he came to them walking on the sea; and he was going to go past them, and when they saw him walking on the sea they thought it was a ghost, and cried out. See also John 6:21 [Riesenfeld; Schrenk] 24. John 1:43 [Riesenfeld; Schrenk; Joüon, ‘Θέλειν’] Τῇ ἐπαύριον ἠθέλησεν ἐξελθεῖν εἰς τὴν Γαλιλαίαν καὶ εὑρίσκει Φίλιππον. καὶ λέγει αὐτῷ ὁ Ἰησοῦς, Ἀκολούθει μοι. The next day he was about to go out into Galilee and he finds Philip. And Jesus says to him, ‘Follow me.’ 25. Acts 14:13 [Riesenfeld; Schrenk] ἐκάλουν τε τὸν Βαρναβᾶν Δία, τὸν δὲ Παῦλον Ἑρμῆν, ἐπειδὴ αὐτὸς ἦν ὁ ἡγούμενος τοῦ λόγου. ὅ τε ἱερεὺς τοῦ Διὸς τοῦ ὄντος πρὸ τῆς πόλεως ταύρους καὶ στέμματα ἐπὶ τοὺς πυλῶνας ἐνέγκας σὺν τοῖς ὄχλοις ἤθελεν θύειν. ἀκούσαντες δὲ οἱ ἀπόστολοι Βαρναβᾶς καὶ Παῦλος, διαρρήξαντες τὰ ἱμάτια αὐτῶν ἐξεπήδησαν εἰς τὸν ὄχλον κράζοντες καὶ λέγοντες, Ἄνδρες, τί ταῦτα ποιεῖτε; … καὶ ταῦτα λέγοντες μόλις κατέπαυσαν τοὺς ὄχλους τοῦ μὴ θύειν αὐτοῖς. They called Barnabas Zeus and Paul Hermes, since he was the leader in the speeches. The priest of Zeus Before the City brought bulls and garlands to the gateway and together with the crowd was about to offer sacrifice. The apostles Barnabas and Paul, when they heard of it, tore their clothes and leapt into the crowd crying out and saying, ‘Men, what is this you are doing? …’ With their words they barely stopped the crowd from offering sacrifice to them. 26. Acts 19:33 ἐκ δὲ τοῦ ὄχλου συνεβίβασαν Ἀλέξανδρον, προβαλόντων αὐτὸν τῶν Ἰουδαίων· ὁ δὲ Ἀλέξανδρος κατασείσας τὴν χεῖρα ἤθελεν ἀπολογεῖσθαι τῷ δήμῳ. ἐπιγνόντες δὲ ὅτι Ἰουδαῖός ἐστιν, ϕωνὴ ἐγένετο μία ἐκ πάντων ὡς ἐπὶ ὥρας δύο κραζόντων, Μεγάλη ἡ Ἄρτεμις Ἐϕεσίων. Some of the crowd instructed(?) Alexandros, whom the Jews put forward. Alexandros motioning with his hand was about to make a defence to the popular assembly. But when they realised he was a Jew, with one voice they all cried out for about two hours, ‘Great is Artemis of the Ephesians!’ 27. Test XII Patr, TReub 1.7 (ed. De Jonge) (c.50 AD?) [Sophocles; Jannaris; Lampe] λέγω γὰρ ὑμῖν, ὅτι ἐνέπληξέ με πληγὴν μεγάλην ἐν ταῖς λαγῶσί μου ἐπὶ μῆνας ἑπτά· καὶ εἰ μὴ Ἰακὼβ ὁ πατὴρ ἡμῶν προσηύξατο περὶ ἐμοῦ πρὸς Κύριον, ὅτι ἤθελε Κύριος ἀνελεῖν με.

250

COLLECTED ESSAYS ON THE GREEK BIBLE

I tell you that he struck me with a great affliction in my flanks over seven months, and if Jacob our father had not prayed for me to the Lord, [I tell you] that the Lord would have destroyed me.24 28. POxy 55.3806.7 (15 AD) [ἠ ἠ ]ἠ ωνίου ἀναπλέοντος ἀναγκαῖον ἔγνων ἀ[σπ]άǀ[σα]σθαί σε διὰ γραπτοῦ καὶ παρακαλέσσαι σε γράϕειν ǀ μοι περὶ ὧν ἐὰν θέλῃς. ἥδιστα γὰρ ποιήσωι. τῶι ǀ ἀδελ(ϕῷ) σου κατὰ ἑκάστην ἡμέραν παρεδρεύωι, ǀ μὴ θέλει ἐπιστολάς σοι πέμψαι. τὸ δῖγμα τοῦ ǀ [ἐ]ριδίου δῖξον Φιλοῦτι καὶ γράψον μοι ἠ (= εἰ) ἀρέσǀκει αὐτῆι ἢ οὔ. Since ...-onios is sailing up I decided I must greet you in writing and ask you to write to me about what you want, for I will gladly do it. I apply to/attend on your brother every day in case he will send you letters. Show Philous the specimen of the wool and write and tell me if she likes it or not. 29–31. LXX 29. Tobit 3:10 GII (II BC?) [Schrenk] ἐν τῇ ἡμέρᾳ ἐκείνῃ ἐλυπήθη ἐν τῇ ψυχῇ καὶ ἔκλαυσεν καὶ ἀναβᾶσα εἰς τὸ ὑπερῷον τοῦ πατρὸς αὐτῆς ἠθέλησεν ἀπάγξασθαι. καὶ πάλιν ἐλογίσατο καὶ λέγει Μήποτε ὀνειδίσωσιν τὸν πατέρα μου. . . . χρησιμώτερόν μοί ἐστιν μὴ ἀπάγξασθαι, ἀλλὰ δεηθῆναι τοῦ κυρίου ὅπως ἀποθάνω. On that day she was grieved in her soul and wept, and she went up to her father’s upper room and was going to hang herself. And she considered again and said, ‘Never let them reproach my father. … Better for me not to hang myself but to pray to the Lord that I might die.’ 30. Tobit 6:15 GII ἤκουσα ὅτι ἑπτὰ ἤδη ἐδόθη ἀνδράσιν, καὶ ἀπέθανον ἐν τοῖς νυμϕῶσιν αὐτῶν τὴν νύκτα, ὁπότε εἰσεπορεύοντο πρὸς αὐτήν, καὶ ἀπέθνῃσκον. καὶ ἤκουσα λεγόντων αὐτῶν ὅτι δαιμόνιον ἀποκτέννει αὐτούς. καὶ νῦν ϕοβοῦμαι ἐγώ – ὅτι αὐτὴν οὐκ ἀδικεῖ, ἀλλ᾽ ὃς ἂν θελήσῃ ἐγγίσαι αὐτῆς, ἀποκτέννει αὐτόν. I have heard that she has already been given in marriage to seven men and they died in their bridal chamber, on the night when they went in to her they would die. And I heard it said that a demon kills them. And now I am afraid, because [the demon] does not harm her, but whoever comes near/wants to come near her it kills. Cf. GI ὅτι δαιμόνιον ϕιλεῖ αὐτήν, ὃ οὐκ ἀδικεῖ οὐδένα πλὴν τῶν προσαγόντων αὐτῇ. 24

In their translation, Hollander and De Jonge, Testaments of the Twelve Patriarchs, supply ‘I would have died’ before the ὅτι clause, rendered ‘because the Lord wanted to kill me.’ Such a supplement is unnecessary if θέλω has the auxiliary use (which they may not have considered).

AUXILIARY THELO

251

31. Exod 2:14 (III BC) περιβλεψάμενος δὲ ὧδε καὶ ὧδε οὐχ ὁρᾷ οὐδένα, καὶ πατάξας τὸν Αἰγύπτιον ἔκρυψεν αὐτὸν ἐν τῇ ἄμμῳ. ἐξελθὼν δὲ τῇ ἡμέρᾳ τῇ δευτέρᾳ ὁρᾷ δύο ἄνδρας Ἐβραίους διαπληκτιζομένους καὶ λέγει τῷ ἀδικοῦντι Διὰ τί τύπτεις τὸν πλησίον; ὁ δὲ εἶπεν Τίς σε κατέστησεν ἄρχοντα καὶ δικαστὴν ἐϕ᾽ ἡμῶν; μὴ ἀνελεῖν με σὺ θέλεις, ὃν τρόπον ἀνεῖλες ἐχθὲς τὸν Αἰγύπτιον; He looked this way and that, and saw no one, so he struck the Egyptian and buried him in the sand. Going out the next day he saw two men, Hebrews, fighting. He said to the one hurting the other, ‘Why do you strike your neighbour?’ He said, ‘Who appointed you ruler and judge over us? Are you going to kill me the way you killed the Egyptian yesterday?’ 32. PHib 1.65.25 (c.265 BC) [Mayser] [θέλ]ομεν οὖν ἐγ δηǀ[μο]σίου τὸν λοιπὸν ǀ [συν]αγοράσαι σῖǀ[το]ν ἵνα μη[θὲ]ν ǀ [εἰς ἐ]μὲ ὑστερήσηι. [The writer has asked the recipient for help in supplying part of the grain he owes and explains how he will obtain the rest:] I want to/am going to purchase the remainder of the grain from the state, so that there might be no arrears against me. 33–35. Plato (IV BC) 33. [Plato], Alkib. I 122d [Wifstrand] τοῦτο μὲν γὰρ εἰ ἐθέλεις τοὺς Λακεδαιμονίων πλούτους ἰδεῖν, γνώσῃ ὅτι πολὺ τἀνθάδε τῶν ἐκεῖ ἐλλείπει. For in this matter, if you consider the wealth of the Lacedaemonians, you will recognise that things here are very much inferior to those there. 34. Protag. 334b [Wifstrand] … οἷον καὶ ἡ κόπρος, πάντων τῶν ϕυτῶν ταῖς μὲν ῥίζαις ἀγαθὸν παραβαλλομένη, εἰ δ᾽ ἐθέλοις ἐπὶ τοὺς πτόρθους καὶ τοὺς νέους κλῶνας ἐπιβάλλειν, πάντα ἀπόλλυσιν. … as for example dung, which when applied to the roots of any plants is a good thing, whereas if you were to put it on the young shoots and twigs it destroys them all. 35. Rep. 423b [LS7, LS8] Τίς, ἔϕη, ὅρος; Οἶμαι μέν, ἦν δ᾽ ἐγώ, τόνδε· μέχρι οὗ ἂν ἐθέλῃ αὐξομένη εἶναι μία, μέχρι τούτου αὔξειν, πέρα δὲ μή. ‘What measure?’ he said. ‘In my opinion this one,’ I said, ‘To the point where [the city], while increasing in size, will still be one, to that point will [the guardians] let it grow, and no further.’ See also Pl., Theaet. 162e; Parm. 158c; Alkib. I 122b, c; Charm. 174c; Lysis 217a, c; Meno 71a; Hipp. Mi. 373a; Rep. 581c [all Wifstrand]; 436b [LS7, LS8].

252

COLLECTED ESSAYS ON THE GREEK BIBLE

36. Aristophanes, Wasps 536 (422 BC) [LS7, LS8; Jannaris; LSJ]25 ὁρᾷς γὰρ ὥς σοι μέγας ἐστὶν ἁγὼν καὶ περὶ τῶν πάντων, εἴπερ – ὃ μὴ γένοιτο – νῦν ἐθέλει κρατῆσαι. [Chorus to Bdelykleon:] You see how great the contest is for you and how everything is at stake, if he wins now – which I hope won’t happen. 37. Antiphon 4.2.7 (V BC) [Rödiger] ὡς μὲν οὖν οὐ δικαίως κατηγοροῦμαι, ἐπιδέδεικταί μοι· ἐθέλω δὲ τοὺς κατηγοροῦντάς μου πᾶσιν οἷς ἐγκαλοῦσιν ἐνόχους αὐτοὺς ὄντας ἀποδεῖξαι. That I am unjustly accused I have demonstrated; but I will show that my accusers are themselves liable to all the accusations they bring against me. 38–40. Herodotus (V BC) 38. 1.109.4 [LSJ] εἰ δ᾽ ἐθέλει τούτου τελευτήσαντος ἐς τὴν θυγατέρα ταύτην ἀναβῆναι ἡ τυραννίς, τῆς νῦν τὸν υἱὸν κτείνει δι᾽ ἐμεῦ, ἄλλο τι ἢ λείπεται τὸ ἐνθεῦτεν ἐμοὶ κινδύνων ὁ μέγιστος; If on his death the sovereignty passes to this daughter, whose son he is now killing by my hand, am I not hereafter in extreme danger? 39. 2.14.1 [Waddell] εἴ σϕι θέλοι, ὡς καὶ πρότερον εἶπον, ἡ χώρη ἡ ἔνερθε Μέμϕιος (αὕτη γάρ ἐστι ἡ αὐξανομένη) κατὰ λόγον τοῦ παροιχομένου χρόνου ἐς ὕψος αὐξάνεσθαι, ἄλλο τι ἢ οἱ ταύτῃ οἰκέοντες Αἰγυπτίων πεινήσουσι . . . ; If, as I said before, the land below Memphis (this is the part that is increasing) should increase in height at the same rate as in the past, isn’t it inevitable that the Egyptians who live there will go hungry . . . ? 40. 9.89.2 ὁ δὲ ᾽Αρτάβαζος γνοὺς ὅτι, εἰ ἐθέλει σϕι πᾶσαν τὴν ἀληθείην τῶν ἀγώνων εἰπεῖν, αὐτός τε κινδυνεύσει ἀπολέσθαι καὶ ὁ μετ᾽ αὐτοῦ στρατός. … Artabazos, realising that if he were to tell them the whole truth about the battles, he and his army would be in danger of destruction, … See also Hdt. 1.32.3; 2.11.4 [LSJ]; 2.99.3; 3.12.1; 7.10d.2 [Rödiger]; 7.49.4 [LSJ].

25

The source lies earlier: this example is noted in the Paris ed. (1831–65) of Stephanus’s Thesaurus.

AUXILIARY THELO

253

41. Sophocles, Ant. 1040 (V BC) [Rödiger] τάϕῳ δ᾽ ἐκεῖνον οὐχὶ κρύψετε, οὐδ᾽ εἰ θέλουσ᾽ οἱ Ζηνὸς αἰετοὶ βορὰν ϕέρειν νιν ἁρπάζοντες ἐς Διὸς θρόνους [Creon to Teiresias:] You shall not bury that man, not even if Zeus’s eagles seize him as food and carry him to Zeus’s throne. 42. Aeschylus, Cho. 851 (458 BC) [Rödiger] Χο. ἠκούσαμεν μέν, πυνθάνου δὲ τῶν ξένων ἔσω παρελθών. οὐδὲν ἀγγέλων σθένος ὡς αὐτὸν αὐτῶν ἄνδρα πεύθεσθαι πάρα. Αἰ. ἰδεῖν ἐλέγξαι τ᾽ εὖ θέλω τὸν ἄγγελον, εἴτ᾽ αὐτὸς ἦν θνήισκοντος ἐγγύθεν παρών, εἴτ᾽ ἐξ ἀμαυρᾶς κληδόνος λέγει μαθών· Chorus: We heard the story; but you pass inside and learn [it] from the strangers. There is no strength in messengers’ reports like learning in person from them. Aegisthus: I will see and examine well the messenger, whether he himself was present close by when [Orestes] died, or tells by learning from an uncertain report. See also Sophocles, OC 1291 (end of V BC); Euripides, Or. 770 (408 BC); Aeschylus, Cho. 904 [all Rödiger].

6.2 θέλησον, θελήσατε, ‘(please) do’ PAnt 3.197.2 (VII AD) θελήσατε στρῶσαι τῷ παρόντ(ι) ἀνθ(ρώπῳ) ǀ ἀνερχο(μένῳ) ἱππάρ(ια) τρία. Please saddle three ponies for this man who comes up. PCairMasp 1.67061.1 (VI AD) θελήσατε οὖν παραχρῆμ[α τοῦτο] ǀ ποιῆσαι. POxy 10.1300.5, 8, 9 (V AD) θέλησον οὖν, κύρα μήτηρ, ἀποστῖλε … θέλησον οὖν, κύρα μήτηρ, ἀγοǀράσε … καὶ θέλ[η]σον λαβῖν τὸ κοῦκλιν. POxy 16.1941.5 (V AD) θέληǀσον ἀποστῆναι τῆς γεωρǀγίας μηχανῆς Στύμονος. SB 6.9158.20 (V AD) θέλησον δὲ ἐν τάǀχι ἐξαποστῖλαι ǀ τὸ σημῖον ǀ μὴ ὀκνουμένη, ǀ ἀμμή, οὐκ ἔχοǀμεν εἰς ἡπερεǀσίαν (= ὑπηρ-). Do send the signet quickly without delay, mama: we don’t have one to use. BGU 3.948.11, 13, 18 (IV–V AD) θέλησον οὖν πέμψεν μοι … θέλησον οὖν ǀ ποιῆσεν μοι … θέλησον … ἀγοράσιν (= -ειν) ǀ μοι. POxy 14.1776.6 (late IV AD) θέλησον ὅσον ǀ χρῄζουσι ἔν τε σίτῳ ἢ ǀ ὄξους παράσχου αὐτοῖς. Aesopi Fab. 40 (ed. Chambry) θέλησον οὖν τοὺς ἐμπροσθίους πόδας ἐρεῖσαι τῷ τοίχῳ, ὀρθῶσαι δὲ τὰ κέρατα, ἀναδραμοῦσα δὲ ἐγὼ καὶ σὲ ἀνασπάσω.

254

COLLECTED ESSAYS ON THE GREEK BIBLE

Just rest your front feet against the wall and hold your horns up, and I will run up [out of the well] and pull you up too.

6.3 μὴ θελήσῃς, μὴ θελήσητε, ‘(please) don’t’ POxy 59.4005.1 (VI AD) μὴ θελήσῃς μηδεμίαν ϕροντίδαν ἔχειν περὶ τῆς ἡμῶν ἀδελϕῆς Μαρίας, σοῦ δὲ ǀ συμβίου. οὐκ ἐοῦμεν (= ἐῶμεν) γάρ, τοῦ θεοῦ θέλοντος, αὐτὴν δεηθῆναί τινος. Don’t have any concern about our sister Maria, your wife. For we will not, God willing, allow her to want for anything. Acta Conc. Ephes. 1.1.2, p. 68 l. 14 (ed. Schwartz) (431 AD) Εἰ θέλετε ἀκοῦσαι, ἡσυχάσατε καὶ μανθάνετε· μὴ θελήσητε ἐμποδίσαι τοῖς λεγομένοις, ἀλλὰ μακροθυμήσατε, ἵνα τὰ ῥήματα ἀκριβῶς ἀκούσητε. If you are willing to hear, keep quiet and learn; do not obstruct what is being said, but be patient. … PAmh 2.143.16 (IV AD) μὴ θελήǀσῃς οὖν, κύριε, μῖνε ἐκǀτὸς ἡμῶν αὔριον διὰ τὴν ǀ ἀϕορμὴν τοῦ ὕδατος ǀ εἵνα δηνηθῶμεν ǀ ποτίσαι τ[ὸ]ν μέγαν κλῆǀρον. Don’t stay away from us tomorrow, sir, for the release of the water, so that we can irrigate the large allotment. PWisc 2.74.5, 20 (III–IV AD) καὶ πάνοι (= πάνυ) χρίαν ἔχω{.ι}μεν τῇ παρουσίᾳ σου. μὴ θεǀλήσῃς οὖν παραμῖναι παρά σοι τοῦ σαι ἀπαντῆǀσαι πρὸς ὑμᾶς καὶ ἀπολέσωμεν τὴν ǀ οἰκίαν ἡμῶν. … μὴ θέλῃς ἀμαιλῆσαι καὶ ὕστερα μεταμελή‹σ›ῃς. PFlor 2.210.9 (255 AD) ἀλλὰ μὴ θελήσῃς μοι πέμǀψε ὄξος· καὶ γὰρ ἐντολὰς ἔσχον ǀ διὰ τοῦ ἐπιστολιδίου παρὰ τοῦ ǀ ε[ὐσ]χήμονος. Tobit 4:5 GI, simil. GII πάσας τὰς ἡμέρας, παιδίον, κυρίου τοῦ θεοῦ ἡμῶν μνημόνευε καὶ μὴ θελήσῃς ἁμαρτάνειν καὶ παραβῆναι τὰς ἐντολὰς αὐτοῦ·

Postscript In Markopoulos, The Future in Greek, the development of θέλω is discussed several times (40–2, 76–7, 80–1, 104–12, 164–77) and examples of auxiliary θέλω are noted. Our collections overlap: some of his examples are missing from mine, and some of mine from his. A consolidated collection would be useful. Markopoulos discusses θέλω in requests at 81–2, with PWisc 2.74.20 and POxy 14.1776.6 (quoted above) as examples, and at 110, with PPrag 2.193.2 (V AD) θέλησον, εἴπερ βούλει, ἐκμισθῶσαι αὐτὸ Καναίῳ τῷ ἁλιεῖ. Another relevant study is Tronci, ‘Future Forms in Postclassical Greek,’ with a discussion of ‘future-oriented’ periphrases using μέλλω and θέλω in the LXX and NT (133–7), including nos. 22, 29–31 above. Tronci takes the LXX exx. to be ‘ingressive,’ as allegedly implied in my translations, concluding that ‘the

AUXILIARY THELO

255

periphrases with θέλω … can never be translated as plain futures in the LXX’ (134). Her conclusion is similar for the NT (137). La Roi, however, disagrees about Exod 2:14, arguing that ‘this example is best interpreted as a future use in this context’ (‘Diachronic Analysis,’ 222). In observing the use of θέλω in polite requests (§4, examples at §6.2, 3), I recorded only aorist imperatives and did not notice that the present imperative can be used in the same way (‘please do/don’t). Aitken’s discussion in ‘Literary Attainment,’ 116–7 alerted me to this. He records two cases in Sirach (Sir 6:35 θέλε; 7:13 μὴ θέλε) and two in non-biblical sources (Vita Aesopi [‘P, 247’]26 μὴ θέλε; Life of Adam and Eve 31 μὴ θέλε), with the remark that these are ‘among many koine examples’ (117). A search in brings to light this nice example: BGU 4.1205.7 (28 BC) κομ|[ψῶ]ς ἐμοὶ θέλετε τὴν αἰτίαν | γράψαι το[ῦ] φακοῦ καὶ πίσον | [. .] οὐδὲ γὰρ σύνστοιχοι ἑατῶν | [γ]ίνεσθε γεγραφηκότες Πα|[νίσ]κωι, ὅτι πεπράκαμεν, ὅτε | [. . .] οὐ πεπράκαμεν. Do kindly write to me the reason for the lentils and peas … for you are not even consistent with yourselves, having written to Paniskos that we sold (them), when [you know?] we have not sold (them).

26

This is a ref. to ‘Vita Planud.’ (diff. || Vita G 32.7). I have doubts about this one.

17 RELEASING LSJ FROM ITS PAST 20101

Abstract What is the future of Liddell and Scott’s lexicon to be? That was the main question posed for the Cambridge Colloquium in 2002. This paper addresses the topic by examining the history of the entry for one word, ἀγαπητός, from the first edition of the lexicon in 1843 to LSJ. The treatment of the word in NT lexicons and commentators is part of the story. Six characteristics of the tradition emerge from the analysis, most of them inimical to satisfactory lexicographical practice. The paper moves on to propose a new approach, utilising the potential of an electronic database. The possible content of such a database is outlined and a revised analysis of ἀγαπητός is offered.

INTRODUCTION ‘Liddell and Scott’ is an institution, the great lexicon of Greek not only familiar to every scholar and student of Greek and Latin, but known to the world at large, if only by name. For more than a century and a half it has been the chief lexical resource for ancient Greek in the Englishspeaking world, and far outside it. The first edition, prepared by the two young Oxford scholars on the basis of the lexicon of Franz Passow,2 was published in 1843. After a succession of editions, each incorporating new material and revisions, it reached an eighth in 1897. A more extensive revision was undertaken by Stuart-Jones and McKenzie to produce the current ‘new edition’ or ninth, issued in parts from 1925 to 1940. Numerous ‘Addenda et Corrigenda’ had already accumulated at the end of the last part. There followed a supplement in 1968, and another in 1996, gathering, and sometimes correcting, all the earlier additional material and adding more. Offspring of the larger lexicon were also produced, the medium-sized 1

2

This is an expanded version of a paper delivered at the Colloquium on Ancient Greek Lexicography held in Cambridge, 7th–9th July, 2002. I am grateful to Anne Thompson for reading it and suggesting a number of improvements. Passow, Handwörterbuch (1831).

258

COLLECTED ESSAYS ON THE GREEK BIBLE

‘Intermediate’ in 1889, based on the 7th edition, and the smaller ‘Abridged’ in 1843, followed by a 2nd edition in 1846. There was also a little-known anonymous derivative in 1850, the ‘Copious Greek–English Vocabulary.’3 Now at the beginning of the twenty-first century the great lexicon has reached a decisive stage in its history. It is evident that the work cannot stand still and remain as it is, but must move forward. It already requires further revision, not only because of the build-up of new material, but because of the pressing need for correction of many of the existing entries. That the latter is true is not readily conceded by all, especially those who have not moved beyond implicit trust in what a lexicon tells them. But the fact is well known to those who have had occasion to study LSJ’s entries closely, and has often been demonstrated, in recent years especially by John Chadwick.4 What is LSJ’s future to be? This paper is designed to address that question and consider what part an electronic database might play in the process. In the first half of this paper I focus on LSJ’s entry for one word, not primarily to show that LSJ have got something wrong – that is easy to do for any number of entries – but to examine the history of the entry. This will enable us to draw out the reasons why the entry is now as it is, namely, in a state of disarray. Its present state will be seen to be as much a result of its history, and the methods and constraints imposed on the lexicographers as they worked on it, as of poor judgement and scholarship on their part. I will go on to suggest what will be required to cure the problems in the treatment of this word, and by extension all the words in the lexicon, and to argue that these processes will be best carried out in an entirely new medium. By this means LSJ will be enabled to escape the mistakes and pressures of the past and go on to a new (and better) stage in its life. THE HISTORY OF AN

ENTRY

The word I have chosen as an illustration is ἀγαπητός. I have recently had occasion to examine this word in a study of the history of lexicons of the NT, a history lasting five hundred years.5 Its treatment there connects directly with the way it has been treated in the main lexicons of 3

4

5

Details of the editions of the larger LS(J) may be seen in Lee, History, 355–6. The compiler of the ‘Copious’ was George Marshall, as also of the ‘Abridged.’ I thank Chris Stray for this information. Chadwick, ‘Semantic History’; idem, ‘Replacing Liddell and Scott’; idem, Lexicographica Graeca. See Lee, History.

RELEASING LSJ FROM ITS PAST

259

ancient Greek, and in a special way with LSJ itself, as we shall see. Let me begin with some remarks on the semantic history of this word as I understand it. The point of most interest is the supposed meaning ‘only,’ whether alone or in combination with some other notion, to give meanings like ‘only-beloved, only-begotten.’ Current NT lexicons put forward such a meaning in a number of NT occurrences and adduce parallels in other texts from Homer onwards. The Septuagint is also supposed to offer support. I have set out in detail elsewhere the case for rejecting ‘only’ as a false meaning.6 It can be traced back through a long line of commentators and lexicon entries, with a few doubters along the way. Its leading proponent in the twentieth century was the Oxford NT scholar C. H. Turner, who argued the case fully in a 1926 article.7 The ultimate source turns out to be a Homeric gloss (on Il. 6.401), preserved in Hesychius and elsewhere. Scrutiny shows that neither this nor any of the other alleged examples provides a sound basis for recognising such a meaning. I am not alone in my scepticism: Chadwick also rejected this meaning, though he did not pursue its history in LS and elsewhere.8 In what follows I trace the progress of this meaning into LSJ and its effect on the shape of the entry for ἀγαπητός. The line of semantic development in ἀγαπητός seems clear. The primary meaning is ‘beloved.’ From ‘beloved’ come weakenings to ‘likable,’ and ‘to be acquiesced in’; the first sense often has an intensified nuance like ‘especially loved, precious.’ The pattern is parallel to that in the verb ἀγαπάω and in its equivalent στέργω, i.e., ‘love’ > ‘feel affection for’ > ‘be content with.’9 There is no meaning ‘only.’ Still less can it be the primary meaning, as Turner would have us believe. Let us turn to the entries at four points in the history of Liddell and Scott’s lexicon. We begin with LS1 (1843): ἀγαπητός, ή, όν, verb. Adj. from ἀγαπάω, beloved, Hom.; μοῦνος ἐὼν ἀγαπητός (the only, and so) doubly beloved son, Od. 2. 365: longed for, desired, Pind. N. 8. 6: worthy of love, 6 7 8 9

Lee, History, 193–211. Turner, ‘Ὁ υἱός μου.’ Chadwick, Lexicographica Graeca, 32–4. The early semantic history of ἀγαπάω is complicated by the presence of ἀγαπάζω alongside it: both are found in Homer, ἀγαπάζω only in the meaning ‘treat/receive with affection,’ ἀγαπάω once in that sense and once as ‘be content’ (with/that). The problem of how these denominatives (as they would normally be) could be formed before the appearance of ἀγάπη (earliest attested c.500 BC, as a name) remains unsolved. Cf. Chadwick, Lexicographica Graeca, 32.

260

COLLECTED ESSAYS ON THE GREEK BIBLE

lovable, Plat. Alc. 1. 131 E. II. neut. ἀγαπητόν (ἐστι), one must be content, εἰ–, ἐάν–, Plat. and Arist. Adv. -τῶς, lovingly, readily, contentedly, Plat.; ἀγαπητῶς ἔχειν, to be contented, like ἀγαπᾶν : – but in Att. prose, so as only just to content one, i. e. only just, barely, scarcely, = μόλις, Heind. Plat. Lys. 218 C; ἀγαπητῶς σωθείς Dem., cf. Meineke Menand. p. 108.

There is little to disagree with here; at any rate, the general structure is right. Beloved is primary and the other senses are developments of that. ‘Only’ is far from being put forward as a meaning: it appears as the equivalent of μοῦνος in Od. 2.365, where ἀγαπητός is rightly treated as beloved. The suggestion that it might in fact mean doubly beloved in this sort of context seems to me to have merit. (This was derived from Passow.)10 What ἀγαπητός is most unlikely to mean in Od. 2.365 is ‘only.’ Yet this is exactly the meaning Turner wishes to give it, apparently on the basis that it refers to an only child. But as Chadwick says, ‘the epithet μοῦνος would be otiose, if ἀγαπητός meant the same. It is evident that [the phrase] means “being greatly loved because you are an only son”.’11 Next let us go to LS4 (1855), where things begin to go wrong: ἀγαπητός, ή, όν, verb. Adj., beloved, μοῦνος ἐὼν ἀγαπητός the only (and so doubly) beloved son, Od. 2. 365; hence without μοῦνος, of an only son, Ἑκτορίδην ἀγαπητόν Il. 6. 401, cf. Od. 4. 817; so in Att., Νικήρατος . . ὁ τοῦ Νικίου ἀγ. υἱός Dem. 567. 24; cf. Arist. Rhet. 1. 7, 41. 2. desirable, delightful, ἀγαπατά (ἐστι), c. inf., Pind. N. 8. 6: – freq. in Att. Prose, worthy of love, loveable, Plat. Alc. 1. 131 E, etc. II. to be acquiesced in (as the least in a choice of evils), Andoc. 26. 15: – hence, ἀγαπητόν (ἐστι) one must be content, εἰ . . , ἐάν . . , Plat. Prot. 328 A, Xen. Oec. 8. 16, Dem. 302. 1, Superl., -ότατος, Plat. Phil. 61 E. So Adv. -τῶς, readily, contentedly, Plat. Legg. 735 D, etc.; ἀγαπητῶς ἔχειν, to be contented, like ἀγαπᾶν: – but also in Att. Prose, so as only just to content one, i.e. only just, barely, scarcely, = μόλις, Heind. Plat. Lys. 218 C: ἀγαπητῶς σωθῆναι Lys. 107. 16, so, ἀγαπητόν, Menand. μέθη 1.

The general structure is still much the same, with beloved as the startingpoint. There are additional examples here and there, and an improvement in the definition of sense II. But in ‘hence without μοῦνος, of an only son’ LS not only make a false deduction, but confuse reference and lexical meaning. ‘Of an only son’ would be a correct statement, i.e., that the word 10 11

[Passow4 (1831): ‘der einzige, darum doppelt geliebte Sohn.’] Chadwick, Lexicographica Graeca, 33.

RELEASING LSJ FROM ITS PAST

261

refers to an only son; but by italicising ‘only’ they turn only into the lexical meaning. Now they have set running, almost by accident, the meaning only, in addition to beloved. Do they really mean that the examples that follow mean nothing more than only? Possibly they do, but it is not quite clear, and one wonders if they have thought through the consequences of writing ‘of an only son.’ Now let us see what shape the entry had reached in LS8 (1897), before the major revision for the ninth edition was undertaken: ἀγαπητός, ή, όν, Dor. -ατός, ά, όν, verb. Adj., beloved, μοῦνος ἐὼν ἀγαπητός the only dearly beloved son, Od. 2. 365; more commonly without μοῦνος, of an only son, Ἑκτορίδην ἀγαπητόν Il. 6. 401, cf. Od. 4. 817; so in Att., Ar. Thesm. 761; Νικήρατος . . ὁ τοῦ Νικίου ἀγ. παῖς Dem. 567. 24, cf. Arist. Rhet. 1. 7, 41, al.; Comically, δαπίδιον ἓν ἀγ. Hipparch. Ἀνασ. 1. II. of things, worthy of love, loveable, desirable, dear, Plat. Alc. 1. 131 E, etc.; Sup. -ότατος Id. Phil. 61 E; τὸ ἀγαπητόν an object of desire, Arist. Rhet. 1. 7, 41, al. 2. to be acquiesced in (as the least in a choice of evils), Andoc. 26. 15: – hence, ἀγαπητόν [ἐστι] one must be content, εἰ . . , ἐάν . . , Plat. Prot. 328 A, Xen. Oec. 8. 16, Dem. 302. 1, Arist., etc.; c. inf. Eth. N. 9. 10, 6. III. Adv. -τῶς, readily, gladly, contentedly, Plat. Legg. 735 D, Dem. 409. 7, etc. 2. to one’s heart’s content, Diphil. Incert. 4. 3. just enough to content one, only just, barely, scarcely, = μόλις, Plat. Lys. 218 C; ἀγαπητῶς σωθῆναι Lys. 107. 16; so also, ἀγαπητόν Menand. Μέθη 1.

There has been more tinkering since the fourth edition, most of it not beneficial. Od. 2.365 is now clearly (and rightly enough) stated to mean dearly beloved. But ‘of an only son’ for the next group of examples survives unscathed, with the minor change of ‘hence without μοῦνος’ to ‘more commonly without μοῦνος,’ presumably because two more examples have been added. Other changes are serious missteps. The new position of ‘II’ separates what was sense [I.]2 desirable . . . worthy of love, etc. into a new major category away from where it belongs with [I.] beloved, and to be acquiesced in is demoted to a subcategory of that, surely not an improvement. Worse, ‘of things’ has been inexplicably introduced ahead of Pl., Alc. 1.131e in II.1, which it will be useful to see at once: οὔτ᾽ ἐγένεθ᾽, ὡς ἔοικεν, Ἀλκιβιάδῃ τῷ Κλεινίου ἐραστὴς οὔτ᾽ ἔστιν ἀλλ᾽ ἢ εἷς μόνος, καὶ οὗτος ἀγαπητός, Σωκράτης ὁ Σωφρονίσκου καὶ Φαιναρέτης. Alkibiades son of Kleinias, it seems, neither had nor has any lover except one only, and that very much loved, namely Socrates, son of Sophroniskos and Phainarete.

262

COLLECTED ESSAYS ON THE GREEK BIBLE

Socrates, it seems, is a thing. What caused this lapse we cannot say; what we can say is that making any change to a sense without checking the examples cited under it is fraught with risk. Finally we come to the result of the revisions made to produce the ninth edition, the one we rely on today. LSJ (1925–40) reads as follows: ἀγαπ- … … … … … … … … ητός, ή, όν, Dor. -ατός, ά, όν, that wherewith one must be content (cf. ἀγαπάω III), hence of only children, μοῦνος ἐὼν ἀ. Od.2.365; Ἑκτορίδην ἀ. Il.6.401, cf. Od.4.817, Sapph. 85, Ar. Th.761, Pl.Alc.1.131e; Νικήρατος . . ὁ τοῦ Νικίου ἀγ. παῖς D.21. 165, cf. Arist.Pol.1262b23, EE 1233b2; αὕτη μονογενὴς αὐτῷ ἀγαπητή LXX Jd.11.34, cf. To.3.10, Ev.Marc.12.6 (but cf. also II.2), etc.; ἀγαπητός· μονογενής, Hsch.: – so of things, Arist.Rh.1365b16; δαπίδιον ἓν ἀ. Hipparch.Com.1; προβάτιον Men.319.3. 2. to be acquiesced in (as the least in a choice of evils), And.3.22, J.BJ 5.10.3: – hence, ἀγαπητόν [ἐστι] one must be content, εἰ . . , ἐάν . . Pl.Prt.328b, X.Oec. 8.16, D.18.220, Arist.Metaph.1076a15, etc.; c. inf., EN 1171a20. II. of things, desirable, ἤθη X.Mem.3.10.5; βίος Pl.Phlb.61e (Sup.). 2. of persons, beloved, ἀδελφὲ ἀγαπητέ LXX To.3.10: in letters, as a term of address, Ep.Rom.12.19, cf. PGrenf.2.73, etc. III. Adv. -τῶς, gladly, contentedly, Pl.Lg.735d, D.19.219, etc. 2. just enough to content one, barely, scarcely, Pl.Lys.218c; ἀ. σωθῆναι Lys. 6.45, cf. Diph.89.2, etc.

The semantic structure of the entry, and with it the presumed semantic history of the word, has been reversed. The meaning previously taken to be primary, beloved, is relegated to sense-category II. Here, apparently, it begins as a weaker sense desirable applied to things, before developing to beloved, which is allegedly not found before the book of Tobit in the LXX. As the primary meaning we now have the impressive that wherewith one must be content. This seems a carefully crafted and insightful definition. That was certainly how it struck me as an undergraduate when I acquired (with some excitement) my own copy of the big LSJ and saw this entry. But mature reflection suggests otherwise. And I can now trace where it originated – in a source outside LSJ. To begin with, that wherewith one must be content is the definition of a noun, not an adjective. A definition in this form is unjustified and unhelpful. The more serious question is, is this meaning really intended to apply to all the examples that are listed in [I.1]? Since no other meaning is stated, and ‘hence of only children’ indicates reference only, not lexical meaning, we must suppose it is. Let us take Od. 2.365. Do LSJ really mean that Eurykleia says to Telemakhos, ‘you being an only son, one with whom his parents must be content’? Or again that Pl., Alc. 1.131e (above) . . . εἷς μόνος, καὶ οὗτος ἀγαπητός means ‘Alkibiades had one lover only,

RELEASING LSJ FROM ITS PAST

263

one that he had to be content with’? Or that Il. 6.401 παῖδ᾽ . . . Ἑκτορίδην ἀγαπητόν can be tortured into meaning ‘the boy, Hektor’s son, the one he had to be satisfied with’? It seems unlikely.12 The fact is that it is difficult or impossible to make the meaning that wherewith one must be content work in the examples cited (with one possible exception, Men. 319.3 = 224.3 Kock-Austin: cf. the lexical analysis below). It appears rather that LSJ intend this as the base meaning from which sense [I.1] is supposed to derive. But what is that sense? They have failed to say. Chadwick thought they did not mean to say it was ‘only.’13 I strongly suspect they did. The Hesychius gloss quoted is a good clue; and we know they were working from LS8, with its ‘of an only son,’ followed by many of the same examples. Perhaps they just forgot to italicise ‘only.’ But it is very unsatisfactory that the user of the lexicon is reduced to guessing what meaning is being stated.14 Some further details. The reference to ‘To. 3.10’ appears twice, in [I.1] and II.2, the first preceded by ‘cf.,’ an unhelpful and irritating device which leaves the reader guessing, and may be suspected to cover up uncertainty or indecision on the part of the editor. Are we meant to suppose that ἀδελφὲ ἀγαπητέ might possibly mean ‘brother I have to accept,’ or even ‘only brother,’ as well as ‘beloved brother’? It gets worse. The reference in II.2 to To. 3.10 ἀδελφὲ ἀγαπητέ, the alleged first instance of the meaning beloved, is wrong: Tob 3:10 (GII text) reads μία σοι ὑπῆρχεν θυγάτηρ ἀγαπητή. It is in Tob 10:12 (GI text) where ἄδελφε ἀγαπητέ is found. Clearly the reference in II.2 should be changed to 10:12; but what of the other in [I.1]? Should it stay as it is or be changed also? The separation of meaning [I.]2 to be acquiesced in . . . hence . . . one must be content from the first meaning was justified, indeed required, in the structure that prevailed in preceding editions. But to retain it as a separate category after introducing as sense [1.1] that wherewith one must be content is asking for trouble: the two overlap in meaning and therefore a reassessment of the second definition and its examples is called for. The point in all this is that not only are LSJ’s meanings unsatisfactory, but the placing of the examples is in disarray. We can see that this has 12

13 14

LSJ Suppl. (1996) adds two more to this group, ‘Ge. 22.2, cf. Am. 8.10.’ Again, it is impossible to see how the definition can apply. Rather, the editor seems to be thinking of the meaning ‘only.’ On these two examples see further Lee, History, 203–6. Chadwick, Lexicographica Graeca, 33. It is curious that in Lampe, PGL, s.v. ἀγαπητός, LSJ is quoted thus: ‘(cf. LS: with which one must be content, hence of only children),’ i.e., with only italicised and the definition adjusted to adjectival form.

264

COLLECTED ESSAYS ON THE GREEK BIBLE

come about through the process of revision, with its attendant difficulties. A reversal of the semantic structure along with the introduction of a new meaning is a particularly hazardous operation to carry out. All the citations in the entry need to be checked if it is to be done without introducing mistakes. One cannot simply strike out one definition and insert another and assume that all will be well. I said I could trace the origin of that wherewith one must be content. C. H. Turner, as already mentioned, published his influential article in January 1926, and in it he refers to part one of LSJ, which had just been published in 1925. Turner in fact uses the new shape of LSJ’s entry to bolster his case for taking υἱὸς ἀγαπητός to mean ‘only son’: It is most instructive here to compare the new edition of Liddell and Scott – the first part is happily available for the subject of this lecture – with the previous one, and to note the change in the treatment of the word.

He goes on to extract some of LSJ’s Classical examples and discuss them, before drawing the conclusion that the assertion may safely be hazarded that when ἀγαπητός is used in connexion with υἱός, θυγατήρ [accent sic], παῖς, or similar words, no Greek of pre-Christian times would have hesitated in understanding it of an ‘only child,’ or would for a moment have thought of any other meaning as possible.15

Apart from offering the latter highly questionable statement, Turner has been a little less than candid with us. Before the 1926 paper, there appeared in 1919 an anonymous article containing the same argument and much of the same material as in 1926. There is no clue to the author’s identity in the 1919 paper; but a footnote in the 1926 paper contains a clear indication that it was Turner.16 In the 1919 paper we find the following statement: With regard to pre-Christian usage, it may be noted that in classical Greek the word ἀγάπη is unknown . . . and that ἀγαπάω means mainly ‘to be contented with’ . . . : ἀγαπητός would therefore properly be ‘what one has to be contented with,’ and so ‘all that one has,’ and then finally ‘the exclusive object of interest or affection.’17 15 16

17

Turner, ‘Ὁ υἱός μου,’ 115, 117. Turner, ‘Ὁ υἱός μου,’ 125 n. 1: ‘Most of the patristic passages have already been collected in an article on the word ἀγαπητός which appeared in J.T.S. xx 339–344 (1919) as a preliminary draft for the Lexicon of Patristic Greek. I should like to acknowledge here the help given me on the classical side on that occasion by my colleague Prof. J. A. Smith and by the late Charles Cannan.’ Turner, ‘Ἀγαπητός,’ 339 n. 2. The statement that ‘in classical Greek . . . ἀγαπάω means mainly “to be contented with”’ is wide of the mark. See the survey of usage of Classical

RELEASING LSJ FROM ITS PAST

265

So Turner had already proposed ‘what one has to be contented with.’ It is extremely likely, then, that Turner was the source, by whatever channel, of LSJ’s new definition for sense I.18 Thus Turner himself had a significant influence on the entry which he then used to support his case. The interest of all this, apart from the tale itself, is in being able to rediscover, and thus assess, the origin of this major shift in the LSJ entry. The revisers of LS8 appear to have relied, with insufficient caution, on a NT scholar’s enthusiastic advocacy of a meaning that he had set his sights on for the sake of its exegetical value in the NT, but that needed more objective assessment by someone less involved.19 CHARACTERISTICS OF THE

TRADITION

One could go on unpacking the detail, but that is quite enough. LSJ’s entry for ἀγαπητός is a mess. Why? One might simply answer that the editors of successive editions have not done their job properly; but that only touches the surface. There are deeper reasons, most of them beyond the control of the editors. They lie in the characteristics of the tradition, both methodological and practical. These have helped to bring about, indeed ensured, the result we now have. Let me summarise these features as I see them. (1) Control of the evidence. By ‘evidence’ I mean the raw data for the lexicographer’s task, the occurrences of the word being treated; by ‘control’ I mean gathering and assessing them adequately. How many occurrences of ἀγαπητός are there? LS1 cites 3 (not counting broad categories like ‘Hom.’); by LSJ there are 27 (not counting ‘etc.’). In DGE there are about 70. Electronic searching via the TLG and PHI7 reveals a tally of more than 4,000 examples of this word in literature up to 1453 and in documentary texts.20 No lexicographer of the past could hope to find all these and

18

19

20

authors in Joly, Vocabulaire de l’amour, 14–9. In Isocrates, for example, there are 50 occurrences, of which 38 show the sense ‘love’; in Plato there are 10, of which 8 have that sense; in Demosthenes 27, of which 12 (all figures from Joly). The likely channel, as Chris Stray suggests to me, was Cannan, who was Secretary to the Delegates of OUP (1898–1919) and whose help is acknowledged by Turner (see n. 16). Patrick James suggests that Alexander Souter may have had a hand in it too. Cf. mention of him in the LSJ preface (p. x) and his short note (Souter, ‘ Ἀγαπητός’) supporting Turner: cf. Lee, History, 207 n. 1. Further critique of Turner’s methods and alleged examples of the meaning ‘only’ can be seen in Lee, History, 197–202 and Chadwick, Lexicographica Graeca, 32–4. Chadwick (ibid. 32) already implied, somewhat cryptically, that Turner might have been behind the change in LSJ, but did not elaborate. PHI7 = The Packard Humanities Institute, CD ROM 7 Greek Documentary Texts (1991– 1996).

266

COLLECTED ESSAYS ON THE GREEK BIBLE

assess them. Even those they had were not assessed each time they were re-used in LS, as is evident from the case of Pl., Alc. 1.131e. I am not suggesting that Liddell and Scott or their successors ought to have tried to find all this material, and then to have checked every example before writing their entry, which would somehow have encompassed it all. Given the physical difficulty of finding such a vast body of data by the methods available to them and of examining it in a reasonable amount of time, it was impossible. My point is that we now know there is much more abundant evidence than the small amount LSJ had to rely on, and that that evidence not only exists but is readily accessible and could be brought to bear on the lexical analysis. There is a further reason, besides the practical and time constraints, why LSJ based their entry on a small sample, and that relates to the next point. (2) Dependence on predecessors. Lexicographers in general (with some rare exceptions) rely for most of their material on preceding lexicons. This is not only because it is easier to reuse what someone has already assembled and analysed (or not), but because it is impossible to do otherwise unless one is willing to spend, say, 50 years gathering and assessing all the data afresh. In the case of a long-running series like LS(J), it was natural each time to trust the material in the previous edition and simply make revisions to it. But there is a risk that dubious lexical decisions remain unquestioned and so are perpetuated. A case in point is the perpetuation of ‘of an only son’ through several editions, without anyone noticing the confusion of reference and meaning; another is the questionable survival of the long-standing to be acquiesced in as a separate sense in LSJ after something similar was newly introduced to replace the existing sense I. (3) Piecemeal revision. At the same time as reusing the bulk of the material of a previous lexicon or edition, editors of lexicons make piecemeal changes and additions. This spasmodic intervention may cause more trouble than leaving the entry alone. This is because any change may impact on the rest of the entry but tends not to be accompanied by a reexamination of the entry as a whole. A glaring case in the LSJ tradition is the introduction of a new definition for sense I, as we have seen, without adequate reflection on how it is meant to work in the examples cited for that sense. (4) Control of secondary literature. Lexicons do not proceed in a vacuum. A vast body of discussion in books and journals accompanies their progress. In that literature all sorts of additions, corrections, reactions, and comments can be found. Editors of lexicons attempt to take account of this material. No doubt the revisers who produced LSJ did so to the best of their capacity. But what means do we have of knowing? And what

RELEASING LSJ FROM ITS PAST

267

reporting of it might usefully have been provided, both to show the sources of information and as an aid to further research? Such questions have not entered LSJ’s horizon, nor at any stage in the history of LS. In the world of NT lexicons some coverage of the secondary literature is almost de rigueur.21 It would of course have been impossible to do this in LSJ without doubling the size of the work. But it is a gap we are entitled to feel, and that we may look for a way of remedying. In the case of ἀγαπητός, it would have been helpful indeed to find in LSJ a reference to the source that inspired the change to sense I of ἀγαπητός. (5) Physical constraints. This is a factor of major importance. I refer to the limitations of presenting lexicon entries on the printed page. All the way from the simple ancient glossary made up of lemmata in the form ‘word x = word y’ through to LSJ and modern printed lexicons, there is a single line of development. From one point of view all that has happened over time is the accumulation of further information up to the limits of the printed book. The ultimate size of the book has always constrained the amount of material that could be presented in each entry and determined the format of presentation, including the need for abbreviation. For the lexicographer at work on compiling the entry, there is always space pressure. A selection has to be made from a larger body of data and whatever is finally included has to be compressed as much as possible. The actual tasks of lexical analysis and description, difficult enough, are made more difficult: the lexicographer is worrying about how to keep the size of the entry down, and material that would have been helpful or illuminating has often to be omitted to save space. To this of course is linked the matter of ease of reading and comprehension by the user of the lexicon. The user tends to be on the losing side of the contest. In the case of LSJ, besides the general compression, we have had to accept the practice of shortening headwords by the use of hyphens and combining groups of headwords into large paragraphs. The motive was obviously to save space; the result is very inconvenient to the reader.22 None of this would have been necessary if the content of entries had not been governed by what can be presented in a printed work. 21

22

Cf. esp. BDAG, building on earlier material gathered by Bauer from 1928 onwards. Older lexicons, such as Thayer’s at the end of the nineteenth century, provided similar bibliographical matter. See Chadwick, ‘Replacing Liddell and Scott,’ 2 for caustic comments on the innovation and the name of the person who had the ‘effrontery’ to claim credit for ‘this astonishing lack of perceptivity.’ Another device is the use of cross-references to other words, introduced by ‘=’. This too is presumably to save space, but may end up not doing so, and simply wastes the reader’s time. See, e.g., s.v. σκήνωμα: first we are told ‘= σκήνημα,’

268

COLLECTED ESSAYS ON THE GREEK BIBLE

Under this heading is also to be noted the sheer physical difficulty of preparing copy. This impacts on the work of the editors and their decisions about how to arrange, or re-arrange, the entries. LSJ provides a telling illustration. The preface to LSJ gives a description of how the revisions to LS8 were done in the 1920s. Corrections and insertions were written in the margins of a paste-up of the existing entries.23 This procedure would have been intricate and prone to error, and a discouragement to major changes. For Chadwick it was a ‘basic fault’ in preparing the new edition; it ‘effectively prevented any attempt to modernise the style or to recast the structure of the articles.’24 Anyone who uses a computer for word-processing knows how radically different things are today. (6) Method of indicating meaning. This is a deeper methodological problem, one that will not be fixed in a day. The lexicon of Liddell and Scott, from LS1 through to LSJ, like all its predecessors from time immemorial, has relied primarily on the unsatisfactory gloss method. The alternative is the ‘definition’ method, which was pioneered by Murray in the Oxford English Dictionary and handed on to the Oxford Latin Dictionary. This development came too late to have an impact on LS(J). In the definition method of indicating lexical meaning, rather than simply using one or more translation equivalents, an attempt is made to define the area of meaning in an exact and unambiguous way, and to distinguish it from other areas of meaning. In the NT sphere, there has been lengthy discussion of the problems associated with the use of glosses and the advantages of a shift to definitions. In the case of ἀγαπητός there is, as it happens, considerable use by LSJ of what might pass for definition; but the usual glosses are also seen, as e.g., in III.2 just enough to content one, barely, scarcely, earlier only just, barely, scarcely. These last three are simply possible translations in context. Similarly in II.[1] desirable. This gloss shows the potential for the method to be misleading, even with the qualifier ‘of things.’ One might be tempted to think, wrongly, of ‘wanted, sought after,’ or even ‘attractive.’ (Cf. below for an attempted definition).25

23

24 25

then under the latter we find ‘= σκηνή.’ The meaning tent nevertheless appears further on under σκήνωμα and might just as well have been given in the first place. More seriously, this device fails to tell which meaning of the cross-referenced word is the relevant one. Thus in the case of ‘σκήνημα = σκηνή,’ when we look up σκηνή we are left to take our pick from no less than nine senses of this word. LSJ, ‘Preface 1925,’ x. [The marked-up galley proofs of the sixth edition (1869), reproduced in Imholtz, ‘Liddell and Scott,’ 130, vividly illustrate the horrors of such a process.] Chadwick, Lexicographica Graeca, 7–8. For discussion of definition versus glossing, see Lee, History, 15–29, with references esp. to the work of Louw and Nida on the NT; see also BDAG, viii. The gloss method

RELEASING LSJ FROM ITS PAST

269

A NEW APPROACH These six points indicate the constraints under which the editors of LSJ worked. No lexicographer of the past could escape them. What ought to be done next? I do not believe an old-style revision of LSJ is worth doing now. That is, the sort in which piecemeal changes and additions are made without a complete overhaul, and the result is another printed book of the same size and scope as LSJ. My reasons for saying this are two. First, to carry out such a revision would be to continue the methods of the past and accept the same constraints; it would not cure the problems of LSJ, but would make them worse. Secondly, there is now an opportunity to do things in an entirely different and better way, that is, by using the electronic medium, and transferring the whole operation to electronic form.26 The rest of my paper is concerned with exploring what this would mean. I should make it clear that I am concerned with the conception rather than the technical detail. The technical expertise to carry out the task is already available; lexicographers need to consider how they would like to see it harnessed to their needs. What I have in mind is not simply to make LSJ available on screen, but to reconceptualise our approach to the tasks of lexicography and the production of lexicons.27 The text of LSJ has already been entered electronically, and this could be used as a first step, but it would be no more than that. What needs to be developed is a database in which everything can be stored: not only the meanings and citations that we currently consult the lexicon for, but all the data that lie behind them and are in any way relevant to them. Moreover, as well as being a storehouse of information, it would be a record of work carried out to produce it. There is of course no limit on the size of an electronic storage system, so one of the problems of the past would simply disappear. Obviously this is a big and long-term enterprise. It may seem to some to be an unrealistic ideal. The practical difficulty of actually getting started and then continuing with the work over many decades certainly looks

26

27

is of course the staple of smaller lexicons such as Morwood and Taylor, Pocket Oxford, in which the meanings given for ἀγαπητός are: ‘beloved; amiable; welcome; sufficient.’ These are derived unchanged from Feyerabend, Langenscheidt’s Pocket Greek Dictionary (1910). The potential for misunderstanding is evident. [John Chadwick had already arrived at a similar view in 1996. In Lexicographica Graeca, 28 he wrote: ‘There is no case at all to be made for an indefinite expansion of the book [i.e., LSJ] in printed form. … Thus we have arrived at the parting of the ways: the total lexicon must henceforth be kept in electronic form. …] The proposals here for a database are similar to those made for NT lexicography in Lee, History, 182–8.

270

COLLECTED ESSAYS ON THE GREEK BIBLE

forbidding; it may seem better not to start. If this is so, it is because we are still in transition from the old way of doing things, that is, working in the medium of print. Sooner or later, in my opinion, an electronic database for Greek lexicography will be seen to be inevitable and will be created, and then maintained indefinitely. To take a simple illustration. Many small revisions and additions that might be made to LSJ have steadily accumulated since the publication of the last Supplement in 1996. Would anyone wish to see another printed supplement to LSJ to take account of them? Rather, it will seem natural to gather such material electronically and make it available in that form. In Stuart-Jones’s 1925 preface to LSJ, the tale is told of how the idea of a ‘Thesaurus Graecus’ emerged in 1904 but was quickly abandoned as unachievable, not least because of the likely bulk of the resulting work.28 It was indeed impossible – in printed form, which was the only one that could be envisaged at the time. The thinking then naturally ran along the lines of a vast collection of handwritten slips, like those for the TLL and OED, which would eventually lead to a printed work. Diels’s guess that it might run to 120 volumes was enough to frighten anyone. In some ways the preservation of this story in our leading lexicon has had an unfortunate effect. It has cemented the idea that ancient Greek is too immense to do more than dip into as far as lexicography is concerned, and a complete collection of examples is neither possible nor necessary. But by using the electronic medium it can be done. The texts of Greek literature are now available in the TLG, and the majority of published documentary texts in PHI7; it is only a matter of re-sorting the data to arrive at a complete list of occurrences of every word, the first step towards a ‘thesaurus’ of Ancient Greek. Individual researchers are already doing this every day on a small scale when they carry out electronic searches for words they are studying.29 What then might the ideal database contain for a word such as ἀγαπητός, and by extension every word in LSJ? What information would one like to be able to find and see presented on screen, if one were to look up this word? It is not difficult to say, at least as regards the primary information. As to the rest, I do not see why our horizons should be limited in any way. There is nothing to stop us including all the data related to a given 28

29

LSJ, ‘Preface 1925,’ v–vi. A fuller account and discussion of the ‘Thesaurus’ proposal may be found in Cohn, ‘Griechische Lexikographie,’ 724–30. It should be understood that the TLG is not a lexicon: it provides raw data, without meanings or lexical analysis of any kind.

RELEASING LSJ FROM ITS PAST

271

word, much of which is currently consigned to separate reference works and has never been assembled in one place. A suggested list of categories of information follows, with discussion and, in the case of the ‘main entry,’ a sample layout. The question of the electronic structure that would organise the data is one I leave to others. The categories have of necessity been placed in linear order here, in accordance with logic and convention, but of course in the electronic form the order is immaterial. CATEGORIES IN THE DATABASE MAIN ENTRY This would be like the lexicon entry for any word in a larger lexicon such as LSJ, with lexical meanings defined and classified, each illustrated by selected examples. Ample quotation of the context would be included with each citation. Translations would also be given. I regard the latter as not just an aid to the user, but a valuable discipline for the lexicographer. It is remarkable how often one’s understanding of the meaning of a word is sharpened, even completely revised, by the attempt to set down a translation of the passage. Since there is no limit on space, all this could be presented in tabular form with the utmost clarity as the objective. The setting-out in the electronic OED is a model of how well this can be done. Besides tabulation and different font sizes, colour could be used to catch the eye. There would of course be links from this view to every other part of the database. The word itself could be the lead-in to most links; in addition, some other items, such as the citations, could be linked to further data on them. All these other data would be material of potential interest, but not essential to grasping the semantic contours of the word. What one saw in the main entry would be complete in that respect; all the rest would be amplification or support. I indicate possible links by the standard convention of underlining. The format is rudimentary; the final product would be visually much more attractive as well as space-efficient. The lexical analysis and selection of examples are illustrative and not intended to be final. The selection is small and confined to the Classical period; in the final product it would be larger and include texts of later dates. I have expanded LSJ’s abbreviations a little. With the space constraint lifted, there is no need for extreme abbreviation of authors’ names and works; in fact there may be no need for any. Nevertheless, a standard

272

COLLECTED ESSAYS ON THE GREEK BIBLE

system of abbreviation for all Greek authors and works, not forgetting documentary texts, remains a desideratum, and could be supplied somewhere in this database.30 It is important to note the difference between the definitions and the translations given below. The definition is an attempt to state succinctly and unambiguously the lexical meaning seen in the examples listed under it; the translation of a given example is simply a translation in context, and does not necessarily use the exact words of the definition, though its general import is the same. The test of the definition is that it fits the context in place of the defined word without difference in meaning in all the cited occurrences. ἀγαπητός etymology, formation, orthography, morphology statistics, full evidence lexical analysis, semantic history, lexical structure, syntax, bilingual data literature mini entry

1 regarded with special love and affection Homer, Iliad 6.401 ἥ οἱ ἔπειτ᾽ ἤντησ᾽, ἅμα δ᾽ ἀμφίπολος κίεν αὐτῇ παῖδ᾽ ἐπὶ κόλπῳ ἔχουσ᾽ ἀταλάφρονα, νήπιον αὔτως, Ἑκτορίδην ἀγαπητόν. She met him then, and with her went the maidservant holding the tender boy in her arms, just a child, the beloved son of Hektor. author work date more context text discussion Sappho 132.2 ἔστι μοι κάλα πάις χρυσίοισιν ἀνθέμοισιν ἐμφέρην ἔχοισα μόρφαν Κλέις ἀγαπάτα, ἀντὶ τᾶς ἔγωὐδὲ Λυδίαν παῖσαν οὐδ᾽ ἐράνναν. . . . I have a beautiful daughter, who looks like golden flowers, Cleis my darling, whom I would not exchange for all Lydia or lovely. . . . (Page) author work date more context text discussion 30

Documentary texts, i.e., inscriptions, papyri, ostraca, etc., currently suffer from another problem, the difficulty of identifying the same text published in different editions and collections over the past century and more. This could be easily remedied by assigning to each text a unique number, which would thereafter be quoted whenever the text was re-edited or used in any way. A database could then collect all information on each (numbered) text. Since this was written, the ‘Trismegistos’ database constructed by Mark Depauw (www.trismegistos.org) has adopted just such a procedure.

RELEASING LSJ FROM ITS PAST

273

Plato, Alc. 1.131e οὔτ᾽ ἐγένεθ᾽, ὡς ἔοικεν, Ἀλκιβιάδῃ τῷ Κλεινίου ἐραστὴς οὔτ᾽ ἔστιν ἀλλ᾽ ἢ εἷς μόνος, καὶ οὗτος ἀγαπητός, Σωκράτης ὁ Σωφρονίσκου καὶ Φαιναρέτης. Alkibiades son of Kleinias, it seems, neither had nor has any lover except one only, and that much loved, namely Socrates, son of Sophroniskos and Phainarete. author work date more context text discussion more examples of sense 1

2 regarded with pleasure and satisfaction Xenophon, Mem. 3.10.5 πότερον οὖν, ἔφη, νομίζεις ἥδιον ὁρᾶν τοὺς ἀνθρώπους δι’ ὧν τὰ καλά τε καὶ ἀγαθὰ καὶ ἀγαπητὰ ἤθη φαίνεται ἢ δι’ ὧν τὰ αἰσρχά τε καὶ πονηρὰ καὶ μισητά; Which then, he said, do you consider more pleasing to see, men in whom fine and good and likable characteristics are displayed, or those in whom shameful and bad and detestable ones? author work date more context text discussion Plato, Protag. 328a ἀλλὰ κἂν εἰ ὀλίγον ἔστιν τις ὅστις διαφέρει ἡμῶν προβιβάσαι εἰς ἀρετήν, ἀγαπητόν. But if there is someone who excels us even a little in leading [us] on towards virtue, it is something to be glad of. author work date more context text discussion31 Plato, Philebus 61e οὐκοῦν εἰ τἀληθέστατα τμήματα ἑκατέρας ἴδοιμεν πρῶτον συμμείξαντες, ἆρα ἱκανὰ ταῦτα συγκεκραμένα τὸν ἀγαπητότατον βίον ἀπεργασάμενα παρέχειν ἡμῖν, ἤ τινος ἔτι προσδεόμεθα καὶ τῶν μὴ τοιούτων; So then, let us see, when we have first mixed the truest segments of each, are these mixed together sufficient to effect and provide us with the most agreeable life, or do we still need something more that is of a different kind? author work date more context text discussion more examples of sense 2

31

This example has been understood as ‘one must be content’ since LS2 (1845), but some rethinking may be appropriate.

274

COLLECTED ESSAYS ON THE GREEK BIBLE

3 to be tolerated as the best attainable Andocides 3.22 μετὰ δὲ τοῦτο ὅρκους ὀμόσαντες αὐτοῖς καὶ τὴν στήλην εὑρόμενοι ‹παρ᾽› αὐτῶν στῆσαι, κακὸν ἀγαπητὸν ἐν ἐκείνῳ τῷ χρόνῳ, σπονδὰς ἤγομεν ἐπὶ ῥητοῖς. After this, when we had sworn oaths with them and had secured their permission to set up the stele, a misfortune to be accepted at that time, we made a truce on specified terms. author work date more context text discussion Xenophon, Oec. 8.16 ἀπειλεῖ γὰρ ὁ θεὸς καὶ κολάζει τοὺς βλᾶκας. ἐὰν δὲ μόνον μὴ ἀπολέσῃ τοὺς μὴ ἁμαρτάνοντες, πάνυ ἀγαπητόν· For God threatens and punishes stupid persons; and if he only refrains from destroying those who are innocent of any wrong, it is something to be glad of. author work date more context text discussion Menander 224.3 (Kassel-Austin) εἶτ᾽ οὐχ ὅμοια πράττομεν καὶ θύομεν; ὅπου γε τοῖς θεοῖς μὲν ἠγορασμένον δραχμῶν ἄγω προβάτιον ἀγαπητὸν δέκα, αὐλητρίδας δὲ καὶ μύρον καὶ ψαλτρίας, . . . μικροῦ ταλάντου. . . . Then don’t we fare in proportion as we sacrifice? When I bring to the gods a poor sheep just good enough, bought for ten drachmas, but flute-girls and perfumed oil and harp-players, . . . worth nearly a talent, . . . author work date more context text discussion32 more examples of sense 3

MINI ENTRY The purpose of this view is of course to supply essential information quickly. The many small (‘manual,’ ‘pocket,’ ‘concise’) dictionaries are 32

This example illustrates the need for discussion; in fact it is a good candidate for the ‘pool of unplaced or problematic occurrences’ mentioned below. The meaning is not obvious, and opinions will differ. LSJ place it in [I.1], where, as we have seen, it is hard to be sure what meaning they intend. DGE places it in I.2, muy amado, apreciadísimo, estimadísimo. Gomme and Sandbach (Menander, 698) remark that προβάτιον is ‘depreciatory’; cf. the translations of Edmonds (‘a lean lamb’) and Allinson (‘a scrawny sheep’). If this is right, sense 3 seems appropriate. Another example in a comic fragment, Hipparch. Com. 1.3 (Kock-Austin) δαπίδιον ἓν ἀγαπητὸν ποικίλον, cited since LS7, is also hard to place.

RELEASING LSJ FROM ITS PAST

275

the model. Meanings would be given, but not citations (unless in special cases). More information about forms, to help the beginner, could be included here. Glosses might be offered as well as the definitions established in the main entry. ETYMOLOGY;

WORD FORMATION; ORTHOGRAPHY; MORPHOLOGY

These might possibly be treated together, since there is much overlap. Some words would call for considerable detail in each category, others only routine information or possibly none at all. In the case of ἀγαπητός, one would like to see displayed the etymology (unknown); how the word is formed (verbal adj. in -τός from ἀγαπάω); a full list of related words; dialectal forms; declensional forms (regular).33 FULL

EVIDENCE

Here would be stored and accessible on screen all occurrences of the word in surviving texts up to whatever date was chosen as the cut-off. Every citation would be provided with the same kinds of information as indicated above for the citations in the main entry, and all would be classified semantically on the basis of the analysis there. A pool of unplaced or problematic occurrences, linked to ongoing discussion, would be a natural adjunct. The data would of course be searchable in a variety of ways, such as by author, date, genre of work, meaning, textual variants. Summaries and statistics would be available. WORK

AND AUTHOR DATA

Here would be found a list like the TLG Canon, which could be consulted for itself, but would also be linked to author names and citations occurring anywhere in the database. LEXICAL ANALYSIS It would be helpful if the database offered an explanation of the thinking behind the definitions and allocation of occurrences, and provided an opportunity for ongoing discussion and fine-tuning of them. Definitions are never arrived at easily, and those given at first might call for improvement; similarly the decision about the meaning of an individual occurrence 33

Much of the information is of the kind found in Chantraine, DELG. In the case of verbs, the database could record not only ‘principal parts’ but a complete list of verb forms, as provided by a number of handbooks today.

276

COLLECTED ESSAYS ON THE GREEK BIBLE

is not always easy to make and could sometimes change. As assessment of the data on the word went on, it would be useful to have a record of such decisions, especially on contested questions (such as the existence of the meaning ‘only’). Unsolved problems would also be noted and offered for consideration. The earlier history of the lexical analysis, i.e., its treatment in previous lexicons of all ages, even as far back as antiquity, could find a place here. DIACHRONIC ANALYSIS An outline of the semantic history of the word (and its field) from its origin onwards, ideally as far as Modern Greek. This goes hand in hand with the lexical analysis and the ordering of senses: implicit in them is a reconstruction of the word’s semantic evolution.34 LEXICAL–STRUCTURAL DATA Here would be an opportunity to take account of the insights of modern linguistics and attempt to locate the word in its place in the lexical structure of the language. The obvious first requirement would be to list the near synonyms and antonyms, and attempt to discern the differences, if any, between them. In the case of ἀγαπητός, one would want to learn how it related semantically to φίλος, φίλιος, προσφιλής, στερκτός, μισητός, ἠγαπημένος, and so on. Data on connotation and register would form part of the picture. SYNTAGMATICS Data on collocation. Our lexicons are already strong on this, partly, one suspects, to provide an aid to prose composition. Comprehensive data could now be given. For instance, one could find a breakdown of all occurrences of ἀγαπητός by application (to persons, animals, things, situations). BILINGUAL DATA I am thinking of words where bilingual interference or interaction between Greek and another language is relevant to the semantics of the Greek word, as when φίλος functions in the same way as Lat. amicus to indicate a courtier of the Roman emperor. In the case of ἀγαπητός, the questions raised by its use as a translation of Hebrew yahid in the Septuagint would be pursued.35 34 35

For examples, see Chadwick, ‘Semantic History,’ and Lexicographica Graeca, passim. On this see Lee, History, 203–6.

RELEASING LSJ FROM ITS PAST

277

SECONDARY LITERATURE Some remarks on this topic have been made above, under (4), among the constraints affecting LSJ. It is safe to say there exists some discussion, whether in special articles or monographs or in commentaries, of almost every word in the lexicon. It would be most helpful to have all this material gathered together in one place. Moreover, this would not only be a collection of what had been done in the past, but an ongoing record of discussion as it proceeds in the future. A mere list of titles is not enough; summary of the content, if not reproduction in full, is essential. In the case of ἀγαπητός, there are nine items listed in RBLG, and more can be found in the NT lexicons. That is only the recent material, of the last century or so: there is a lot more in older works, all of it potentially relevant to improving our understanding of the word.36 Everyone who works on this word has to chase them up; then the next person does it all again. This duplication of effort can be brought to an end. CONCLUSION This then is an outline of the possible content of the ideal electronic database for Greek lexicography. The question of the printed form is in my opinion a non-issue. Whatever printed book is desirable can be produced from the database. Any such book can select the right material from the larger store, and thus be properly targeted to the intended user, unlike most older works, which try to meet the needs of many at once. One printed book that will remain a desideratum is an equivalent to LSJ, on a somewhat reduced scale, the book for general use by more advanced undergraduates and scholars. This will offer a certain degree of detail, but not attempt to supply everything. A lot of what is in LSJ now is superfluous to the needs of the general user. As John Chadwick said, ‘much of the information . . . ought to be consigned to a special lexicon of record.’37 The database I am envisaging is that ‘special lexicon of record.’ In it all the information on each word will be stored and readily accessible to those who really want it. If we go along this path, the attempt to put a sort of digest of all the data on the printed page can be given up and be recognised to be a thing of the past. Most of the problems I have listed will be overcome, and LSJ will be released from the confines of its history. 36 37

See Lee, History, 193–211 passim. Chadwick, Lexicographica Graeca, 19. Cf. also this remark (ibid.): ‘Part of the trouble with LSJ, and even more with its Supplements, is that no attempt has been made to define the user for whom the work is created.’

278

COLLECTED ESSAYS ON THE GREEK BIBLE

Postscript Recent work on LSJ includes a notable volume of essays, Liddell and Scott, edited by Stray, Clarke, and Katz, published in 2019. This book offers many insights into the history of Liddell and Scott’s Lexicon, its methods, and possible future directions. Among the contributions, Anne Thompson’s paper ‘Βάπτω: An Illustration’ forms a strong complement to my essay above. Thompson presents a close analysis of the entries for βάπτω in all the major lexicons and in Chadwick’s Lexicographica Graeca, and then offers a new lexical analysis of her own based on their source material.38 The need for a complete overhaul of LSJ is again evident. Further, a lucid explanation of the definition method as opposed to the gloss method is offered in Thompson’s ‘Response’ to the remarks of a reviewer of Liddell and Scott (in BMCR). The chapter by John Henderson, ‘A1–Zythum,’ in Stray, Classical Dictionaries, though on the history of the OLD, is both entertaining and pertinent. Much of the history of the OLD is a ‘re-run’ of that of the OED. The familiar themes appear: underestimation of time, cost, and staffing, overestimation of abilities,39 production quandaries, and intrusion by non-experts. Future planners of major lexicon projects would be well advised to read before acting. My enthusiastic embrace of a lexicographical database that would include everything was mainly a ‘scoping’ exercise to show what could be done. I did not and do not imagine it can be done easily or quickly, or in its entirety in the near future. The main point is that the move into some kind of database cannot be avoided and a start ought to be made as soon as possible. As far as I know, there has been no progress in that direction. The future of LSJ, and with it Ancient Greek lexicography, remains tied to the past. Meanwhile new lexicons have appeared in print or electronic equivalent, following traditional methods: GE, its predecessor GI, and the recent CGL (2021); DGE has reached ἔξαυος in Vol. 7 (2007).40

38

39

40

To this analysis (with 16 senses) is added a ‘concise outline’ listing seven senses of βάπτω (388–9). This would form the ideal basis of an entry in a shorter lexicon, with minimal illustrations added. I do not disagree with Anne’s remarks at 393–4: ‘There is … no need initially for exhaustive collection of source material and bibliography. A relatively small sample of texts can be used to set up reasonably accurate preliminary models that can be added to, corrected, and modified.’ Souter (in 1921) asserted that ‘Such a scheme [his plan to revise Lewis and Short] requires no genius or group of geniuses to carry it out’ (Henderson, ‘A1–Zythum,’ 148). Yet his own attempt at a specimen entry was so poor that his unsuitability as editor of the OLD was plain to see. The proposal for a database appears also in Essays 9 (delivered 1995) and 13 (delivered 2006).

18 WHY DIDN’T ST BASIL WRITE IN NEW TESTAMENT GREEK? 20101

Abstract St Basil’s Greek is noticeably different from that of the NT, though both fall within the Koine period of ancient Greek. This paper sets out to explain the reason. St Basil’s Greek and the Greek of the NT are briefly characterised, then the discussion surveys the rise of Atticism, the ancient education system, and St Basil’s education. The conclusion is reached that St Basil’s choice of the higher language and style of the pagan Classics was the natural one for him, though paradoxical in that he, as a Christian, rated the content of the biblical books as superior.

PREAMBLE Saint Basil the Great was a prolific author, and there is a large body of his works surviving. There are over three hundred letters alone; other works number at least a hundred. Besides the count of his works, the tally of words in some of the longer works gives an idea of how prolific he was: the Homiliae in Hexaemeron numbers 35,834 words, and the De Spiritu Sancto 26,309, according to the TLG electronic count.2 All of this was written in Greek, of course. But what sort of Greek? To say that it was Greek as written in the fourth century AD provides only a partial answer. True enough, that locates it within the Koine Greek period (300 BC to 600 AD) of ancient Greek, but the Greek of that period is a complex phenomenon, affected by currents and influences not obvious without investigation. 1

2

This paper was presented at the first Patristic Symposium of St Andrew’s Greek Orthodox Theological College in 2009, dedicated to St Basil the Great. I am grateful to His Eminence Archbishop †Stylianos and St Andrew’s College for the invitation to present in the Symposium, and to Fr Doru Costache and Edwin Judge for valuable suggestions. [The paper has been reprinted in Costache and Kariatlis (eds.), Cappadocian Legacy, 63–77.] TLG Canon, 81–4.

280

COLLECTED ESSAYS ON THE GREEK BIBLE

ST BASIL’S GREEK As my starting-point I want to quote just the first sentence of one of St Basil’s most famous works, To Young Men, Πρὸς τοὺς Νέους:3 Πολλά με τὰ παρακαλοῦντά ἐστι ξυμβουλεῦσαι ὑμῖν, ὦ παῖδες, ἃ βέλτιστα εἶναι κρίνω, καὶ ἃ ξυνοίσειν ὑμῖν ἑλομένοις πεπίστευκα. Many are the things encouraging me to advise you, children, what I judge to be best and what I am convinced will be of benefit to you when you make your choice.

In this one short sentence, we already have striking indications of the kind of Greek we are dealing with in St Basil’s works; and this is not an isolated instance, but typical of his style everywhere. To begin with, the opening phrase imitates the first words of a speech attributed to the famous Athenian orator Demosthenes, who flourished in the fourth century BC. His sentence begins Πολλά με τὰ παρακαλοῦντα ἦν, ὦ ἄνδρες Ἀθηναῖοι, γράψασθαι Νέαιραν τὴν γραφὴν ταυτηνί (‘many were the things encouraging me, men of Athens, to bring this suit against Neaira’).4 The resemblance is unlikely to be accidental. St Basil knew the Classics (as we shall see), and the sentence includes an idiomatic feature of Classical Greek, the placing of the enclitic pronoun (in this case με) as second word.5 Other significant features in the sentence are: the choice of the Attic, or rather old Attic, form ξυν- instead of συν- in two of the verb compounds; the idiomatic Classical use of the perfect πεπίστευκα describing a state; and the skilful use of the aorist participle ἑλομένοις attached to ὑμῖν, meaning literally ‘to you (at the point of) choosing.’6 These are not natural features of the everyday Greek of St Basil’s time but the result of conscious effort by the writer. There is something else, a subtlety that one would hardly notice unless alerted to it, the use of clausulae. This involves choosing or contriving a certain pattern of syllables in the closing phrase of a sentence or clause: 3

4

5

6

In full Πρὸς τοὺς νέους, ὅπως ἂν ἐξ Ἑλληνικῶν ὠφέλοιντο λόγων, ‘To young men, on how they might benefit from Greek literature.’ The exact title is in fact uncertain: see Wilson, Saint Basil, 8–9. The TLG Latin title is De Legendis Gentilium Libris; others, such as Rousseau (Basil of Caesarea, e.g., 49), use Ad adulescentes. [D.,] Against Neaira 59.1; Wilson, Saint Basil, 37. Modern and ancient scholarship has doubted the attribution to Demosthenes, but the speech dates from his time and was just as much a model for later generations as the genuine speeches. See Wilson, Saint Basil, 37. Stasinopoulos (Μορφές, 277) detects further Demosthenic phrasing in ἃ βέλτιστα … and ἃ ξυνοίσειν … in St Basil’s opening sentence. On πεπίστευκα see Wilson, Saint Basil, 37.

WHY DIDN’T ST BASIL WRITE IN NEW TESTAMENT GREEK?

281

the last two stressed syllables are separated by an even number of unaccented syllables, as here in ἑλομένοις πεπίστευκα. Another illustration, this time with four syllables between the accented syllables, may be seen further on in the same work, at 5.9: καὶ τελευτήσαντι παραμένουσα.7 The effect of this device is to give a pleasing balance and smoothness as the sentence comes to a conclusion. Numerous other linguistic features found generally in St Basil’s works could be mentioned, all tending in the same direction, that is, towards old-fashioned, Classical Attic, of the kind that was no longer spoken in St Basil’s day. Examples ready to hand are forms in -ττ-, as in φυλάττω, instead of -σσ-; the frequent use of obsolete particles such as μήν, γε, δή; and recourse to the full range of optative forms and uses in the verb.8 These are all connected with the phenomenon of ‘Atticism,’ on which more will be said shortly. Besides using individual features adopted from the Classical language, St Basil refers frequently to the works of Classical authors, both by quoting them and alluding to them. In To Young Men can be found quotations and reminiscences from a list of authors including Homer, Hesiod, Theognis, Plato, Aristotle, and Solon. Some are direct quotations, as when three lines of Solon are quoted (5.9), or an expression is borrowed verbatim from Plato (ὅτι μὴ πᾶσα ἀνάγκη, 9.2).9 Others are more subtle, as when a Homeric phrase, ἔνθεν ἑλών (‘taking it up from this point’) is slipped into a sentence (1.5).10 There is even an allusion (9.6) to the famous maxim on the temple of Apollo at Delphi, γνῶθι σεαυτόν, but this is done without actually quoting the words.11 In 5.6–8 St Basil uses the episode of the meeting of Odysseus and Nausicaa in Odyssey 6 to draw a lesson about virtue, introducing it without even mentioning the names: he is able to assume that his educated hearers will be as familiar with the story as he is. In his works as a whole, the same Classical authors and many others are cited or alluded to with great frequency.12 7

8

9 10

11 12

The rule is set out in Maas, Greek Metre, 17; cf. Wilson, Saint Basil, 13. St Basil was not alone in following this rule; his practice ‘corresponds closely to that of other well known writers of the same date’ (Wilson). As in the title of To Young Men: ὠφέλοιντο (see n. 3). Cf. the list of features in Fabricius, ‘Der sprachliche Klassizismus,’ esp. 192. Wilson, Saint Basil, 64. This was ‘a favourite phrase with Atticists’ (Wilson, Saint Basil, 40). The sentence as a whole reads: τίνα οὖν ἐστι ταῦτα, καὶ ὅπως διακρινοῦμεν, τοῦτο δὴ καὶ διδάξω ἔνθεν ἑλών. See Wilson, Saint Basil, 65. See the discussion and index of references in Deferrari, Basil: Letters 249–368, pp. 366– 8, 458–61.

282

COLLECTED ESSAYS ON THE GREEK BIBLE

St Basil’s writing style in general has attracted high compliments. Of the Hexaemeron, Quasten expressed the view that ‘there is no work in late Greek literature which could be compared with these homilies in rhetorical beauty.’13 More tellingly, his friend and contemporary St Gregory of Nazianzus regarded St Basil’s letter-writing skills as far superior to his own.14 The playful exchange between St Basil and the renowned rhetorician Libanius in Letters 338, 339 and 340 indicates as surely as anything that St Basil was on an equal footing with Libanius in mastery of refined writing. Each tries to outdo the other in εἰρωνεία, Libanius claiming that on reading St Basil’s letter he declared he had been defeated in beauty of letter-writing (ἐν κάλλει μέν, ἔφην, ἐπιστολῶν ἥττημαι), to which St Basil responds by describing his letter as ‘that shabby letter of mine’ (τὴν ἐπιστολὴν ἐκείνην τὴν ῥυπῶσαν), and so on.15 We have then, from this brief sketch, an indication of the style and language of St Basil, and the impression can be quickly confirmed by reading any of his works. St Basil writes a high level, literary Atticistic Greek, densely constructed with all the devices and subtleties of language and rhetoric available to the trained writer.

THE NEW TESTAMENT Three centuries before St Basil, the NT authors also wrote in Greek, within the same time-period of ancient Greek as St Basil, that is, the Koine Greek period, mentioned before. But the contrast with the language and style of St Basil is marked. The NT is written, in general, in ordinary Koine Greek, and the style of writing, though varying from book to book, in no case reaches the same high plane as St Basil’s. This is the general picture, within which there is a considerable range; this range we must now briefly explore. To focus first on language. Simple, low-level Koine Greek is characteristic of the Gospels of Mark and John; the Gospel of Matthew is similar but on a slightly higher level. The two books by St Luke, the Gospel 13

14 15

Quasten, Patrology, 3:216–7. In the ninth century St Photius offers a glowing tribute to St Basil’s style and rhetorical skill in the Hexaemeron (Bibliotheca, 141). See Gallay (ed.), Saint-Grégoire, 53. Cf. Bardy, ‘Basilius von Caesarea.’ On the question of the authenticity of the correspondence, see Rousseau, Basil of Caesarea, 57 n. 130; Deferrari, Basil: Letters 249–368, xiii–xv. There seems to be no doubt that some are genuine, among them the three mentioned. For further remarks on St Basil’s rhetorical skills see Kinzig, ‘Greek Christian Writers,’ esp. 646, with bibliography.

WHY DIDN’T ST BASIL WRITE IN NEW TESTAMENT GREEK?

283

of Luke and the Acts of the Apostles, show a distinct effort to write more literary Greek than the other Gospel authors. Many features indicative of ‘better’ Greek can be noticed, as, for example, certain choices in vocabulary, in which popular words are avoided in favour of more Attic ones. This is noticeable when St Luke rewrites material derived from the Gospel of Mark. St Luke can also write a showpiece of high level Greek in the preface to his Gospel (1:1–4). But he does not keep it up. He quickly turns to a different, ‘biblical’ sort of language (and style), which owes a great deal to the Greek of the LXX. This is his model for a ‘classic’ style, not the authors of Classical literature; and features of Atticistic Greek, though present, are not maintained with the consistency seen in St Basil. The other books of the NT, apart from Revelation, which is below even Mark and John, are rather different and show middle-range educated Greek. The Epistles of St Paul are a definite step above Mark and John. This is seen in a variety of features – forms, syntactical structures, choice of vocabulary – that contrast, sometimes sharply, with the simpler Greek of Mark and John. At the top among the books of the NT, in terms of language and style, is the Epistle to the Hebrews, long recognised as the work of a highlyeducated writer. We are already touching on style, which is a somewhat different matter from level of language, though linked. The NT writers have their individual styles. In trying to characterise these different styles, an important question is the influence of rhetorical techniques. A good deal of modern study has turned attention to this topic, which has much relevance to the comparison with St Basil, since St Basil’s works are the product of advanced training in rhetoric. Among the NT books, the Gospels and Acts reveal very little evidence of rhetorical techniques, and in the case of Mark none at all.16 The case is different with the Pauline Epistles. The question of St Paul’s training in rhetoric and the extent of his use of rhetoric in his letters has perplexed ancients and moderns alike. He appears not to have been formally trained in rhetoric, but it is clear that he was not ignorant of the art; he claims to be a ‘layman in speech’ (ἰδιώτης τῷ λόγῳ, 2 Cor 11:6), but this cannot be taken at face value; later exponents saw that he was no expert by Classical standards, yet the power of his eloquence was unmistakable; his writings can be analysed to a considerable extent according to formal rhetorical categories, but the results have been diverse and inconclusive.17 16 17

For a thorough exploration see Burridge, ‘The Gospels and Acts.’ On Paul see Judge, ‘Paul’s Boasting’; Porter, ‘Paul of Tarsus.’ Judge’s paper, delivered as long ago as 1967, was full of insights and set out the parameters of the debate. Porter covers all recent discussion without reaching any firmer conclusions.

284

COLLECTED ESSAYS ON THE GREEK BIBLE

The situation is clear, however, with regard to the Epistle to the Hebrews: there is general agreement that it reveals a high level of rhetorical training and competence.18 As regards reference to Classical authors, the strong characteristic of St Basil’s writing that we noted above, the NT authors have little to offer, naturally enough. It is true that Paul quotes Menander (1 Cor 15:33 φθείρουσιν ἤθη χρηστὰ ὁμιλίαι κακαί, ‘bad company corrupts good morals’), but even this was a popular epigram and unlikely to derive from direct acquaintance with Menander.19 Paul’s speech in Athens, as recorded by Luke (Acts 17:22–31), contains, appropriately in that learned setting, a quotation from Aratus (Acts 17:28 τοῦ γὰρ καὶ γένος ἐσμέν, ‘for we too are his offspring’), but there is little more of that nature in the NT. What the authors do quote extensively is the Christians’ own ‘classical’ text, the OT in Greek, the LXX. In sum, the books of the NT show a range of language, from higher to lower, but the majority remain ordinary Koine Greek, and are not in the same league as St Basil and most of the Fathers. Those who looked back to them from the high point of the fourth century saw writing that struck them as inelegant and uneducated – ‘als stilistische Monstra,’ as Norden put it – and from their point of view it was true.20 St Basil himself speaks of the material communicated by the biblical authors as ‘in sense true, but in style unlearned’ (νοῦν μὲν ἀληθῆ, λέξιν δὲ ἀμαθῆ).21 Several other Fathers observe and discuss this problem, sometimes at length.22 One feature alone is sufficient to sum up the gulf between the Hochsprache of St Basil and the language of the NT. The form ξύν, which we observed in St Basil’s sentence at the beginning of this paper, is not used by a single NT writer.23

18

19 20

21

22

23

See Thurén, ‘New Testament Writings,’ esp. 589–92. There is a brief but useful survey of NT rhetoric in BDF, §§485–96 (‘The Arrangement of Words: Figures of Speech’). Fee, First Epistle to the Corinthians, 773 n. 61. Norden, Die antike Kunstprosa, 2:517; see further 516–25; Kinzig, ‘Greek Christian Writers,’ 634–6. Letter 339. St Basil means this playfully (cf. Norden, Die antike Kunstprosa, 2:529, n. 1), so that he can pretend that his own style has been infected by theirs and it shows. The passages are set out in full in Triantaphyllides, Γραμματική 1: Ιστορική Εισαγωγή, 418–22. Cf. Kinzig, ‘Greek Christian Writers,’ 635. While noting the poor quality of NT writing, the Fathers offer good defences and explanations. See BDF, §34.4.

WHY DIDN’T ST BASIL WRITE IN NEW TESTAMENT GREEK?

285

KOINE GREEK AND ATTICISM Why is St Basil’s Greek so different from what we read in the NT? The short answer is: St Basil was an educated man. But what did that mean in the fourth century AD? This is what I want to explore next. To understand what it meant we need to go back, well beyond St Basil’s time, all the way back to Classical Athens. As is well known, there was a great flowering of Greek culture in Athens of the Classical age. Art, science, drama, historiography, philosophy, oratory – everything – reached a pinnacle in that one small πόλις in the fifth and fourth centuries BC. When one looked back from a later time in the ancient world, the achievements seemed unsurpassed, and indeed unsurpassable. Even today it seems so in many respects. The language of the texts of that period was the dialect of Athens, that is, primarily Attic. Other dialects and styles were on the scene, especially poetic genres; for example, Homer, the foundation text for all study of literature, was in Ionic dialect combined with other elements. But Attic was the primary vehicle of the literature of Classical Athens. When Alexander’s conquests carried Greek and Greek culture all over the Eastern Mediterranean at the end of the fourth century BC, a new phase of the language had already begun to emerge. This was Koine Greek, based on Attic, but with numerous differences that were characteristic of the vernacular. Koine Greek became the spoken language wherever Greek was used, all over the area of Alexander’s empire. At this time, works of literature could be written in a literary Koine, without adhering to strict Attic or attempting to follow Classical models. But as time went on, a new outlook arose.24 This was the tendency to view the early post-Classical period as one of decline in Greek language and culture, and to look back to the golden age for inspiration and for models to imitate, not only in literature but in everything else – art, architecture, and so on. It became a matter of importance not only for Greeks but for Romans, the new masters of the ancient world, who looked to the Greeks for education and culture. This tendency emerged in the first century BC and led to the phenomenon of ‘Atticism’ in language, that is, the attempt to revive Attic of the fourth century BC as the instrument of good writing. The great classics of the literature of Athens in the fourth and fifth centuries BC became the models, and Greek with Attic features came to 24

On the subject of this paragraph, see Browning, Greek, 44–50; Kazazis, ‘Atticism,’ 1200– 17, and esp. 1200-6; Lee, ‘Atticist Grammarians,’ 283–5, with further references.

286

COLLECTED ESSAYS ON THE GREEK BIBLE

be taught and learnt in the schools, as opposed to Koine Greek, the spoken language. The two of course are not completely separate, but overlap. What Atticism involved, however, was trying to revive and maintain Attic usage wherever the spoken language had moved on and changed. It was essentially a purifying process, and it must be admitted, artificial. To take a simple example: a popular word for ‘bed,’ κράβαττος, appeared in early Koine Greek, but Atticists recommended avoiding it because it had no Classical precedent. In the NT it is found several times, but Matthew and Luke, influenced by the puristic tendency that they had been introduced to in school, remove Mark’s examples or replace them (with κλίνη or κλινίδιον).25 This is only a brief glimpse of the history of the period, but it gives us the background we need to appreciate the character of ancient education, our next topic.

AN

ANCIENT EDUCATION

Let us turn now to the education system in antiquity, especially in the Greco-Roman world. What was it like? It is easy for moderns to be unaware that there was any such thing, still less that it was highly developed and systematic. This is partly because we have never had reason to give the matter any thought; but it is also because everything before a couple of centuries ago is often imagined as primitive, enveloped in a sort of dark age. This of course is entirely mistaken. Education already existed in the Classical period and developed further in the Greco-Roman period; by the time we reach St Basil it had been in place for centuries, and was an organised and confident system, sure of its goals and methods and its role in training elites. (It was, it may be added, not a public, free education system such as we know today, but was available only to those who could afford to pay for it; even so, most people learnt at least the rudiments.) The full program involved three stages or levels.26 Stage one, naturally, aimed at teaching the pupil to read and write. To learn the rudiments, recitation, not just of the alphabet, but of syllables combining all the consonants and vowels, was a staple technique. Fragments of Classical authors, 25

26

Cf. the anecdote in Browning, Greek, 47 n. 34. [Bp Triphyllios of Ledra is said to have been rebuked by St Spyridon for replacing κράβαττος with σκίμπους in a Gospel reading (Mark 2:11 / John 5:8). For σκίμπους see Phryn. etc. in Essay 19, §8.] See Cribiore, Gymnastics of the Mind, 2–3, 160–244; Whitmarsh, Second Sophistic, 13– 5. For a full study see Morgan, Literate Education.

WHY DIDN’T ST BASIL WRITE IN NEW TESTAMENT GREEK?

287

especially Homer, were introduced even at this stage; they were copied out as exercises and learnt by heart. So a start was made on Classical Greek literature here. At stage two the curriculum covered γραμματική (τέχνη), loosely equivalent to our ‘grammar,’ though embracing more than what we mean by that. It involved analysis of texts, with close study of details of language – vocabulary, grammar, etymology, and so on. Letter-writing, regarded as an important skill, was introduced and practised here. There was further study of the poets and the prose authors, the ‘Classics,’ as models of expression and sources of themes. Most students would not advance even to this stage, but there was one more beyond it. Stage three was the equivalent of our tertiary level today, but based on different premises. The training offered here brought the student to the pinnacle.27 To get into stage three there was a sort of entrance exam, as Libanius tells us.28 As well as learning to write in the best style, the student learnt the art of rhetorical performance, which was just as important. The aim was to train the student in Classical models so well that he could offer a complete oration (μελέτη) in a similar style. This skill was regarded as ‘the crowning achievement of the student of rhetoric.’29 To get to that point, the student advanced through a series of increasingly demanding and complex exercises, called προγυμνάσματα. In these he was expected to read, copy, and memorise large portions of text, and to immerse himself in the works of orators, poets, and historians of the Classical age. Finally, after years of intense practice, he would be able to create his own display, weaving together elements he had learnt and adding his own. The full course lasted five or six years, though not necessarily all stayed the distance. Many would not have felt the need to: it is said that after two years the student could have acquired enough skill to plead in court and take up legal advocacy as a profession.30 We know that there were public displays of oratory before an audience, sometimes before the Roman Emperor himself. The top performers, the ‘Sophists,’ as they were known then, attracted large audiences. We have descriptions of these performances.31 They were clearly highly-charged occasions, demanding the utmost of the speaker’s skill and training. The speech was often delivered on a theme proposed by the audience on the 27 28 29 30 31

Whitmarsh, Second Sophistic, esp. ch. 2; Cribiore, Gymnastics of the Mind, 220–44. Cribiore, Gymnastics of the Mind, 224. Cribiore, Gymnastics of the Mind, 223; cf. Whitmarsh, Second Sophistic, 20–1. Cribiore, Gymnastics of the Mind, 224. E.g., in Philostratus, Lives of the Sophists: see Whitmarsh, Second Sophistic, 27–9.

288

COLLECTED ESSAYS ON THE GREEK BIBLE

day.32 Thus the speaker at this level had not only to compose, on the spot, an oration displaying his familiarity with Classical texts and themes, but also to be careful not to make any slips in purity of Attic in his own speech. The audience would be keeping a close watch on every word, and failure to impress was a real possibility. But success brought renown. It was towards this ultimate prize that the third stage of education led, for those who could reach it.

ST BASIL’S

EDUCATION

What we have described is indeed the course of training that St Basil had been through. A good deal is known about his education and career. He was born in Caesarea in Cappadocia, and received his elementary training from his own father, himself a distinguished rhetorician. He then attended schools in Caesarea, Constantinople, and finally the great ‘University’ of Athens. It was not possible to go higher than that. Athens the city was still at this time well preserved and the centre of the ancient inheritance of literature, philosophy, and culture. It was, as Deferrari puts it, ‘the pattern of excellence to a world that elected to see in Atticism the cultural ideal.’33 It seems certain St Basil was acquainted with the greatest pagan rhetorician of the fourth century, Libanius, whom we have mentioned before, though whether he actually studied under him (in Constantinople) is less certain.34 We do know that in Athens one of his teachers was Prohaeresius (276–366 AD), who held the chair of rhetoric and was another celebrated rhetorician.35 All this is important, because the school you attended and the quality of your teachers mattered – just as it does today. Finally, St Basil himself came to hold a ‘chair of rhetoric,’ that is, a chair in pagan literature and learning, in his home town of Caesarea. It was only after all this rigorous training, and the career culminating in the achievement of a chair of rhetoric, that St Basil turned to the Church and was ordained. Then began a different life, his celebrated career in the 32

33 34

35

It was rather like a great organist improvising on a theme given to him or her just before the performance. Deferrari, Basil: Letters 1–58, xviii. Wilson, Saint Basil, 52; Basil: Letters 1–58, xvii–xviii. A large quantity of Libanius’s writings survives: 64 orations, 51 declamations, 1544 letters, a Progymnasmata, plus lesser works (TLG Canon, 245). Joyal, et al., Greek and Roman Education, 257–8. Stasinopoulos (Μορφές, 99) says he was also a Christian.

WHY DIDN’T ST BASIL WRITE IN NEW TESTAMENT GREEK?

289

Church, during which he established his famous monastic Rule, became Archbishop of Caesarea, and so on. In doing that, he left behind the world of pagan learning – but he did not forget what he had learnt. A fitting conclusion to this section is the statement that can be found in the Menaion on St Basil’s feastday (1 January), in the short biography recorded there: Ἐν δὲ λόγοις, οὐ μόνον τοὺς καθ᾽ αὑτόν, ἀλλὰ καὶ τοὺς παλαιοὺς ὑπερέβαλε· διὰ γὰρ πάσης ἐλθὼν ἰδέας παιδεύσεως, ἐν ἑκάστῃ τὸ κράτος ἐκτήσατο. In Classical learning he surpassed not only his contemporaries but even the ancients; for passing through every kind of training, he acquired mastery in each.36

This sentence sums up the first phase of his life succinctly and perfectly, and ‘every kind of training’ fits exactly the educational curriculum of the time. The author of this summary knew his subject – and also how to write the sort of Greek St Basil himself would have been comfortable with. ST BASIL’S CHOICE We can see now ‘where St Basil was coming from’ – to use an expression of today – and we have come to the answer to our question: St Basil writes as he does because that was the way an educated man naturally wrote in his time. The choice was as it were predetermined. That was especially the case for one who had been educated right up to the top level. He used, to the full extent of his ability, the Atticising Greek and style that he had learnt and that was the status form of his time. It was the style expected of him by his contemporaries, and it would have been difficult for him to abandon it. This does not mean he could not speak everyday Greek: Attic Greek was learnt in the schoolroom, not the nursery. It also does not mean that he was incapable of preaching and writing in everyday Greek if he chose. 36

Μηναῖον τοῦ Ἰανουαρίου (Ἐν Ἀθήναις· Ἀποστολικὴ Διακονία, 1991), 24. The meaning of λόγοι here can be demonstrated from other examples, some in St Basil himself: see Deferrari, Basil: Letters 337, 338; To Young Men, title (see n. 3); 4.8 (quoted below); 5.1; see also Lampe, PGL, s.v. λόγος I. A.13. It is clear that (οἱ) λόγοι (pl.) refers to learning in Classical pagan literature, sometimes specifically to the literature itself, as in 4.8, 5.1. It is easy, as Rousseau does (Basil of Caesarea, 41, 49), to miss the full significance of the term and translate simply as ‘words’ or ‘rhetoric.’ In To Young Men, 10.4 St Basil refers by way of contrast to οἱ ἡμέτεροι λόγοι, ‘our (Christian) literature,’ and to ἱεροὶ λόγοι (2.5, quoted below).

290

COLLECTED ESSAYS ON THE GREEK BIBLE

The authenticity of certain of St Basil’s works has been called into question on this very issue. The two homilies On the Origin of Humanity 1 and 2 (Περὶ τῆς τοῦ ἀνθρώπου κατασκευῆς) are written in a style that differs noticeably from St Basil’s other writing, and specifically from the other nine homilies of the Hexaemeron, with which these two seem to belong.37 The style ‘lacks the refinement and clear, balanced syntax of his usual rhetorical prose’ (Harrison). Some take this as an indication that they were not St Basil’s work; others that he did compose them but did not have time to polish them for publication. While it is not intended to resolve the issue here, it is worth noticing the possibility, as Fr Doru Costache suggests to me, that St Basil intentionally composed them in that style for delivery to a less educated audience; the question of polishing was separate, and he may not have had the time or inclination to attend to it. We may thus have a glimpse in these works of the simpler Greek of which St Basil was capable in the appropriate setting.38 So St Basil writes in the language and style of the great pagan Classics and rejects the simple Koine Greek of the NT as the vehicle for his published teaching and thoughts. But there is a paradox at the heart of this choice, one already hinted at in St Basil’s remark on the biblical texts quoted above (‘in sense true, but in style unlearned’). The basis of faith and moral guidance for St Basil, as for all Christians, is the NT and the Scriptures (which means the OT). The pagan Classics are not without value in this respect, as the work To Young Men is written to demonstrate, but only where they have good to offer: a higher truth governs what may be selected from them as valuable.39 As St Basil himself puts it (2.5): ‘To this [other, that is, eternal, life] the Holy Scriptures lead the way, teaching us through secrets’ (εἰς δὴ τοῦτον ἄγουσι μὲν ἱεροὶ λόγοι, δι᾽ ἀπορρήτων ἡμᾶς ἐκπαιδεύοντες). Yet despite this, St Basil’s chosen linguistic vehicle for his writing is not biblical Greek but the elevated language of the pagan Classics. He cannot envisage a writing style based on the NT (still less on the LXX), or an education system having only the Christian texts as its models.40 37

38

39 40

Harrison, Saint Basil the Great, 14–5; Rousseau, Basil of Caesarea, 318 n. 1. I am grateful to Fr Doru Costache for drawing this issue to my attention and for assistance with bibliography. While the style is certainly simpler, it seems to me the language remains Atticising and rather demanding for the uneducated listener. If the story in Browning, Greek, 50 can be trusted, St John Chrysostom once demonstrated his capacity to preach in everyday Greek when called upon to do so. See To Young Men 1.5, 4.6–7; Quasten, Patrology, 3:214; Jaeger, Early Christianity, 81. On this last point cf. Wilson, Saint Basil, 9; Jaeger, Early Christianity, 83–4; Louth in Young, et al., Cambridge History, 294: St Basil ‘clearly envisages no alternative way

WHY DIDN’T ST BASIL WRITE IN NEW TESTAMENT GREEK?

291

CONCLUSION I end with a paragraph from the work with which we began. In this passage we see the paradox played out, as it were, as St Basil makes his point about selective use of Classical literature while at the same time demonstrating his mastery of both the content of that literature and its linguistic form. Κατὰ πᾶσαν δὴ οὖν τῶν μελιττῶν τὴν εἰκόνα τῶν λόγων ἡμῖν μεθεκτέον. ἐκεῖναί τε γὰρ οὔτε ἅπασι τοῖς ἄνθεσι παραπλησίως ἐπέρχονται, οὔτε μὴν οἷς ἐπιπτῶσιν ὅλα φέρειν ἐπιχειροῦσιν, ἀλλ᾽ ὅσον αὐτῶν ἐπιτήδειον πρὸς τὴν ἐργασίαν λαβοῦσαι, τὸ λοιπὸν χαίρειν ἀφῆκαν· ἡμεῖς τε, ἢν σωφρονῶμεν, ὅσον οἰκεῖον ἡμῖν καὶ συγγενὲς τῇ ἀληθείᾳ παρ᾽ αὐτῶν κομισάμενοι, ὑπερβησόμεθα τὸ λειπόμενον.41 We should, then, partake of pagan literature (λόγοι) exactly in the manner of the bees: for they do not approach all the flowers equally, nor do they try to carry off the whole of those on which they land, but taking as much of them as is suitable for their work, they bid the rest goodbye; so we, if we are wise, receiving from them (i.e., pagan literature) as much as is suitable (οἰκεῖον) to us and related (συγγενές) to the truth, will pass over the remainder.

The simile of the bee is itself not new but a commonplace going back to Isocrates, and the passage ‘bears a close resemblance’ to one in Plutarch.42 In vocabulary we see at one point a choice, surely intentional, of words with suggestive dual meanings: οἰκεῖον can mean ‘suitable, proper’ but also ‘of one’s own family, household’; similarly συγγενές has both a figurative sense (‘related’), as well as a literal one (‘of the same kin, family’). Thus is implied that the truth by which the pagan Classics are to be measured is the truth found within the Christian family. The passage is replete with the expected linguistic features marking it as high style based on Attic models: the Attic -ττ- in μελιττῶν; a ‘gnomic aorist’ in ἀφῆκαν; controlled use of particles, δὴ οὖν, τε, γάρ, μήν; Ionic/Old Attic ἤν (= ἐάν); and the verbal adjective μεθεκτέον, a rarity in everyday language. Finally, the sentences exhibit a flow and balance (every sentence-end and

41

42

of learning Greek than using pagan literature.’ This topic is connected with the issue of tension between ‘two sets of standards’ in St Basil’s thought over time, explored at length by Rousseau (Basil of Caesarea, 40–57). Rousseau goes so far as to conclude that ‘Basil consistently rejected, probably from the earliest period following his departure from Athens, the claims of the traditional curriculum’ (56). To Young Men 4.8. Text from Wilson, Saint Basil. My translation, based on Deferrari and McGuire, Basil: To Young Men, pp. 391–3. Wilson, Saint Basil, 48; Deferrari and McGuire, Basil: To Young Men, 391 n. 2; Stasinopoulos, Μορφές, 280.

292

COLLECTED ESSAYS ON THE GREEK BIBLE

most clause-ends obey the rule of clausulae), that achieves a pleasing effect without showing quite how it is done. Thus St Basil puts on display his omnivorous training in Classical literature at the same time as he cautions his young readers not to partake of all of its flowers.

19 THE ATTICIST GRAMMARIANS 2013

Abstract The first aim of this essay is to introduce the Atticist grammarians and explain their purpose; how they present their material; and how they can be used as evidence for the history of Greek. The point is made that their assertions are not statements of fact but require interpretation. The Atticists (or works) covered are: Phrynichus, Moeris, the Antiatticista, Aelius Dionysius, the Philetaerus, Thomas Magister, Ammonius, and the Onomasticon of Pollux. The discussion then turns to the relevance of the Atticist grammarians to the NT. The grammarians’ evidence helps to prove that the language of the NT was by and large the Koine Greek of their day. There is, however, some influence of Atticism on the NT, mainly in Luke–Acts and the Epistles. Specimens are presented. Finally, the phenomenon of Atticising variants in the MS tradition and the questions raised for NT textual criticism are briefly outlined.

1. ATTICISM Atticism was a complex phenomenon that has been described and explained in numerous ways, but its essential element was the tendency to look back to the language and literature of a former era as the model to follow, from a later time when the spoken language had changed and original composition of that literature was in the past. The former era was the Classical period, and access to it was through its texts, which were studied and imitated in the education system. The later time was almost the whole of the post-Classical period: the phenomenon appeared in the first century BC, reached a peak in the second century AD, and continued its influence through the Byzantine era, with effects still today. The results were seen not only in writing but in the shape of the language as a whole, that is, in a differentiation of the spoken and written varieties, or a ‘diglossy/ diglossia,’ that affected Greek for the rest of its history.1 1

For an accurate definition of the term (too long to repeat here), see Ferguson’s in Hudson, Sociolinguistics, 54. I am grateful to Michael Curran for reading this paper and offering many useful comments.

294

COLLECTED ESSAYS ON THE GREEK BIBLE

Atticism was not simply a linguistic phenomenon but part of a larger enterprise to recover the Classical past. The cultural world of fifth and fourth century BC Athens was to be recreated, in literature, rhetoric, the arts, and philosophy. Powerful cultural and social forces fostered the enterprise, and even emperors supported it. The whole was an exercise in imitation, but it had a deep and enduring – some would say disastrous – impact.2 The attempt to turn the clock back created complexity for users of the language, especially when they wished to commit anything to writing, or, at a more advanced level, engage in rhetorical display. Anyone who had been to school had been introduced to the notion that some forms, words, and uses were ‘better’ or more ‘correct’ than others. They were the ones that matched the Classical models, contrasting with new features that had arisen as a result of change in the spoken language. The contemporary spoken language was disparaged, the older, obsolete features were approved and accorded prestige. The need for guidance in finding one’s way through these artificial shoals soon led to the creation of works offering instruction in how to do it. The authors of these works were the Atticist grammarians.3 There is a long list of names of Atticist grammarians, though most of their works survive today only in fragments. The effects of their efforts, however, and of the whole movement to treat the Greek of the past as the model of good Greek, are to be seen in almost all the written remains of post-Classical Greek. Most works of literature from that period exhibit artificial Atticising features to varying degrees, and even the lower levels of everyday writing display some influence.4 Although the Atticising movement had marked success, and writers at the top of the scale could write a whole work that appeared to reproduce the Attic of fifth-century Athens, this Atticising Greek was not a spoken language or dialect separate from Koine Greek; it was a stylistic variety, or rather group of varieties, added on to the living language, Koine Greek. 2

3 4

The literature is extensive. Recommended are: Anderson, Second Sophistic, esp. 86– 100; Bowie, ‘Greeks and Their Past’; Browning, Greek, 44–50, 104–13; Horrocks, Greek, 78–86, 151–3; Kazazis, ‘Atticism’; Swain, Hellenism and Empire, esp. 17–41; Whitmarsh, Second Sophistic; Cribiore, Gymnastics of the Mind, esp. ch. 8. It is worth noting that ‘Asianism,’ against which Atticism was supposedly a reaction, is largely a fiction: see Whitmarsh, Second Sophistic, 7–8, 50–1. The epithet ‘grammarian’ is not quite appropriate, but it is hard to find a better alternative. This last topic has hardly been touched. There are a limited observations in Mandilaras, Verb (mostly on the optative); Schmid, Atticismus, 3:19–33 passim; Kotsevalov, ‘Koine Syntax.’ The only extensive study known to me is an unpublished work: Connolly, ‘Atticism in Non-literary Papyri’ (see bibliography).

THE ATTICIST GRAMMARIANS

295

The latter was the mainstream that went on being spoken and evolving over the course of nine centuries, from 300 BC to 600 AD, before passing into the next phase termed Byzantine Greek. Atticism is to be viewed as one constituent of the wide range of variation of which Koine Greek was capable.5

2. THE ATTICIST GRAMMARIANS This essay focuses on a selection of three Atticist grammarians whose works survive in substantial remnants, Phrynichus, Moeris, and the Antiatticista. The aim is to introduce them, with specimens of their work, and to show how they can be used as evidence for the progress of the language and for the linguistic character of a text such as the NT. We begin with some general observations about them. The aim of the Atticist grammarians was to identify what was acceptable ‘Attic.’ Such an aim always had a corollary, that something else was not acceptable, whether this was stated openly or merely implied. One feature was approved, another condemned. Since their purpose was to revive features that were no longer alive and did not come naturally to speakers of the language, it follows that the condemned feature was in fact the live one used in the language of the day; the recommended one was the feature that was obsolete. Evidence from the texts of the time, when brought to bear, confirms this. We can, then, as a general principle, take the Atticists’ statements as good information about what was actually spoken Koine Greek, though this was not their intention.6 Their opinions about what was good Attic are another story. For various reasons they vary in reliability: they always need to be checked against the evidence of the texts. The features favoured by the Atticists are what they thought was good Attic of the fifth century BC, rather than what actually was. Furthermore, the Atticists had their own ideas about which authors provided models of good Attic. Not all the authors of the Classical period were accepted by any means. Controversies developed between Atticists over the choice of authors. The strictest Atticists worked with a limited canon of approved writers, regarding the rest as below standard and not to be imitated. They even went so far as to correct or rebuke an ancient 5

6

Cf. Kazazis, ‘Atticism,’ 1208–9; Horrocks, Greek, 81; Frösén, Prolegomena, 99–100, 177–8, 229. In Hatzidakis, Einleitung, 285–303 numerous features condemned by the Atticists are shown to descend into the modern language.

296

COLLECTED ESSAYS ON THE GREEK BIBLE

author for lapsing into a use that they did not accept. They also mixed the evidence of prose and poetry, and of authors of widely varying dates and styles, so that their picture of Classical Attic prose usage was, overall, a distorted amalgam.7 Many of the features that they focus on seem trivial in the extreme, but this is not surprising in a movement requiring great attention to detail if error was to be avoided; everything written and spoken could be subjected to pedantic scrutiny by others. On the other hand, we find no remark in the Atticists on many of the changes between Classical and Koine Greek. This may be partly because the remains of their works are not complete; but some features, especially broader trends, simply slipped under their radar.8 A key point about use of the Atticists’ material is that their assertions are not to be taken at face value as objective statements of fact; they require interpretation. They were made with a purpose, and the purpose was prescriptive rather than descriptive. Interpretation needs to take into account two things: the purpose of the Atticist grammarian and the evidence of usage. The discussion that follows is intended to illustrate these in practice. Since the NT is an important specimen of Koine Greek, the illustrations often involve it in some way. The history of the features over the course of time, from Classical to Koine Greek and beyond, is always part of the picture also.

3. A FIRST SPECIMEN Let us consider an example now, as a general illustration of what has been said so far. In one of the statements by the Atticist Phrynichus (of whom more shortly), the concern is over so small a matter as the form and gender of the word for ‘flea.’ Phrynichus says: ψύλλος βάρβαρον, ἡ δὲ ψύλλα δόκιμον ὅτι καὶ ἀρχαῖον. [The form] ψύλλος (masc.) is barbarous; but ψύλλα (fem.) is approved because it is also ancient. (Phryn. 306)

The assertion that the masculine form is ‘barbarous’ is of course not a statement of fact, but an opinion, a derogatory description of something 7

8

Cf. Anderson, Second Sophistic, 88; Kazazis, ‘Atticism,’ 1206. To be fair to Phrynichus, he was well aware of the ancient dialects and that Herodotus, for instance, has Ionic features that are not Attic: see, e.g., Phryn. 235. For example, the long-term change from a three-voice to a two-voice system in the verb, with absorption of the middle functions into the active; see, e.g., Browning, Greek, 30.

THE ATTICIST GRAMMARIANS

297

that aroused the Atticist’s distaste. The approved form is the feminine, which is described as ‘ancient,’ a not very exact term, but evidently referring to Attic Greek of the Classical period. The Atticist instructs us to avoid the masculine form and use the feminine. The masculine must of course have been in use in his time, otherwise there would have been no reason to condemn it. This much we can deduce from the statement itself. We also have a parallel statement in Moeris (ψ 2), saying the same thing more baldly: ψύλλα θηλυκῶς Ἀττικῶς. ἀρσενικῶς Ἕλληνες. The evidence of actual usage in surviving texts, incomplete though of course it is, fills out the picture and gives us food for thought. It is true that ψύλλα is the original formation and attested early in Aristophanes and Xenophon, then in Aristotle.9 But according to the Suda, ψύλλος was used by Epicharmus, a Sicilian comic poet older than Aristophanes.10 There is also an example in Aristotle (HA 537a.6, referring to a sea-flea). The next appearance of ψύλλος is in the LXX, at 1 Kgdms 24:15, a useful piece of evidence; further, we have evidence of Ψύλλος as a name in Herodotus and Menander and in documents from the third century BC onwards.11 The form approved by the Atticists, ψύλλα, is common in literary sources from the second century AD onwards.12 Finally the modern language comes in with unequivocal evidence of ψύλλος as the normal spoken form.13 What does all this tell us? We can confidently say that the original form ψύλλα had competition as early as the Classical period from the masculine ψύλλος, which may have been in Attic already in the time of Aristotle; it was well established by the early Koine (cf. the LXX) and continued to be used in the spoken language from then until modern times; the Atticists revived ψύλλα and caused it to be used artificially by educated writers of their time; it then persisted in puristic writing for centuries. The actual attestation of the words in this case is meagre, but we can do a lot with it. What we are doing is reconstructing the full history of the forms from whatever clues and evidence we have. The statements of the Atticists are, or may be, useful clues to the general direction, even if they are not fully accurate as to the facts. 9 10 11

12

13

See Chantraine, DELG, s.v.; LSJ, s.v. Suda ψ 152 (Adler); ‘καὶ ἕτεροι,’ says the Suda. Hdt. 4.173; Men. 272 (Körte); Fraser, LGPN, IIIA, s.v. The fem. Ψύλλα is known from Lycophron (IV BC) onwards. A TLG search produces over 200 occurrences, from Lucian and Galen in II AD till well into the Byzantine period. See, e.g., Lex. Proia, s.v. The fem. (no surprise!) is retained in puristic Greek.

298

COLLECTED ESSAYS ON THE GREEK BIBLE

4. PHRYNICHUS Phrynichus was a famous strict Atticist who lived in the second century AD. His major work, the Ecloga, can be dated to 178 AD. He was born, not in some part of old Greece such as Athens, as one might have thought, but in Bithynia, in NW Asia Minor. He is said to have been a competitor for the chair of rhetoric at Athens, against his successful rival Pollux. His birthplace and career are a good reflection of the cultural landscape of the Greco-Roman world.14 The standard up-to-date edition of the Ecloga is that of Fischer,15 but the 1820 edition of the fine scholar Lobeck, which also incorporates his predecessors’ observations, retains much value.16 The later edition of Rutherford is still useful at times, though he adopts the role of a sort of latter-day Atticist.17 Another work of Phrynichus, called the Praeparatio sophistica (Σοφιστικὴ Προπαρασκευή), survives in a sizeable epitome. The current edition is that of de Borries;18 the text can also be found in Bekker’s Anecdota.19 Its purpose is much the same as that of the Ecloga, but the material more specialised. Often the aim is to suggest or explain a useful Attic expression, or give the tick of approval to some feature, without necessarily naming any equivalent to be avoided. The Ecloga is primarily a list for the budding Atticist of what to avoid and what to use instead, mostly with the former first, but sometimes the reverse. Most of the entries are quite short, consisting of a line or two, though a few run to five or more. The material is unordered, with only traces of alphabetical order. As to length, there are 424 lemmata in Fischer’s edition. The simplest type of entry is: δυσὶ μὴ λέγε, ἀλλὰ δυοῖν. Do not say δυσί [dat. plural] but δυοῖν [dat. dual]. (Phryn. 180)

Here we see Phrynichus’s typical peremptory style and the extreme nature of the Atticist program. He tries to revive the use of the dual, which 14

15

16 17

18 19

On Phrynichus see Strout and French, ‘Phrynichos’; OCD, s.v.; Dickey, Scholarship, 96–7. Photius gives Phrynichus the epithet ‘Arabius,’ the import of which is unclear. Fischer, Die Ekloge des Phrynichos. Fischer’s introduction (37–50) deals fully with the questions of date, sources, and whether the Ecloga is abridged (no, says F., 37). Lobeck, Phrynichi Eclogae. Rutherford, The New Phrynichus. For his attitude see, e.g., ix–x (Phrynichus has ‘scholarly nerve and wholesome masculine common sense’), 115, 144, 203. De Borries, Phrynichi Sophistae Praeparatio Sophistica. Bekker, Anecdota Graeca, 1:1–74.

THE ATTICIST GRAMMARIANS

299

even in the word for ‘two’ had begun to be replaced before the Koine period, and was obsolescent throughout the noun and verb morphology of Attic by the fourth century BC. The Atticists’ Attic was the most conservative variety, more conservative than Attic itself by the end of the Classical period.20 Like this one, many entries deal with morphology; just as many condemn a word entirely, for the simple reason that it cannot be found in a Classical author. Even the occurrence of a related form is not enough. In the following example Phrynichus concedes that the Attic writers do use the verb, but the noun is not found and is therefore not acceptable: προκόπτειν λέγουσιν, τὸ δὲ ὄνομα προκοπὴ οὐκ ἔστιν παρ᾽ αὐτοῖς. They [the ancient authors] say προκόπτειν (‘to make progress’), but the noun προκοπή (‘progress’) is not found in them. (Phryn. 58; cf. Philet. 84)

Particular uses of words are also freely condemned for the same reason, though a less rigid observer might have seen that the extension is natural and could have occurred at any time. So in the next example, in which a nicety of usage presents an inviting target for the pedant: τέμαχος κρέως ἢ πλακοῦντος ἢ ἄρτου οὐκ ὀρθῶς ἐρεῖ τις, ἀλλὰ τόμος κρέως ἢ πλακοῦντος· τὸ δὲ τέμαχος μόνον ἐπὶ ἰχθύος λέγεται. τέμαχος (‘slice’) of meat or cake or bread is not correct for one to say, but τόμος (‘piece’) of meat or cake; τέμαχος is said only of fish. (Phryn. 12)

Another feature is that Phrynichus does not accept just any occurrence in a Classical author as proof of Attic status; he sets the bar high: the author must be one of the select group of which he approves. This is why we sometimes find him declaring a feature unacceptable even if someone has pointed out an example in an ancient author, as in: ἀκμὴν ἀντὶ τοῦ ἔτι· Ξενοφῶντα λέγουσιν ἅπαξ αὐτῷ κεχρῆσθαι, σὺ δὲ φυλάττου χρῆσθαι, λέγε δὲ ἔτι. ἀκμήν (‘still’) instead of ἔτι: they say Xenophon once uses it, but you beware of using it and say ἔτι. (Phryn. 93; cf. Antiatt. 77.27; Moeris α 149)

Here Phrynichus rejects the authority of Xenophon, who frustrated the Atticists because he refused to fit the mould of an Attic writer.21 When 20 21

See LSJ for attestation of the two forms. The NT has only δυσί (× 9). Cf. Phryn. 62, where Xenophon is accused of an outrage against his πάτριον διάλεκτον in saying ὀδμή instead of ὀσμή. But Antiatt. 92.26 counter-attacks by calling him ὁ καλὸς Ξενοφῶν.

300

COLLECTED ESSAYS ON THE GREEK BIBLE

he says ‘they say’ he is referring to the work of other Atticists, less strict than himself, who sought to legitimise this use on the basis of an occurrence in Xenophon (cf. below on the Antiatticist). This use of ἀκμήν as an adverb meaning ‘still’ was well established in Koine Greek, from the time of Polybius on; there is an example in the NT at Matt 15:16. In the modern language ακόμα/ακόμη is its descendant. But Phrynichus and others oversimplified the ancient evidence. As Lobeck showed, in the Xenophon example (Anab. 4.3.26) and elsewhere, ἀκμήν has the sense of ‘just then,’ a step before the shift to ‘still.’ So in genuine Attic it is not simply a matter of avoiding ἀκμήν and replacing it with ἔτι.22 Phrynichus may even rebuke an ancient writer for not being more careful: ἐμπυρισμός· οὕτως Ὑπερείδης ἠμελημένως, δέον ἐμπρησμὸς λέγειν. ἐμπυρισμός (‘burning’): so Hyperides carelessly, when one ought to say ἐμπρησμός. (Phryn. 311; cf. Antiatt. 97)

Menander comes in for a good deal of disapproval of the same kind (he is mentioned 17 times in the Ecloga). Sometimes the remarks suggest that Phrynichus actually thought of the ancient writers as guided by a scale of acceptability just like his own: αἰχμαλωτισθῆναι· τοῦθ᾽ οὕτως ἀδόκιμον ὡς μηδὲ Μένανδρον αὐτῷ χρήσασθαι. διαλύων οὖν λέγε αἰχμάλωτον γενέσθαι. αἰχμαλωτισθῆναι (‘to be captured’): this is so unacceptable that not even Menander uses it. Separate the words, therefore, and say αἰχμάλωτον γενέσθαι (‘to become captive’). (Phryn. 411)

Menander, it seems, might have been tempted to use this verb, but ‘even’ he, who set such low standards, knew it was beyond the pale and avoided it. A certain amount of background knowledge is often necessary to appreciate an Atticist prescription fully, as in: ἔνδον εἰσέρχομαι βάρβαρον· ἔνδον γάρ ἐστι καὶ ἔνδον εἰμὶ δόκιμον. δεῖ οὖν εἴσω παρέρχομαι λέγειν. ἔσω δὲ διατρίβω οὐκ ἐρεῖς, ἀλλ᾽ ἔνδον διατρίβω. ‘I go ἔνδον’ (‘within’) is barbarous; for ‘he is ἔνδον’ and ‘I am ἔνδον’ are approved. One must say ‘I pass εἴσω’ (‘inside’). But you shall not say ‘I spend time ἔσω,’ but ‘I spend time ἔνδον.’ (Phryn. 99; cf. Antiatt. 91.31; Ammon. 169)23 22

23

Lobeck, Phrynichi Ecloga, 123–4. See further Shipp, MGE, 51–2, with more on the Modern Greek forms. [It is possible that ἔσω δὲ διατρίβω … is a separate lemma: see Fischer, Ekloge, app. crit.]

THE ATTICIST GRAMMARIANS

301

The need to prescribe the correct use of ἔνδον and εἴσω arose out of a general trend of the language, the weakening of the distinction between rest and motion in adverbs and prepositions. This had already begun in the Classical period and gathered pace in Koine Greek. Thus ἔνδον originally applied to rest ‘in’ a place, εἴσω to motion ‘into’; but the difference gradually ceased to be felt and the words were often interchanged.24 Phrynichus and the other Atticists tackle the trend piecemeal rather than in a systematic way. Another instance they notice is ποῖ versus ποῦ,25 and Ammonius was trying to counteract the same trend with his distinction between ἐκεῖ (‘there’) and ἐκεῖσε (‘thither’).26 But the wide reach of the phenomenon and many other possible instances are not noticed by the Atticists, at least in their surviving works, as, e.g., the running together of εἰς and ἐν, which had the end result that (ει)σ- serves for both ‘in’ and ‘into’ in Modern Greek. Finally, an example that illustrates the hazards of interpreting the Atticists’ material and how easy it is for faulty argumentation to arise: αὐθέντης μηδέποτε χρήσῃ ἐπὶ τοῦ δεσπότου, ὡς οἱ περὶ τὰ δικαστήρια ῥήτορες, ἀλλ᾽ ἐπὶ τοῦ αὐτόχειρος φονέως. Never use αὐθέντης for the master, as the orators at the lawcourts do, but for the murderer with his own hand. (Phryn. 89)

This word occurs in the Greek book of Wisdom at 12:6, with the meaning ‘murderer’ (αὐθέντας γονεῖς ψυχῶν ἀβοηθήτων, ‘parents [who are] murderers of defenceless souls’). Winston in his commentary sketches the attestation of the word and its meanings from ‘murderer’ (in Hdt. etc.) through ‘perpetrator, author’ (in ‘late prose’) to ‘master’ (in Hermas and Pap.Graec.Mag.), noting that the last usage was condemned by Phrynichus. He goes on: A. Dihle has deduced from this that Wisdom was written in the first century CE: ‘My point is that in this book the word authentês is used in the meaning of ‘murderer,’ and this is the Attic meaning of the word which no Hellenistic writer would have applied in the first century BCE.’27

Is this a valid deduction? I do not think so. It assumes that the meaning ‘murderer’ disappeared altogether and was not known to anyone until it was revived in the first or second century AD by the Atticists. This is 24

25 26 27

[See Lobeck, Phrynichi Ecloga, 128 for Class. evidence of εἴσω/ἔσω for ἔνδον, none of which (he says) ‘deterred’ (absterruit) the Atticists.] Phryn. 28; Moeris π 49; Ammon. 392; Philet. 271, 272. Ammon. 159; also ἐνταυθοῖ etc. at 170; cf. Schmid, Atticismus, 1:91. Winston, Wisdom of Solomon, 240–1 (with ref. to source of Dihle’s statement).

302

COLLECTED ESSAYS ON THE GREEK BIBLE

an improbable scenario. Old uses, like old words, may reappear at any time in Greek. They therefore cannot be used to establish a date post quem; only new uses and words can do that. Furthermore, Dihle overlooks an important fact: Wisdom is written in literary Greek full of rhetorical features and learned vocabulary. It is just the sort of Greek in which an older use of αὐθέντης might be employed, even if the author was a ‘Hellenistic’ writer composing in the first century BC; in fact it could be due to Atticistic influence already at work then.28

5. MOERIS Little is known of Moeris himself, only that he left a sizeable lexicon, variously titled, listing Attic features and their non-Attic counterparts.29 His date cannot be determined exactly, but since the work has been shown to draw on Phrynichus and most of the other Atticists it is probably to be dated in the third century AD. A welcome new edition, by Hansen, has appeared quite recently, superseding the older works, which however offer useful material.30 Moeris’s work, like Phrynichus’s, is clearly a handbook for the wouldbe Atticist. Although Moeris mostly does not condemn or recommend features outright as Phrynichus does, it is clear what the purpose is. The method of presentation is compact and formulaic. It works with labels, mostly the same two, Ἀττικοί (= Attic writers) and Ἕλληνες (= later Greeks). Another label, κοινόν, appears from time to time, but exactly what it is intended to convey is open to debate.31 The great majority of the entries deal with forms and words; very few relate to uses of words (e.g., ε 15). Sometimes entries record other information, such as an occurrence in an Attic writer (as δ 33 below) or the meaning of an Attic term (e.g., ο 36). Occasionally an entry does not simply label features but 28

29

30

31

On the linguistic character of Wisdom see Reese, Hellenistic Influence. Thackeray (Grammar, 13) labelled Wisdom ‘Literary and Atticistic.’ The meaning ‘murderer’ is still in use in Ap. Rhod. 2.754 (Epic poetry, III BC); PCairZen 4.532.15 (poetic epitaph for a dog, III BC) = Lloyd-Jones and Parsons Supplementum Hellenisticum, 977.15. See Shipp, MGE, 115 [for futher details, including the modern descendant effendi.] On Moeris see Wendel, ‘Moiris’; OCD, s.v.; Dickey, Scholarship, 98; Hansen, Moeris, 36–56. In Sakalis, ‘Ἀνάττικα καὶ Ψευδαττικά’ the inaccuracy of some of Moeris’s statements about Attic is demonstrated. Hansen, Moeris. The older editions are those of Pierson (1759), Koch (1830), and Bekker (1833). Cf. Maidhof, Begriffsbestimmung; Wendel, ‘Moiris,’ 2504. Maidhof claims to show that κοινόν means ‘colloquial speech,’ but I have doubts: cf. ε 59 and η 6 below.

THE ATTICIST GRAMMARIANS

303

comments in some way and in so doing gives us an unmistakable glimpse of the Atticist program (see δ 33 below). Some specimens to illustrate: ἄρρενα Ἀττικοί· ἄρσενα Ἕλληνες. (α 45) οἶσθα χωρὶς τοῦ σ Ἀττικοί· οἶδας Ἕλληνες. (ο 24) εἴσεται Ἀττικοί· γνώσεται κοινόν. (ε 59) ἤρεσέ με Ἀττικοί· ἤρεσέ μοι κοινόν. (η 6) νώ δυϊκῶς Ἀττικοί· ἡμεῖς Ἕλληνες. (ν 2) πάντοτε οὐδεὶς τῶν Ἀττικῶν. (π 57) None of the Attic authors [uses] πάντοτε (‘always’). διαφορότητος Πλάτων Θεαιτήτῳ· παρ᾽ ἄλλῳ οὐχ εὗρον. (δ 33) διαφορότης (‘difference’; in gen. sing.) [occurs in] Plato’s Theaetetus; I have not found it in any other [ancient author].

Some brief notes may be helpful. In ἄρρενα, we see the characteristic Attic change of original -ρσ-, in contrast to retention in other ancient dialects, from which ἄρσενα etc. entered Koine Greek. The budding Atticist would need to learn to make the change to -ρρ- in all the words where -ρσ- was usual in spoken Koine. The next item relates to the morphology of the verb οἶδα ‘I know,’ whose irregular conjugation in Attic had been regularised in later Greek; the next relates to its irregular future. The lemma on ἤρεσε concerns a very minor point of syntax that would seem hardly worth the trouble. In both these last we have the label κοινόν, which here at least seems to bear the meaning ‘common,’ i.e., to both Attic and later Greek, since this is true to the evidence: γνώσεται and ἤρεσε + dative are well attested in Attic as well as later. Next comes another attempt to revive the dual, as in our first example in Phrynichus; here it is the dual first person pronoun (‘we two’). The word for ‘always’ tended to be replaced in Greek: the newcomer πάντοτε had already appeared in fourth-century BC Attic (Aristotle, Menander, Philemon), but these writers were not acceptable models to the Atticists. Moeris does not tell us what to say instead. Phrynichus recommends ἑκάστοτε and διὰ παντός, without mentioning ἀεί, the other old word that πάντοτε was replacing.32 It is no surprise that πάντοτε (with πάντα) has come down as the modern word. 32

Phryn. 74. I think Phryn. is giving acceptable equivalents for πάντοτε as meaning ‘at all times, on every occasion’; he takes ἀεί for granted. Shipp (MGE, 439) notes the ‘tendency to use strong expressions for the idea.’

304

COLLECTED ESSAYS ON THE GREEK BIBLE

The final lemma gives us a picture of Moeris searching the Classics for the word διαφορότης, of which he was evidently suspicious, and coming up with examples only in one dialogue of Plato.33 I think the message he wants to convey is that the word remains under a cloud and the careful Atticist had better be wary of it.

6. THE ANTIATTICISTA The author of the work entitled Ἀντιαττικιστής (Latin Antiatticista) is unknown and the title is not ancient.34 It is also something of a misnomer: the compiler was not opposed to Atticism as such; he only adopted a less strict standard than Atticists like Phrynichus. This meant allowing a wider canon of approved authors and accepting a single occurrence in a Classical author as enough to make a feature good Attic, something not conceded by Phrynichus and others like him. The enterprise belonged fully within the framework of the Atticist program and reflected controversies and differences of opinion between Atticists, not a complete rejection of the program.35 The surviving text of the Antiatticista is a very reduced epitome of a much ampler work. This is clear from the fact that many of the lemmata lack the supporting reference to a Classical author that would have formed part of the original. One might suppose that it was compiled as a reaction to Phrynichus’s Ecloga, but it emerges from a study of the lemmata that Phrynichus made use of the Antiatticista.36 The date of the work, then, must be the second century AD. The current edition remains that in Bekker’s Anecdota of 1814.37 A new edition, fully annotated, is a desideratum.38 33

34

35

36 37 38

Plato, Theaet. 209a.5; d.1; e.7; 210a.4, 9; three are gen. sing. There are four more in Plato, but none in other Classical authors; it is not exactly rare later (see LSJ). It appears the name was given by Ruhnken (1723–1798): see Naber, Photii Patriarchae Lexicon, 1:95; cf. Bekker, Anecdota Graeca, 3:1074. There is also something wrong with it: Greek ἀντι- would not normally mean ‘against.’ Ruhnken was thinking in Latin. On the Antiatticista see Dickey, Scholarship, 97–8; Lee, ‘Future of ζῆν,’ 290–2 [= Essay 4] (but my description ‘a rebel against the Atticizing fashion of his time’ was overstated). See Fischer, Ekloge, 39–41. Bekker, Anecdota Graeca, 1:75–116; textual and other notes 3:1074–7. De Gruyter reports (email 4 Feb 2009) that they have no edition in prospect. [An edition by S. Valente (2015) has appeared since the above was written.] See Arnott, ‘Note on the Antiatticist,’ for an instance of further work to be done on the text.

THE ATTICIST GRAMMARIANS

305

The work is no less valuable for the study of the Greek language than the works of Phrynichus and the others, but its purpose, and hence the layout of the entries, is somewhat different and needs to be understood if it is to be interpreted correctly.39 The author gives a form, word, or use censured as un-Attic by other Atticists and then cites an occurrence in a Classical author to show that it does in fact occur, with the clear implication that it is acceptable, at least by the standards set by this Atticist. There are about 800 lemmata in the list, in alphabetical order, with the condemned feature usually as headword, though not invariably. Many entries are very abbreviated because of the excision of material, but they are still readily understood. Where the Classical authority has been omitted, the intention is still plain; in some extreme cases there is nothing left but a headword, but even these yield a meaning. But let us begin with an entry that shows what is going on: ἤθη· πληθυντικῶς φασι μὴ δεῖν λέγειν, ἀλλ᾽ ἑνικῶς. ἀλλ᾽ Ἀντιφάνης εἶπε πληθυντικῶς. ἤθη (‘habits’): they say you should not use this in the plural, but the singular. But Antiphanes used it in the plural. (Antiatt. 98.15; η 4 V.)40

‘They’ are of course the stricter Atticists, and in this case we actually have the statement of Phrynichus, who says one must guard against using ἤθη in the plural, for οἱ δόκιμοι (‘the approved authors’) have the singular (Phryn. 344). An occurrence in the fourth-century BC comic poet Antiphanes is cited to prove that it occurs in a Classical author, though this would not have been enough to satisfy Phrynichus. The author of the Antiatticista might also have cited Menander’s Thais, quoted by Paul in 1 Cor 15:33 (φθείρουσιν ἤθη χρηστὰ ὁμιλίαι κακαί), but this would have cut no ice with Phrynichus either (see above on his attitude to Menander). A more abbreviated type is readily interpreted: δεῦρο· ἀντὶ τοῦ ἔρχου. Πλάτων Πολιτείας τρίτῳ. (Antiatt. 88.19; δ 4 V.)

This means: ‘δεῦρο (‘hither’), when used in the meaning ἔρχου (‘come’; ‘come here’), is condemned by some, but there is an example in Plato’s Republic, third book’ (he refers to 4.445c.1). One would not have expected 39

40

A misinterpretation of the Antiatticista by G. D. Kilpatrick is taken up in Lee, ‘Future of ζῆν.’ For Kilpatrick’s response see ‘Atticism and the Future of ζῆν.’ Nigel Turner completely misunderstands the Antiatticista in ‘Literary Character,’ 113–4. References to page and line of Bekker’s edition, Anecdota Graeca, 1:75–116. [Refs. to Valente’s edition (‘V.’) are added; there are no substantial differences.]

306

COLLECTED ESSAYS ON THE GREEK BIBLE

this use of δεῦρο to be under a cloud, since there are Classical examples, but we learn that it was. To a rigorous Atticist it was the original function of δεῦρο, as an adverb, that mattered, and any departure, even if ancient, was not allowed. The condemned use of δεῦρο is of course widespread in Koine Greek and occurs in the NT, for example in Mark 10:21 (δεῦρο ἀκολούθει μοι). A lemma from Moeris related to noun morphology: κλεῖν ἀξιοῦσι λέγειν, οὐ κλεῖδα. Δίφιλος Εὐνούχῳ. (Antiatt. 101.29; κ 17 V.; cf. Phryn. PS 82.17; Moeris κ 45; Philet. 116)

The accusative κλεῖδα, which regularised the anomalous declension, was rejected by some Atticists, but can be cited from a play of Diphilus, a comic poet of the fourth to third centuries BC, and is therefore acceptable. Next an entry on a significant newcomer to the Greek vocabulary: τρώγειν οὔ φασι δεῖν λέγειν τὸ ἐσθίειν, ἀλλὰ τὸ τραγήματα ἐσθίειν. They say that one ought not to use τρώγειν to mean ‘to eat,’ but ‘to eat nuts, fruit, etc.’ (Antiatt. 114.15; τ 2 V.)

The question relates to the use of the word, not to τρώγω itself, which is as old as Homer. No ancient authority for the meaning ‘to eat’ is cited, but we can assume that one was originally included, since that is the point of the entry.41 Nevertheless it is true that τρώγω was not originally the word for just ‘eat’: that did not become established until Middle Koine.42 τρώγω then became the standard vernacular and finally modern word (only in the present tense: the aorist remained ἔφαγον). John’s Gospel in the NT provides significant evidence along the way: τρώγω, Shipp observed, ‘is John’s word for “eat,” one of the features in which he anticipates modern Greek.’43 My last illustration shows how a short and seemingly cryptic entry may be unexpectedly useful: μετάβα· ὥσπερ καὶ ἀνάβα καὶ κατάβα. Ἄλεξις Ἀμφώτιδι. (Antiatt. 108.10; μ 25 V.)

‘In compounds … -βηθι and -βᾱ alternate [in the NT] without very clear rationale,’ says Moulton.44 The lemma in the Antiatticista shows that 41

42

43

44

Cf. Philetaerus 231, with an example from Eupolis, which is perhaps the one, though not very germane. For a proposed division into Early (III–I BC), Middle (I–III AD) and Late Koine (IV– VI AD) see Lee, ‘Ἐξαποστέλλω,’ 113 n. 31 [= Essay 12]. Shipp, MGE, 541. Note that John does not have pres. ἐσθίω/ἔσθω. BDAG, s.v. τρώγω, still reflects the obsolete idea that τρώγω differs from just ‘eat’ in the NT: ‘to bite or chew food, eat (audibly).’ MH, Grammar, II, 209.

THE ATTICIST GRAMMARIANS

307

the -βα form was disapproved of by the Atticists, and was therefore the popular form; the approved form is not named in the lemma, but we can be certain it was the archaic -βηθι. This gives us the clue to how these alternatives were perceived by speakers in the first century AD, and may help to dispel the mystery.45

7. OTHER

SOURCES

There were many more Atticist grammarians than those mentioned so far: producing handbooks was something of an industry once the Atticising movement took hold. But their works mostly survive only in fragments, if at all.46 A selection will be covered briefly in this section. The first three, Aelius Dionysius, the Philetaerus, and Thomas Magister, offer lists similar in character to the works we have just seen; the other two, Ammonius and Pollux, compiled works of a quite different shape, though they still reflect the aims of Atticism. We will also need to notice another important aspect of our topic, that is, the works of the Atticist writers themselves, which form a major body of evidence of Atticist practice. 7.1. Aelius Dionysius Aelius Dionysius, of the second century AD, produced an early lexicon of Attic, entitled Ἀττικὰ ὀνόματα. The full work is lost, but substantial fragments can be recovered from the works of later authors.47 A sample: εὐώνυμον λέγουσιν, οὐ μόνον ἀριστερόν. καὶ παρὰ Θουκυδίδῃ πλεῖστον. They [the ancients] say εὐώνυμον (‘left’), not only ἀριστερόν. And it is most frequent in Thucydides. (Ael. Dion. ε 77; cf. Antiatt. 97.1)

This note alerts the would-be Atticist to the fact that ancient Attic had two words for ‘left,’ not just the one familiar in his own time; it implies 45

46

47

Details of NT occurrences in MH, Grammar, II, 210; BDF, §95.3; most are affected by variants (cf. ‘Atticism and the Text of the NT’ below). BDF, §2 (similarly BDR) confusingly describe κατάβα as an ‘Atticism.’ The reason is that BDF use ‘Atticism’ both for original Attic elements surviving in the Koine (as in this case) and for later Atticising features revived by the Atticists (as, e.g., in BDF, §3 n. 4). Since -βα forms appear already in Attic (see MH, Grammar, ibid.; Schwyzer, Grammatik, I, 676 n. 1), BDF can call it an Atticism in the former sense; but it is the opposite of an Atticism in the latter sense, since it is the popular form. See Cohn, ‘Griechische Lexikographie,’ 693–5; Kazazis, ‘Atticism,’ 1205; Wendel, ‘Moiris,’ 2505–9; Dickey, Scholarship, 94–6, 98–100. Edition: Erbse, Untersuchungen. Cf. Hansen, Moeris, 42–6; Dickey, Scholarship, 99.

308

COLLECTED ESSAYS ON THE GREEK BIBLE

that he should not forget to use εὐώνυμος in his writing or speech-making. Its value for us is that it suggests that εὐώνυμος was dying out and the ordinary word for ‘left’ was ἀριστερός. It follows that where writers use εὐώνυμος it may be artificial or have an archaic flavour; the original euphemistic force (‘well-named’) might also still be in play. This seems to be relevant to the LXX, where there are 19 examples of εὐώνυμος to 65 of ἀριστερός, and to the NT, where εὐώνυμος is found 9 times in contexts with a special flavour, besides 4 of ἀριστερός.48 7.2. The Philetaerus The work known as the Φιλέταιρος (Latin Philetaerus) is more substantial, and worth consulting alongside the works of Phrynichus and Moeris.49 It is attributed to Herodian (Aelius Herodianus Grammaticus, II AD) in the MS tradition, but there is doubt about this.50 The date, too, is uncertain. Dain dates the surviving redaction to the third century, or even the fourth or early fifth.51 Two samples of the Philetaerus will suffice: τῇ ἐπιούσῃ ἐρεῖς μὴ προστιθεὶς ἡμέρᾳ· τῆς δὲ ἐπιούσης ἡμέρας. You shall say τῇ ἐπιούσῃ (‘on the next day’) without adding ἡμέρᾳ (‘day’); but τῆς ἐπιούσης ἡμέρας (‘of the next day’) [is acceptable]. (Philet. 151)

A concern for such a nicety as the omission of ‘day’ – but only in the dative! – shows the same pedantic spirit as the best of them – and another subtle pitfall for the novice Atticist. It is interesting to compare the NT evidence. The usual NT expression is τῇ ἐπαύριον (17 times); but Luke has τῇ ἐπιούσῃ three times (Acts 16:11; 20:15; 21:18) and τῇ ἐπιούσῃ ἡμέρᾳ once (7:26). Luke scores three out of four in the Atticist game; and it may not be fanciful to think that in the last instance ἡμέρᾳ is a residue from Exod 2:13 (τῇ ἡμέρᾳ τῇ δευτέρᾳ). ἄρκτος σὺν τῷ τ. (Philet. 314) 48

49 50

51

See Chantraine, ‘Les mots designant la gauche,’ in Μνήμης χάριν, 1:61–9. NT occurrences of εὐώνυμος: in the Gospels only in the formula ἐξ εὐωνύμων, ‘de résonnance probablement religieuse’ (69); Acts 21:3 euphemistic in a nautical setting; Rev 10:2 vision of an angel. Edition: Dain, Philétaeros. Cf. Hansen, Moeris, 52–4. Dain, Philétaeros, 9–13. Argyle (‘A New Greek Grammarian’) makes a good case for attribution to Cornelianus, Imperial Secretary, to whom Phrynichus addresses his Ecloga. Dain, Philétaeros, 15. Dain’s arguments seem persuasive to me; but if Argyle is right, II AD is required.

THE ATTICIST GRAMMARIANS

309

Though the entry is very terse, the interpretation is not difficult if we bear in mind that the Philetaerus is a guide to Atticism. The entry means that ἄρκτος (‘bear’) with the τ is the correct Attic form to be maintained; it is taken for granted there is another form without τ to be avoided. Indeed ἄρκος exists and the distribution fits: ἄρκτος is the ancient form in Homer, Attic, and later literary sources; ἄρκος, clearly a more popular form, is attested first in a third century BC papyrus,52 then in the LXX and a few other Koine texts. It is no surprise that ἄρκος is what we read in Rev 13:2.53 7.3. Thomas Magister Next is Thomas Magister, whose date is in the fourteenth century AD. An Atticist grammarian still at work so late in the day may seem strange and irrelevant, but not so. Guidance in writing in Attic style was still needed because there were Byzantine writers doing just that; and Thomas’s material is based on and continues the work of his much older predecessors.54 This means that an observation of his may be as useful for the history of the language as one from a second century Atticist. Consider this: βρέχειν οὐδεὶς τῶν ἀρχαίων εἶπεν ἐπὶ ὑετοῦ, ἀλλὰ ὕειν. None of the ancients said βρέχειν (‘to wet’) with reference to rain, but [they said] ὕειν (‘to rain’). (Th. Mag. 57.8)

The entry relates to Koine Greek developments in the word for ‘to rain.’ The old word ὕειν was replaced in everyday language by βρέχειν, which, though old, had had different senses. Phrynichus has a note of similar import, warning against this use (Phryn. 255). The new word is standard in the NT; in the LXX Pentateuch there is an interplay between the two, based on the difference of formality.55 Thus Thomas Magister’s note is relevant to a text a millenium and a half older. 7.4. Ammonius A work of a different character is the Περὶ ὁμοίων καὶ διαφόρων λέξεων (De adfinium vocabulorum differentia) attributed to Ammonius, but probably by Herennius Philo. In its original form it is datable to the 52 53

54 55

PMich 1.86.5 (Philadelphia) περὶ τῆς ἄρκου. Technically ἄρκος shows reduction of a cluster of three consonants (-rkt-) by dropping one, a common phenomenon: cf. artic in older Engl. and Lat. articus. See Shipp, MGE, 97, for modern evidence. Edition: Ritschl, Thomae Magistri ecloga. See Lee, Lexical Study, 122–4.

310

COLLECTED ESSAYS ON THE GREEK BIBLE

late first or early second century AD.56 It is quite extensive (525 lemmata) and many entries contain several lines of discussion. The aim of this work is to draw distinctions between apparent synonyms, mostly two, but sometimes three or even four.57 The distinctions are mostly well done, but it is important to realise that they describe Attic norms of an earlier time. The purpose is didactic, to set out and maintain distinctions (sometimes even words) that had been lost, or were fading away, in the current language. The statements are evidence for the loss of these distinctions or words in the first century AD, not a real picture of how the words were used at that time. There is also a tendency to find distinctions whether genuine or not, and some are not true, or are too absolute, even for Attic. It is impossible to do justice here to the riches of this source; I offer only one specimen, in which the words for ‘wake up’ and ‘get up’ are the subject of attention: ἀναστῆναι καὶ ἐγερθῆναι διαφέρει. ἀναστῆναι μὲν ἐπὶ ἔργον, ἐγερθῆναι δὲ ἐξ ὕπνου. ἀναστῆναι (‘to get up’) and ἐγερθῆναι (‘to wake up’) differ. [You say] ἀναστῆναι for work, but ἐγερθῆναι from sleep. (Ammon. 50; similarly 216)

What makes it necessary for a distinction to be made between ἐγείρομαι (aor. ἠγέρθη), originally ‘I wake up,’ and ἀνίσταμαι (aor. ἀνέστη), ‘I stand/get up’? It is the fact that in Koine Greek the former increasingly invaded the territory of the latter, that is, ἐγείρομαι was used to mean ‘I stand/get up’ (as well as ‘I wake up’).58 Examples are found all over Koine Greek, as in Mark 2:12 (ἠγέρθη καὶ εὐθὺς ἄρας τὸν κράβαττον ἐξῆλθεν, ‘he got up and …’). Ammonius is aware of this development and is resisting it by defining the usage of Classical Attic; it is implicit that contemporary usage is condemned. The diligent Atticist will take note. 7.5 Pollux The Onomastikon of Pollux (Julius Pollux/ Πολυδεύκης) is well known and often cited in lexicons and elsewhere. The work is a topical dictionary, the first of its kind. It is a list of all, or most, of the words in the Greek vocabulary, arranged by topics or subjects. Meanings of words, as in a lexicon, are not the primary goal and are not usually indicated. The original 56

57

58

Edition: Nickau, Ammonii Vocabulorum Differentia. Cf. Cohn, ‘Ammonios’; Hansen, Moeris, 51–2; Dickey, Scholarship, 94–6. Authorship and date: Nickau, lxvi–vii. Rarely homonyms, such as 390 πεῖρα and πήρα, 399 πόμα and πῶμα, despite Dickey, Scholarship, 95 (her examples are not homonyms). Cf. Shipp, MGE, 75–80.

THE ATTICIST GRAMMARIANS

311

was compiled in the second century AD and survives in a reduced though still very extensive form.59 The true nature of this collection is seldom spelt out and often not understood. It is not a record of the Greek vocabulary as it stood in the second century AD, but a list of acceptable Attic words and turns of phrase from the golden age of Athens, compiled by a later scholar looking back. Ordinary Koine Greek vocabulary also appears, but only to the extent that it is the same as the older vocabulary. The book is a resource for those wishing to write or speak in the ancient manner; for this reason it is also in part encyclopedic, including not just words, but information on the life and culture of ancient Athens, all derived from the ancient texts. Pollux was an Atticist; he was not an objective observer of the language of his own time. That this is so can be readily shown by a few samples, beginning with some of the words discussed previously. In the list of words for eating, ἐσθίω appears (6.39) but not τρώγω. Ἀεί is in (1.151), but not πάντοτε. The word ἀκμή appears several times in various uses (1.60, 61, 177, 180; 2.10; 5.158) but not the adverbial use ἀκμήν, ‘still.’ The form ἄρκτος is listed (4.157, 159; 5.81, 15), but not ἄρκος. Words condemned by Atticists may be listed, but with a comment. So κράββατος (‘bed, stretcher’), condemned by Phrynichus (cf. next section, §8), is entered like this (10.35): καὶ κράββατον εἰρῆσθαι λέγουσιν· ἐγὼ δ᾽ οὐκ ἐντετύχηκα τοῖς δράμασιν (‘they say κράββατος was used [by ancient writers] but I have not met it in the dramatists’). Most tellingly κοράσιον (‘girl’), also condemned by Phrynichus, attracts this comment (2.17): τὸ γὰρ κοράσιον εἴρηται μέν, ἀλλὰ εὐτελές, ὥσπερ καὶ τὸ κορίδιον (‘κοράσιον is said, but it is cheap, as is κορίδιον’). As for εὐώνυμος, it appears a number of times (along with ἀριστερός), as one would expect; but once it appears (1.88) in a context that reveals the purpose of the collection clearly: ὀνομάσαις δ᾽ ἂν τοῖχον δεξιὸν καὶ εὐώνυμον (‘and you may call the side[s] [of the ship] δεξιὸν καὶ εὐώνυμον’). This is in a list of the vocabulary of the trireme, the warship of another era.60 7.6 Atticist writers Finally in this section we must take note of the fact that the works of Atticising writers are themselves a major source of data on the practice of Atticism. As already mentioned, most of the literature from the first century BC onwards was affected by Atticism. The degree of effect and the 59

60

Edition: Bethe, Pollucis Onomasticon. Cf. Bethe, ‘Pollux’; OCD, s.v. Pollux; Dickey, Scholarship, 96. For an extended example from Pollux, see Lee, History, 200.

312

COLLECTED ESSAYS ON THE GREEK BIBLE

exact features deployed vary from one author to another, though with a solid core of common Atticisms. The complexity of the situation is well illustrated by Lucian, who was an expert Atticist himself but can vary his style up and down the scale and even make fun of Atticist pretensions. The control of this vast body of data is difficult for us for obvious reasons. Schmid studied some important representatives in great detail in his classic work, but it is not exhaustive.61 The language of Dionysius of Halicarnassus has not yet been examined in depth, though he is key evidence for the onset of the Atticist phenomenon in the first century BC. A legion of other Atticising writers over many centuries is available for further attention.62 I take one brief illustration of how Atticist writers provide evidence on Atticising features, namely, the substantival neuter adjective with τό functioning as an abstract noun. Though not explicitly remarked on by any Atticist grammarian to my knowledge, it is one of the ‘hallmarks’ of Atticist usage.63 There is an instance early on in Dionysius’s De compositione verborum 1.7 (τὸ περὶ τὰς λέξεις φιλόκαλον, ‘the love of fine language’). Phrynichus, in the preface to his Ecloga, where he shows off his technique, has οὐ γάρ τις οὕτως ἄθλιος ὡς τὸ αἰσχρὸν τοῦ καλοῦ προτιθέναι (‘for no one is so pitiful as to prefer the ugly to the beautiful’). In the NT there is a concentration in [Luke–]Acts and the Epistles. A suitably contrived example is heard in the mouth of Paul before Agrippa in Acts 26:7 (τὸ δωδεκάφυλον ἡμῶν = ‘our twelve tribes’). This, Haenchen aptly says, is ‘meant to sound solemn.’64 8. THE ATTICIST GRAMMARIANS AND THE NEW TESTAMENT The Atticist grammarians did not, of course, comment directly on the NT, which would have been beneath their notice if they had even heard of it. Conversely, the NT writers could not have been directly influenced 61

62

63 64

Schmid, Atticismus. Cf. Anderson, Second Sophistic, 263: ‘the daunting scale of Schmid … should not blind us to the fact that the linguistic map of the second century AD, for example, is still far from complete. It is disconcerting to be confronted with the hundreds of words in Galen unreported in LSJ, for example.’ Besides the work of Schmid, short lists of features may be seen in Kazazis, ‘Atticism,’ 1207–8; Horrocks, Greek, 83–4. [See Kim, ‘Literary Heritage,’ 472–4 for useful observations in D.H.] Horrocks, Greek, 83. Haenchen, Acts, 683 n. 7 [I would add ‘learned’]. On this feature cf. MHT, Grammar, III, 13–4; BDF, §263; Schmid, Atticismus, 4:608; Lee, ‘Some Features,’ 23–4 (τό + substantival participle). There is no complete list of exx. in the NT, it appears.

THE ATTICIST GRAMMARIANS

313

by any of our Atticist grammarians, because the NT writers preceded them. The grammarians’ material is nonetheless of great relevance to the NT. To begin with, a comparison between Atticist strictures and the language of the NT provides proof of its generally Koine Greek character. Time and again the feature disapproved of by the Atticists is the one that appears in the NT. Here is just a small selection of condemned words and forms found in the NT, set alongside the recommended equivalents, which do not appear in the NT: γλωσσόκομον γρηγορῶ ἐμπτύει κλίβανος κοράσιον κράββατος

γλωττοκομεῖον ἐγρήγορα καταπτύει κρίβανος κόριον, κορίδιον, κορίσκη σκίμπους, ἀσκάντης

μύλος οἰκοδεσπότης σίναπι τάχιον

μύλη οἰκίας δεσπότης νᾶπυ θᾶττον

φάγομαι

ἔδομαι

(Phryn. 70) (Phryn. 88) (Phryn. 8) (Phryn. 150) (Phryn. 50) (Phryn. 41; Moeris α 119, σ 33; Th. Mag. 333.2)65 (Moeris μ 5) (Phryn. 349; Th. Mag. 259.3) (Phryn. 252; cf. Moeris ν 16) (Phryn. 52; Moeris θ 18, τ 7; Philet. 18) (Phryn. 300; Th. Mag. 126.13)

None of this is by accident: the writers of the NT, for the most part, were using the Koine Greek of their day, and the evidence of the Atticist grammarians helps to prove it.66 The statements of the Atticist grammarians may also further our understanding of NT usage of words. We have seen some straightforward examples already (such as τρώγω), but it can also happen in quite unexpected ways. Without the clear statement of Moeris (α 132) that ἀπάτη means τέρψις (‘enjoyment’) in later popular Greek, such a meaning might not have been suspected in Matt 13:22, Mark 4:19 and 2 Pet 2:13. A more remarkable instance arises from an entry of Ammonius in which he differentiates between παρρησία and παρουσία. As Shipp has shown, the modern evidence confirms the possibility of semantic confusion of these seemingly dissimilar words. Ammonius needs to explain the difference 65

66

Luke and Matt avoid κράββατος where it occurs in Mark, but Luke uses it himself in Acts 5:15; 9:33; see Shipp, MGE, 106 for a probable explanation of the exx. in Acts. Similar lists are offered in Triantaphyllides, Γραμματική 1: Ιστορική Εισαγωγή, 411–8 (very extensive), and Browning, Greek, 47–8.

314

COLLECTED ESSAYS ON THE GREEK BIBLE

between them because the confusion was occurring in his time. The NT may reflect this confusion in that it is possible that adverbial παρρησίᾳ (‘openly, publicly’) has rather the sense of ‘(present) in person,’ in, e.g., John 11:54 (οὐκέτι παρρησίᾳ περιεπάτει ἐν τοῖς Ἰουδαίοις).67

9. ATTICISM IN THE NEW TESTAMENT That the language of the NT is basically Koine Greek is clear. Though the 27 books of the NT differ in position on the scale from vernacular to literary, none can be properly described as Atticistic, with the possible exception of 2 Peter.68 But it is not true that the NT is untouched by Atticism. As we have seen, the phenomenon was well under way in the first century AD and making its influence felt in the education system; it would be no surprise if the NT books reflected to some degree what their authors had learnt in school. Atticistic features do in fact appear in the NT, in a range of environments. The identification of these features is greatly aided by the works of the Atticist grammarians, though it is not dependent on them alone: they may be recognised from the practice of Atticist authors or from general developments in the language. Important also for understanding the NT phenomena is distribution, that is, the kinds of books and contexts in which significant features are found. Let us begin with Luke’s Gospel, where the instances are prominent. Luke, in his rewriting of Mark, frequently makes changes that are in line with Atticist precepts. Many were noted, along with other changes towards more stylish or elegant wording, by Cadbury in his notable study of Luke’s language.69 In the following small selection, the word or expression replaced by Luke is one censured by the Atticist grammarians: Mark 5:41 τὸ κοράσιον || Luke 8:54 ἡ παῖς Mark 5:42 τὸ κοράσιον || Luke 8:55 omitted70 Mark 5:23 ἐσχάτως ἔχει || Luke 8:42 ἀπέθνῃσκεν71 67

68

69

70 71

Ammonius 401; see Shipp, MGE, 441–2. This example has not yet had any impact on the lexicons. [An Engl. ex. of the present day (2021) may be mentioned, the confusion of wonder and wander, probably a result of the misleading spellings.] Cf. Norden, Die Antike Kuntsprosa, 2:479–510; Lee, ‘Some Features,’ 9 (with references); on 2 Peter MHT, Grammar, III, 30, 126–7. Cadbury, Style of Luke, esp. 182–8, 196–7, 201. Earlier Norden, Die Antike Kunstprosa, 2:485–92. Phryn. 50; Cadbury, Style of Luke, 186. Phryn. 369; Th. Mag. 136.4; Cadbury, Style of Luke, 182.

THE ATTICIST GRAMMARIANS

315

Mark 10:25 ῥαφίδος || Luke 18:25 βελόνης72 Mark 1:44 + 5 ὕπαγε || Luke 5:14 + 5 all omitted or changed73

Apart from these instances of avoidance, reflections of Atticism can be found lightly spread through the NT books that are of a higher literary level, namely, Luke–Acts, the Epistles, and to a lesser extent Matthew. The features appear generally, but some may also occur in special contexts, where the style has been raised to suit the speaker or the situation. What follows are just a few samples. The control of particles, whether by employing those that were obsolete or by skilled handling of those still available, is a sign of literary taste inspired by Atticism. On this large topic a small selection must suffice. τοίνυν appears only in Luke, 1 Corinthians, and Hebrews, once each. The even more elevated τοιγαροῦν makes two appearances only, in 1 Thessalonians and Hebrews. Rare δήπου is used only in Heb 2:16 (‘classical, literary,’ BDF).74 Beside these very infrequent particles, let us notice τε, ‘an indication of stylistic pretension,’ as Turner puts it, which is very frequent in Acts, found occasionally elsewhere, and not at all with certainty in Mark, John, and Revelation.75 Next a short list of Atticistic features of various kinds occurring only in literary books or special contexts: ἴσασι Acts 26:4 Paul’s speech before Agrippa: Koine οἴδασι × 7.76 ἴστε (indic.) Heb 12:17: Koine οἴδατε × 63.77 χρή Jas 3:3.78 μέγιστος 2 Pet 1:4.79 72

73

74 75

76 77 78 79

Phryn. 63; Th. Mag. 56.16; Antiatt. 113.14 (ρ 7 V.); Cadbury, Style of Luke, 187. Elliott (Essays and Studies, 66), rejects the reading βελόνη, even in Luke, because there is a v.l. ῥαφίς: he does not allow for Luke’s Atticising tendency. He takes the opposite tack (ibid., 71) with ἦσθα in Matt 26:69, where there is no v.l. (see below). Antiatt. 114.31 (υ 1 V.); Cadbury, Style of Luke, 173; Shipp, MGE, 35. I consider Markan priority inescapable. Those who reject it face the task of explaining why an author would revise the language of his source text downwards, against the trend of the time, and that too not only by replacing ‘correct’ words with vernacular ones (e.g., βελόνη) but by introducing vernacular words that had no counterpart in the source (e.g., some of ὕπαγε). For further telling evidence, see Luke’s handling of the participle (Cadbury, Style of Luke, 133–7). On these three, see MHT, Grammar, III, 339–40; on δήπου also BDF, §441.3. MHT, Grammar, III, 338–9; BDF, §443; Kilpatrick, ‘Atticism and the Text,’ 135; BDAG, s.v. Cf. μέν: Lee, ‘Some Features,’ 4–6. Moeris ι 22; BDF, §3 n. 4. Phryn., PS 53. BDF, §358.2; Schmid, Atticismus, 4:592. MHT, Grammar, III, 29–30.

316

COLLECTED ESSAYS ON THE GREEK BIBLE

μεταξύ × 8 in Matt, Luke, Acts, Rom + John × 1 (ἐν τῷ μεταξύ).80 ἀμῶ Jas 5:4: Koine θερίζω × 21.81 καθότι Luke–Acts × 6; καθό Rom, 2 Cor, 1 Pet × 4; καθάπερ Rom, 1 and 2 Cor, 1 Thess, Heb × 13: Koine καθώς × 182.82

A final point here. The fact that a NT author such as Luke may also use words and forms censured by the Atticists does not invalidate these observations. As a general principle, the more popular feature can reappear at any time, since it is the default form, even if a more ‘correct’ one is usually deployed. But the reason is often likely to be that the rejection of some features belonged to higher, stricter levels of Atticism than that to which the NT authors aspired. It may also be that some of these features had not attracted Atticist attention in the first century AD. I take three examples from among many. Both βασίλισσα and ῥύμη (in the meaning ‘lane’), are deprecated by the Atticist grammarians, but both are used in Matt and Luke–Acts (besides βασίλισσα once in Rev). The future form φάγομαι, lambasted by Phrynichus as ‘barbarous,’ occurs in the NT not only in Revelation where we would expect it, but also twice in Luke and once in James; probably Attic ἔδομαι was beyond the ken of Luke and James.83

10. ATTICISM AND THE

TEXT OF THE

NEW TESTAMENT

Atticism has relevance to the NT in another way, one that relates not to the choice of words at the time of composition but to the transmission of the text. There is good reason to believe that copyists made changes to Koine Greek words and forms to match Atticist standards. The process was not systematic or complete; it shows itself only in the fairly frequent but sporadic appearance of manuscript variants that have an Atticising character. It may then be argued that in these cases – whether some or all is debatable – the Atticising variant is secondary. The matter is of course 80 81 82

83

MHT, Grammar, III, 277 (‘literary and even atticizing’). Moeris α 158; Shipp, MGE, 63. Moeris κ 17; Phryn. 399; Th. Mag. 195.8; MHT, Grammar, III, 319–20. There are many other features, such as optatives: see scattered observations in MHT, Grammar, III; BDF; Voelz, ‘Language of the NT.’ βασίλισσα Phryn. 197, 231; Moeris β 16; Antiatt. 84.26 (β 16 V.); Philet. 121. NT: Matt 12:42; Luke 11:31; Acts 8:27; Rev 18:7. ῥύμη Phryn. 383; Antiatt. 113.6 (ρ 2 V.). NT: Matt 6:2; Luke 14:21; Acts 9:11; 12:10. φάγομαι Phryn. 300, cf. 325. NT: Luke 14:15, 17:8; Jas 5:3; Rev 17:16; also καταφάγεται John 2:17 (LXX). ἔδομαι is not used by any NT author. [See Adams, S. A., ‘Atticism, Classicism, and Luke–Acts’ for a discussion of the placing (and labelling) of Luke’s Greek in the development of Atticism.]

THE ATTICIST GRAMMARIANS

317

of great importance to the textual critic of the NT and raises questions that have caused sharp debate. The recognition of this phenomenon was the work especially of G. D. Kilpatrick, followed by J. K. Elliott, both of whom have published extensively on the topic.84 To illustrate the point let us take two examples. ἦς was the Koine form of the second person sing. impf. of the verb ‘to be,’ replacing Attic ἦσθα. The Atticists naturally ruled the former unacceptable and required the latter.85 In the NT ἦς occurs six times without variant (Matt 25:21, etc.); but in Mark 14:67 we find ἦς in a small number of MSS and ἦσθα in the majority. It may be, in fact it looks rather likely, that ἦς is original and ἦσθα a scribal change. The former would be expected in Mark’s low-level Koine; the latter could be due to Atticism as well as assimilation to the parallel, namely Matt 26:69, where ἦσθα is without variant and likely to be original in Matthew’s more literary Greek.86 In the case of ἄρρην versus ἄρσην, quoted from Moeris above, we find that in seven of the nine NT occurrences of ἄρσην there is a variant ἄρρην, exactly the form recommended by the Atticists and probably a scribal alteration in the course of transmission of the NT text.87

11. CONCLUSION Atticism was a phenomenon of the utmost significance for the history of post-Classical Greek. Once it came into the picture, every writer could not avoid taking, consciously or unconsciously, a position in relation to it, from complete indifference to complete obedience to its dictates. So we today cannot fully appreciate any text unless we are aware of its potential effects and the presence or absence of the features Atticism tried to impose. The specimens we have seen show that this applies to the books of the NT just as much as to other texts of the period. Both the Koine features noted above in §8 ‘The Atticist Grammarians and the NT’ and the Atticising features in §9 ‘Atticism in the NT’ acquire their significance against the background of Atticism. The Atticist grammarians, suitably interpreted, provide a direct window into the phenomenon. They not only alert us to 84

85 86

87

See esp. Kilpatrick, ‘Atticism and the Text.’ Elliott’s work is conveniently consulted in Elliott, Essays and Studies; on Atticism see 30–2, 65–77, 106–11. For the textual debate see, e.g., Epp and Fee, Studies, esp. chs. 7, 8; Elliott, Essays and Studies, ch. 1. Phryn. 118; Moeris η 4; cf. Ammon. 220. Elliott, Essays and Studies, 70–1. The argument is not accepted by NA27, which reads ἦσθα in Mark 14:67. [NA28 likewise.] For full NT details see Elliott, Essays and Studies, 75.

318

COLLECTED ESSAYS ON THE GREEK BIBLE

Atticising features, but assist us, along with other evidence, in reconstructing the history of those features and their Koine Greek counterparts. They are also a guide to assessing the literary standing of any text of the period. The study of Atticism may seem a dead subject, exhausted by older researches. But this is not so. There is much room for future work. I suggest the following topics for further attention. (1) A new edition of the text of the Antiatticista. [Valente’s edition (2015) has met this need.] (2) More linguistic commentary on the Atticist grammarians, especially in the light of evidence from other sources; modern editions provide only the text, and for commentary we have to go back to older scholarship, some from the eighteenth century. (3) Further study of reflections of Atticism in the vernacular documentary texts; there is a yawning gap here. (4) A systematic study of the relation of Atticism to the NT; this would lead to a more accurate description of the stylistic character of each book than what we have at present. (5) The compilation of a list or register of stylistically marked features, based on both the Atticists’ statements and the observed practice of Atticising authors;88 this would be a valuable tool for achieving objectives (3) and (4), as well as a fuller understanding of the complex phenomenon known as Atticism.

88

But not by ‘consulting … Attic authors of the fifth and fourth centuries B.C. to form our own picture of Attic use,’ as suggested by Kilpatrick (‘Atticism and the Text,’ 134): this would give us real Attic, not the Atticists’ Attic.

20 ETYMOLOGICAL FOLLIES: THREE RECENT LEXICONS OF THE NEW TESTAMENT 20131

Abstract Three smaller lexicons of NT Greek have recently appeared, all introducing etymological material of certain kinds. The lexicons are those of Trenchard (2003), Danker (2009), and Newman (rev. ed., 2010). This paper examines the etymological material in each lexicon and points out serious problems arising from dependence on obsolete works as well as other weaknesses. In conclusion the value of including etymologies in NT lexicons is assessed and the future direction considered.

1. INTRODUCTION For most of the past century we have not been accustomed to see information on etymology or derivation of words in our lexicons of the Greek NT, nor does it seem to have been greatly missed. In three recent smaller lexicons a trend towards inclusion of such material has emerged. In principle there is no objection to it, and it could prove useful to the beginner and experienced scholar alike, if suitably handled. But for the lexicographer it involves excursions outside familiar territory into a specialised field that may prove hazardous. Before proceeding to a discussion of these new lexicons, it will be useful to set out a brief analysis of what might be referred to by ‘etymology’ in a Greek lexicon. There are, as it were, three potential levels, corresponding to the depth of penetration into the history of the word or form. The 1

Parts of this paper were presented at the SBL Annual Meeting, New Orleans 2009 (on Trenchard; also Souter-House), and the Senior New Testament Seminar, Faculty of Divinity, Cambridge, 2011 (on Trenchard and Danker). I am grateful to Michael Curran, Anne Thompson, and Steve Walton (see n. 34) for reading this paper, and especially to Anne for many helpful points.

320

COLLECTED ESSAYS ON THE GREEK BIBLE

examples I give simply illustrate the kind of information that might come under each heading; the manner of presentation varies from one lexicon to another. a. Immediate derivation E.g., λόγιος < λόγος, καταλέγω < κατά + λέγω. Here we are dealing with simple word-formation and composition within Greek, and the relationships tend to suggest themselves.2 b. Relationships between stems of the same ‘root’ E.g., λόγος ~ λέγω. The example involves vowel gradation (or ‘Ablaut’); this one is a common pattern and easy to recognise. But things can get complicated, as in τέκνον ~ τίκτω, where we have to deal with not only a zero grade of the root but reduplication and a metathesis (τίκτω < *τί-τκ-ω). Further, the derivatives of a root may branch off in different directions semantically, as in ἐκλογή, ‘selection,’ relating to λέγω not in its common meaning ‘say,’ but in an older sense ‘collect, choose.’ c. Ultimate etymology, i.e., origins and cognates in other languages E.g., λέγω < IE *leg- : Lat. lego. That is to say, λέγω descends from Indo-European (IE) into Greek and a cognate is found in Latin lego. This one is easy, but most of them are not; all sorts of complications can arise and the etymology often remains obscure. Also, relationships in type b. often involve going back into much earlier etymology of type c. to find the link (e.g., εὐδία ~ Ζεύς: see below in 2.b). Loanwords belong here too under origins (cf. below in 4.b). The recent history of etymological material in the NT lexicons may be briefly outlined now. The major lexicon of our time, Bauer (1928–2000) has not included any until BDAG (2000), which offers very limited data, mainly with a semantic purpose, in some entries.3 Nor is it found in the small lexicons of the past century, such as Newman (1971), Gingrich (1965), Souter (1916), and before that Hickie (1893).4 In Abbott-Smith’s 2

3

4

Symbols: ‘ Trenchard: ‘(opp. to right)’ > Newman: ‘(opposite right).’ 35 θησαυρός [orig. uncertain and apparently primitive] < Boisacq: ‘θησαυρός. … Prob. composé primitif …; origine inexpliquée.’ Boisacq’s ‘primitif’ refers to the compound; ‘primitive’ will convey nothing to the beginner. Cf. Beekes: ‘No etymology, but probably a technical loanword, without a doubt from Pre-Greek.’ 36 καθαρός [etym. in dispute (note: no connection with κατά)] Danker seemed to be warning against a non-existent error, but now we see it in Newman (3 above). There is no known etymology for καθαρός. Beekes thinks Pre-Greek. 37 σελήνη [IE] Formation within Greek, < *σελασ-να. Etymol. of σέλας obscure, not established as IE. 33

See esp. Beekes, EDG, xiii–xlii.

336

COLLECTED ESSAYS ON THE GREEK BIBLE

38 χάρτης [orig. uncertain, cp. χαράσσω ‘engrave’] Danker has fallen into the same trap as Trenchard (see 2.a above). These words are unrelated. Where did Danker get this obsolete etymology from? It is not even in Boisacq. Did he happen to glance at Trenchard?

Danker’s foray into etymology is not a disaster, as are those of Trenchard and Newman, but it is ill-advised. The main problem, apart from inaccuracies, is the randomness of the information given. Danker simply chooses something, without any consistent system. One can see the difficulty that he faced: to summarise in a few words a complex and highly variable body of data on each word. But one could wish he had defined his system and objective more clearly, for himself as much as for us. Part of doing so would have been to decide what purpose the etymological information serves and how it would impact on the user.

5. CONCLUSION It looks as if there is a trend under way to reintroduce ‘etymologies’ into NT lexicons. The idea has already caught on and gained momentum in less than ten years. It is likely to continue and grow in strength. There is also the likelihood, or rather certainty, that etymological material, from these three lexicons or others yet to come, will make its way into electronic resources. All this will be accompanied by the usual copying of material, of whatever quality, and without scrutiny, from one source to another, as we see already in Trenchard and Newman. Mistakes, once in the tradition, will be extremely hard to dislodge. At this point in the process some reconsideration is desirable. First, there is the question of what the etymological information is for. What need does it serve and is it useful? The purpose of information of type a. is presumably to help the beginner to learn vocabulary by seeing derivations and linking new words with those already known. This has obvious pedagogical value and may be useful even for the non-beginner. The value is clear in Abbott-Smith’s presentation of, for example: ἀπιστέω (< ἄπιστος), ἀπιστία (< ἄπιστος), ἄ-πιστος (< ἀ- neg., πιστός); Danker’s is the same. Trenchard and Newman on the other hand reduce the value by linking the three words to verbs, Trenchard to πείθω and Newman to πιστεύω: the beginner has to do more work to get anything out of it. So even the simplest type a. can run into problems if it is not handled carefully.

ETYMOLOGICAL FOLLIES

337

The problems with type b. are well illustrated by the same group. When we look to πιστός we find it linked with πείθω in Abbott-Smith, Danker and Trenchard (and with πιστεύω in Newman). Is this useful? The beginner, and non-beginner too, will struggle with both the phonetics (πισ-τός < *πιθ-τός, zero grade of root) and the semantics (how does ‘faithful’ relate to ‘persuade’?). Our earlier instance of τέκνον (τίκτω), given like that in all four lexicons, is another illustration. In the light of these examples I personally have doubts that even basic information of types a. and b. is worth including in a lexicon. If it is to be included, thought must be given to how to present it in a way that is actually helpful and not misleading or puzzling. Type c., etymology proper, is even more problematic. There is a practical difficulty here to begin with, that is, the near impossibility of summarising in a short space the etymology of any word, given the complications and unpredictable features that constantly turn up, as we have seen so well illustrated. More serious still is the problem inherent in the whole exercise, that the beginner, and even the advanced NT scholar, has no framework of knowledge in which to fit the information and interpret it. A full course of training is barely enough to prepare one for it. For this reason I consider the exercise pointless and potentially harmful. But if such material does continue to be included, it is obvious that it must be prepared competently and on the basis of the latest etymological authorities, not simply drawn from predecessors. There is a further reason for caution about inclusion of etymological material, and that is the risk of fostering the use of etymology to draw conclusions based on the ‘original’ or ‘real’ meanings of words. The ‘etymological fallacy’ endures despite repeated warnings from scholars, and is not confined to students and preachers, or the biblical field. Even the seemingly innocuous statement ‘ἐκκλησία [ἐκ, καλέω],’ as in Danker, will continue to feed one of the classics of false exegesis: ἐκκλησία = ‘those called out’; etymologies of type c. offer still more alarming possibilities. The trend towards inclusion of etymologies in NT lexicons may have the unintended consequence of adding fuel to this fire.34 This article has had an entirely negative aspect, but that is unavoidable. There is good in these lexicons, but the mistakes in them vitiate the good. If some of the etymologies cannot be trusted, then all of them must be treated with caution. The beginner simply does not have the experience 34

For good discussions see Louw, Semantics, 26–31; Silva, Biblical Words, 44–51; Carson, Exegetical Fallacies, 28–33. I thank Steve Walton for reminding me of this important issue.

338

COLLECTED ESSAYS ON THE GREEK BIBLE

to sort them out. There is however a positive suggestion with which to end. The massive etymological reference works mentioned throughout this discussion are difficult even for the experienced scholar. What we need, and have long needed, is a concise etymological dictionary of Greek, one that summarises the information in a targeted, efficient way for the nonexpert. It could cover even basic derivations within Greek, as is already done in the big works, as well as the difficult questions of ultimate origin. If this challenge could be met, NT lexicographers could refer to or draw on this work and be relieved of the burden of dealing with something outside their competence.

21 THE ONOMA RULE 2014

Abstract The paper presents an unnoticed ‘rule’ governing the presence or absence of the article with ὄνομα in naming-constructions of the type ‘the name of x is N.’ It arises from the survival of an archaic pattern with the dative and no article, alongside a newer pattern with the genitive and the article. The ‘rule’ is shown to operate not only in the LXX and NT but in ancient Greek generally, from Homer to Koine Greek. Observance of the ‘rule’ indicates the LXX translators’ and the NT authors’ familiarity with idiomatic Greek. The ‘rule’ can also contribute to resolving textual questions, notably Luke 1:63.

1. THE LXX The grammatical ‘rule’ presented here was first noticed in the LXX Pentateuch, and then found to operate elsewhere in the LXX, in the NT, in Koine Greek generally, and in Classical Greek, where it originates. As far as I can discover, it has not been noticed before.1 Aside from its intrinsic interest, the ‘rule’ can be used as a proof of natural Greek and as a tool of textual criticism. I begin with the example that pointed the way to this discovery: Exod 15:3 κύριος συντρίβων πολέμους, κύριος ὄνομα αὐτῷ. The Lord is (the) shatterer of wars, the Lord is his name. :‫יהוה אישׁ מלחמה יהוה שׁמו‬

In κύριος ὄνομα αὐτῷ we have a ‘possessive’ dative αὐτῷ, meaning literally: ‘Lord (is) name to him.’ The Hebrew ‫ שׁמו‬is of course somewhat different, with a possessive suffix, meaning ‘his name.’ While the rendering is hardly strange as Greek, a comment by Wevers shows that some puzzlement has been felt: ‘Exod renders ‫ שׁמו‬by ὄνομα αὐτῷ, though one 1

Cf. KG, Grammatik, II.1, 45, 417; Schwyzer, Grammatik, II, 66, 86; Smyth, Grammar, §1478; Mayser, Grammatik, II.2, 113.

340

COLLECTED ESSAYS ON THE GREEK BIBLE

might have expected a genitive pronoun rather than a dative of possession.’2 In a recent paper Perkins is similarly puzzled, remarking that ‘the translator is quite careful normally to render the Hebrew-suffixed pronoun by a genitive form,’ going on to suggest that the rendering perhaps reflects an earlier passage, Exod 3:13 τί ὄνομα αὐτῷ;3 But a search of all occurrences of ὄνομα in the Pentateuch shows that the combination with the dative is not uncommon.4 Besides three with the pronoun αὐτῷ (Exod 3:13; 15:3 quoted above, and Gen 2:19 τοῦτο ὄνομα αὐτῷ), there are no less than 28 other instances, by my count, of ὄνομα with dative of the possessor of the name.5 Examples: Gen 4:19 ὄνομα τῇ μιᾷ Ἀδά, καὶ ὄνομα τῇ δευτέρᾳ Σελλά. :‫שׁם האחת עדה ושׁם השׁנית צלה‬ Gen 22:24 καὶ ἡ παλλακὴ αὐτοῦ, ᾗ ὄνομα Ῥεημά. ‫ופילגשׁו ושׁמהּ ראומה‬

As can be seen, the choice of the dative case here is not dependent on the Hebrew, and this is true of all the instances: the Hebrew has either ‫ שׁם‬with the possessive suffix (‫שׁמו‬/‫ )שׁמהּ‬or ‫ שׁם‬in the construct state + possessor of the name. The search confirms that the construction with genitive of the possessor occurs too, as expected, though less often than with dative – as Wevers and Perkins might not have expected: I find only 12 instances (with ὄνομα in the singular).6 Here is one of this type, side by side with one of the other: Exod 18:3 ὄνομα τῷ ἑνὶ αὐτῶν Γηρσάμ … 4 καὶ τὸ ὄνομα τοῦ δευτέρου Ἐλιέζερ ‫ ושׁם האחד אליעזר‬4 … ‫שׁם האחד גרשׂם‬ 2 3 4

5

6

Wevers, Exodus, 228. Perkins, ‘The Lord is Warrior,’ 129–30. All LXX and NT searches were made using the program ‘Accordance.’ LXX refs. were confirmed in the Göttingen ed. where available, otherwise Rahlfs. Full list of ὄνομα + dat. (total 31): Gen 2:11, 13; 2:19; 4:19, 19, 21; 10:25, 25; 11:29, 29; 16:1; 22:24; 24:29; 25:1; 28:19; 29:16, 16; 36:32, 35, 39, 39; 38:1, 2, 6; Exod 1:15; 3:13; 15:3; 18:3; Num 11:26, 26; 25:15. [I do not include Gen 26:33. The Göttingen ed. has διὰ τοῦτο ἐκάλεσεν ὄνομα τῇ πόλει Φρέαρ ὅρκου, which is not the onoma-rule pattern; Rahlfs’s text without ἐκάλεσεν is. See Wevers, Genesis, 415: ‘Admittedly, the verb is somewhat awkward, but the lectio difficilior must here be original.’ In my opinion, the onoma rule resolves the question in favour of Rahlfs’s reading.] Full list of τὸ ὄνομα + gen. (total 12): Gen 17:5, 15; 32:27(28), 28(29); 35:10, 10; Exod 1:15; 18:4; Lev 24:11; Num 25:14; 26:30(46); 26:59. There are also 17 of plural τὰ ὀνόματα + gen. in the expression ταῦτα τὰ ὀνόματα τῶν/αὐτῶν …: Gen 25:13 etc.

THE ONOMA RULE

341

So the case of the possessor can be varied by the translator, regardless of the Hebrew, which is the same each time. But there is something else that varies. In the verses just quoted, we see that ὄνομα has no article in the first instance, but has it in the second. Is this accidental? Is it just the translator exercising free choice? Investigation shows that there is a pattern here that is adhered to all through the Pentateuch. It is, put simply: in name-statements of the type ‘the name of x is N,’ if the possessor of the name is expressed in the dative, ὄνομα has no article; if the possessor of the name is expressed in the genitive, ὄνομα has the article. This may seem unexpected, but it turns out to be true; and the likely explanation will be seen in due course. The result of an enquiry into other books of the LXX is the same. The rule is followed in this selection: Joshua, Psalms, Isaiah, XII Prophets, and Ruth (in 1–4 Maccabees or Tobit there are no relevant cases), with the dative tending to be more common, as in the Pentateuch. Full data need not be given here. I simply illustrate with further instances of the two types, some in close proximity: Gen 32:27(28) Τί τὸ ὄνομά σού ἐστιν; Exod 3:13 Τί ὄνομα αὐτῷ; Exod 1:15 καὶ εἶπεν … ταῖς μαίαις τῶν Ἐβραίων, τῇ μιᾷ αὐτῶν, ᾗ ὄνομα Σεπφωρά, καὶ τὸ ὄνομα τῆς δευτέρας Φουά Num 25:14 τὸ δὲ ὄνομα τοῦ ἀνθρώπου τοῦ Ἰσραηλίτου … Ζαμβρί Num 25:15 καὶ ὄνομα τῇ γυναικὶ τῇ Μαδιανίτιδι … Χασβί Judg 1:23AB τὸ δὲ ὄνομα τῆς πόλεως (αὐτῶν) ἦν ἔμπροσθεν Λουζα. Judg 18:29A καὶ ἦν Λαις ὄνομα τῇ πόλει τὸ πρότερον. Ruth 2:1 καὶ ὄνομα αὐτῷ Βόος. Ruth 2:19 τὸ ὄνομα τοῦ ἀνδρὸς … Βόος. Isa 42:8 ἐγὼ κύριος ὁ θεός, τοῦτό μού ἐστι τὸ ὄνομα. Isa 51:15 κύριος σαβαωθ ὄνομά μοι.

In these cases we are dealing with name-statements of the form ‘the name of x is N.’ Outside of such name-statements the phrase τὸ ὄνομα + gen. of the possessor of the name can of course occur, and does so frequently in the Pentateuch, when τὸ ὄνομα is object or subject of a verb. A small sample out of many: Gen 38:4 ἐκάλεσεν τὸ ὄνομα αὐτοῦ Αὐνάν. Exod 23:21 τὸ γὰρ ὄνομά μού ἐστιν ἐπ᾽ αὐτῷ. Num 11:3 καὶ ἐκλήθη τὸ ὄνομα τοῦ τόπου ἐκείνου Ἐμπυρισμός.

These show the standard pattern of noun + possessive gen. in which the article is usual, as in ὁ θεός μου, ἡ ψυχή μου etc. In name-statements,

342

COLLECTED ESSAYS ON THE GREEK BIBLE

then, the translators are simply going over to this pattern when they depart from the pattern with the dative. It is the latter that is somewhat unusual in its syntax. We may already suspect that it was an older, special idiom. Yet the data show that it was normal for the translators, and that sometimes they use the other pattern, τὸ ὄνομα + gen., only as a varier from it (Exod 1:15; 18:3; Num 25:14–15 above).7 2. LXX PERIOD Was this ‘rule’ a peculiarity of the LXX translators’ Greek? It is hardly likely. But to answer this question it is imperative to investigate Greek outside the LXX. Contemporary documents are a good place to start. In documentary papyri of III–I BC, at least 30 name-statements can be found (including plurals).8 Nearly all show dative of the possessor and conform to the rule, that is, ὄνομα has no article. Expressions with the relative pronoun, like Gen 22:24 ᾗ ὄνομα Ῥεημά above, are commonest, providing a close parallel to Pentateuch usage (ᾧ/ᾗ ὄνομα Ν occurs nine times there). The pattern τὸ ὄνομα + gen. of the possessor is also found once (in plural) in a name-statement; but it is mostly found as subject or object of a verb. A small selection to illustrate: PCairZen 1.37.7 (258/7 BC) ὄνομα δʼ ἐστὶ αὐτῶι Ἡδύλος· PCairZen 2.148.2 (III BC) παῖδα … ὧι ὄνομα Ὀνήσιμος. PMich 1.103.22 (III BC) γράψον οὖν τῶι ἀρχιφυλακείτηι· ὄνομα δʼ αὐτῶι Ἁρμυῦσις. PCairZen 3.379.11 (III BC) ἔστι[ν] τὰ ὀνόματα τῶν [ὑ]φειρημ[έ]νων [τὰ] ἱερεῖα Κ … Ψε[ν]πεσῶτος … PTebt 3.793ii.16 (183 BC) … τοῦ Δημητρίου οὗ σοι ὑπόκειται τὸ ὄνομα

Inscriptions tell the same story.9 The first 500 matches to ὄνομα in inscriptions of all periods follow the expected pattern: when ὄνομα is accompanied by a dative it has no article. The type with the relative pronoun, ᾧ/ᾗ ὄνομα Ν, is overwhelmingly frequent, being formulaic in manumissions. Other examples following the rule do occur, such as Ἀντί[οχος] ὄνομα αὐτῶι ἐστι.10 The pattern τὸ ὄνομα + gen. as object of a verb can also be exemplified. 7

8 9 10

I owe this point to Michael Curran. My thanks to him and to Trevor Evans and Terry Roberts for their helpful comments on this paper. Accessed via ‘Papyri.info.’ Cf. Mayser, Grammatik, II.2, 113. Accessed via ‘PHI Inscriptions.’ Rizakis, Achaïe III.2.8 (II BC) = SIG 530.8.

THE ONOMA RULE

343

This evidence already indicates that LXX usage was in accordance with the Greek of its time. But let us sample the literature of the same period by looking at Polybius (II BC) and Diodorus Siculus (I BC).11 There are not many instances in either, but they follow the expected pattern. (Diodorus much prefers parenthetical ὄνομα + name.) Examples in namestatements: Plb. 13.7.6 τοῦτο δ’ ἦν ὄνομα τῇ γυναικὶ τοῦ Νάβιδος. D.S. 20.33.5 τοῦτο γὰρ ἦν ὄνομα τῇ γυναικί.

Examples with τὸ ὄνομα + gen. as object of a verb: Plb. 15.24.4 πᾶσι προτείνουσι τὸ τῆς ἐλευθερίας ὄνομα. D.S. 9.2.3 ἀνεβόα συνεχῶς τὸ τοῦ Σόλωνος ὄνομα.

3. CLASSICAL PERIOD It is time now to look back into what came before, that is, into Greek of the Classical period. Where better to start than with Odysseus’s famous trick answer to the Cyclops: Homer, Od. 9.366 Οὖτις ἐμοί γ᾽ ὄνομα· My name is Nobody.

Here is the pattern ὄνομα + dative already established in the oldest Greek we have after Mycenaean, with absence of the article as in earliest Greek usage.12 There are some other examples in Homer and the Homeric Hymns.13 We may reasonably conclude that this was an archaic pattern, maintained in Greek long after it first appeared. In fact there is a good chance the pattern originated in Indo-European: there is evidence of it in Hittite and Latin (though not in Indo-Iranian).14 To fill in the picture of usage in the Classical period after Homer I have tested (wholly or in part) the works of these authors: Herodotus, Aristophanes, Xenophon, and Plato. In Herodotus books 1 to 3, I find 56 instances of οὔνομα (Ionic form) + dat. following the rule, of which 11 12

13 14

Searches conducted via TLG. Cf. Chantraine, Grammaire homérique, 2:158–66. The dat. in Od. 9.366 and like instances Chantraine only says is ‘assez proche du génitif exprimant la possession’ (71). H., Od. 19.183, 409; 24.306. H. Aphrod. 198; Dem. 122. (There are none in Hesiod.) Hahn, Naming-Constructions, 10, 24–5, 116–7, 124–5. It also occurs in Old English (Hahn, 169–71, 178–80) but I am not certain if any of the exx. are independent of translation of biblical texts. I am grateful to James Clackson and John Sheldon for assistance here.

344

COLLECTED ESSAYS ON THE GREEK BIBLE

a specimen is: 1.34.2 οὔνομα δέ οἱ ἦν Ἄτυς.15 There are two apparent exceptions.16 Aristophanes offers a much smaller number of instances (8) but all follow the rule without exception, as in Thesm. 1200 ὄνομα δέ σοι τί ἐστιν; Xenophon similarly has a small number (7), all following the ὄνομα + dat. pattern (he has more of parenthetical ὄνομα + name).17 Plato provides important evidence, with some 273 examples of ὄνομα.18 The Cratylus has more than half the examples and much variation in syntax in Plato’s playful discussion of ‘names’ of things. Even so, I find that the great majority of the name-statements behave exactly as predicted. By my count there are 42 in Plato, of which 39 obey the rule. Of these the usual pattern ὄνομα + dat. prevails (36 times), but there are three instances of τὸ ὄνομα + gen., confirming the alternative pattern. (Occurrences of the latter as object of a verb can also be given.) The remaining three are anomalous, involving οὗ ὄνομα, but they may have explanations. I offer brief illustrations of each type: ὄνομα + dat.: Euthyd. 275b.1 ὄνομα δ᾽ αὐτῷ Κλεινίας. Symp. 209a.7 … ᾗ δὴ ὄνομά ἐστι σωφροσύνη τε καὶ δικαιοσύνη. Ion 531e.8 τίς οὗτος; τί ὄνομα αὐτῷ; τὸ ὄνομα + gen.: Phileb. 12b.8 … τὸ δ᾽ ἀληθέστατον αὐτῆς ὄνομα Ἡδονὴν εἶναι.19 τὸ ὄνομα + gen. as obj. of verb: Theaet. 144c.8 τὸ δ᾽ ὄνομα οὐκ οἶδα τοῦ μειρακίου. anomalous ὄνομα + gen.: Parm. 147e.2 … ἢ ἐκεῖνο οὗπερ ἦν ὄνομα.20

4. THE NT We now know that the LXX translators were following established patterns in their observance of ‘the onoma rule.’ Next it will be interesting 15 16

17

18 19 20

οἱ = dat. sing., masc./fem. of old pers. prn. Hdt. 1.1.3 … τοῦ βασιλέος θυγατέρα· τὸ δὲ οἱ οὔνομα εἶναι … Ἰοῦν τὴν ᾽Ινάχου. 1.179.4 ἔστι δὲ ἄλλη πόλις …· Ἲς οὔνομα αὐτῇ. ἔνθα ἐστὶ ποταμὸς οὐ μέγας· Ἲς καὶ τῷ ποταμῷ τὸ (v.l. τῶ) οὔνομα. For τὸ οὔνομα + gen. as obj. of a verb see, e.g., 1.51.4; 2.115.2; 3.64.5. I am not sure if X., Oec. 1.1 (similarly 6.4), with an abstract possessor, is really anomalous: ἆρά γε ἡ οἰκονομία ἐπιστήμης τινὸς ὄνομά ἐστιν, ὥσπερ ἡ ἰατρικὴ …; ‘Is home management the name of a certain science, like medicine …?’ Including ‘Spuria’ but not ‘Dubia.’ TLG search 5.09.13. The two others of this pattern: Crat. 430a.7; Theaet. 144d.1 (sc. αὐτοῦ). Is the text sound?: contrast 147d.4 οὗπέρ ἐστι τοὔνομα, e.6 ἧσπερ ἦν τοὔνομα. The other two exceptions: Crat. 432e.2; Meno 74e.11.

THE ONOMA RULE

345

to go forward to the NT and see what we find.21 Will things have changed? Will the NT authors know these patterns? There turn out to be 19 namestatements of the kind we are looking for, of which no less than 14 adhere to the old pattern with ὄνομα + dat., while four use the alternative with τὸ ὄνομα + gen., again confounding expectations. I include one with plural ὀνόματα, which also follows the rule. The one exception (Luke 1:63) will be discussed below. It will be useful to see them all, the datives first, then the genitives: Mark 5:9 τί ὄνομά σοι; Mark 5:9 καὶ λέγει αὐτῷ· λεγιὼν ὄνομά μοι Luke 1:26 εἰς πόλιν τῆς Γαλιλαίας ᾗ ὄνομα Ναζαρέθ Luke 1:27 πρὸς παρθένον ἐμνηστευμένην ἀνδρὶ ᾧ ὄνομα Ἰωσὴφ ἐξ οἴκου Δαυὶδ Luke 2:25 ἄνθρωπος ἦν ἐν Ἰερουσαλὴμ ᾧ ὄνομα Συμεών Luke 8:30 τί σοι ὄνομά ἐστιν; Luke 8:41 ἦλθεν ἀνὴρ ᾧ ὄνομα Ἰάϊρος Luke 24:13 εἰς κώμην … ᾗ ὄνομα Ἐμμαοῦς John 1:6 ἐγένετο ἄνθρωπος, ἀπεσταλμένος παρὰ θεοῦ, ὄνομα αὐτῷ Ἰωάννης· John 3:1 ἦν δὲ ἄνθρωπος ἐκ τῶν Φαρισαίων, Νικόδημος ὄνομα αὐτῷ John 18:10 ἦν δὲ ὄνομα τῷ δούλῳ Μάλχος. Acts 13:6 εὗρον ἄνδρα τινὰ … ᾧ ὄνομα Βαριησοῦ Rev 6:8 καὶ ὁ καθήμενος ἐπάνω αὐτοῦ ὄνομα αὐτῷ [ὁ] θάνατος Rev 9:11 ἔχουσιν … βασιλέα τὸν ἄγγελον τῆς ἀβύσσου, ὄνομα αὐτῷ Ἑβραϊστὶ Ἀβαδδών Matt 10:2 τῶν δὲ δώδεκα ἀποστόλων τὰ ὀνόματά ἐστιν ταῦτα· Mark 14:32 καὶ ἔρχονται εἰς χωρίον οὗ τὸ ὄνομα Γεθσημανί Luke 1:5 καὶ τὸ ὄνομα αὐτῆς Ἐλισάβετ Luke 1:27 καὶ τὸ ὄνομα τῆς παρθένου Μαριάμ.

It is remarkable that these are spread through different books rather than being the achievement of one author, and that the ‘rule’ is followed so well by writers supposedly more familiar with Aramaic and, in the case of Revelation, barely competent in Greek. Attempts to find a Hebraism in ὄνομα αὐτῷ in Rev 6:8; 9:11 (and in John) can hardly convince, when the dative is Greek idiom and the supposed Hebrew model ‫( שׁמו‬and Aram. ‫ )שׁמהּ‬has the possessive suffix (= αὐτοῦ).22 In Luke 1:27 Luke achieves an elegant variation by using both patterns, first dat., then gen. 21

22

The standard ΝΤ treatments do not remark on the article with ὄνομα: Robertson, Grammar, 460; BDF, §§128.3, 144; MHT, Grammar, III, 230, 295–6, 304–5, 310. See MHT, Grammar, III, 295; Charles, Revelation, 169 n. 1, 246 n. 2; Thompson, S., Apocalypse, 89; Burney, Aramaic Origin, 30–2.

346

COLLECTED ESSAYS ON THE GREEK BIBLE

5. NT PERIOD Evidence of Greek contemporary with the NT is to be brought into the picture here. I have checked three authors of the first to second centuries AD, Josephus (AJ only), Plutarch, and Epictetus, and found confirmation that the patterns continued to apply. Both Josephus and Plutarch use ὄνομα + dat. quite often (I count 27 and 21 each) in accordance with the rule, though they much prefer the parenthetical ὄνομα + name pattern, probably because of its more literary tone. I do not find the other pattern τὸ ὄνομα + gen. in Plutarch, but there is one nice example in Josephus: AJ 1.36 τὸ δ᾽ ἐκείνης ὄνομα τῆς γυναικὸς Εὔα ἦν.23 Epictetus has only two relevant examples and both show ὄνομα + dat. as expected. The evidence from papyri of the first to third centuries AD confirms the persistence of the ὄνομα + dat. pattern, mainly in the formula ᾧ/ᾗ ὄνομα Ν (I count about 30 instances); there are none of the pattern τὸ ὄνομα + gen., except as subj./obj. of a verb (frequent). Plurals match the pattern too: I find several each of ὀνόματα + dat. and τὰ ὀνόματα + gen. But there is something new emerging in this documentary evidence. I find not one – which might not mean much – but five instances of ὄνομα + gen. without article, even in the formulaic structure with the relative pronoun: BGU 2.423.22 + BL 8.27 (II AD) ἔσ[τ]ι[ν] μου ὄνομα Ἀντῶνις Μάξιμος. POxy 50.3555.5 (I/II AD) θεραπαινίδιόν μου οἰκογενέ[ς], οὗ ἔστιν ὄνομα Πεῖνα SB 22.15614.9 (154 AD) \ἀπὸ κοπρίας δο[υ]λ[ι]κὴν/ ἧς ὄνομα Θαήσιος24

I interpret this as an indication that the archaic pattern ὄνομα + dat. was beginning to break down. The more intuitive genitive was replacing the dative, though the anarthrous ὄνομα remained from the old pattern. Our documentary examples point to II AD as the time when the shift was under way in non-literary Greek. The progressive loss of the dative case is sure to have been a factor.25 There is interesting confirmation of the shift from another quarter. In Vita Aesopi G, a popular work datable to I AD, there is one name-statement, which reads (24.2): τί ὄνομά σου; Notable is the contrast with Mark’s τί ὄνομά σοι; 23

24 25

Plutarch has two exceptional exx. of ὄνομα + gen. without article: Is. Os. 354c.8; 357e.8 (both in or. obl.). Cf. the papyrus exx. following. The others are: PLond 2.360.9 (p. 216) (II AD); StudPal 22.40.10 (150 AD). Cf. Browning, Greek, 36–8.

THE ONOMA RULE

347

6. NT TEXT What of the NT exception to the rule? It is a case where the ‘onoma rule’ can bring something to textual criticism: Luke 1:63 ἔγραψεν λέγων· Ἰωάννης ἐστὶν ὄνομα αὐτοῦ. τὸ ὄνομα αὐτοῦ ‫ א‬A B2 C D W Θ Ψ … ै ὄνομα αὐτοῦ ो4 B* L Ξ 565. 579. 700. …

Thus NA27, where according to the rule we expect τὸ ὄνομα αὐτοῦ as in the majority reading, or ὄνομα αὐτῷ. The decision to adopt the minority reading seems to have been made initially by Westcott and Hort, and the discovery of ो4 (III AD) would have seemed strong confirmation. We must resist the temptation simply to restore τὸ ὄνομα αὐτοῦ to the text in accordance with the rule. If this reading was original, we need to explain how the other, clearly early, reading ὄνομα αὐτοῦ arose. Simple omission of τό is possible but not likely. The evidence we have seen of the breakdown around II AD of the archaic pattern, leading to ὄνομα + gen. instead of dat., might support ὄνομα αὐτοῦ as original, were it not for the consideration that the well-educated Luke, whose Greek accords with the ‘rule’ elsewhere, is unlikely to have made this ‘mistake’: it is more likely to be the error of a later copyist affected by the new pattern. This argument leads to the conclusion that the original reading was in fact ὄνομα αὐτῷ. The attested readings are then explained: ὄνομα αὐτῷ was altered to ὄνομα αὐτοῦ for the reason suggested, then ὄνομα αὐτοῦ was corrected to τὸ ὄνομα αὐτοῦ in line with the normal, ‘correct’ pattern of a noun with a possessive genitive.

7. LXX TEXT Let us return to where we began, the LXX Pentateuch. The question of the text arises there too in some instances: Exod 3:15 τοῦτό μού ἐστιν ὄνομα αἰώνιον καὶ μνημόσυνον γενεῶν γενεαῖς.

There is a variant μοι in a few MSS (and Patristic citations), which would remove the anomaly. Wevers thinks it might have arisen ‘under pressure of the τί ὄνομα αὐτῷ of v. 13.’26 It could also be pressure of the standard pattern. But the reason for the anomaly may lie in the nature of the 26

Wevers, Exodus, 34.

348

COLLECTED ESSAYS ON THE GREEK BIBLE

sentence: τοῦτο refers back to ὁ ὤν of v. 13 and does not so much state the name – which may not be a name but a description or appellation – as tell us something further about it, as NETS has it: ‘This [ὁ ὤν] is an everlasting name of mine and a memorial of generations to generations.’ In Gen 2:19 τοῦτο ὄνομα αὐτῷ, there is a variant αὐτοῦ, which Rahlfs accepted in his text. The rule supports Wevers’s decision not to adopt it (‘the A + variant αυτου is not to be preferred’).27

8. CONCLUSION Far more significant than its value as a tool of textual criticism is what the ‘onoma rule’ indicates about the LXX translators’ command of Greek. An archaic pattern still current in the language of their time is carefully adhered to and not confused with a more recent pattern also used, and the former is used contrary to the structure of the original Hebrew. Their familiarity with this subtlety is strong evidence of their knowledge of idiomatic Greek. It is on a par with their intermittent adherence to another unobtrusive pattern of Greek syntax, the forward placing of an enclitic personal pronoun (‘Wackernagel’s Law’) against the Hebrew order. This is significant enough to have led one recent scholar to conclude that they had nativespeaker competence.28 Their observance of the present rule is as significant and even more consistent. The NT phenomena similarly indicate the NT authors’ familiarity with Greek idiom. I conclude with a summary of the ‘onoma rule.’ In name-statements of the type ‘the name of x is N,’ with the components ὄνομα + possessor of name + name (+ copula), if the possessor is expressed in the dative, ὄνομα is without the article; if the possessor is expressed in the genitive, ὄνομα has the article. Thus: ὄνομα αὐτῷ Ν (ἐστιν) / τὸ ὄνομα αὐτοῦ Ν (ἐστιν). The word-order is variable and the copula is optional. The rule originates in an archaic pattern with the dative, probably descended from Indo-European, and operates in Greek from the earliest times to at least Middle Koine (I–III AD), when it begins to break down. Observance of this rule by writers of Greek is of course to be understood as largely intuitive, not the product of conscious application of a rule learnt as such. 27 28

Wevers, Genesis, 32. [See above, n. 5 for another LXX textual question.] Janse, ‘Aspects of Bilingualism,’ 383. [See further Lee, Greek of the Pentateuch, 123– 7.]

THE ONOMA RULE

349

Postscript Sollamo, in ‘Usage of the Article’ (2020), includes the onoma rule among her evidence for concluding that the Genesis translator ‘must have been a native Greek speaker’ (53). In the same paper, Sollamo refers (51 n. 20, 52 n. 21) to a forthcoming article on a related subject. It will have the title ‘The Onoma Rule, the Vocative, and Some Other Free Renderings of Hebrew Nouns in the Status Constructus Defined by Another Noun in the Status Absolutus or by a Pronominal Suffix in the LXX Genesis.’ (I thank Raija Sollamo for this information.)

22 THE LITERARY GREEK OF SEPTUAGINT ISAIAH 20141

Abstract The Greek of LXX-Isaiah was characterised by Thackeray many years ago as ‘Good κοινή Greek.’ This label has been accepted with little question generally and been very influential among Isaiah scholars. The present paper first sets out to establish that to classify the books of the LXX we need to give up Thackeray’s mixed and imprecise categories and to recognise three main criteria of classification, namely: ‘translation method,’ ‘style,’ and ‘level of Greek.’ The aim of the paper is then to present evidence to demonstrate that LXX-Isaiah is appropriately classified as ‘literary’ under the third category, along with other books so labelled by Thackeray. The implications for a better understanding of the translator and his methods are explored, especially in the light of the depictions by A. van der Kooij and R. Troxel.

1. PREAMBLE The Greek translation of Isaiah has been the subject of intense activity in recent research, as is well known to any who have taken an interest in the book. Van der Kooij’s publications have been prominent over more than a decade and are ongoing.2 Notable monographs have appeared recently, by Baer, Troxel, De Sousa, and Wagner.3 There is a new collection of essays by leading scholars in the field,4 and of course numerous other articles have been written. Work is in progress on the translation and commentary in the series Bible d’Alexandrie. All these build on the efforts of 1

2 3

4

The material of this paper was first presented as the fifth Grinfield Lecture in Oxford in February, 2012, and again in the lecture series ‘La Bible grecque des Septante’ in Paris in April, 2013. Van der Kooij, Oracle of Tyre, and, e.g., ‘Septuagint of Isaiah.’ Baer, When We All; Troxel, LXX-Isaiah; De Sousa, Eschatology; Wagner, Sealed Book. Another I have not yet seen is Ngunga, Messianism. Van der Kooij and van der Meer (eds.), Old Greek of Isaiah.

352

COLLECTED ESSAYS ON THE GREEK BIBLE

earlier scholars, notably Ottley, Ziegler, and Seeligmann.5 So it is a field already well occupied. This paper is about one aspect of the book that has interested me for some years but that I see as neglected by the Isaiah specialists. The Isaiah translator’s work has greatly engaged the interest of LXX scholars – and provided enormous scope for inventive interpretation – because of his method of translation. It is characterised by inconsistency: he translates his Hebrew original both accurately and freely, in an unpredictable way. Older responses reflect the disquiet felt at this. Swete said the Isaiah translation (along with Psalms) shows ‘obvious signs of incompetence.’6 Walters called the translator ‘an ingenuous blunderer,’7 and bequeathed us the very quotable remark: ‘he was completely unequal to his task.’8 Of course our response is: what was his task? How did he view it? We ought, surely, to be wary of putting our own expectations onto him. In recent scholarship this kind of criticism has been dropped and much more sensitive attempts have been made to understand the translator’s approach. The focus of much of this research has been on identifying the translator’s exegetical tendencies. Minute attention has been given to deviations from a strictly literal rendering and significance extracted from them. The results have not led to agreement, however, but to ongoing debate.

2. THE

CLASSIFICATION OF

LXX-ISAIAH

This is merely background, though not irrelevant, to the main subject of this paper, which is the Greek of LXX-Isaiah. By this I mean the Greek viewed as a specimen of the written language in its own time. Its time I take to be the second century BC, following the general consensus. It is necessary to begin with Thackeray’s well-known classification of the LXX books, in which he put Isaiah in his group 1, along with the Pentateuch, ‘Joshua (part),’ and 1 Maccabees. He labelled this group ‘Good κοινή Greek.’9 The more I have looked at this description the more puzzling, 5

6 7 8

9

Ottley, Book of Isaiah; Ziegler, Untersuchungen; Seeligmann, Septuagint Version. Cf. bibliography in van der Kooij and van der Meer (eds.), Old Greek of Isaiah, 211–23. Swete, Introduction, 315–6. Walters, Text, 242. Katz, ‘Septuagintal Studies,’ 200. Troxel (‘βουλή and βουλεύειν,’ 155 n. 6) quotes the first description as ‘an ingenious blunderer,’ which looks like an ingenuous improvement. Neither works for me. Thackeray, Grammar, 13. I do not know what Thackeray meant by ‘Joshua (part).’

THE LITERARY GREEK OF SEPTUAGINT ISAIAH

353

even deceptive, it seems. It is not at all clear. ‘Koine Greek’ we understand; good Koine can mean almost anything you want it to mean. It could be ‘good’ in the sense of idiomatic everyday Greek; ‘good’ in the sense of high-quality learned Greek; ‘good’ in the sense of solid middle-of-theroad grammatical Greek; and perhaps more. The other groups that follow in Thackeray’s classification are different in character from group 1 and each other. Group 2 ‘Indifferent Greek,’ is, it appears, about the quality of the Greek as Greek, but the term is very vague and might well refer to translation method; Thackeray himself says this group contains ‘books of various styles’ [my emphasis].10 Then follows group 3, ‘Literal or unintelligent versions,’ which must refer to translation method. Then the two groups 4 ‘Literary,’ and 5 ‘Literary and Atticistic,’ concern quality of Greek on a scale of ‘literariness,’ and are at least clear, as is the final group 6, ‘Vernacular,’ indicating a position on the same scale. Just before he gives his table, Thackeray himself confuses things by saying that he classifies the books ‘into groups from the point of view of style’ (p. 12). This is hardly true, except insofar as ‘style’ is allowed to have multiple meanings. He goes on to say, ‘Isaiah, considered as a translation, would certainly not be placed in the first class’: now he seems to be referring to translation method. Similarly, in the preceding discussion of items of vocabulary used as a test (pp. 9–12), it seems clear (mostly) that ‘style’ refers to translation method. All this is getting us nowhere. The problem is that Thackeray is mixing the categories, as everyone knows and others have pointed out before.11 As I see it, to fix the problem we need to distinguish three main categories or criteria of classification, which are: (a) ‘translation method’; (b) ‘style’; (c) ‘level of Greek.’ To these might be added one or two others, notably ‘control/use of idiomatic Greek,’ but this would not change the main groupings. The intended meaning of these categories will be more fully explained in the discussion to follow. Thackeray’s groups do not separate these properly, and confusion has ensued. The whole matter would not need to be pursued in such detail if it were not that Thackeray’s classification has been very influential and is still quoted as the primary authority. LXX-Isaiah scholars seem to have generally accepted Thackeray’s ‘Good κοινή Greek’ label trustingly, and I think it has been unhelpful to them. Let us take an example from a recent paper by van der VormCroughs, a very good one as it happens, on the presence of rhetorical figures in Isaiah. Her argument that the Isaiah translator was familiar with 10 11

Thackeray, Grammar, 12. E.g., Sollamo, ‘Some “Improper” Prepositions,’ 775; Barr, Typology, 283.

354

COLLECTED ESSAYS ON THE GREEK BIBLE

Classical rhetoric is already convincing, but she adds that it is ‘supported by the fact [my emphasis] that LXX Isaiah’s translator was writing in good Koine Greek, as was pointed out by Thackeray more than a century ago.’12 She takes it as a fact; and she evidently understands Thackeray’s description as referring to style: rhetorical devices are a feature of style (as van der Vorm-Croughs well knows). But if Thackeray’s group 1 is a style group, then what about the other books in that category, the Pentateuch, Joshua and 1 Maccabees? Are they similar in style to Isaiah? Clearly not.13 The aim of this paper is not to offer a new classification of the LXX books – though that is certainly needed – but to demonstrate something about the Greek of Isaiah that has hitherto not been appreciated by all parties, including Thackeray. My aim, in brief, is to show that LXX-Isaiah has features of language that justify classifying it as ‘literary’ Greek. That of course falls under my category (c) ‘level of Greek’ proposed above. Before I come to the evidence under that heading, it will be useful to say a little about the other two categories, (a) ‘translation method’ and (b) ‘style,’ to show what I mean by them. The discussion of these will be illustrative rather than a full enquiry, and does not claim to be saying anything new. I will then move on to (c), my main focus.

3. TRANSLATION

METHOD

Description of translation method in the LXX has traditionally been framed in terms of ‘literal’ versus ‘free.’ This will serve for the present purpose. It at least gives us a means of indicating what we are talking about under this heading and differentiating the category from the other two. Nevertheless, the whole matter is complicated. The terms ‘literal’ and ‘free’ are inexact, but it has been hard to replace them so far. Barr in his much-admired Typology says at the start that they are unsatisfactory, but at the end of his investigation has nothing better to offer.14 There has been renewed discussion lately, in an attempt to refine them in some way, for example by Lemmelijn.15 Barr, while not changing the terminology, showed that literalism is a complex phenomenon and there are ‘variations 12 13

14

15

Van der Vorm-Croughs, ‘LXX Isaiah,’ 175. Other reiterations of Thackeray’s description: Seeligmann, Septuagint Version, 43; van der Kooij, ‘Isaiah in the Septuagint,’ 515, 518; De Sousa, Eschatology, 18. Barr, Typology, 6; cf. 49–50. The attempt to identify criteria for literalness made in Tov, Text-Critical Use (1981), 53–66 illustrates the difficulty. Lemmelijn, Plague of Texts?, 108–14; see esp. Boyd-Taylor, ‘Classification of Literalism.’ Cf. Lee, ‘Accuracy and Idiom’ for a different angle.

THE LITERARY GREEK OF SEPTUAGINT ISAIAH

355

within a basically literal approach.’16 In addition to the aspects of literalism that he identified, it seems to me there is another aspect that needs to be allowed for, that is, whether the translation in any given place is natural or unnatural Greek. This will be illustrated from Isaiah in a moment. Another difficulty is that to label or characterise a whole book accurately, one really needs to analyse every rendering throughout the book. But it has proved almost impossible to do that (so far), despite the extensive analysis of selected features that has been achieved, especially by scholars of the Finnish school. So we still rely on a combination of hard data and impression. My purpose here is not to solve these problems, but simply to sum up the Isaiah translator’s translation method. In terms of ‘literal’ versus ‘free,’ it is obvious that the work cannot be characterised overall as ‘literal’ because ‘free’ renderings occur with great frequency. But this translator can also be ‘literal.’ Van der Louw has rightly said, ‘literal translation constitutes his most important tool.’17 The fact is that the Isaiah translator is both ‘literal’ and ‘free’ – sometimes both within the one sentence. So it is actually quite difficult to sum him up.18 In addition it seems clear that the translator has sometimes had problems understanding his Hebrew, and has simply got an impression and worked from that. This could be the reason for many of the ‘free’ renderings, rather than a deliberate intention to depart from a ‘literal’ rendering and introduce an interpretation.19 I offer just two examples to illustrate these remarks and the translator’s method. Let us begin with an example of ‘literal’ rendering, in a famous locus: (1) Isa 6:3 καὶ ἐκέκραγον ἕτερος πρὸς τὸν ἕτερον καὶ ἔλεγον Ἅγιος ἅγιος ἅγιος κύριος σαβαωθ, πλήρης πᾶσα ἡ γῆ τῆς δόξης αὐτοῦ.

‫וקרא זא אל־זה‬ ‫ואמר קדושׁ קדושׁ קדושׁ‬ ‫יהוה צבאות‬ :‫מלא כל־הארץ כבודו‬

The translation is very ‘literal,’ that is, word-for-word or ‘isomorphic’: the Greek word-order and parts of speech match the Hebrew; and the Greek has the same meaning as the original. But it is not totally literal. The translator has changed the verbs (ἐκέκραγον, ἔλεγον) to plural, and 16 17 18

19

Barr, Typology, 7. Van der Louw, Transformations, 243. The same point has of course been made by others: see Troxel, LXX-Isaiah, 87–8; Wagner, Sealed Book, 113. See Wagner, Sealed Book, 30 n. 151 and 49–50 for a good summary of the difficulties faced both in reading and in understanding the Hebrew text.

356

COLLECTED ESSAYS ON THE GREEK BIBLE

has turned ‫‘( זא אל־זה‬this one to this one’) into natural Greek by using the idiom ἕτερος πρὸς τὸν ἕτερον, yet without changing the meaning (‘they cried out to one another’). So we see one of the ways in which ‘literal’ can vary in itself, in this case between more natural and less natural Greek, and we see how the Isaiah translator may be literal and idiomatic at the same time.20 My second example illustrates the Isaiah translator’s ‘freedom,’ combined with literal rendering: (2) Isa 2:16 :‫… ועל כל־אניות תרשׁישׁ ועל כל־שׂכיות החמדה‬ [The day of the Lord will come] … and on all the ships of Tarshish and on all the ships(?) of desire. LXX … καὶ ἐπὶ πᾶν πλοῖον θαλάσσης καὶ ἐπὶ πᾶσαν θέαν πλοίων κάλλους. [The day of the Lord will come] … and upon every ship of the sea and upon every spectacle of beauty of ships (or: of ships of beauty).

How exactly he arrived at this rendering we need not analyse in detail.21 What is interesting is that he follows the general shape of the original, but with the freedom he allows himself, he creates a meaningful and, as I see it, almost poetic rendering. The turning of the two last Hebrew words into three (θέαν πλοίων κάλλους) achieves this effect. We notice also the alliteration in π, the repetition of πλοῖον, and possible hyperbaton in the final phrase (θέαν … κάλλους). These elements belong to style, which is our next category.

4. STYLE Every writer has his or her own individual style. It can be hard to say what constitute the elements of style, and even harder to say what is good or bad style. It is an elusive category, and very subjective; but it is real. Style relates to the features which depend on the personal choice of the writer, and can include almost anything. Even literal translation can be seen as a style choice; but style is a separate entity from translation method. 20

21

For the idiom see LSJ, s.v. ἕτερος I.3. In Muraoka, Lexicon, s.v. ἕτερος 5, it is mistakenly marked as unattested before the LXX, perhaps from a misreading of Thackeray, Grammar, 45. It occurs three more times in Isa at 13:8; 34:14, 16, with the varier ἀλλήλων in 34:15. Cf. Ottley, Book of Isaiah, 2:114. Aramaic is likely to be involved in the interpretation of ‫שׂכה‬.

THE LITERARY GREEK OF SEPTUAGINT ISAIAH

357

My own impression of the Isaiah translator is that he is a skilled stylist, very conscious of what he is doing, and that the style of his version is a most important, perhaps overriding, concern to him. His skill as a stylist has of course been noted by others. What are the indicators of an interest in style? One sure indicator is the use of rhetorical devices. I mean ancient Greek rhetorical devices (though we still use many of them today). In the context of translation we mean those not in the original but introduced or contrived by the translator. These are clearly present in LXX-Isaiah. Van der Vorm-Croughs has noted examples of inclusio, epiphora, anaphora and parallelism, as well as avoidance of excessive use of some of these, especially repetition.22 I made some slight observations myself some years ago of variatio, alliteration and asyndeton in Isaiah.23 I have not seen the dissertation by Philippe Le Moigne, which seems to be an important contribution; I know that van der Kooij refers to it for a treatment of chiasmus in Isaiah, which of course is a notable stylistic tool.24 Van der Louw in his study of Isaiah chapter 1 has identified a remarkable instance of double chiasmus purposely created by the translator (Isa 1:25).25 I select here just one passage to illustrate the Isaiah translator’s interest in style: Isa 30:13–14 διὰ τοῦτο ἔσται ὑμῖν ἡ ἁμαρτία αὕτη ὡς τεῖχος πῖπτον παραχρῆμα πόλεως ὀχυρᾶς ἑαλωκυίας, ἧς παραχρῆμα πάρεστι τὸ πτῶμα, (14) καὶ τὸ πτῶμα αὐτῆς ἔσται ὡς σύντριμμα ἀγγείου ὀστρακίνου, . . . Therefore will be to you this sin as a wall falling immediately, of a strong city captured, of which immediately comes the fall, and its fall will be like the breaking of an earthen jar.26 ‫ לכן יהיה לכם העון הזה‬13 ‫כפרץ נפל נבעה בחומה נשׂגבה‬ :‫אשׁר־פּתאם לפתע יבוא שׁברהּ‬ ‫ ושׁברהּ כשׁבר נבל יוצרים‬14 22

23 24 25

26

Van der Vorm-Croughs, ‘LXX Isaiah,’ 174–88. More exx. of anaphora (repetition of a word or phrase at the beginning of successive clauses): Isa 14:5; 32:5; 59:19–20. For terminology one can consult Smyth, Grammar, 671–83; Rowe, ‘Style,’ as well as van der Vorm-Croughs, ibid. Lee, ‘Translations,’ 777–8, 780. Van der Kooij, ‘Septuagint of Isaiah,’ 34. Le Moigne, ‘Le livre d’Esaïe.’ Van der Louw, Transformations, 219–22; cf. 244–5 for summary of stylistic features in Isaiah 1. My translations are deliberately literal: stylish English is not an issue here.

358

COLLECTED ESSAYS ON THE GREEK BIBLE

Therefore will be to you this iniquity like a breach falling, bulging in a wall raised high, (of) which suddenly in an instant will come its breaking; and one will break it like the breaking of a jar of potters.

The Greek is a loose rendering of the original, conveying a meaning not quite the same but similar. The steps by which the translator reached this result are not clear; but it seems to me that the style is as important to him as the meaning. He certainly does not aim at an exact rendering. The stylistic features we can notice are: (a) alliteration in π, achieved by choosing certain words instead of other equivalents; (b) anastrophe, that is, when a word at the end of a clause is repeated at the beginning of the next, in ‘τὸ πτῶμα, καὶ τὸ πτῶμα’: this of course twice matches the Hebrew (‫)שׁבר‬, but the third time the translator declines to repeat it, using instead σύντριμμα; (c) the chiasmus ‘πάρεστι τὸ πτῶμα … τὸ πτῶμα αὐτῆς ἔσται’; (d) repetition of παραχρῆμα; (e) hyperbaton in ‘τεῖχος … πόλεως.’ None of this is accidental: the translator has contrived this outcome, at the expense of a completely literal rendering. And the result is a pleasing one. This matter of style is very relevant to our third and main topic, the ‘literary level’ of the Greek. The Isaiah translator clearly had some training in the rhetorical arts. Such a person would also have had a training in writing good-quality Greek: the two go together. So we would not be surprised to find evidence of skill in the latter.

5. LEVEL OF GREEK Level of Greek is about the kind of Greek a writer uses. Greek, like every language, offers a range of variations from vernacular to learned. Post-Classical Greek was particularly sensitive to this kind of differentiation. Every text reveals degrees of ‘literariness,’ that is, the presence (or absence) of features that reflect the level of education of the writer and his success in deploying them. A Greek education beyond the basics encouraged the use of features of good-quality or ‘correct’ Greek, and taught the elements of writing in a good style. The level aimed at became higher as the student progressed. Every aspect of the language was involved: phonology, morphology, syntax, vocabulary, and at the higher levels, rhetorical figures. The approved features contrasted with those of everyday spoken Greek, which were disparaged. The educational model was of course not static but developed over time. The second century BC was

THE LITERARY GREEK OF SEPTUAGINT ISAIAH

359

not yet the era of Atticism, which began to exert influence in the first century BC, calling for the use of strict features of Attic (like -ττ-). But Atticism had its roots in earlier approaches to education and ideas of what constituted ‘good’ Greek, and these are reflected in the texts of the third and second centuries BC. The labels we use are inevitably imprecise, and the details of the features involved are still the subject of study, but the phenomenon is real. Thackeray himself, as alluded to above, classified certain books as ‘literary’ or ‘literary and Atticistic.’ He never explained or justified his decisions, but his judgement was generally first-rate, based as it was on a deep knowledge of Classical and later Greek. Experience with the LXX confirms his decisions about the books that he did place in these categories, but I believe that in the case of Isaiah he was not quite on the mark (as also with Sirach).27 It is interesting to see that in an earlier article, before he published his Grammar, Thackeray said that the Isaiah translator ‘employs a Greek which much more nearly approaches the classical style than the Greek of the more painstaking translators of the other prophetical books.’28 He went on to illustrate this by the translator’s use of ‘connecting particles.’ But for some reason Thackeray did not follow these remarks through to their natural conclusion. Since that time, no one seems to have looked into the question.29 A study of the language of Isaiah many years ago (in 1934) by Ziegler showed, from observation of parallels in documentary papyri, that the Isaiah translator was very familiar with contemporary Egyptian Greek vocabulary and drew on it extensively.30 More recently van der Meer has pursued similar enquiries with good results.31 But in some ways this work has been misleading. It has allowed the impression to be gained that the Greek of Isaiah is everyday, even vernacular Koine Greek, rather than anything else. While the parallels are good, this is not the full story by any means.

27

28 29

30 31

On Sirach see Aitken, ‘Literary Attainment.’ The list of ‘literary’ books then is: 1 Esd, Dan (LXX), Esth, Job, Prov, Wis, Ep Jer, Bar, 2–4 Macc, Isa, Sir (+ Eccl?). Thackeray, ‘Greek Translators,’ 583. Olley (Righteousness, 9) perceptively remarked that ‘little attention has been given to Thackeray’s conclusion that “he employs a Greek which … the other prophetical books”’ [full quotation as above]. Ziegler, Untersuchungen, 175–212 (conclusions, 211). See van der Meer, ‘Papyrological Perspectives,’ 107–33. Note the observation (109–10) that Ziegler’s study covered only five percent of the whole vocabulary of LXX-Isaiah and was based only on the lexicon of Preisigke.

360

COLLECTED ESSAYS ON THE GREEK BIBLE

6. EVIDENCE FOR

A HIGHER LEVEL OF

GREEK IN ISAIAH

The evidence for my case follows now, under a number of headings. Only the strongest examples are included: others have been considered and dropped. Even so, I am sure that the list could be extended. A full discussion is not given for every item but could readily be added. 6.1. A Tragic phrase: ὦ τάλας ἐγώ Isa 6:5 καὶ εἶπα Ὦ τάλας ἐγώ, ὅτι κατανένυγμαι, ὅτι ἄνθρωπος ὢν καὶ ἀκάθαρτα χείλη ἔχων ἐν μέσῳ λαοῦ ἀκάθαρτα χείλη ἔχοντος ἐγὼ οἰκῶ. I said, ‘O wretched am I! for I am cut to the heart, that being a man and having unclean lips in the midst of a people having unclean lips I dwell’ … ‫ואמר אוי־לי כי־נדמי כי אישׁ טמא־שׂפתים אנכי‬ I said, ‘Woe is me! for I am destroyed/struck dumb, for a man unclean of lips am I …’

The phrase is highly recognisable as an expression from Attic Tragedy, found in Sophocles and Euripides. It is sure to have been encountered at some point in the course of a Greek education. Let us see two examples from popular plays: Sophocles, Phil. 744 δύστηνος, ὢ τάλας ἐγώ. | ἀπόλωλα, τέκνον· Euripides, Hippol. 1090 ὦ τάλας ἐγώ, | ὡς οἶδα μὲν ταῦτ,’ οἶδα δ’ οὐχ ὅπως φράσω.

This expression cannot be found in contemporary documents. Even the word τάλας is rare and used only in poetic epitaphs, as one would expect. In the LXX τάλας occurs only in Wis 15:14; 4 Macc 12:4, both literary books. This does not amount to proof that the translator knew ὦ τάλας ἐγώ from reading Tragedy, though I myself think it probable. But even if the phrase was sometimes used in speech by the educated, its literary provenance would have been known. It is notable that the only other example in the LXX is in the literary 4 Maccabees (8:17 ὦ τάλανες ἡμεῖς). No other LXX translator thought of using it; and of course it is independent of the Hebrew (the common rendering of ‫ אוי‬is οὐαί).32 32

The connexion with Tragic language goes unnoticed in: Ziegler, Untersuchungen, 108; Ottley, Book of Isaiah, in loc.; Seeligmann, Septuagint Version, 53; LXX.D Komm., in loc. But Walters (Text, 230) spotted it: ‘an isolated classical reminiscence, well fitted to the sublime note of this passage.’ Schleusner (Lexicon in LXX, s.v. τάλας) was best of all: after Isa 6:5 he immediately quoted a parallel in Euripides, Phoenissae 1345.

THE LITERARY GREEK OF SEPTUAGINT ISAIAH

361

6.2. A proverbial expression Isa 44:4 … ἐπὶ τὰ τέκνα σου, (4) καὶ ἀνατελοῦσιν ὡσεὶ χόρτος ἀνὰ μέσον ὕδατος καὶ ὡς ἰτέα ἐπὶ παραρρέον ὕδωρ. … upon your children, and they will spring up like grass in the midst of water and like a willow beside water flowing by. :‫וצמחו בבין חציר כערבים על־יבלי־מים‬ and they will spring up in the midst of grass, like willows by streams of water.

The Greek rendering ἐπὶ παραρρέον ὕδωρ does not follow the original exactly but catches the sense of it while converting it into a pleasing poetic phrase. The verb παραρρέω is rare in the LXX, occurring only once elsewhere (Prov 3:21), in a figurative, developed sense (intrans. in pass.) ‘drift away,’ hence ‘be neglectful’ (of something), also found in the NT (Heb 2:1). It is a word with Classical and literary antecedents, in which the figurative uses already predominate. So the choice of word is interesting to begin with. But why this phrase? Is it his own creation? A search in Greek literature establishes that there was a proverb, first attested in a fragment of the Old Comedian Cratinus (V BC), and explained as referring to those who promise to undertake something whatever the odds. The proverb reappears centuries later in the renowned rhetorician Libanius (IV AD).33 In between, the phrase recurs at long intervals, in a scholiast on Homer, then in Plutarch, then much later in Patristic literature.34 All this suggests that it was a set expression with a literary provenance, known to those with a good education, not one made up by the Isaiah translator. He takes it and puts it to work for his own purpose, without allusion to the sense of the proverb. 6.3. Literary or poetic vocabulary Here are grouped a number of words, and a phrase, with literary or poetic antecedents. It is not intended to be a complete list. A full study of the vocabulary of LXX-Isaiah, which is a desideratum, would produce more examples like these and interesting results overall.35 33 34

35

Cratinus, Fr. 64 ὕδωρ παραρρέει. Libanius, Ep. 109.2 κἂν ὕδωρ, φασί, παραρρέῃ. Schol. in Hom., Il. 2.307 . . . Πλάτων πλάτανον καὶ τὸ παραρρέον ὕδωρ φησί (cf. Pl., Phaedr. 230b). Plut., Cato Ma. 19.1 (I–II AD) τὸ παραρρέον δημόσιον ὕδωρ. In Patristic lit., e.g., Asterius, Homil. 2.3.3 (IV–V AD) τὸ παραρρέον ὕδωρ. Cf. Baer, When We All, 281.

362

COLLECTED ESSAYS ON THE GREEK BIBLE

(1) κῦδος, ‘renown’: only here in LXX. Isa 14:25 καὶ ἀφαιρεθήσεται ἀπ’ αὐτῶν ὁ ζυγὸς αὐτῶν, καὶ τὸ κῦδος αὐτῶν ἀπὸ τῶν ὤμων ἀφαιρεθήσεται. And will be removed from them their yoke, and their renown from their shoulders will be removed. :‫וסר מעליהם עלו וסבלו מעל שׁכמו יסור‬ And will fall off/depart from them his yoke and his burden from on his shoulder will fall off/depart.

The Greek rendering differs somewhat from the original, and it is not clear why the translator used κῦδος to translate ‫‘( סבל‬burden’), but that is not the point.36 This is quite an unusual choice of word. κῦδος is used only here in the LXX; it is an Epic and poetic word, obviously not an everyday one; and in contemporary documents it is found only in inscriptions, often epitaphs and other poetic texts.37 So it seems to me a clear indicator of the translator’s command of the poetic ‘register’ of vocabulary. The context (verses 24–25) might be relevant to his choice of κῦδος here: God is speaking, in a noticeably lofty style. (The elegant chiasmus is in the original and the translator naturally does not disturb it.) (2) ἐν ᾅδου, ‘in Hades’: Isa 38:18, 18. Other LXX × 4: Job 1; Sir 3. The similar expression εἰς ᾅδου (‘to Hades’) appears first in the LXX in the Pentateuch and then occasionally in other books following their lead. It originates in Homer and then appears frequently in Classical authors, especially Tragedy. It contains an archaism (εἰς + gen.) that could not have been created by the translators. It seems to me a sign of the Pentateuch translators’ Greek education. The Isaiah translator also uses εἰς ᾅδου (14:11, 15, 19), and ἕως ᾅδου (57:9), but he goes a step further, as it were, and shows knowledge of another form of it that similarly appears in Attic Tragedy (and Plato) and goes back to Homer’s εἰν Ἀΐδαο. It is significant that the only other translators to use it are Job and Sirach. (3) ἀγαυρίαμα, ‘pride’: Isa 62:7. Other LXX × 3: Job 1; Jer 1; Bar 1. Not attested outside LXX. Cf. ἀγαυρός Hesiod, Hdt., Nicander, rare; ἀγαυριῶμαι Job 1; Aquila. 36

37

See an attempted explanation in Ottley, Book of Isaiah, 2:181–2; Seeligmann, Septuagint Version, 50. E.g., IEgVers 38.8 (Egypt, late Ptol.) καὶ γὰρ ἐνὶ φθιμνοισι θουδείης γέας ἔσχον, | ἡμιθέων ἀλόχων κῦδος ἐνεγκαμένη·

THE LITERARY GREEK OF SEPTUAGINT ISAIAH

363

(4) ἀπαμαυρῶ, ‘make dim, make fade away’: Isa 44:18. Not elsewhere in LXX. Rare: Hymns to Isis (I BC), Orphica (date?), + later. (5) ἀποσκορακίζω, ‘reject with contempt’: Isa 17:13. Other LXX × 2: 1 Macc 1; Ps 1. First in LXX, then Plutarch +. Cf. σκορακίζω, Demosth., Phld., Philo, + later; (6) ἀποσκορακισμός, ‘rejection with contempt’: Isa 66:15. Not attested again. Cf. σκορακισμός Sir 41:19, then Plut.38 (7) κύω, metaph. ‘conceive in the mind’: Isa 59:4, 13. Not elsewhere in LXX. The word Homer (κυέω) +; this use Plato, Xen., rare. (8) μακρόβιος, ‘long-lived’: Isa 53:10. Other LXX: Wis 3:17. Hdt., Hippocr. +. Cf. μακροβίωσις Bar 3:14, only here. (9) ὁμίχλη, ‘mist, fog’: Isa 29:18. Other LXX × 9: Ps 1; Job 2; Wis 1; Sir 2; XII Proph 3. Homer +. (10) πυρίκαυστος, ‘burnt by fire’: Isa 1:7; 9:4; 64:10. Not elsewhere in LXX. Homer +. 6.4. New creations LXX-Isaiah presents a number of compounds that are hapax legomena not only in the LXX but in all of Greek (as so far attested). It is always risky to assert that a word is entirely new, but these certainly have the look of learned neologisms coined by the Isaiah translator to meet his purpose at certain points. They indicate a confident command of the processes of forming literary compounds. (1) (2) (3) (4)

38

39

40

ἄνισχυς, ‘strengthless’: Isa 40:30. βαθύφωνος, ‘deep-voiced’: Isa 33:19.39 κατάκλιτος, ‘downward-flowing’: Isa 3:23.40 πριστηροειδής, ‘saw-like’: Isa 41:15.

The origin of this group is in the Attic expression (Ar. +) ἐς κόρακας ‘(go) to the ravens’ = ‘go to hell’: see Beekes, EDG, s.v. κόραξ. The full list of -φωνος compounds (about 125) contains only a few common words; most are poetic or learned creations, with limited attestation: see Buck-Petersen, Reverse Index, 287–8. As a descriptor of θέριστρα: see van der Meer, ‘Trendy Translations,’ 592, 596.

364

COLLECTED ESSAYS ON THE GREEK BIBLE

6.5. Particles The reduction of the Classical range of particles in post-Classical Greek and the significance of particles as a sign of education are well known.41 The following is a list of those occurring in LXX-Isaiah that contribute to my case, though they are not all of the same weight. (1) τοιγαροῦν Isa 5:26 οὐκ ἀπεστράφη ὁ θυμός, ἀλλ’ ἔτι ἡ χεὶρ ὑψηλή. τοιγαροῦν ἀρεῖ σύσσημον ἐν τοῖς ἔθνεσι τοῖς μακρὰν καὶ συριεῖ αὐτοῖς ἀπ’ ἄκρου τῆς γῆς His anger has not turned back, but his hand is still raised high. For that reason he will raise a standard among the nations that are far off and will whistle to them from the end of the earth.

(2) τοίνυν: Isa 3:10 + 3. Other LXX × 16: literary books 14; others 2. (3) μέν: Isa 6:2; 41:7. Other LXX × 216: Pent. 19; literary books 184; others 13. (4) οὖν: Isa 30:8; 40:25; 57:6. Other LXX × 243: Pent. 79; literary books 124; others 40. (5) τε: Isa 37:12. Other LXX × 265: Pent. 54; literary books 189; others 22. (6) καὶ γάρ: Isa 23:10. Other LXX × 81: Pent. 5; literary books 42; others 34 (1, 2 Chr 8; Ps 23; Zeph 1; Ezek 2). (7) δή: Isa 3:1 + 5. Other LXX × 280 +: Pent. 4. The word that stands out in this list is τοιγαροῦν. It is rare elsewhere in the LXX and then only in literary books (Prov × 2; Job × 2; Sir × 1; 2–4 Macc × 5). No examples can be found in Ptolemaic papyri. Thackeray noted it (and τοίνυν) in his list illustrating the ‘classical style’ of Isaiah.42 Here in Isaiah it corresponds to a simple ‫ו‬, so it is the translator’s choice. Also quite significant is τοίνυν, which again is a rarity elsewhere. The other particles listed are more common, and the Pentateuch, in middle-level Greek, uses μέν, οὖν, and τε quite often. But their use there is an indicator of education too, and they still occur most often in the literary books. καὶ γάρ is also favoured in literary books, except in the case of Psalms, where some stereotyping as an equivalent of ‫ גם‬is involved (13 times out of 23). I add δή to the list with caution, because it had become a stereotyped equivalent of ‫ נא‬for many translators. It is suggestive however that the 41 42

See, e.g., Blomqvist, Greek Particles, 132, 144–5. Thackeray, ‘Greek Translators,’ 583. Cf. Blomqvist, Greek Particles, 130.

THE LITERARY GREEK OF SEPTUAGINT ISAIAH

365

Isaiah translator introduces it independently of ‫( נא‬or any Hebrew word) four out of six times. I see this collection of particles as evidence of the Isaiah translator’s intention to compose his work in Greek of a higher literary level. To see τοίνυν and τοιγαροῦν only as showing ‘the translator’s interest in marking sequences of thought,’ as Troxel does, is to miss half the story. Of course they do that, but why does he choose these particular words?43 6.6. Literary word-choices There are many instances in which the Isaiah translator chooses the more literary equivalent of some everyday word. This is not to say that he will never use the everyday one; in fact he may do so, and no particular significance should be seen in it when he does. But it is the use of the higherlevel, ‘better’ word that is significant. Contrast is the key, that is, between the ‘better’ word and the more everyday one. In the list I have given the latter in each case (with ‘≃’ indicating practical equivalence rather than complete synonymy). For νοῦς and οἶμαι, there is no single equivalent: it is rather a choice between using the word and not using it by saying it in a different way (or not at all). For example, instead of νοῦς he could have used καρδία, the more Semitic expression of the idea, which he does use often. I have listed below the vocabulary that seems to me significant; there is probably more. Some are selected for brief remarks after the list. ἀεί, ‘always’: Isa 42:14; 51:13. Other LXX × 12: literary books 10; Judg 1; Ps 1. ≃ διὰ παντός Isa × 10; LXX × c.130. ἀμῶ, ‘I reap’: Isa 17:5; 37:30. Other LXX × 3: Pent. 2; Mi 1. ≃ θερίζω Isa × 0; LXX × c.27. ἀροτήρ, ‘ploughman, farmer’: Isa 61:5. ≃ γεωργός Isa × 0; LXX × 9. αὔξω, ‘I increase’: Isa 61:11. ≃ αὐξάνω. βούλομαι, ‘I wish, want’: Isa 1:11 + 12. Other LXX × 108: Pent. 14; literary books 39; others 55. ≃ θέλω Isa × 9; other LXX × 144. 43

Troxel, LXX-Isaiah, 93, in a discussion of ‘conjunctions.’ Similarly on γάρ (92) Troxel notes an ‘unusually high’ frequency in Isaiah, but only concludes that it ‘attests his interest in creating smoother connections between clauses.’ That is what γάρ does if you use it; but using it a lot indicates literary intent. Wagner (Sealed Book, 30) distils Troxel into a bland: ‘He fosters cohesion in the translated text with his adept use of Greek particles.’

366

COLLECTED ESSAYS ON THE GREEK BIBLE

δέδοικα, ‘I am afraid’: Isa 60:14. Other LXX: Job 7. ≃ φοβοῦμαι. ἐπίσταμαι, ‘I know’: Isa 29:11 + 7. Other LXX × 50: Pent. 14; literary books 18; others 18. ≃ γινώσκω, ἐπιγινώσκω, οἶδα. ἐρυθρός, ‘red’: Isa 63:2 ἐρυθρὰ τὰ ἱμάτια. Other LXX × 29, all ἐρυθρὰ θάλασσα. ≃ κόκκινος Isa × 2; other LXX x 41. νοῦς, ‘mind’: Isa 10:7 + 3 (41:22 ἐφίστημι τὸν νοῦν). Other LXX × 25: literary books 22; others 3. οἶμαι, ‘I suppose’: Isa 57:8 (ᾤου). Other LXX × 21: Gen 4; 1 Macc 2; literary books 15. πόρρω, ‘far off’: Isa 17:13 + 4. Other LXX × 12: literary books 8; others 2. ≃ μακράν Isa × 7; other LXX x c.65. πόρρωθεν, ‘from afar’: Isa 10:3 + 7. Other LXX × 8: literary books 4; others 4. ≃ μακρόθεν Isa × 2; other LXX × 35. πράσσω, ‘I do, act’: Isa 57:10. Other LXX × 36: Pent. 1; literary books 33; Josh 1; 1 Macc 1. ≃ ποιῶ Isa × c.153; other LXX × 3000+. ἀεί. It may seem surprising that ἀεί is significant, but changes were happening in words for ‘always’ in early post-Classical Greek. ἀεί was in decline except in set expressions or more stylish Greek, while διὰ παντός was coming in strongly at this time. The frequency of διὰ παντός in the LXX itself reflects this, and the Isaiah translator himself uses it more often than ἀεί. The word that prevailed later, πάντοτε, had only recently appeared.44 ἀμῶ. This word is the subject of an observation by the Atticist Moeris: ἀμᾶν Ἀττικοί· θερίζειν Ἕλληνες.45 Though much later, this indicates the direction in which things were going. Shipp says that ἀμῶ was ‘already rare in Attic.’46 It is true that ἀμῶ also occurs in the Pentateuch (Lev 25:11; Deut 24:19) and in Micah (6:15), and that does not put those books in the ‘literary’ category; it only shows that their translators had some education. It is a more significant choice in Isaiah when combined with the other evidence.

44 45 46

Cf. Lee, ‘Atticist Grammarians,’ 293–4 [= Essay 19]; Meecham, Letter of Aristeas, 204. Moeris α 158 (ed. Hansen). Shipp, MGE, 63.

THE LITERARY GREEK OF SEPTUAGINT ISAIAH

367

ἐρυθρός. The point is that a new word κόκκινος had come in to replace ἐρυθρός, and ἐρυθρός was confined to the set phrase ἐρυθρὰ θάλασσα, the ‘Red Sea.’ The Isaiah translator uses the old-fashioned word on its own to mean ‘red’ in this one instance.47 πράσσω. This is another word that was going out of use, except in restricted senses.48 In papyri of the third century BC it is nearly always ‘exact payment.’ When it does occur as a word for ‘do’ it is a more formal word, and usually refers to bad actions, as in the one example in the Pentateuch (Gen 31:28). In Josh 1:7 it occurs in God’s speech. The figures for the LXX show how infrequent it is compared with the standard word ποιῶ, and how much it is favoured by the literary books. 6.7. Other features Noted here with short discussion are miscellaneous items: an Attic form, two features of syntax, a poetic expression, and a special item of vocabulary. (1) ἡττῶμαι Isa 8:9 + 11. Other LXX: 2 Macc 1; Sir 1; Jer 1; Dan 2. The word is an official, sometimes legal term that is rare in the LXX outside literary books, but used often in Isaiah.49 I think it is significant that it appears in the Attic form ἡττῶμαι, instead of ἡσσῶμαι. It used to be thought that only the Attic form was available in early Koine Greek, but we now have documentary evidence of both forms. So it was possible to make a choice.50 (2) Comparative optative, mostly with ὡς, ὡς ἄν: Isa 11:9; 21:1; 66:20. Other LXX: Pent. 9; JudgAB 2; Ps 7 (5 in one verse); Prov 3; Ezek 1. The comparative optative is a special and rare use in Greek. It is best explained as derived from Homeric language.51 47

48

49

50

51

Evidence in Lee, Lexical Study, 111–2. Another documentary ex. of κόκκινος can be added: PCairZen 4.696.4 (III BC). Cf. Shipp, MGE, 325. See Shipp, MGE, 461–8; cf. Gerleman, Job, 15: πράσσω ‘is much more literary than the ordinary ποιεῖν.’ [See also Lee, Greek of the Pentateuch, 114–7.] A good study of its usage in Isaiah, esp. in the light of documentary papyri, is found in van der Meer, ‘Papyrological Perspectives,’ 113–20. Verb forms: -σσ- in PHal 1.54, 115, 118, 188 (259 BC); PHib 2.197.53 (III BC); -ττ- in PZenPestm 51.27 (257 BC). In the adj., ἧσσ- (60+) is commoner than ἧττ- (5) in III–I BC papyri (as at 23.1.14). Obsolete discussion in: Thackeray, Grammar, 121 n. 2; Walters, Text, 260. See Evans, Verbal Syntax, 190–7; cf. Joosten, ‘Elaborate Similes,’ 230–3. No support here for Baer’s (unconvincing) argument that ch. 66 is by a different hand: Baer, ‘What Happens in the End?’

368

COLLECTED ESSAYS ON THE GREEK BIBLE

(3) Conditional optative with εἰ: Isa 49:15. Other LXX: Pent. 0; 1 Kgdms 1; 4 Macc 8; Job 2. The conditional optative also looks significant. It is rare and favoured in Job and especially ‘the literary essay’ 4 Maccabees, as Thackeray calls it in remarks on rare uses of the optative.52 This and the type of optative in (2) are remnants left over from the older fuller range of use, as the mood became desystematised in early Koine Greek.53 The volitive optative persisted longest, but these two types were certainly obsolescent and are therefore signs of a higher education. (4) βουλεύομαι βουλήν: Isa 3:9; 7:5; 8:10; 31:6 (cf. ἡ βουλὴ /τὴν βουλὴν ἣν βεβούλευται 14:26; 19:17). Elsewhere Ezek 11:2. The Isaiah translator has a fondness for this particular figura etymologica, with its solemn, poetic effect. In one place (8:10) it can be said to be motivated by the Hebrew (‫ ;)עצו עצה‬but in the other three he contrives to introduce it himself. A search in Greek literature finds the combination seven times in Homer and once in Plato, not elsewhere.54 It does not seem unlikely that acquaintance with Homer contributed to the Isaiah translator’s inclination to use it.55 (5) Isa 43:14 Ἕνεκεν ὑμῶν ἀποστελῶ εἰς Βαβυλῶνα καὶ ἐπεγερῶ πάντας φεύγοντας, καὶ Χαλδαῖοι ἐν πλοίοις δεθήσονται. In Greek of an earlier period there was a distinct use of φεύγω to mean ‘be in exile,’ especially in the participial form. It is almost the default meaning when (οἱ) φεύγοντες is encountered in Classical texts.56 It seems to me that it is intended here: ‘I will rouse up all exiles’ (not ‘all who are fleeing’).57 It is triggered naturally by the mention of Babylon, the place of exile; the obscurity of what comes next (‘and Chaldaeans will be bound in boats’) does not matter. In this use of φεύγω the translator reveals familiarity with Classical usage. 52 53 54

55

56 57

Thackeray, Grammar, 193. See Evans, Verbal Syntax, 175–8. Homer, Il. 9.75; 10.147, 327, 415; 23.78; 24.652; Od. 6.61 (βουλεύω act. in all, βουλάς pl. except βουλήν once); Plato, Politicus 298b.8 εἰ δὴ ταῦτα διανοηθέντες βουλευσαίμεθα περὶ αὐτῶν βουλήν τινα. Troxel, in ‘βουλή and βουλεύειν,’ seeks reasons for the ‘insertion’ of βουλή but works entirely within the framework of LXX-Isaiah and its Hebrew original. See LSJ, s.v. φεύγω III. So NETS (‘all who are fleeing’); similarly Ottley, Book of Isaiah, 1:235 (‘all that flee’). But cf. LXX.D: ‘alle Verbannten aufwecken.’ Muraoka, Lexicon does not separate this sense. The Modern Greek translations of Kolitsaras (1928) and Giannakopoulos (1986) interpret as ‘I will stir up all (the Babylonians) to flee,’ which [does not seem] possible.

THE LITERARY GREEK OF SEPTUAGINT ISAIAH

369

7. CONCLUSION A strong case can be made for classifying LXX-Isaiah as ‘literary’ in its level of Greek. It should be bracketed with other ‘literary’ books. This is not to say that all those labelled ‘literary’ are exactly the same: there are certainly differences between books, and further study might place Isaiah somewhat lower than, say, Job and Proverbs. Nor does it mean that highlevel Greek is consistently maintained in Isaiah: there is of course variation downwards on the scale and there is interference in the quality of the Greek from the Hebrew original.58 But I think it is mistaken to put Isaiah into a vague ‘good Koine Greek’ category, as Thackeray did, and helpful to know that its place is in the ‘literary’ category. More work needs to be done, certainly, to refine what we mean by ‘level of language,’ to securely identify ‘literary’ features, and to establish criteria for classifying books on that basis. But we have some definite evidence in front of us now for Isaiah, beyond Thackeray’s mainly intuitive placing of books in the ‘literary’ category. This result may seem unimportant, but it has value for linguistic study of the LXX. To appreciate properly the Greek of any LXX translated book (or free composition), and to connect it with other Greek of the time, we need to know where it stands on the literary scale manifested in the whole of post-Classical Greek. We then understand better the choices the translator makes from the range of features available, and we can gain a better understanding of his aims and of the translator himself. In the case of a book where the translator and his aims are the subject of so much debate, an appreciation of the kind of Greek employed may have a significant contribution to make. The finding that the Greek of LXX-Isaiah is high on the literary Greek scale tells us something important about the translator. It indicates that he had had a Greek education to an advanced level. We do not have exact knowledge of that education, but it cannot be doubted that a Greek education was accessible to Alexandrian Jews of the second century BC and that they availed themselves of it. If we had nothing else, the Exagoge of the Jewish tragedian Ezekiel and the polished Greek of the Letter of Aristeas and the Prologue to Sirach would be sufficient evidence.59 How well does 58

59

I am grateful to Cameron Boyd-Taylor for input here and other comments on the paper. My thanks also to Michael Curran for valuable feedback and improvements. I believe the author of the Prologue refers to his education in the knotty εὑρὼν οὐ μικρᾶς παιδείας ἀφόμοιον (l. 29), which I read as ‘having acquired the likeness/shape of no small education.’ This is modesty clothed in artistry.

370

COLLECTED ESSAYS ON THE GREEK BIBLE

a translator with an advanced level of Greek education accord with the portraits of the Isaiah translator that have been propounded in recent study? It is not possible for me to deal with this question in a thorough or even adequate way; I only put forward some observations that I think are worth considering. I will take as representative the portraits drawn by two leading Isaiah scholars. Arie van der Kooij has led the way in characterising the Isaiah translator as a ‘scribe.’60 I have to admit to some difficulty understanding the reason for choosing this term and the precise meaning van der Kooij intends it to have.61 Variations such as ‘scribe-translator’ do not help. ‘Scribe’ seems to be no more than a convenient label to cover whatever it is the Isaiah translator does. Wilk expands its scope by reference to ‘scribal artistry,’ and even extends this description to Paul’s handling of quotations from Isaiah.62 Nevertheless it is fairly clear what van der Kooij’s conception of this translator is: he is a Jewish scholar trained in the reading and interpretation of the Hebrew text. The observations made in this paper may raise some questions about van der Kooij’s picture. In regard to translation method, one might have expected a consistent (even ‘slavish’) literal method from one accustomed to a close reading of the Hebrew, rather than a method that is basically literal but takes great liberties within that framework. Similarly, in the style category, one does not expect to see Greek stylistic and rhetorical devices playing such a significant role, often at a tangent to the meaning of the original. And thirdly, the choice of the literary Greek register requires some explanation if the translator’s focus was on conveying the meaning of the Hebrew text. The person who made this translation seems rather to have been focused on turning the text not just into meaningful Greek, but into stylish ‘biblical’ Greek, that is, Greek with a ‘biblical’ flavour on the pattern of the Pentateuch, but in a better style and on a higher level. Such a translation would appeal to the Greek-educated Jewish reader and display the translator’s own Greek learning. To me it seems that this was an aim that at times overrode even the basic purpose of rendering the meaning of the Hebrew text. Whether the translator also reveals himself to be an 60

61

62

Van der Kooij, Oracle of Tyre, esp. 112–23. Cf., e.g., Troxel, LXX-Isaiah, 20; Wilk, ‘Between Scripture and History,’ 202–4; De Sousa, Eschatology, 157. Is it equivalent to ‫ ?ס ֵֹפ ר‬In what sense? As in Ezra ch. 7? Or is it the γραμματεύς (‘scribe’) of the NT? Perhaps the rendering γραμματικοί in 33:18 plays a role too in its meaning: cf. van der Kooij, Oracle of Tyre, 19: ‘it is reasonable to assume that our translator is to be seen as a scholar (‘scribe’).’ Wilk, ‘Between Scripture and History,’ 204–8.

THE LITERARY GREEK OF SEPTUAGINT ISAIAH

371

exegete with certain tendencies such as ‘actualisation,’ is a question for others to decide; personally I have difficulty being persuaded.63 Another leading LXX-Isaiah scholar, Ronald Troxel, offers a portrait that seems at first sight to accord rather well with the findings of this paper. Troxel rejects van der Kooij’s ‘scribe,’ but on the basis of the rendering of ‫ ספר‬by γραμματικοί in Isa 33:18, which van der Kooij sees as a selfreference, Troxel builds an argument of his own. While he explicitly rejects calling the translator a γραμματικός, he uses the term to develop a connexion with the γραμματικοί of the Alexandrian Museum as a direct influence on the translator’s work, and to locate him within ‘the literati of the ancient world.’64 We might expect then that the translator’s high-quality Greek would be used to support this proposal. Troxel comes close to it, but then qualifies it away: ‘The Isaiah translator appears to have been familiar enough with literary Greek to formulate sentences in its style, suggesting that he was among the well educated, although we must suspect that to be true, in any case, for a person undertaking the task of translation.’65 In that case all the LXX translators were ‘well educated.’ Earlier, the discussion of particles, as already noted, is used by Troxel only to show that the translator was interested in creating ‘smoother syntax for his Greek-reading audience.’66 Furthermore, the link to the methods of the Homer scholars of the Museum seems strained and is not really necessary to explain the translator’s procedure.67 The idea at the heart of all this, that the choice of rendering in 33:18 is significant and tells us something about the translator’s attitude to γραμματικοί, seems to me very speculative.68 In the end Troxel’s portrait of the Isaiah translator is not well focused: a better appreciation of the language could do much to sharpen it. My main aim has been to show that the Isaiah translator’s Greek should be classified as ‘literary’ and his language connected with other works composed in that kind of Greek, both in the LXX and outside it. If I have succeeded, the observation may also contribute to unfolding the secret of the Isaiah translator and his methods.

63 64 65 66 67 68

Cf. Wagner, Sealed Book, 34 for a summary of proponents and critics. Troxel, LXX-Isaiah, 290–1. Troxel, LXX-Isaiah, 132. Troxel, LXX-Isaiah, p. x. Cf. Hugo, Review of Troxel, LXX-Isaiah (2008). I agree with Pietersma’s discussion of the rendering in ‘A Panel Presentation,’ 354–5.

23 THE SAYINGS OF THE DESERT FATHERS: THEIR EVIDENCE FOR LATE KOINE GREEK 20151

Abstract The Sayings of the Desert Fathers, in their Greek form, date from the fourth to fifth centuries AD and are written in the Greek language of their time. They contain both sayings and simple narratives, composed for the most part in everyday, not literary Greek. These texts therefore have the potential to provide useful evidence for the ongoing development of Greek in the late Koine period (IV– VI AD). There is a shortage of evidence of this kind and the Sayings have received little notice. The aim of this paper is to draw attention to the linguistic value of the Sayings, concentrating on the Alphabetic collection. One Saying (Paphnoutios 2) is sampled first, with observations on features of interest there, then other features found elsewhere in the corpus are listed with brief comments. The list is extensive but not exhaustive or systematic: a full study still needs to be done. The objective of the paper is linguistic, not theological, nor does it aim to resolve textual questions pending until a critical text of the Alphabetic collection is achieved.

1. INTRODUCTION The Desert Fathers were those first Christian ascetics who withdrew from the world and went out into the Egyptian desert to focus on the spiritual life in solitude. Abba Antonios is still commemorated today as the one who led the way. The Sayings are a collection of both memorable sayings and simple narratives about the Fathers that were thought worth remembering. They always contain some point of significance related to 1

Most of the material of this article was first presented at the ‘Images of Egypt’ conference at Macquarie University, Sydney in April, 2004. I am grateful to Trevor Evans for the invitation to present at the conference, and also to the students of St Andrew’s Theological College who over the years 1988–2000 joined me in reading the Sayings in the ‘Advanced Koine’ course.

374

COLLECTED ESSAYS ON THE GREEK BIBLE

the spiritual life and practice. They range in length from a line or two to stories extending over two pages. In the form of the collection studied here, the Greek Alphabetic, there are some 130 named monks, and the number of Sayings attached to each ranges from one to over 200 in the case of Abba Poimen. In most cases, the material has an Egyptian setting, that is, almost all of the Fathers are located in Egypt and the various sayings and narratives take place in Egypt. The texts survive in a number of languages, but it appears that the Greek form was the earliest, from which translations were made into Syriac, Armenian, Sahidic, and so on. We know that some of these monks were Egyptians and may not have spoken Greek, so some translation into Greek is likely to have occurred at an early stage. The Greek collection is in three forms, the Alphabetic, the Anonymous, and the Systematic, with a lot of material in common. The Alphabetic arranges the material by the names of the Fathers (or Mothers) with whom the saying or story is associated; the Anonymous collects sayings that have no name attached; and the Systematic arranges much of the same material by topics or themes, 21 in all. It appears that the Alphabetic collection came first. That is the one on which this paper is based.2 Who first put this material together, and who did any necessary translating, say from Coptic into Greek, we do not know. But the nature of the Greek itself indicates something. It is not learned or literary Greek, nor is it entirely vernacular, but it is middle-level Koine Greek of a century before the beginning of early Medieval Greek (600–1100).3 There is no evidence of interference from Coptic, but biblical Greek, that is, the Greek of the NT and the Greek OT (‘the Scriptures’), has a significant influence on the language and there is much use of scriptural quotations and allusions.4 Much of this influence could come, and in some instances clearly does come, from liturgical Greek, which itself is deeply influenced by the Greek of both Testaments.5 Apart from these influences, the narrative framework and most of the Sayings material are in ordinary Greek of the period and as a result are of great interest linguistically.

2

3 4 5

For the foregoing points see: Guy, Recherches; idem, Les Apophtegmes: Collection systématique; Ward, The Sayings, xvii–xxxi; Wortley, ‘How the Desert Fathers’; Rönnegård, Threads and Images, 5–11, 197–8; Butler, Lausiac History, 1:208–15. See Browning, Greek, 53, 55 for the term (‘early middle ages’) and dates. See Rönnegård, Threads and Images. See Lee, ‘LXX in Liturgy.’

THE SAYINGS OF THE DESERT FATHERS

2. TEXT

375

AND TRANSLATIONS

The ruling edition of the Greek text of the Alphabetic collection is that found in Migne’s great collection of patristic texts.6 It is basically a reprint of an edition of 1677 by Jean-Baptiste Cotelier, derived from one twelfthcentury manuscript.7 A modern critical text of the Alphabetic is obviously needed, but in the meantime we have no choice but to use Migne’s text. There is also a more recent edition by Paschos, which is a convenient work to use but is in essence the same text as in Migne.8 In the following discussion, Migne’s is the text quoted, with a reference to Paschos as well (the numbering sometimes differs). Despite the lack of a proper critical edition, linguistic studies can and should proceed on the basis of the text available (as the translations have done), on the understanding that when a critical text is produced, earlier comments on language might need some correction or revision. The alternative is to delay such study indefinitely. Furthermore, observations on the language may themselves contribute to establishing the text when the task is undertaken. There are modern translations of the Alphabetic collection, into English by Sr Benedicta Ward; into French by Jean-Claude Guy and Lucien Regnault; and into Modern Greek by Basileios Pentzas.9 In Migne’s edition there is a Latin parallel version. The author is not stated, but presumably it was Migne or Cotelier. This is not the same version as the sixth-century translation of the Systematic collection to be noticed in a moment. 3. DATE The date of the material is important for linguistic study, though precise dating is not necessary: a date within a century or even longer is adequate. 6 7

8

9

Migne, PG, vol. 65. Text also available electronically in TLG (no. 2742). Guy, Recherches, 8 n. 2, 13, 18–19. The MS is Paris grec 1599; Guy gives the date ‘XIIe siècle’ without comment. Migne gives the title of his source at PG 65:71–72 (below heading). Paschos, Τὸ Γεροντικόν (1981). As far as I am aware, there is no work in progress on a critical edition of the Alphabetic. Guy’s critical edition of the Systematic reached completion in 2005 (Les Apophtegmes: Collection systématique). This includes an index of words (3:251–463). [The research for this paper was done before Guy’s edition of the Systematic appeared; otherwise it would have been possible (and perhaps better) to base my study on it. Even so, there is not a great deal of difference in the sayings themselves in the two texts.] Ward, The Sayings; Guy, Les Apophtegmes: Série alphabétique; Regnault, Les sentences; Pentzas, Εἶπε Γέρων. Ward’s translation, on which everyone relies, is not as accurate as it is assumed to be (for an example, see n. 32). A revision is needed.

376

COLLECTED ESSAYS ON THE GREEK BIBLE

We have one indication of date in the fact that the MS of the Alphabetic collection on which the first printed edition was based is from the twelfth century. Moreover, there are still older MSS of this collection dating from the tenth to eleventh centuries AD.10 This is useful to begin with, as establishing that there can have been no intrusion of late Medieval or Modern features that arose in the language after the eleventh century. But we can narrow the date much further. A terminus ante quem is provided by the Latin translation of the Systematic collection made by two Popes, Pelagius I and John III.11 This was made in the early sixth century, so the date of compilation of the Systematic must precede the translation, and the compilation of the Alphabetic preceded the Systematic. Guy finds a date around 530 AD for both collections assured, and argues further that both can be dated to 480–490 AD.12 The history of the material before then is hazy. We do not know how it came to be written down or how much adjustment (and translation) there was between the original saying and its appearance in written form, or at the final stage of compilation. Oral transmission is obviously likely to have had a role. But we know that the first generation of monks belong to the fourth century. Antonios is said to have gone out into the desert around 285, and he died in 356. Makarios the Egyptian, the founder of Scetis, overlapped with Antonios and died in 390. The last generation, which included Arsenios (d. 449) and Poimen (d. after Arsenios), are in the fifth century.13 So we can say with some confidence that the material originates in the fourth to fifth centuries and that the compilation date is late fifth or early sixth century, let us say around 500 AD. The Sayings therefore could provide good evidence of Greek of the fifth century, right in the middle of ‘Late Koine,’ that is, Greek of IV–VI AD.14 There is however the question of changes in the course of transmission, as the texts were copied between 500 and the surviving MSS. First as regards modernisation, that is, intrusion of more recent features of the popular language, it is evident that this has had no major effect. To take specific cases, the phonological change of the loss of unaccented initial vowels (except ἀ-), which took place along with related changes in early 10 11

12 13 14

Guy, Recherches, 16. On their dates see Guy, Les Apophtegmes: Collection systématique, 1:80, and for details of editions see, e.g., Wortley, ‘How the Desert Fathers,’ 328. Guy, Les Apophtegmes: Collection systématique, 1:80–3. For these dates, see ibid., 47–8, 76, 78. For the term see Lee, ‘Ἐξαποστέλλω,’ 113 [= Essay 12].

THE SAYINGS OF THE DESERT FATHERS

377

Medieval Greek, is not in evidence; nor is the loss of final -ν.15 The later morphological changes to the noun and verb that might have intruded are also not to be seen.16 All this does not rule out the possibility, even likelihood, that minor spelling changes or slips occurred. ‘Improvement’ of the text, that is, from vernacular to more educated, is a different possibility to consider, and eliminate. The presence of so many lower-register items of vocabulary suggests that no, or at least very little, ‘improvement’ of this kind occurred in transmission: words like νερόν (‘water’) and περνῶ (‘cross over’) and others noticed below would be strong candidates for replacement in that case. It seems rather that the character of the text was fixed at the time of compilation around 500 AD. Further evidence that the Greek text goes back to the early sixth century is found in the fact that the sixth-century Latin translation mentioned above appears to have been made from the same Greek text as the one we have. That translation is of the Systematic collection, but much of the same material appears in the Alphabetic, though in a different order, and the Greek texts can be compared. In the Paphnoutios text discussed below, for instance, the text in Migne (Alphabetic) differs only in very minor details from that in Guy’s edition of the Systematic;17 the Latin version can be seen to translate this Greek.18 The translation therefore indicates the existence in the early sixth century of a Greek text of the Alphabetic like the one we now have, even though the current MSS are later. 4. AIM The aim of this paper is philological, to draw attention to the value of the Sayings as a witness to Late Koine Greek. Their potential contribution for this purpose has been generally overlooked in study of the history of the Greek language. While some of the vocabulary has been noted in Lampe’s lexicon,19 and is often well observed in Sophocles’s,20 there 15 16 17 18

19 20

Browning, Greek, 57–8; cf. Horrocks, Greek, 274–7. Browning, Greek, 58–60; 63–6; cf. Horrocks, Greek, 284–323. Guy, Les Apophtegmes: Collection systématique, 3:19–20; XVII 15. Migne, PL, 73:975A–B. The first few lines: Dicebant de abbate Paphnutio, quia non cito bibebat vinum. Ambulans autem aliquando iter, supervenit in conventu latronum, et invenit eos bibentes: cognovit autem eum qui primus erat latronum, et sciebat quia non biberet vinum. Videns ergo eum de multo labore fatigatum, implevit calicem vini, et in alia manu tenuit gladium evaginatum, et dicit seni: Si non bibis, occidam te. Lampe, PGL. Sophocles, Lexicon. See, e.g., s.vv. νερόν, φαγίον.

378

COLLECTED ESSAYS ON THE GREEK BIBLE

is no mention of the Sayings in a standard work such as Browning and there appears to be just one in Horrocks.21 But they offer riches that ought to be better known. In a short paper such as this, the thorough study that needs to be done is not possible. My paper is to be seen as a first foray into a subject that awaits a much larger enquiry. Other sources of Greek of the same period may be mentioned in passing. There are of course numerous patristic works contemporary with the Sayings. But their Greek is usually of a much higher literary level, often following Attic models, and provides little information about spoken Greek. One work of a similar linguistic level and similar in content, a little later in date, is the Meadow of John Moschos.22 One known parallel between its language and that of the Sayings is the occasional use of θέλω as a future auxiliary.23 It too could do with more thorough study. But the other main source of everyday Greek in this period, as in all earlier periods from the time the evidence begins in the third century BC, is the body of documentary texts preserved on papyrus and stone. They of course vary in linguistic level, but parallels with the evidence of the Sayings are evident (e.g., νερόν, μισθοφορία, below). Making the links and exploiting the full value of the combined evidence is something still to be done.24 In the following discussion, I start with a sample text and note the features of interest occurring in it. This shows at a glance what a typical saying is like and its linguistic potential. Then follow lists of features from elsewhere in the Sayings, with brief remarks on their significance. A full discussion of each item and its place in the longer-term history of the language cannot be attempted here. Many of the items will be recognised by students of various aspects of Koine Greek usage, including that of the NT and even 21

22

23 24

Browning, Greek; Horrocks, Greek, 292 on κἂν ἕν at Migne, PG, 65:261B [not 45] (the book has no Greek index). I do not find the Sayings (Ἀποφθέγματα …) in the list of works covered in the Mega Lexikon. Text in Migne, PG, vol. 87 (TLG no. 2856). See Horrocks, Greek, 253–6 for an extract with comments. The Historia Monachorum (TLG no. 2744), is another similar text, earlier in date (V AD), but somewhat more literary in language: text in Festugière, Historia Monachorum; Engl. translation Russell, Lives of the Desert Fathers. Lee, ‘Auxiliary θέλω,’ 23. Millar (Greek Roman Empire, 16) presses the claims of the records of the fifth-century Church Council debates as representing ‘for spoken Greek by far the best evidence for the Greek language that survives from Antiquity.’ A record of what was spoken in the Councils they may be, but Millar overlooks the educated quality of the Greek used by the learned bishops. For better representatives of ‘spoken Greek,’ see the letter of condolence, POxy 16.1874 (VI AD) or the catalogue of complaints against a husband, POxy 6.903 (IV AD) – or the Sayings. (I am grateful to Trevor Evans and Genevieve Young-Evans for bringing Millar’s remark to my attention.)

THE SAYINGS OF THE DESERT FATHERS

379

the LXX. Some are early instances of developments that were to become established later in Greek and to continue into the modern language; others look back to older usage on which they throw new or additional light.

5. A SAMPLE TEXT WITH

FEATURES OF INTEREST

5.1 Paphnoutios 1/2 (P105, M377C–380A)25 Ἔλεγον περὶ τοῦ ἀββᾶ Παφνουτίου, ὅτι οὐ ταχέως ἔπινεν οἶνον. Ὁδεύων δέ ποτε εὑρέθη ἐπάνω κοληγίου λῃστῶν, καὶ εὗρεν αὐτοὺς πίνοντας οἶνον. Ἐγνώριζε δὲ αὐτὸν ὁ ἀρχιλῃστής, καὶ ᾔδει ὅτι οὐ πίνει οἶνον. Καὶ θεωρῶν αὐτὸν ἀπὸ μεγάλου κόπου, ἐγέμισε ποτήριον οἴνου, καὶ τὸ ξίφος ἐν τῇ χειρὶ αὐτοῦ, καὶ λέγει τῷ γέροντι· Ἐὰν μὴ πίῃς, φονεύω σε. Γνοὺς δὲ ὁ γέρων ὅτι ἐντολὴν Θεοῦ θέλει ποιῆσαι, βουλόμενος αὐτὸν κερδῆσαι, ἔλαβε καὶ ἔπιεν. Ὁ δὲ ἀρχιλῃστὴς μετενόησεν αὐτῷ, λέγων· Συγχώρησόν μοι, ἀββᾶ, ὅτι ἔθλιψά σε. Καὶ λέγει ὁ γέρων· Πιστεύω τῷ Θεῷ, ὅτι διὰ τὸ ποτήριον τοῦτο ποιεῖ μετὰ σοῦ ἔλεος καὶ ἐν τῷ νῦν καὶ ἐν τῷ μέλλοντι αἰῶνι. Λέγει ὁ ἀρχιλῃστής· Πιστεύω τῷ Θεῷ, ὅτι ἀπὸ τοῦ νῦν οὐ μὴ κακοποιήσω τινά. Καὶ ἐκέρδησεν ὁ γέρων ὅλον τὸν κολήγιον, ἀφεὶς τὸ θέλημα αὐτοῦ διὰ τὸν Κύριον. They said concerning Abba Paphnoutios that he drank wine reluctantly. Once when he was on a journey he found himself face to face with a gang of robbers, and he found them drinking wine. The robber chief recognised him and knew that he did not drink wine. Seeing that he was tired from great exertion, he filled a cup of wine and with his sword in his hand said to the elder, ‘If you don’t drink, I’ll kill you.’ The elder, knowing that he was about to perform a command from God, and wishing to win over the robber, took it and drank. The robber chief made a metania to him, saying, ‘Forgive me, Abba, for distressing you.’ The elder said, ‘I trust in God that because of this cup he will show you mercy both now and in the age to come.’ The robber chief said, ‘I trust in God that from now on I will do no harm to anyone.’ And so the elder won over the whole gang, by giving up his own will for the Lord.

This is a simple narrative in ordinary Greek, with occasional biblical flourishes. It is literate but not literary. Even a short extract like this is full of interesting linguistic material. Some features are routine Koine usage, 25

Key: name of Abba + saying no. (x/x if different in P and M); P = Paschos page; M = Migne, PG 65, col. no. This saying is also found in the Systematic collection at Guy, Les Apophtegmes: Collection systématique, XVII 15.

380

COLLECTED ESSAYS ON THE GREEK BIBLE

though useful as confirming the continuance of the word or use to this date; others open up a whole line of enquiry. The notes indicate the point of interest, without going into full detail. 5.2 Features of Paphnoutios 1 οὐ ταχέως. This is evidence of ταχέως in the sense of ‘readily, without hesitation,’ a weakening of the original sense ‘quickly.’ It could support a similar interpretation of NT Gal 1:6; 2 Thess 2:2; 1 Tim 5:22. The attestation noted in BDAG is quite limited. εὑρέθη, ‘found himself.’ That is, ‘he came to be, arrived at,’ as in the similar English expression, and French se trouver (cf. Guy’s translation il se trouva). There is no actual finding involved. This has relevance to the debated occurrence in Acts 8:40, where even Semitic influence has been suggested: see BDAG, s.v. 1.b. There are others in the Sayings, e.g., Milesios 2: P77, M297B. ἐπάνω is normally ‘above’; here it is like Engl. ‘on top of,’ in effect ‘face to face with.’ Similarly ἦλθόν ποτε ἐπάνω αὐτοῦ τρεῖς λῃσταί (Theodoros of Pherme 29: P41, M196A). Cf. from further afield: LXX Gen 18:2 ἰδοὺ τρεῖς ἄνδρες εἱστήκεισαν ἐπάνω αὐτοῦ. POxy 6.903.20 (IV AD) καὶ ἐποίησεν | τὰς ἔξω θύρας αὐτοῦ ἐνκλισθῆναι ἐπάνω μου, ‘he had the outer doors of it (the church) closed in my face.’26 There is more to find out about the history of this interesting word, now standard Mod. (ε)πάνω. κολ(λ)ήγιον, a Latin word (collegium, ‘college’; ‘guild,’ etc.) borrowed into Greek. LSJ cite two older examples; Lampe, PGL has others contemporary. The meaning ‘gang,’ not in Latin covered by the OLD (to end II AD), is clearly attested here. ἐγνώριζε. The modern meaning ‘recognise’ (= ἀνα-) fits best; similarly Kronios 5: P60, M249C. No instances are recorded in LSJ; Lampe, PGL; BDAG. θεωρῶν. The standard Koine use continues. ἀπό. A special use of the preposition: ‘just arrived from’; similarly ἦν γὰρ ἀπὸ καμάτου πολλοῦ (Makarios 4: P65, M264C). This is the same as in Mark 7:4 ἀπ᾽ ἀγορᾶς ‘when they come in from the agora,’ on which there has been (unnecessary) debate: see BDAG, s.v. ἀγορά. ἐγέμισε. Standard Koine use of the verb applied to filling anything, not just ships, as originally. 26

I was led to this last example by LSJ (s.v. ἐπάνω I.3), though their meaning in the presence of is not right.

THE SAYINGS OF THE DESERT FATHERS

381

φονεύω. Present with future meaning: ‘I kill you’ = ‘I will kill you.’ (Cf. below.) θέλει. ‘Wanted, wished’ does not fit. Rather, the sense is: ‘was about to do/would be doing.’ It is an instance of θέλω beginning to function as a future auxiliary. (Cf. below on the future tense.) βουλόμενος. At an earlier time (e.g., in the NT) this was the more formal or stylish equivalent of θέλω in the meaning ‘wish, want.’27 While βούλομαι is not uncommon in the Sayings (c.18 times), θέλω remains far commoner (200+). A fuller study of the two in this period would be interesting. Perhaps βούλομαι had made a comeback under learned influence. κερδαίνω (twice). What meaning is intended? Originally the verb means ‘gain (money), profit, get for oneself’; hence ‘gain (for God)’ > ‘win over, convert’ (as 1 Cor 9:19, 20, 22); or ‘win over’ could be taken simply to mean ‘win the favour of, avert the hostility of.’ The meaning ‘convert’ is perhaps more likely. A study of the semantic history of the word, with its modern descendant κερδίζω, would shed more light.28 μετενόησεν. Not ‘expressed repentance to him,’ but a physical action, ‘made a metania to him.’ Cf. the frequent expression ἔβαλε/ ἔβαλον μετάνοιαν, ‘made a metania’ (Kronios 5: P60, M249B +). Similarly Lampe, PGL, s.v. μετανοέω C. συγχωρῶ, ‘forgive’ is post-NT usage, but becomes significant in liturgical and patristic language (see Lampe, PGL, s.v. 5; Sophocles, Lexicon, s.v.). θλίβω. Here it shows a weakened meaning, like ‘trouble, bother,’ not ‘afflict’ etc., as earlier. Ward’s ‘made you unhappy’ is on the right track. Modern θλίβω may have a similarly mild sense. ποιεῖ. Another example of present in future sense. ποιεῖ μετὰ σοῦ ἔλεος. An OT Greek expression derived from Hebrew. ‘Do mercy with’ = ‘show mercy to.’ This and the following καὶ ἐν τῷ νῦν καὶ ἐν τῷ μέλλοντι αἰῶνι are biblical and liturgical in origin and flavour. ἀπὸ τοῦ νῦν is also a biblical phrase: in the LXX c.20 times, notably in the Psalms, and in the NT 8 times including the extremely familiar ἀπὸ τοῦ νῦν μακαριοῦσί με πᾶσαι αἱ γενεαί (Luke 1:48). οὐ μή + fut., aor. subj. This construction was originally Classical, but is found only sporadically in Koine Greek. Its adoption for use in the LXX, 27 28

See Shipp, MGE, 166–7. Ward (The Sayings, 202) translates the first occurrence of κερδαίνω in our text as ‘to win the confidence of’ and the second as ‘converted’: but the meaning ought to be the same in both. I do not think the meaning ‘avoid, spare oneself,’ that is found in BDAG (s.v. 2) for Acts 27:21, and in LSJ (s.v. III), and might fit here, is soundly established.

382

COLLECTED ESSAYS ON THE GREEK BIBLE

however, where it is very frequent as a solemn-sounding negative, ensured its entry into the NT and liturgical Greek.29 It probably continues to have a ‘biblical’ flavour in the Sayings, though it is not a ‘Hebraism.’ There are many examples. διὰ τὸν Κύριον. Here διά = ‘for the sake of,’ a good example of extension in the meaning of διά, leading to Mod. για, with a range as wide as Engl. ‘for.’ Cf. already in Mark 2:27 τὸ σάββατον διὰ τὸν ἄνθρωπον ἐγένετο.

6. SIGNIFICANT FEATURES ELSEWHERE IN THE

SAYINGS

6.1 Morphology ἀναβάζω, ‘bring up’ = ἀναβιβάζω, cf. Mod. ανεβάζω ‘bring up’ (Makarios 30: P69, M273C).30 ἀφίω (or ἀφιῶ), δίδω (-ῶ), τίθω (-ῶ) are frequent, providing further evidence of the new thematic forms that appeared early in Koine Greek.31 ἔσο ‘be’ (imper.): ἅλατι δὲ ἠρτυμένος ἔσο ἐν Κυρίῳ, ‘be seasoned with salt in the Lord’ (Or 13: P127, M440C; also Antonios 6: P2, M77A). This echoes Col 4:6 ὁ λόγος ὑμῶν... ἅλατι ἠρτυμένος.32 Only rare glimpses of this replacement for ἴσθι are available. Some few examples occur in literature, starting with Plutarch.33 PMert 1.45r.6 (V–VI AD) ἔσο δὲ καὶ αὐτὸς μέρος τῆς καθ[- - -] is a possible instance. ληκύνθιον, ‘flask,’ for ληκύθιον: not recorded in LSJ or Lampe, PGL, but in Sophocles, Lexicon (Makarios 3: P64, M261A). There is no reason to doubt it is a real form, influenced by other words ending in -ίνθιον.34 μελανός, ‘black,’ for older μέλας, continued in Modern Greek (Moyses 4, 8: P72, M284B, 285A). See LSJ; Lampe, PGL; Sophocles, Lexicon. ῥιγᾶτε, ‘shiver,’ originally ῥιγέω, -όω (Makarios 2: P64, M260D).35 29 30 31 32

33

34 35

See Lee, ‘Some Features,’ 18–23. On loss of reduplication see Gignac, Grammar, 2:242–3. Gignac, Grammar, 2:380–3. Ward has gone quite wrong: ‘Let your salvation be founded in the Lord.’ This seems to be a flawed version of the Latin transl. in Migne: Sale autem conditus sis in Domino. LSJ, s.v. εἰμί, introd. In Thackeray, Grammar, 257 n. 1, it is noted in Symmachus (without ref.). See Buck-Petersen, Reverse Index, 73. See Gignac, Grammar, 2:363–5 on the merging of inflectional types in contract verbs.

THE SAYINGS OF THE DESERT FATHERS

383

6.2 Vocabulary: common Koine αἴρω, ‘take’: δεῦρο σὺ ἆρόν με πρὸς τὸν βασιλέα (Poimen 109: P95, M349A +). ἀκμήν, ‘still’ (Sisoes 7: P110, M393B; Arsenios 24: P7, M93C +). ἀφανίζω, ‘ruin, spoil’ (Karion 2: P61, M252C). δέρω, ‘hit’ (Zacharias 4: P36, M180B). διαφέρω, ‘belong to’ (Karion 2: P60, M252A). εὑρίσκω, ‘be able’ (Poimen 111: P95, M349C +).36 ἴδιος as simple possessive (e.g., Karion 2: P60, 61, M249D, 252A). καιρός, ‘period of time’ (Karion 2: P60, M249D). παραβάλλω, ‘come to, visit’ (Antonios 12: P2, M77C + many). πλήν, ‘but’ (Matoes 7: P75, M292B +). πληροφορῶ, ‘inform’ (Eucharistos: P32, M169A). τρώγω, ‘eat’ (Sisoes 31: P113, M401C +).37 φεύγω, ‘go away’ (Sisoes 27: P113, M401A; Kopris 3: P61, M252D). φθάνω, ‘reach, arrive at’ (Agathon 13: P13, M113A +). 6.3 Vocabulary: new examples of uncommon or unknown words and uses ἀρχάριος, ‘novice, beginner’ (Esaias 1: P36, M181A). βράζω / βράσσω, ‘boil’ (food) (Esaias 5/6: P37, M181C). γεοῦχος, ‘landowner’ (Esaias 4/5, P36, M181B). γραΐς, ‘old woman’ (Sarra 4: P119, M420C +). δράξ, ‘handful’ (Esaias 5/6: P37, M181C). ἑαυτόν as 1st pers. sing. reflexive, ‘myself’ (Simon 2: P117, M412D). ἐπαίρω, ‘take away’ (Antonios 5: P1, M77A). A further contribution to the evidence for the derivation of Mod. παίρνω from ἐπαίρω (not ἀπαίρω).38 θεωρία, ‘vision’ (Zacharias 4: P36, M180B). καθάπαξ / κατὰ ἅπαξ, ‘each time’ (Kronios 5: P60, M249B; Poimen 93: P93, M344D). κρούω, ‘run,’ ‘spring up’: οἴδατε καὶ τὸ χωρίον τοῦ δεῖνος ὅπου ὁ ποταμὸς κρούει; (Makarios 30: P69, M273C). There is no record of this use elsewhere. Cf. κρουνός, ‘spring.’ 36 37 38

Cf. Lee, ‘A Non-Aramaism.’ Cf. Shipp, MGE, 540–1. See Shipp, MGE, 243. Exx. from other sources in Lampe, PGL, s.v. ἐπαίρω B, and Sophocles, Lexicon, s.v. ἐπαίρω.

384

COLLECTED ESSAYS ON THE GREEK BIBLE

λαγύνιον, ‘flagon’ (Moyses 13: P73, M288A). μισθοφορία, ‘rent; payment,’ here a special payment in return for freedom (Mios 2: P78, M301C). Lampe, PGL (service for wages) and Ward (‘his wages’) are on the wrong track. The evidence in late documentary texts (BGU 12.2139.12, 13 [432 AD], 17.2696.27 [616 AD] +) helps a better understanding of this old but rare word. Cf. LSJ, Suppl. (1996), s.v. (rent, for BGU 12.2139), Sophocles, Lexicon (‘payment by a freedman to his former master,’ for Mios 2). νερόν, ‘water’ (Ioannes o Kolobos 7: P45, M205B). The standard Mod. word νερό already in place. If Ioannes’ own words are preserved, we could date this example to early V AD (he died in 409, according to Guy). Other evidence around this time is secure but meagre.39 περνῶ (-άω), ‘cross over’; ‘pass through’: φύσει οὐ περνᾷ τὴν φάραγγα ταύτην, καὶ βλέπει με (Or 6: P126, M440A). Here we have a rare early attestation of what would become a common Modern verb. ποῖον; ‘which? what?’ (= older τί;): ἕν ἐστι μόνον ἐν ᾧ νικᾷς με. Λέγει … ποῖον τοῦτο; ‘There is only one thing in which you beat me.’ He said, ‘What’s that?’ (Makarios 11: P66, M268C +). Evidence of progress towards the standard modern use. The beginnings are much earlier: see LSJ, s.v. ποῖος IV; BDAG, s.v. ποῖος 2. σαλός, ‘half-witted’ (Moyses 8: P72, M285A).40 σπλαγχνίζομαι, ‘have pity’ (Moyses 18: P74, M289A). συλῶ (-άω), ‘steal; rob (someone)’ (Saio: P119, M420B). τελῶ pass., ‘become ripe’ (Arsenios 19: P6, M92C). τίποτε, ‘something, anything’ (Sisoes 7: P110, M393B). Mod. τίποτα. φαγίον, τό, ‘food’ (Sisoes 52: P115, M408B). ψιχίον, τό, ‘crumb’ (Phokas 2: P124, M433C). Found in the NT (Matt 15:27; Mark 7:28), but not well attested (cf. BDAG; Lampe, PGL). 6.4 Syntax 6.4.1 The future tense. New ways of expressing futurity are appearing: (a) The present indicative, which is frequent (e.g., Paphnoutios 1/2 quoted above). (b) θέλω as future auxiliary: πῶς οὖν θέλεις σῖτον λαβεῖν, μὴ θερίσας; ‘how then will you receive grain if you have not harvested?’ 39

40

See note on POxy 56.3865.35 (late V AD) by Sirivianou, with refs. there. For the date of Ioannes: Guy, Les Apophtegmes: Collection systématique, 67. Cf. POxy 56.3865.57 with Sirivianou’s note.

THE SAYINGS OF THE DESERT FATHERS

385

(Esaias 4/5: P36, M181B); ἤθελε σχίσαι ‘he was about to tear it in two’ (Arsenios 29: P8 M97C; simil. Makarios 1: P64, M260B; Silouanos 1: P115, M408C +).41 (c) ἔχω + aor. infin. (Sisoes 15: P111, M397A; Arsenios 24: P7, M96A +).42 6.4.2 Other auxiliary uses:43 (a) εἶχον + infin. = ‘would have’ (Makarios 21: P67, M272A). (b) ἔχω + pres. part.: ἔχεις πολλὴν ὥραν κρούων; ‘have you been knocking for a long time?’ (Arsenios 27: P7, M96C). (c) εἰμι + part.: ἤμην ἀναβαίνων, ‘I used to go up’ (Poimen 110: P95, M349C). 6.4.3 Aoristic perfect: ἔλεγον περὶ τοῦ Ἀββᾶ Μακαρίου τοῦ μεγάλου, ὅτι γέγονε καθώς ἐστι γεγραμμένον θεὸς ἐπίγειος, ‘They said about Abba Makarios the Great that he was (or became), as it is written, God on earth’ (Makarios 32: P69, M273D; Arsenios 32: P9, M100C; Karion 2: P61, M249D +). 6.4.4 Other features τό as relative: ἐκεῖνο τὸ ἐφάγομεν τῆς ἀγάπης ἦν, ‘that (food) which we ate was for love’ (Silouanos 1: P115, M408C). εἷς = ‘a’: καὶ ἰδοὺ μία χήρα καλαμωμένη ἦν ὀπίσω ἡμῶν, ‘and there was a widow gathering reeds behind us’ (Makarios 6/7: P65, M265A; simil. Sisoes 47: P115, M405C +). ἐπάνω + gen. as prep.: frequent. ὀπίσω + gen. as prep.: frequent. αὐτός as demonstr.: ἀπῆλθεν εἰς ἀγρὸν καὶ ἐκρύβη … οἱ πατέρες … φθάσαντες εἰς αὐτὸν τὸν ἀγρόν …, ‘he went into a field and hid. … The fathers, arriving at that field. …’ (Isaak 1: P51, M224B +).44 ἵνα temporal: ἔρχεται καιρός, ἵνα οἱ ἄνθρωποι μανῶσι, ‘a time is coming when men go mad’ (Antonios 24/25: P4, M84C). Cf. NT examples noted in BDAG, s.v. ἵνα 2.d. ἵνα imperatival or jussive: εἰ δὲ ἔτι ζῶντα, ἵνα μεταλάβω τῆς ἁγίας κοινωνίας, ‘but if (you find me) still alive, let me receive Holy Communion’ 41 42 43 44

Cf. Lee, ‘Auxiliary θέλω.’ Cf. Browning, Greek, 33; Horrocks, Greek, 130. Cf. Browning, ibid. On the development of αὐτός see Horrocks, Greek, 128–9.

386

COLLECTED ESSAYS ON THE GREEK BIBLE

(Phokas 2: P124, M433A +). This phenomenon has a long history, reaching back to NT times and even earlier, and resulting in imperatival uses of να + subj. in the modern language.45 6.5 Idioms and colloquial expressions ἄρχομαι + infin., e.g., ὡς δὲ ἤρξατο ἰδιάζειν αὐτόν, ‘when he came to be/was alone with him’ (Makarios 3: P65, M264A +). This is relevant to the debate about ‘pleonastic’ ἄρχομαι in the NT, allegedly due to Semitic influence. A proper study of the evidence outside the NT has not been done.46 βάλλω + infin. ‘begin (to)’: ἔβαλε κράζειν ‘he began to shout’ (Philagrios: P124, M436A +). Cf. Lampe, PGL, s.v. C.5. εἰς τὰ ἴδια ‘to one’s home’: ἀπῆλθε μετὰ χαρᾶς εἰς τὰ ἴδια, ‘she went home with joy’ (Arsenios 28: P8, M97B). Cf., e.g., John 16:32 ἔρχεται ὥρα … ἵνα σκορπισθῆτε ἕκαστος εἰς τὰ ἴδια. ἐφέτος / ἐπέτος adv., ‘this year’: εὐχαριστοῦμεν τῷ Θεῷ, ὅτι ἔβρεξεν ἐπ᾽ ἔτος [sic M] πολλά (Ioannis o Kolobos 10: P45, M208A; Or 4: P126, M437C). This is the Mod. (ε)φέτος, ‘this year’ making its appearance. Other evidence around this time can be seen in Sophocles, Lexicon, s.v. ἐπέτος, and earlier in LSJ, Supplement (1996), s.v. ἐφέτος, improving the LSJ entry. ἦν ἰδεῖν αὐτὸν ὡς ἄγγελον ‘he was like an angel to look at’ (Kronios 5: P60, M249B). ἰδού is frequent, and not only in the sense of ‘behold, look’ but also ‘here is/was, there is/was’ (Arsenios 36: P10, M101C; Ammonas 6: P15, M120D +). κέλευσον, ‘give the word,’ in effect ‘please allow’: κέλευσον ἵνα κἀγὼ βρέξω ἐμαυτῷ, ‘please let me also soak (some palms) for myself’ (Makarios 4: P65, M264C). λοιπόν, ‘then, therefore’ (Sisoes 3: P110, M392D). μὴ οὐ …; strengthened form of οὐ introducing a question expecting the answer ‘yes’: μὴ οὐκ εἰμὶ μήτηρ ὑμῶν; ‘am I not your mother?’ (Poimen 76: P92, M341A). NT instances like 1 Cor 9:4 are clearly similar: μὴ οὐκ ἔχομεν ἐξουσίαν φαγεῖν καὶ πεῖν; cf. BDAG, s.v. μή 3.a. 45 46

See, e.g., BDAG, s.v. ἵνα 2.g; Mandilaras, Verb, §§585–9. There are numerous examples in the popular I AD text Vita Aesopi G. [Hunkin’s 1924 study ‘Pleonastic ἄρχομαι’ goes a good way towards correcting old misunderstandings.]

THE SAYINGS OF THE DESERT FATHERS

387

ὅλως in several idiomatic senses: ‘at all’ (Agathon 14: P13, M113A; Isidoros the Priest 1: P55, M236A); ‘just’ (Esaias 5/6: P37, M181C; Poimen 154: P99, M360C); ‘really, actually’ (Poimen 76: P92, M341B). The interpretation of the NT examples at 1 Cor 5:1 and 15:29 might be aided by this evidence (cf. BDAG, s.v. ὅλως). οὗτος οὐκ ἔχει πρᾶγμα, ‘he has nothing to do with it’ (Makarios 1: P64, M260A). ποῦ ἔνι μοι …; ‘how can I …?’ (Makarios 3: P64, M261B). πῶς ὁ κόσμος; ‘how’s the world?’ (Makarios 2: P64, M260D).47 πῶς τὰ κατὰ σέ; ‘how are things with you?’ (Makarios 3: P65, M264A). σωθείης, ‘good health!’ (Makarios 3: P64, M261B; +). τί ἔχεις, πάτερ; ‘what’s up, father?’ (Sisoes 47: P115, M405C). τόπον ἐκ τόπου, ‘to one place after another’ (Synklitiki 6: P120, M424A). The remarkable longevity of this expression is demonstrated. It is first attested in a III BC papyrus (PLond 7.2049.8), and sporadically thereafter. The same syntactic pattern is seen in ἡμέραν ἐξ ἡμέρας (LXX etc.) and other phrases.48 φύσει as an interjection or particle (cf. Engl. ‘naturally’), with a range of uses hard to define precisely: ‘in fact,’ e.g., καὶ πόσον χρόνον ἔχεις ὧδε; ὁ δὲ ἔφη· Φύσει, Ἀββᾶ, ἔχω ἕνδεκα μῆνας ἐν τῷ ὄρει τούτῳ, ‘… Actually, Abba, I’ve been on this mountain eleven months’ (Sisoes 7: P110, M393B); ‘really,’ ‘truly,’ e.g., ἠρώτησαν αὐτὸν οἱ ἀδελφοί· Πῶς ἡ πόλις; ὁ δὲ εἶπε· Φύσει, ἀδελφοί, ἐγὼ πρόσωπον ἀνθρώπου οὐκ εἶδον, εἰ μὴ μόνον τοῦ ἀρχιεπισκόπου, ‘… Truly, brothers, I did not see the face of any man except the Archbishop’ (Isidoros 8: P50, M221B; simil. Arsenios 25: P7, M96A +). Cf. Lampe, PGL, s.v. φύσις II.D.2; Sophocles, Lexicon, s.v. φύσις 5. This idiomatic φύσει seems to have arisen in Late Koine, not before. There does not appear to be any direct modern descendant of φύσει; but there is φυσικά.49 A full study would be useful. The ancient ‘onoma rule’ is still being followed, i.e., ὄνομα without article and the possessor of the name in dat., in naming-constructions of the type ‘the name of x is N’: ὄνομα δὲ αὐτῇ Παησία (Ioannis o Kolobos 40: P49, M217B).50 47

48 49 50

This expression (in the form πῶς εἶναι ὁ κόσμος;) has continued in use right up to today. I am grateful to Fr Miltiades Chryssavgis for confirming this and demonstrating it in conversation with others (on 11th April, 2004, Pascha). [See discussion of LXX exx. of this pattern in Lee, Greek of the Pentateuch, 141–5.] I thank Emmanuel Roumanis for drawing my attention to this. See Lee, ‘Onoma Rule.’

388

COLLECTED ESSAYS ON THE GREEK BIBLE

CONCLUSION The linguistic value of the Sayings material is established with little difficulty. We have been able to observe a wide range of features, from basic Koine Greek continuing in use to striking instances of new developments. All of these help to fill out the record of the language in this period, a record that is limited for this level of Greek. Besides completing the record for possible inclusion in lexicons and other reference books, value will be found in this evidence for studies of particular features, such as the development of the future tense. The study of the history of the Greek language is not yet complete by any means. Two further observations may be made. This study shows that there are untapped sources of evidence still available. Periods for which we thought we had little information turn out to be witnessed to by texts into which little or no enquiry has been made. Work needs to continue on these less obvious sources. Essential to further progress is the combining of evidence from all available sources and eras, even those distant in time. Electronic databases and searching tools have changed the nature of such work and made it immensely faster and easier than ever before. The second point concerns the recording of the results. How will it happen? Printed lexicons (and grammars) take many years to be updated and even then cannot include all the material available. Recording of data is haphazard and slow. New observations are scattered through multiple publications and easily missed. Moreover, there is as yet no lexicon that is devoted to the Koine Greek period: Lampe’s lexicon is focused on theological vocabulary and leaves the rest to be taken care of by LSJ; LSJ itself is focused on the Classical period, with haphazard coverage of later Greek;51 the Mega Lexikon covers everything from Classical to Modern, but has many gaps (and is unlikely to be updated); the excellent lexicon of Kriaras begins its coverage later; and the specialist lexicons of the NT and LXX are focused on what concerns them. But the era of these works is coming to an end. Electronic databases will serve the same purpose in a far more efficient way. This is not to say that printed lexicons will cease to exist, only that the major storehouse of lexicographical data will be an electronic database. This is a development not yet achieved, but one to be sought and welcomed.52 51

52

DGE provides better coverage of the post-Classical period than LSJ but still does not cover Koine Greek in depth, and has a long way to go to reach completion. Cf. Lee, ‘Lexicographical Database.’

24 THE VOCABULARY OF THE SEPTUAGINT AND DOCUMENTARY EVIDENCE 2016

Abstract This chapter surveys concisely (a) the history of study of the relation between documentary texts and the LXX; (b) the nature and value of the documentary evidence; and (c) practical aspects of accessing the evidence. Three examples are then presented, namely, ποτίζω, εὐίλατος, and ἀγαπῶ, showing how the documentary evidence impacts on our understanding of these LXX words. A consideration of future directions completes the chapter.

1. INTRODUCTION The Septuagint translators translated their Hebrew Bible into Greek. That Greek was the Greek of their time, the third to first centuries BC. Our knowledge of it depends on evidence in surviving Greek texts. Much of that evidence comes from sources that have long been known and have formed the basis of the material in our lexicons, namely, literary texts of both the translators’ time and the preceding period of ancient Greek (Classical Greek). These texts comprise a range of genres, styles, dialects, and levels of Greek, from the Epic of Homer to the historical prose of Polybius. Everyday language is not well represented, though much can be gleaned from writers like Plato and Aristophanes. The literary sources may provide a fairly accurate picture of the usage of a word in the Greek known to the LXX translators, and then again may not. The discovery in modern times of documentary texts brought forth a different and illuminating source of evidence. Documentary texts are those inscribed or written on stone or papyrus, surviving from the day they were set down. The majority are post-Classical, though there is a large body of inscriptions that are earlier. They are mainly not literary but official or everyday, using the kind of Greek appropriate to their purpose. They thus give access to another dimension of the language, and fill out much of

390

COLLECTED ESSAYS ON THE GREEK BIBLE

the incomplete picture that we worked with before. The absorption of this new evidence into our reference books is still in progress today and will be ongoing as more documentary texts are found and published. They have made, and will continue to make, a major contribution to our knowledge of LXX Greek.

2. HISTORY OF STUDY Investigation of the documentary texts in relation to the LXX began in earnest with Adolf Deissmann, shortly after the first major finds of papyri. But there were forerunners. A full list of those who earlier made the connection would be tedious, but the names of Friedrich Sturz,1 Heinrich Thiersch,2 and Heinrich Anz3 deserve mention. All had little documentary material to work with, but realised its importance for the LXX. John P. Mahaffy, the first editor of the Petrie papyri, contemporary with the LXX Pentateuch,4 was struck by the similarities.5 Henry B. Swete also, following up Mahaffy’s insight, pointed out a number of LXX words found in PPetrie and the just-published BGU, and quoted in full a PPetrie text.6 These were early straws in the wind. Deissmann’s researches were not systematic but a collection of the most telling data he could find to demolish the notion of a ‘Biblical Greek,’ which had long held sway.7 He pointed out documentary parallels to the language of both the NT and the LXX, but the NT was the ultimate target and grows more prominent in the later studies.8 Deissmann drew on what was currently available, both inscriptions and papyri, and though they were few in comparison with subsequent discoveries, they were usually enough to establish his point. His work was a breakthrough at the time: it showed that the documentary texts were the evidence needed to bring the language of the NT and LXX out of the isolation that had been imposed on them. But there was a long way to go to exploit these sources fully, and 1 2 3 4 5 6 7

8

Sturz, De dialecto (1808). Thiersch, De pentateuchi versione (1841), 87–90. Anz, Subsidia (1894). Mahaffy, Flinders Petrie Papyri. See Swete, Introduction, 21 for quotations from Mahaffy and refs. Swete, Introduction, 292–3. Deissmann, Bible Studies (transl. of Bibelstudien, 1895 and Neue Bibelstudien, 1897); idem, Light from the Ancient East (transl. of Licht vom Osten (4th ed., 1923). On the LXX (with NT) see, in particular, Bible Studies, 86–169 (about 70 lexemes); with more focus on the NT, 194–267 (= Neue Bibelstudien), and, generally, Light (LXX 81– 116 passim).

THE VOCABULARY OF THE SEPTUAGINT AND DOCUMENTARY EVIDENCE

391

some of Deissmann’s conclusions have not worn well, especially his view that the Greek found in the NT was the language of ‘simple, unlearned folk.’9 Even his statement that the language of the Seventy was ‘the Egyptian Greek of the period of the Ptolemies’ needs considerable nuancing.10 James Moulton took up the challenge of gathering documentary parallels to the vocabulary of the NT, and with his (eventual) collaborator George Milligan produced the now-famous ‘MM.’11 As Frederick W. Danker puts it, ‘Their work helped open up even more the curtains that Deissmann had drawn aside to expose an exciting new world for New Testament explorers.’12 But MM had limited application to the LXX: its focus was the NT, even if LXX vocabulary was often mentioned along the way. And there were other limitations: it is not a lexicon, and the analyses of meaning are not well done; it deals with a selection of NT words, not the whole vocabulary; it draws mostly on papyri, not inscriptions; and its material is inevitably out of date and incomplete now.13 The ‘new world’ has expanded to undreamt-of horizons. Those who use MM today need to be aware that it must be supplemented by further investigation, and that its conclusions are not necessarily reliable. An ‘MM’ for the LXX was on no one’s agenda when MM reached completion in 1929. Observations on selected parallels may be found in Joseph Ziegler’s 1934 study of the book of Isaiah,14 and in a survey article by Orsolina Montevecchi in 1964.15 That was the scene when the present author’s 1970 dissertation on the LXX Pentateuch was undertaken.16 Its aim was to combat once more the idea of a special ‘Jewish Greek,’ which had shown a resurgence, by gathering evidence for the links between the vocabulary of the Pentateuch and the contemporary language. Use was made especially of third-century BC papyri from Egypt, which had grown in quantity but been barely exploited. Only a selection of words and uses was studied, namely, new developments in Koine Greek, as being especially 9 10

11 12 13

14

15 16

Deissmann, Light, 62; cf. 143–5 (‘Book of the People,’ etc.). Deissmann, Bible Studies, 70. On the one hand, there is variety within the Greek of Ptolemaic Egypt, and it is not a distinct entity from Greek elsewhere; on the other, the LXX translators made adjustments to the language to suit their purpose and chose differing styles and literary levels. MM; originally published in 5 vols. in 1914–1929, with a one-volume edition in 1930. Danker, Multipurpose Tools, 117. See Hemer, ‘Towards a New MM’; Horsley, New Documents, 5:83–92; Lee, History, 123 with n. 9. Ziegler, Untersuchungen, 175–212. See assessment of Ziegler in van der Meer, ‘Papyrological Perspectives,’ esp. 109–10. Montevecchi, ‘Continuità ed evoluzione.’ Lee, Lexical Study.

392

COLLECTED ESSAYS ON THE GREEK BIBLE

good indicators of the translators’ familiarity with the Greek of their time. Widening the field to other vocabulary and other books of the LXX remained to be done. The years following saw minor studies on selected words or parts of the LXX vocabulary by Raija Sollamo,17 Hélène Cadell,18 and the Italian papyrologists Montevecchi, Francesco Vattioni, and Anna Passoni dell’Acqua.19 More recently there has been renewed activity in short studies by this author20 and Trevor Evans,21 and significant results are being obtained in Michael van der Meer’s work on Isaiah, notable for its close engagement with the papyrus documents (and even pictorial remains).22 All this has continued to show the value of the documentary evidence for elucidating the vocabulary of the LXX, but there has been no systematic exploration, and a means of gathering the results of these piecemeal studies into one place has not yet emerged. The two recent lexicons of the LXX, by Lust, Eynikel, and Hauspie and Muraoka, have to some extent performed this role, noting or using any material up to the date of compilation.23 But the rate of production has outstripped the ability of any printed work to keep an up-todate record. These lexicons, moreover, designedly made no attempt at new researches into the documentary evidence, but took in what was on hand, rightly considering further work on that front a task for the future. 3. NATURE AND VALUE OF THE

EVIDENCE

Documentary texts were engraved on stone, or written on papyrus, ostraka, wooden tablets, and other material. They survive from the time when they were written or inscribed, and are in varying states of preservation, from complete to very fragmentary, from perfectly legible to illegible. Decipherment, i.e., simply reading them, is a specialist task, because of differences in the writing of Greek as compared with today, as well as other factors. The range of dates is from the late eighth century BC for inscriptions and from the late fourth century BC for papyri, down to the 17 18 19

20

21 22

23

Sollamo, Semiprepositions, 308–51 (parallels in documents and literary texts). Cadell, ‘Vocabulaire de l’irrigation’; ‘Vocabulaire de la législation.’ Detail in van der Meer, ‘Papyrological Perspectives,’ 110; see also HDM, La Bible grecque, 243. Lee, ‘Thirty Years On’; ‘Ἐξαποστέλλω’; ‘Lexicographical Database’. Note also Lee, ‘Aposkeuê in the Septuagint,’ and O’Connor and Lee, ‘A Problem.’ Evans, ‘Court Function of the Interpreter.’ Van der Meer, ‘Trendy Translations’; ‘Bridge over Troubled Waters?’; ‘Papyrological Perspectives.’ LEH, Lexicon; Muraoka, Lexicon.

THE VOCABULARY OF THE SEPTUAGINT AND DOCUMENTARY EVIDENCE

393

end of the Koine period (600 AD) and beyond. The find locations also range greatly and the extremes are remarkable, but it is sufficient to say that Greek inscriptions have been found in quantity all over ancient Greece, Asia Minor, Egypt, and the rest of the eastern Mediterranean as well as Rome, while papyri have survived mainly in upper Egypt, with a small number in Syria, Palestine, and elsewhere; ostraka also come mainly from Egypt, though not all. The number of inscriptions is in the thousands; papyri number more than a million, of which less than a tenth have been published.24 The nature of the documents is endlessly variable and any generalisation is liable to be contradicted by a particular text, but a summary will be useful. The inscriptions tend to be of a formal and public character – not surprisingly, since they were set up for public viewing. They include: decrees of civic bodies, treaties, honorific decrees, dedications, epitaphs, manumissions, sacred laws, records of benefactions, accounts, catalogues, and even milestones.25 Texts on papyrus may be said to vary even more. At the formal end there is, say, a letter of the emperor, at the other extreme a shopping list or a brief receipt on an ostrakon. In between there is an almost inexhaustible range of public and private documents: legal proceedings, contracts, accounts, petitions, tax receipts, reports, lists, and private letters, the last often the most interesting of all for their human appeal and language.26 The Greek of the documents is almost as varied as their content.27 We must first separate the texts of the Classical period (to 300 BC), which are written in varieties of Classical Greek, including of course local dialects. The inscriptions of Athens are in Attic, those of Delphi in Delphian, and so on. Once Koine Greek, which grew primarily out of Attic, emerged, the dialects were gradually eroded, but not entirely: inscriptions in dialect continue to appear as late as the second century AD, usually with an admixture of Koine features. The poetic genres are also an important part of the picture, beginning in the earlier period and continuing right through the later. Epitaphs and the like will often display metrical forms and poetic vocabulary and morphology, though they may also contain Koine features. The researcher needs to be able to discern these differing genres and the features of language that accompany them. 24 25 26 27

Bagnall, Handbook of Papyrology, 645 (P. van Minnen). [Thousands of ostraka have also been found and are still being found and edited.] For a full survey see McLean, Greek Epigraphy, 181–210. See survey in Bagnall, Handbook of Papyrology, 358–94 (B. Palme). Besides Greek, there are of course documents in other languages, notably Latin, Egyptian Demotic, and Coptic, and also bilingual texts.

394

COLLECTED ESSAYS ON THE GREEK BIBLE

The great majority of the texts of the later period, both inscriptions and papyri, are in Koine Greek, but not of one single variety. It is a mistake to regard them all uniformly as witnesses to vernacular or ‘popular’ Greek. While they may and do witness to that, their language offers a range from everyday to literary, and what features appear will depend on the choice of register and the degree of education of the writer. Even an ‘uneducated’ writer may introduce learned features and use stereotyped expressions containing obsolete forms. A sensitivity to the possible variations of ‘literariness’ is a prerequisite in the researcher. Works of literature, whether texts of Classical or of later authors, also survive on papyrus (less often stone) and may be included in modern editions of documents.28 They are by definition not ‘documentary,’ but in terms of language the line is not so easily drawn. The language of literary works overlaps with that of everyday language. For enquiry into the LXX vocabulary this source of evidence is not to be neglected, both for everyday words and for literary vocabulary if ‘literary’ books or sections of the LXX are in view. In any case, any occurrence of a word in any source may be useful for filling in the history of that word. The value of the documentary evidence for study of the vocabulary of the LXX is essentially that it enables us to understand the LXX better. All evidence is valuable, but this body of additional and growing evidence, as was said above, gives access to a dimension of the language that was inadequately known before. The material that is closest in time to the LXX is the most valuable, and there is by good fortune a large body of Ptolemaic papyri. Elucidation involves a quest for parallels, and this has the additional outcome of enabling us to conclude (up to a point) whether a word or use belongs to normal Greek or is peculiar to the LXX. A further aspect of such researches is important. The traffic is not all one way: the LXX itself, used with due caution, is a witness to Koine Greek. By bringing the LXX and documentary evidence together we may elucidate or support either by the other. 4. PRACTICALITIES If access to the documentary texts were just a matter of consulting a printed lexicon or other tool, this section would be unnecessary. As it is, the constant growth of the material and rapid developments in technology 28

The categories ‘subliterary’ and ‘paraliterary’ are also used: see Bagnall, Handbook of Papyrology, 282–3 (T. Renner).

THE VOCABULARY OF THE SEPTUAGINT AND DOCUMENTARY EVIDENCE

395

have meant that printed works could not possibly keep pace. Scholarship itself cannot keep pace. Consulting any of the existing reference works, such as LSJ, Preisigke, Wörterbuch, or MM, gives only a partial and often misleading view. On the other hand, full access to the documentary texts themselves has become easy, transformed in recent years by electronic databases and search tools. Any word can be searched for and found electronically with great speed. But therein lies a difficulty. The results are raw material, a vast body of it, requiring expertise and sheer hard work to make use of. A list of occurrences is just the beginning. The data need thorough analysis if they are to be useful for any purpose. Quick and easy answers will not be found. Until the arrival of digitisation, documentary texts were accessible only in the printed volumes in which they were published. Searching could be done only via the indexes supplied (or not) by the editors, or by continuous reading. Whatever the drawbacks of this form, the book had, and still has, the advantage that discussion, notes, and translation (if provided) can all be readily consulted in one place on the printed page.29 With the transfer of the texts to electronic form much of this assistance has been omitted (though no doubt it will in time be added). Print publication of texts is however still the norm and is the source from which electronic texts are derived (though this too could change). The upshot at present is that for an accurate report of the text and aids to comprehension, it is necessary, or at least desirable, to consult the printed volume.30 Published volumes of documentary texts number many hundreds, and to manage the abundance, lists and abbreviations have been created.31 The secondary bibliography, itself enormous, will be found listed to some extent in the printed volumes, but complete only up to the date of publication; the electronic media are in the process of bringing this under better control.

29

30

31

Translations are an important aid. Besides those in the editions, there are selections of translated documents, which can usually be located via one of the websites (see notes 32 and 34). This of course is easier said than done: it is a rare library that possesses even a majority of the volumes, let alone all. Epigraphic volumes are especially intractable. [Reliance on the electronic form is increasingly the reality whether it should be or not, as access to the printed volumes (often now in ‘storage’) becomes harder.] For papyri: Oates et al., Checklist. Epigraphic abbreviations are in disarray. A start was made on a unified system in Horsley and Lee, ‘Preliminary Checklist.’ This is adopted in McLean, Greek Epigraphy, 387–472, with additions to 2002. The system in PHI Greek Inscriptions is different; CLAROS () offers another; LSJ and LSJ, Suppl. (1996) inevitably have another. A useful guide to epigraphic works is Bérard et al., Guide de l’épigraphiste.

396

COLLECTED ESSAYS ON THE GREEK BIBLE

Electronic searching, which until recently was possible only via CDs, can now be done online. The current main search locations may be noted here,32 but further change and development are inevitable. There are also multiple sites gathering different types of information.33 A significant development has been the creation of the ‘Trismegistos’ database, which aims to gather ‘metadata’ on all published papyrological and epigraphic texts from Egypt, in the period 800 BC to 800 AD.34 It also sets out to provide links to all the main projects, and has begun the process of assigning a unique number to each documentary text, an essential device for management of the proliferating information. There are good general introductions to the two fields, by B. Hudson McLean in epigraphy35 and Roger S. Bagnall in papyrology.36 For an introduction to Koine Greek, works by Robert Browning37 and Geoffrey Horrocks38 are the best on offer. Important reference grammars are those of Edwin Mayser39 and Francis T. Gignac,40 but vocabulary is treated only incidentally in these works. A lexicon or extended treatment of the Koine Greek vocabulary is non-existent, though MM has a useful collection of data, provided it is not regarded as the final word. As for the LXX, the lexicon of Muraoka does not offer documentary evidence as such, or undertake new research in the field, but simply uses the results of existing studies; but it is the essential guide to the meanings of words in the LXX, based on a fresh and complete lexical analysis. LSJ’s material on the LXX is patchy and often unreliable; but its coverage of documentary evidence, though limited, is more trustworthy, at least as a starting-point. The need for expertise has been mentioned. Of course this can be said of any subject, but work on documentary texts can be especially challenging for those coming to it from other disciplines. There is first the need for familiarity with a wide range of types of Greek, from Classical to late Koine, from learned to colloquial. This includes awareness of the phonetic and other changes in the later period; to approach Koine Greek as 32 33

34 35 36 37 38 39 40

and . See further McLean, Greek Epigraphy, 383–5; Bagnall, Handbook of Papyrology, xv– xvi.

McLean, Greek Epigraphy. See also Woodhead, Greek Inscriptions. Bagnall (ed.), Handbook of Papyrology. See also E. G. Turner, Greek Papyri. Browning, Greek, chapter 2. Horrocks, Greek. Mayser, Grammatik. Gignac, Grammar. Epigraphy lacks any equivalent to Gignac and Mayser.

THE VOCABULARY OF THE SEPTUAGINT AND DOCUMENTARY EVIDENCE

397

a sort of debased Classical Greek is a serious mistake. Even with this knowledge, to understand the technical and other terminology, or even just to grasp the purpose and background of a text, can be demanding. The word being studied will not necessarily yield its meaning quickly, and extensive enquiry may be needed. Some particular difficulties deserve notice. Incomplete texts invite restoration, but not every restoration found in the published editions is to be trusted. The researcher needs to be wary of accepting any word fully or even partially restored, unless it is part of a formula or other predictable expression.41 Another problem is ‘correction’ of originals by editors, that is to say, when the original presents a non-standard spelling or other feature, the editor offers the ‘correct’ form. Some of the corrections go beyond indicating the standard spelling and alter forms or syntax that actually belong to the Greek of the time.42 Finally, dating. This is of vital concern to the lexical enquirer, but there are many obstacles put in the way of ascertaining the date of a document quickly. These include total omission of ‘BC’ and ‘AD’ or equivalents, the use of a variety of dating formulae (some in Latin), and most frustrating of all, failure to offer any date whatever, even within wide parameters.43

5. ILLUSTRATIONS 5.1 ποτίζω Let us begin with a straightforward example of a new use in the LXX paralleled in contemporary Egyptian papyri.44 In the Classical period ποτίζω is well attested, but not in the meaning ‘irrigate’ (land); the word for this was ἄρδω. In the Pentateuch and occasionally in later books ποτίζω occurs in this sense, ἄρδω not at all.45 An example: 41

42

43

44 45

Cf. E. G. Turner, Greek Papyri, 67–70; McLean, Greek Epigraphy, 35–9. Acquaintance with the editorial sigla (square brackets, subscript dots, etc.) is also a requisite: see Bagnall, Handbook of Papyrology, 203 (P. Schubert); McLean, Greek Epigraphy, 27– 35. An example: in PCairZen 2.155 χερός is corrected to χειρός, but the former is a genuine form, not a misspelling. Some variations border on the ridiculous, e.g.: ‘sac’ = saeculum ante Christum, i.e., BC/BCE, but ‘ac’ = after Christ, i.e., AD/CE; but then ‘a.C.’ = ante Christum, i.e., BC/ BCE. Cf. Lee, Lexical Study, 118–22. ποτίζω ‘irrigate’: Gen 2:6, 10; 13:10; Deut 11:10; Sir 24:31; Joel 4:18.

398

COLLECTED ESSAYS ON THE GREEK BIBLE

Gen 2:6 πηγὴ δὲ ἀνέβαινεν ἐκ τῆς γῆς καὶ ἐπότιζεν πᾶν τὸ πρόσωπον τῆς γῆς.

The evidence of the Ptolemaic papyri shows that this use was standard in the time of the LXX translators. There are now thirty or so occurrences, of which one sample will suffice: PZenPestm Suppl. B p. 266.2, 3 (256 or 255 BC) Ζήνων Κλέωνι χαίρειν. τὸ ὕδωρ τὸ ἐ[ν τῆι διώρυγι οὐκ ἀνα]βέβη[κ]εν πλείω ἢ [πῆ]χυν | ὥστε μὴ δύνασθαι ἀπ᾽ αὐτῆς ποτίζε[σθαι τὴν γῆν. καλῶς ἂν ο]ὖν π[ο]ήσαις ἀνοίξας | τὰς θύρας, ἵνα ποτίζηται ἡ γῆ. Zenon to Kleon greetings. The water in [the canal has not] come up more than a cubit, so it is not possible for [the land] to be irrigated from that source. Please, then, open the sluice-gates, so that the land may be irrigated.

In Deut 11:10 the word appears in a special expression, which is worth looking into briefly: ὅταν σπείρωσιν τὸν σπόρον καὶ ποτίζωσιν τοῖς ποσὶν ὡσεὶ κῆπον λαχανείας (‘when they sow the seed and irrigate it with their feet like a vegetable garden’). The documentary evidence does not provide an exact parallel, but it comes tantalizingly close: there are examples of ποτίζω / ποτισμὸς ἀπὸ ποδός (‘irrigate/-ation by foot’) in papyri of the first and second centuries AD.46 This looks like the same practice, but what method of irrigation was it? The debate has been drawn-out, but there is a probable answer.47 The evidence of the papyri makes an important contribution to tracing the history of ποτίζω. Without it the literary record is not very informative: if we trusted LSJ this use appears to occur only in Lucian (II AD).48 The inscriptions also happen to offer only late occurrences of ποτίζω, none of them in the sense of ‘irrigate.’

46

47

48

The current tally of exx. is five: PSoter 1.32 (69 AD), 2.26 (71 AD); PFlor 3.369.7 (139/149 AD); PRyl 2.157.21 (135 AD); PMilVogl 7.308.97 (II AD); similarly Cadell, ‘Vocabulaire de l’irrigation,’ 114–5; this has grown from two in Lee, Lexical Study, 121–2. (SIFC 13.366 in LSJ Suppl. (1996) = PFlor 3.369.) See Lee, Lexical Study, 121–2; Dogniez and Harl, Le Deutéronome, ad loc.; Cadell, ‘Vocabulaire de l’irrigation,’ 110–7; Wevers, Deuteronomy, ad loc.; LSJ Suppl. (1996), s.v. πούς; Muraoka, Lexicon (2002, 2009), s.v. ποτίζω. Cadell examines all the possibilities and returns to the suggestion (in Lee, from Driver) that it involves use of the feet to control the flow of water on to the garden through channels in the earth, also known in modern parallels. Luc. Abd. 27 ‘etc.’ (LSJ, s.v. 3). The truth of course is that there are more, as, e.g., in Strabo the Geographer (I BC/I AD) at 17.1.3, 1.4. These were found by a search in the TLG.

THE VOCABULARY OF THE SEPTUAGINT AND DOCUMENTARY EVIDENCE

399

5.2 εὐίλατος This adjective (‘kindly disposed,’ ‘merciful’) appears twice in the LXX: 1 Esd 8:53 καὶ πάλιν ἐδεήθημεν τοῦ κυρίου ἡμῶν κατὰ ταῦτα καὶ εὐιλάτου ἐτύχομεν. And again we entreated our Lord in these matters and found him merciful. Ps 98:8 ὁ θεός, σὺ εὐίλατος ἐγίνου αὐτοῖς καὶ ἐκδικῶν ἐπὶ πάντα τὰ ἐπιτηδεύματα αὐτῶν. O God, you were merciful to them, even while taking vengeance on all their practices.

What attestation does εὐίλατος have outside the LXX? The results of a thorough search are remarkable. Aside from the LXX itself, it does not appear in literature until much later, namely, Patristic texts from the second century AD onwards (mostly quoting Ps 98:8) and the Byzantine lexicographers;49 its attestation before then is entirely documentary. And this evidence reveals what we would otherwise not have known, that εὐίλατος was in extensive use in the time of the LXX, and not just in Egypt. LSJ (1925–40) were able to cite six occurrences in documents; the tally today is almost five times as many. The bulk of the texts fall in the period from the third century BC to the first century AD; geographically they range from Egypt, through Asia Minor (notably Knidos in Caria), Attica, and Thrace, to Pompeii.50 A small selection must suffice: IKnidos 1.150.A.9 (Knidos, II–I BC) μὴ τύχοι Δάματρος καὶ | [Κ]όρας μηδὲ θεῶν τῶν παρὰ Δάματρος εὐιλάτων May he not find Demeter and Kore nor the gods with Demeter welldisposed.51 IEgVers 175.I.36 (Narmouthis, I BC?) κλῦθι ἐμῶν εὐχῶν, μεγαλοσθενὲς οὔνομ᾽ ἔχουσα | εὐείλατος ἐμοί τε γείνου, λύπης μ᾽ ἀνάπαυσον ἁπάσης. Hear my prayers, O mighty-named one [= Isis-Thermouthis] | and be merciful to me, give me rest from all pain. 49

50

51

Results from TLG search. An example attributed to Antimachus (V/IV BC) in Et. Mag. s.v. Ἀδόροις is to be read as εὐήλατος (Frag. 109, ed. Wyss). There is also a later example in a magical papyrus, PapGraecMag 2.166 (IV AD?): δέσποτα Μουσῶ[ν], ἵλαθί μοι, τῷ σῷ ἱκέτῃ, καὶ ἔσο εὐμενὴς καὶ εὐίλατος, φάνηθί μοι καθαρῷ τῷ προσώπῳ. [There is even an occurrence in Seneca!: Aitken, No Stone Unturned, 101 n. 46.] Doric α for η in the goddesses’ names illustrates the continuation of older dialect: Knidos was an old Doric foundation. There are about ten other similar texts from Knidos. [For παρά + gen., ‘attached to, with’ see Essay 26.]

400

COLLECTED ESSAYS ON THE GREEK BIBLE

IG II2 1365.25 (Attica: Sunium, I AD) = CMRDM 1.12.25 (n. d.) εὐίλατος γένοιτο ὁ θεὸς τοῖς | θεραπεύουσιν ἁπλῇ τῇ ψυχῇ. May the god [Men Tyrannos] be merciful to those who tend him with simplicity of soul. PSI 4.392.6 (242/1 BC) … τοῦ βασιλέως εὐειλάτου γενομένου καὶ ἐπιγράψαντος τὴν ἄφεσιν. … if the king is merciful and signs the release.

Something further is evident. This is a word with an elevated, religious tone, applied in the documentary texts only to gods or King Ptolemy (also a god). The Psalms and Esdras translators knew what they were doing when they chose it; even the expressions match.52 The LXX also has the verb εὐιλατεύω, derived from εὐίλατος, in Deut 29:19, Jdt 16:15, and Ps 102:3, all with the same religious tone. So far there is no other attestation, except in later Patristic texts and lexicographers. But it would be unwise to conclude that it was a neologism of the LXX translators; and if past experience is any guide, there is a good chance that it will turn up. 5.3 ἀγαπῶ A full study of this word would be a large subject. The point to be made here is that documentary evidence supports the view that ἀγαπῶ began to supplant φιλῶ as the word for ‘love’ already in the fourth century BC and continued on the same path later. This was demonstrated by Robert Joly from the literary evidence.53 It follows that the LXX translators used ἀγαπῶ (c.250 times) because it was current usage, not because they intended to reject ‘pagan’ usage and create a new vocabulary for ‘love’ words, passed on in due course to the NT. Documentary examples from the time of the LXX are very few, but they provide evidence that ἀγαπῶ ‘love’ was used not just in Egypt but in the furthest corners of the Greek-speaking world: OGI 1.90.9 (Rosetta, 196 BC) ἐπειδὴ βασιλεὺς Πτολεμαῖος αἰωνόβιος, ἠγαπημένος ὑπὸ τοῦ Φθᾶ, θεὸς Ἐπιφανὴς Εὐχάριστος … 52

53

Talshir (1 Esdras, on 8:53) missed an opportunity: she quotes just one example, taken from LSJ (without the reference, PPetr 2 p. 45). [See Aitken, No Stone Unturned, 98–102 for a full treatment of this word, with interesting results, including that it appears to be a typically Egyptian term and may have an Egyptian resonance even outside Egypt.] Joly, Le vocabulaire de l’amour. A valuable exposition is found in Swinn, ‘Ἀγαπᾶν in the Septuagint.’

THE VOCABULARY OF THE SEPTUAGINT AND DOCUMENTARY EVIDENCE

401

Whereas King Ptolemy, living eternally, loved by Phtha, god Manifest, Gracious …54 IGExtrEast 292.9 (Kandahar, mid III BC)55 ἔδει … διδάσκαλον καὶ πατέρα καὶ μητέρα | ἐπαισχύνεσθαι καὶ θαυμάζειν, φίλους καὶ ἑταίρους ἀγαπᾶν καὶ μὴ διαψεύδεσθαι, | δούλοις καὶ μισθωτοῖς ὡς κουφότατα χρᾶσθαι. They were to … show modesty and respect to their teacher and father and mother, to love their friends and companions and not deceive them, to treat slaves and hired servants as gently as possible.

6. FUTURE DIRECTIONS The time of establishing the relevance of the documentary evidence to the language of the LXX is past: it has been well and truly proven. Numerous studies over the past century have made forays into this resource and produced good findings. But a thorough, systematic exploration is still needed, one that is ongoing, as the material continues to grow and become even more accessible via electronic tools. For vocabulary we now have a basic tool of investigation, namely, a full lexical analysis of the LXX, provided by Muraoka’s lexicon. The next step is systematic searching of the documents for parallels (or not) to the whole gamut of usage of the LXX. We are also in great need of a record of the published discussions on LXX words, kept up to date as they appear. At present there is no reliable way of keeping track, and each researcher begins the quest all over again. A way of gathering the data into one accessible resource will be essential. In the past this might have been attempted by a printed lexicon, as it was for example by Bauer/BDAG for the NT, but the drawbacks of that type of resource are evident. The obvious tool now is an electronic database, able to gather all the material and keep it up to date. This is the new desideratum of LXX studies. In the long run, benefits to all scholarship on the Greek vocabulary will ensue from such a project. Not only will the vocabulary of the LXX be seen more fully in its context, but its own witness to the language of its time will be more evident.56 54

55

56

There are several more instances of this formula in the Rosetta stone and copies. In PCairZen 4.580.3 (mid III BC) ἀγαπῶ occurs without context. The abbreviation problem illustrated. The volume is F. Canali De Rossi (ed.), Iscrizioni dello estremo oriente greco (IK 65; Bonn, 2004). Abbreviation in PHI Greek Inscriptions: ‘IK Estremo oriente’; in CLAROS: ‘IEOG.’ I have suggested ‘IGExtrEast’ on the pattern of the Horsley-Lee list. My thanks to Trevor Evans for useful comments on this chapter.

25 THE PUZZLE OF JOHN 21:15–17: A FORMALITY SOLUTION 2017

Abstract The alternation of the words for ‘love’ in John 21:15–17 is a long-standing puzzle that has defied solution. It is suggested in this note that ἀγαπῶ and φιλῶ carry a slight difference of ‘formality’ as a result of their history. This accounts for the interplay and makes it meaningful compared with simple variation.

Many attempts have been made to explain the mysterious alternation of the two verbs for ‘love,’ ἀγαπῶ and φιλῶ, in John 21:15–17. This short note will not rehearse them but move as speedily as possible to a suggestion that as far as I know has not been made before. In the famous series of questions addressed by Jesus to ‘Simon son of John,’ Jesus twice asks, ‘Do you love me?’, ἀγαπᾷς με; to which Simon replies both times, ‘You know that I love you,’ φιλῶ σε, using a different verb. Jesus asks the question a third time, using now the same verb as Simon, φιλεῖς με; to which Simon replies as before, ‘You know that I love you,’ φιλῶ σε. So Jesus uses ἀγαπῶ twice, then switches to Simon’s φιλῶ, while Simon persists with φιλῶ throughout the exchange. Attempts to find a difference of meaning between the two verbs have proved futile, and rightly so. It is not possible to show that each verb indicates a different type of love, or that such a difference can make the exchange more meaningful.1 Barrett declared it ‘impossible to doubt that they are synonyms.’2 The best alternative explanation, in fact the only one currently available, is that this is a case of simple variation between words of similar meaning, a phenomenon seen elsewhere in John, indeed in this very 1

2

For survey and discussion see Morris, John, 871–3; TDNT, 9:134–5 (Stählin); Brown, John, 2:1102–3. I am grateful to Trevor Evans for reading this paper and offering helpful comments. Barrett, John, 486.

404

COLLECTED ESSAYS ON THE GREEK BIBLE

passage.3 In that case the words have the same lexical meaning, that is, approximately love. Lexical meaning as such, however, is not all that may be involved; there is also ‘connotation’ or ‘tone,’ that is, the additional semantic elements that lie outside lexical meaning and attach to many if not most words. If these two words do not differ in lexical meaning, might they differ in ‘tone’? To find the answer we need to look into their history. The two verbs ἀγαπῶ and φιλῶ were in competition in the Classical period, and in the fourth century BC ἀγαπῶ began to supplant φιλῶ as the word for ‘love.’ This has been demonstrated by Robert Joly from the evidence of Classical and later literature.4 The verb φιλῶ, while retaining its meaning ‘love,’ had already developed two additional meanings, namely, ‘be accustomed (to)’ (VI–V BC +) and ‘kiss’ (V BC +). These competed with the meaning ‘love’ and were maintained through its later history. ‘Kiss’ prevails as the usual modern meaning. In addition, presumably as a development from its meaning ‘kiss,’ φιλῶ shows signs of occasional use as a euphemism for ‘have sexual intercourse with.’5 Joly did not notice this but he came close: he thought that the meaning ‘kiss’ killed (‘a tué’) the meaning ‘love.’6 The sexual meaning would have been even more deadly – the kiss of death, one might say.7 These changes were enough to ensure that a new word for ‘love’ would make its way into the vocabulary. φιλῶ did continue to be used with its old meaning ‘love,’ but not entirely on the same footing as ἀγαπῶ. A highly significant fact is that φιλῶ, while frequent in documentary papyri, occurs mostly in formulae, which are of course a standard location of old-fashioned, formal language.8 Another 3

4

5

6 7

8

Argued with many exx. in Morris, Studies, 293–319. In this passage: βόσκε τὰ ἀρνία μου (15), ποίμαινε τὰ πρόβατά μου (16), βόσκε τὰ πρόβατά μου (17); οἶδας, γινώσκεις (17). Joly, Vocabulaire de l’amour, 10–29. See also Swinn, ‘ἀγαπᾶν in the Septuagint’; Silva, Biblical Words, 96–7. The evidence is limited but persuasive: Hesych. βαίνειν· φιλεῖν (noted in LSJ, s.v. φιλῶ I.3.b ‘of sexual intercourse’); Schol. Theocr. 4.58 μύλλει· ἀντὶ τοῦ φιλεῖ; Schol. Ar., Plutus 1093 καὶ νῦν φιλῶ σημαίνει διακορεύω ἢ βινῶ. See Shipp, MGE, 126–7 with n. 72 for these examples and discussion. Continuation into vernac. mod. Greek is indicated by the saying quoted in Lex. Proia, s.v. φιλῶ: [ὅπου βρίσκει καὶ φιλεῖ, μοῦντζές του κι᾽ ἂν παντρευτῇ, ‘wherever a man finds and makes love (to a woman), mountzes (a gesture of contempt) to him, even if he marries her.’ I am grateful to Tasos Kalogerakis and John Fountotos for help with the translation.] Joly, Le vocabulaire de l’amour, 31. French baiser shows parallel development. The earliest Greek word for ‘kiss,’ κυνῶ, had suffered a similar fate: see Joly, ibid., 33. καταφιλῶ ‘kiss’ was an early addition to the mix (Xenophon +), no doubt to disambiguate φιλῶ. Joly, ibid., 27–9. Most are in the epistolary closing formula ἀσπάζομαι / ἀσπάζου τοὺς φιλοῦντας ἡμᾶς / σε, attested from I AD onwards. Examples: POxy 17.2148.12 (27 AD); OClaud 1.146.7 (100–120 AD); PMich 3.221.19 (296 AD). Also formulaic is the cliché

THE PUZZLE OF JOHN 21:15–17: A FORMALITY SOLUTION

405

observation is that φιλῶ often appears with μισῶ, in a more or less fixed pair.9 Further data pointing in the same direction could be added. This evidence leads to the conclusion that in early Koine Greek, and certainly by the time of the New Testament, φιλῶ was obsolescent in the meaning ‘love.’ The standard word for the idea was ἀγαπῶ. Obsolescent words tend to carry greater formality. Thus we can say that φιλῶ was the slightly more formal word, while ἀγαπῶ had no extra connotation of that kind, but was ‘unmarked’ for formality. It should be emphasised that the difference, though noticeable to a native speaker, would be slight. English offers no parallel in words for ‘love,’ but we might illustrate by ‘Do you wish to go?’ and ‘Do you want to go?’: wish is slightly more formal than want.10 In the conversation in John 21, Simon Peter is being formal and polite. He draws back from using Jesus’s plain word, which might imply a certain familiarity. This does not mean that he loves any less, only that he feels unable to express it so directly. When Jesus the third time uses the same word as Simon, he accedes, as it were, to Simon’s level of formality, but without changing his meaning. Simon persists with his chosen politeness in his final answer. How did this feature come to be part of the conversation now recorded in John’s Gospel? We can only guess, but some things are more likely than others. It is only possible in the Greek form of the exchange. If the conversation was translated from Aramaic, someone introduced it, either at the point of translation or later. Introduction at the point of translation is unlikely, unless Aramaic offers a similar pair of variables.11 If the conversation took place originally in Greek, with the alternation of words as it is now, it is unlikely that it was noted and preserved exactly. More likely the compiler of this Gospel, John, introduced the subtle detail of Peter’s choice of word to portray Peter’s respectful attitude to Jesus.

9 10 11

οἶδα πῶς / ὅτι με φιλεῖς, in PCol 10.279.13 (III AD) etc. In addition, Joly (29) notes some ‘formules religieuses’ with φιλῶ. Joly, ibid., 11, 16, 18, 30. The pairing is found in John 12:25; 15:19. On formality cf. Lee, ‘Some Features,’ esp. 7, 10 [= Essay 5]. Brown (John, 2:1103) notes that it does not (nor does Hebrew).

26 THE BRILL DICTIONARY OF ANCIENT GREEK: NEW CLOTHES FOR AN OLD EMPEROR 2017

Abstract This review article offers a critical assessment of the new Brill dictionary. Several important aspects of the work are considered and illustrated by examples: the provenance of the dictionary, going back through Italian predecessors to LSJ; the structure of verb entries; the lexical analysis; the coverage of documentary evidence; the coverage of the LXX and NT; and the treatment of a sample particle. In the conclusions an overall assessment of GE is given, especially in comparison with LSJ, and remarks are offered on the state of lexicography of Ancient Greek and its future direction.

1. PREAMBLE The publication of a new ‘major’ lexicon of Ancient Greek is a significant event.1 The new Brill dictionary (‘GE’)2 is of the same size and scope as the lexicon of Liddell and Scott, the current standard work. Moreover, the new lexicon has been published complete in one volume, unlike the other major lexicon project of our time, DGE, still progressing through publication in parts.3 Though GE is based, it is true, on a previous lexicon, the completion of the project in just four years is a prodigious achievement. The co-ordination of the team that produced this result – there are at least 15 names mentioned in the preface – was a major undertaking in itself. Similarly, the Italian predecessor (‘GI’) involved at least 60 collaborators under the leadership of Franco Montanari.4 1

2 3 4

This review is a much-expanded version of my presentation in a panel on the new lexicon at the SBL Annual Meeting, San Antonio, 2016. My thanks to Michael Curran for reading it and offering helpful comments. Montanari, et al. (eds.), The Brill Dictionary of Ancient Greek (2015). Adrados, et al. (eds.), Diccionario Griego-Español (7 vols. to 2009). Montanari, (ed.), Vocabolario della Lingua Greca (2013). For the figures mentioned, see GE, p. vii; GI, pp. i–ii.

408

COLLECTED ESSAYS ON THE GREEK BIBLE

The character and merits of the new lexicon are indicated in the preface to GE by the three leaders of the team that produced it.5 Though many statements are made and most are clear, some reading between the lines is required to understand the presuppositions of the editors and what this lexicon attempts to achieve. The editors naturally draw attention to the improvements (new evidence; new lemmata; corrections to GI; inclusion of proper names), but they also appear to recognise that GE is not, and could not possibly be, the equivalent of a completely revised LSJ (cf. below, §8). They do claim that ‘this Greek-English project is presented as an enhancement of lexicography,’ and that the dictionary ‘presents a critical approach to lexicography in and of itself,’ statements that are difficult to interpret, if not meaningless. Two further assertions will be commented on in the next section. It is of course impossible to assess the entire contents of a massive reference tool such as a lexicon: any review necessarily involves sampling. My samples and discussion will have in view three main themes or aspects: (a) the provenance of GE; (b) GE as compared with LSJ; and (c) the current state of the lexicography of Ancient Greek. The samples are presented under various headings, but all will be found to illustrate these three topics.

2. PROVENANCE OF GE Dependence on predecessors is a general phenomenon of lexicography. Rarely has a lexicon been created afresh, that is, on the basis of a new collection of data subjected to a new lexical analysis. Only if the compilers are prepared to spend up to a century can it be done. The Oxford English Dictionary was created in this way (70 years); so was the Oxford Latin Dictionary (50) and the recently completed Dictionary of Medieval Latin from British Sources (100). LSJ was not; neither was GE. Both descend from a line of predecessors, which in GE’s case include LSJ itself. The provenance of GE may be summarised as follows: GE 2015 < GI (1st ed. 1995; 2nd ed. 2004); 3rd ed. 2013 < Rocci (1st ed. 1939; 2nd ed. 1941); 3rd ed. 1943 < LSJ 9th ed. in 10 parts, 1925–1940; + Bailly 1894 < LS 8th ed. 1897 … 5

G. Nagy, L. Muellner, M. Goh; see GE, p. vii.

THE BRILL DICTIONARY OF ANCIENT GREEK

< < <
with acc. IL. 20.210 OD. 1.363 … || to call by crying: κλάειν μάμμας καὶ τιτθάς to cry for mummies and nurses ARR. … 2 mid. A to weep, lament IL. 18.340 … B to suffer, undergo punishment or harm, usually fut. ARISTOPH. Ve. 1327 etc.; … C to weep for, bewail > with acc. … 3 pass. to be bewailed AESCHL. Ch. 687 … | impers. μάτην ἐμοὶ κεκλαύσεται … ARISTOPH. Nub. 1436. … LSJ: κλαίω … I. intr., cry, wail, lament, of any loud expression of pain or sorrow, … Od. 10.201 … 2. αὐτὸν κλαίοντα ἀφήσω I shall send him home crying, howling, i.e. well beaten, Il.2.263: freq. in Att., κλαύσεται he shall howl, i.e. he shall suffer for it, Ar. V.1327 … II. trans., weep for, lament … Od.1.363 … : – Pass., to be mourned or lamented …: impers., μάτην ἐμοὶ κεκλαύσεται … Ar.Nu.1436. 2. cry for, of infants, μάμμας καὶ τιτθάς Arr. … III. Med., bewail oneself, weep aloud, A.Th. 920 … 2. trans. bewail to oneself. … S.Tr. …

The general layout in GE is very similar to LSJ’s, with the usual tumble of glosses, divided by a numbering system that requires concentration. GE gains in clarity by using bold for the meanings; but its numbering system is harder to pick up and somewhat counter-intuitive. The letters A, B, etc. (enclosed in boxes) are subordinate to the numbers 1, 2, etc. (enclosed in black blobs). While the blobs catch the eye, the numbers in them are very small; the boxed letters likewise are quite hard to spot. In addition, there are subdivisions introduced by double and single bars (||, |).16 The explanation of these given in the endpapers is clear with regard to marking tenses and moods in the ‘principal parts section’ of the entry (as in ‘|| fut. κλαιήσω’ above), but is much less clear about why some senses are introduced in this way: the unsatisfying explanation given is that they are ‘further subdivisions of the entries.’ In the case of κλαίω above, it looks rather as if ‘|| to call by crying’ ought to have been another numbered sense. These are somewhat minor matters. More importantly, we see in κλαίω a feature carried through systematically in all verb entries, a division into active, middle, and passive voices (marked by numbers 1, 2, 3). This is 16

There are further subdivisions using a solid arrow, dot, and diamond (not exemplified above): see the endpapers of GE for their explanation.

THE BRILL DICTIONARY OF ANCIENT GREEK

413

a significant innovation, introduced by Montanari in GI. While it may be helpful to the beginner, and even to some advanced scholars, it may not be a wise one from the point of view of lexical analysis. It has the effect of changing the upper level of the analysis from semantic to morphological; that is, lexical meaning is subordinated to the active–middle–passive distinction. The result is that in verbs (though only verbs) the primary divisions into 1, 2, and 3 do not indicate different senses but different voices, and consequently the same meanings may appear, indeed must often appear, twice or even three times in the entry. So in κλαίω, the meanings in 1 act. A, B, C are the same as those in 2 mid. A, B, C (‘weep,’ ‘suffer,’ ‘weep for’); and sense 3 pass. gives a meaning which is just the passive of 1C and 2C (‘weep for’). This innovation can also lead to trouble of a different kind in the many Greek verbs in which form and meaning do not necessarily match.17 Examples: (a) λαμβάνω is normally middle in form in the future (λήψομαι), but has an ‘active’ meaning; so GE has to present this as ‘1 act. (with fut. mid.).’ (b) The usual aorist of βούλομαι is passive in form but not in meaning; GE gives up the division by form for this verb and simply says, in the principal parts section, ‘aor. only pass. ἐβουλήθην or ἠβ-.’ (c) The verb βιάζομαι has the annoying characteristic of being usable in the present middle form in both an ‘active’ and a passive sense; GE’s solution is ‘1 act. … 2 usu. mid. … 3 pass.’; but this means that 2 mixes form and meaning. As for verbs that have suppletives with different voices, major difficulties are unavoidable. GE in some cases simply gives up the division by voice, as in ἔρχομαι and its aorist ἦλθον, which are wrapped together; in others various compromises are required.18 In ἐσθίω, all these problems come home to roost. The entry is simply divided into ‘1 act. … 2 pass.’ Under the first, the active ἔδω is mentioned, but we are not told that the usual future of ἐσθίω is actually the middle ἔδομαι (in Class., but later φάγομαι); the standard aorist ἔφαγον, formed from a different root, is also not mentioned: one has to know that there is a separate entry for ἔφαγον, where its relation to ἐσθίω is noted; and although coverage 17

18

The extent of these variations (just for Classical Greek) can be appreciated from the lists in Smyth, Grammar, §§805–818. In the case of πίπτω, πεσοῦμαι is noted as the fut. form, but the mid. category does not appear. In ὁράω we find ‘1 act. and mid.’ combined, then 2 pass. In ἐρωτάω we have only 1 act., and 2 pass. but no middle; the usual (Attic) aor. to ἐρωτάω, the middle ἠρόμην, is not mentioned (it is found under εἴρομαι). Similarly in πωλέω: no mention of the standard aor. ἀπεδόμην.

414

COLLECTED ESSAYS ON THE GREEK BIBLE

of later Greek is said to be a feature of this lexicon, nothing at all is said about τρώγω, which came in as the present to replace ἐσθίω (cf. below). One would not want to suggest that LSJ has solved all these problems, but its presentation of the voices and suppletives is generally much clearer than GE’s, and its entry for ἐσθίω covers everything with great efficiency. All in all, as regards layout and convenience for the user, there is not much to choose between GE and LSJ. Both present us with a packed entry offering a range of possible meanings, arranged in sense-divisions that may convey little to most readers. The selection of examples is overloaded and mostly from Classical authors. We have become used to finding our way through this in LSJ; we will have to do the same with GE. How are such entries actually used? I suspect that most readers simply pluck out from the offerings a gloss that seems to fit the passage they are reading; whether it is the right one is often a matter of chance (unless they can locate a citation of the passage). Let us follow up τρώγω, just mentioned in relation to ἐσθίω, and see how it has fared: 3.2. τρώγω GE: τρώγω … 1 act. to gnaw on, nibble at, of animals: τ. ἄγρωστιν to nibble on weeds OD. 6.90; … | of sick people … || to eat (raw), chew on, of pers., esp. vegetables and sweets … || later simpl. to eat (= ἐσθίω) … 2 pass. to be eaten: τρώγεται ἁπαλὰ ταῦτα καὶ αὖα the fresh and dry are eaten HDT. 2.92.4; … LSJ: τρώγω … [I.] gnaw, nibble, munch, esp. of herbivorous animals, as mules, τ. ἄγρωστιν Od.6.90; of swine … of cattle … of human beings in disease … II. of men, eat vegetables or fruit … of dessert, eat fruits, as figs, almonds, etc. … of small fish as horsd’oeuvres … III. later, simply eat, serving as pres. to ἔφαγον instead of ἐσθίω …

The basic idea of this verb is ‘gnaw, nibble,’ as stated, an action typical of animals but readily extended to humans. But it is rather the kind of eating, not who is eating or what is being eaten, that characterises τρώγω and distinguishes it from ἐσθίω, as Chadwick pointed out.19 LSJ’s division into I., applied to animals and II., applied to humans (with leakage in ‘of human beings …’ in I.) is not necessary and the two could have been collapsed into one lexical meaning. Later on, from about the 19

Chadwick, Lexicographica Graeca, 288; discussion of τρώγω, 287–90. Chadwick says the human/animal distinction was made in antiquity.

THE BRILL DICTIONARY OF ANCIENT GREEK

415

first century AD, by the same process by which Lat. manducare (‘chew’) developed into Fr. manger (‘eat’), this verb came to mean simply ‘eat,’ making it equivalent to ἐσθίω, as LSJ correctly report. In GE the distinction between animals and humans is kept, with the meaning stated more or less satisfactorily (though ‘to eat (raw)’ is not a good idea); but the insertion of the later meaning ‘eat’ into this section, with only || to introduce it, underplays its significance as a distinct development in meaning, which leads in the long run to the Modern Greek word for ‘eat.’ Then when the active–passive distinction is imposed, it creates problems. It looks to the user as if the passive voice is only used with the (sub-) meaning ‘eat,’ a meaning said to be ‘later’ (in 1.), yet cited from Herodotus (in 2.). In fact this example is the passive of ‘gnaw, nibble, chew’ not ‘eat’: Hdt. 2.92.4 τρώγεται δὲ καὶ ἁπαλὰ ταῦτα καὶ αὖα = ‘and these [edible kernels] are nibbled both fresh and dried.’20 In short, GE’s entry, though ultimately derived from LSJ’s, has lost more than it has gained in the process of revision and translation.

4. LEXICAL ANALYSIS The core of the lexicographer’s task is to state the meanings of words, that is, to analyse the available data for each word into its ‘lexical meanings,’ with suitable definitions. This is a very challenging exercise and good results are not guaranteed; they are usually worse when an entry has passed through many hands in its history. The samples in the previous section have already illustrated problems in GE (especially τρώγω), and others will be found in later sections (notably ὀψάριον, §5). Here I explore a typical noun frequently attested and with a range of meanings. βίος Just as in the verb κλαίω, both GE and LSJ present crowded entries for βίος, with divisions into an array of senses without paragraphing. GE has five major divisions (A–E), and within these, nine further subgroups marked by ||. LSJ’s major sense-divisions run from I to VIII, with subdivisions 1, 2, 3 in the first. Let us summarise and compare the main features of these two entries, with a look back at their predecessors. 20

GE’s translation, ‘the fresh and dry are eaten,’ is not accurate; nor does it appear to be an accurate translation of GI: ‘questi vengono mangiati freschi e secchi.’ For ταῦτα = τρωκτά ‘edible kernels’ see Waddell, Herodotus Book II, 205.

416

COLLECTED ESSAYS ON THE GREEK BIBLE

LSJ begin with a ham-fisted attempt at distinguishing βίος from ζωή: ‘life, i.e. not animal life (ζωή), but mode of life …, manner of living (mostly therefore of men, v. Ammon. …; but also of animals …’. This is difficult enough to grasp; a look at ζωή only makes it harder: there is nothing remotely like ‘animal life’ to be seen there. The distinction is based on a statement in Ammonius, and was added in LS7 (1890).21 Offering a distinction is not a bad thing in itself, but relying on an ancient Atticist grammarian is ill-advised. Ammonius is doing what he does, that is, separating two apparent synonyms (in Classical Greek), by some simple and rigid distinction, in this case that βίος is used only of humans, ζωή of humans and beasts. A more reliable guide to the distinction may be found in Chantraine’s DELG, which in a nutshell is: βίος = life as lived, ζωή = life itself.22 Differences in application or reference are incidental. This opening statement in LSJ makes the whole section I.1 more difficult to follow. The rest of LSJ’s entry, however, is more or less satisfactory and clear. Before turning to GE, it will be useful to note the content of Rocci’s entry. Rocci does not mention or try to apply the human-animal distinction. His division into five meanings a) – e) is simple but good. (One can see clearly the influence of Bailly.) GI’s entry, however, shows enough differences from Rocci’s to indicate that extensive revision was undertaken for GI. Comparison with LSJ shows that LSJ was the main source, not a new analysis. Most of the senses in LSJ have been taken over, mostly in the same places, sometimes rearranged. In GI and hence GE, the attempt to distinguish βίος and ζωή is abandoned (and Ammonius is not quoted); yet LSJ has left its mark in the survival of the human-animal distinction, now set up as two separate senses, with LSJ’s examples relocated: ‘A life, existence, of a human … B rar. life, of other beings.’23 Some other odd things occur and the use of numbered sense-divisions versus subdivision by || is erratic. In particular there is a slip-up at ‘D later a life, biography … || the world, mankind, humanity, people (the surroundings in which one lives),’ which presents the latter as if it were a sub-sense of the former. LSJ rightly separated these on an equal level, as ‘III. the world we live in, ‘the world’’ and ‘V. a life, biography.’ Overall, then, the results 21

22

23

Ammonius 101 (ed. Nickau): βίος μὲν ἐπὶ τῶν λογικῶν τάσσεται ζῴων, τουτέστιν ἀνθρώπων μόνων, ζωὴ δὲ καὶ ἐπὶ ἀνθρώπων καὶ ἐπὶ τῶν ἀλόγων ζῴων. LSJ’s quotation from Plotinus 3.7.11 (III AD) is not a lot of help. Chantraine, DELG, s.v. βίος: ‘Sens, non le fait de vivre, mais la manière de vivre … d’où « moyens de vivre, ressources »’; s.v. ζωή: ‘« propriété d’être vivant, vie » par opposition à « mort ».’ There is a fair bit of overlap in actual usage. A hare that was in LSJ’s I.1 has managed to slip through the net and stay in GE’s ‘A … of a human’ (‘DEMOSTH. 18.263’).

THE BRILL DICTIONARY OF ANCIENT GREEK

417

of the revision in GI and GE are mixed, and it cannot be said that the entry is an advance on LSJ. A close look at the entries for βίος in GE and GI reveals something further: they are identical in content. The only changes are the translations from Italian into English and minor alterations in reference style (e.g., LUC. 51.25 > LUC. Pseudol. 25). It is a fair guess that the editors of GE were able to work from the electronic files of GI, making only minimum changes. There was no need to retype the Greek or the references; only the translations needed to be done, addenda inserted, if any, and some information relocated.24 The consequence of this process was of course that little or no revision to the structure of the entry in GI could be, or was made in GE.25 5. DOCUMENTARY

EVIDENCE

The main purpose of lexicons is to supply meanings, but we also consult them for other information, especially attestation. Innumerable discussions in the scholarly literature offer statements like ‘first found in …,’ ‘common in …’. On what are they based? On a lexicon usually, and usually LSJ. Lexicons thus have an important role to play in this regard. Yet it cannot be said that the present state of play is satisfactory. In the LSJ tradition the attestation or evidence for each word has accumulated unsystematically over centuries. It is primarily employed, as is right, to establish and illustrate the meanings, with the additional aim of supporting the reading of standard texts. But there is no guarantee that the picture given is a reliable guide to the spread and currency of the word. Until there is a systematic gathering of evidence, with some sort of statistical information, this situation will continue. We cannot expect GE to have solved this problem, but we would expect at least to see no loss of existing data, and the addition of new data where significant. Let us take as a test documentary evidence, that is, inscriptions and papyri, of which it is said GI/GE ‘makes substantial use.’26 24

25

26

In verbs, some morphological material is moved from the end of GI’s entry to the ‘principal parts’ section after the headword in GE, leaving only ‘dialectal, poetic … forms’ at the end. For further samples of GE’s handling of lexical analysis, the reader might like to compare ἰσχυρός with Chadwick, Lexicographica Graeca, 165–70, and ἕξις with Lee, ‘Hebrews 5:14’ (summary 165–6). GE, p. vi; see also p. vii: ‘In addition to the updated language of our definitions, the strengths of this volume include the incorporation of new evidence, especially from epigraphical sources and papyri.’

418

COLLECTED ESSAYS ON THE GREEK BIBLE

5.1. εὐίλατος GE: εὐίλατος -ον [εὖ, ἱλάομαι] kind, favorable, merciful with dat. VT Ps. 98.8 etc. GI: εὐίλατος -ον [εὖ, ἱλάομαι] benigno, propizio, misericordioso, con dat. VT. Ps. 98.8 ecc. Rocci: εὐ-ίλατος, ον [ῑ, ἱλάομαι] propizio; pietoso, I.; SET. LSJ: εὐίλατ-ος [ῑ], ον, (ἵλημι) very merciful, of deities, PCair.Zen. 34.19 (iii B.C), IG … GDI … etc.; … Lxx Ps. …; also … PPetr. …; later written … GDI… UPZ 109.6 (i B.C.).

The documentary evidence for this word, mainly in inscriptions, is extensive and important, showing as it does that the LXX was using a current term and how it was used.27 LSJ had quite a good collection. What has happened to it in GI/GE? It has disappeared. Rocci at least had ‘I.’ (= ‘Iscrizioni’) to alert the reader; GI let it go. Hence in GE it looks as if εὐίλατος is a word used only in the LXX. 5.2. ἀποσκευή GE: ἀποσκευή … A baggage … || household goods VT Gen. 14.12, al. … | all members of the house VT Exod. 10.24 | the whole people, aside from the adult males VT Exod. 12.37 | in the army attendants VT Iud. 7.2 || excrement, filth STRAB. … B suppression, elimination IOS. A.I. 18.2.4(41). LSJ Suppl. (1996): … add ‘2 soldier’s encumbrances, i.e. family, PBaden 48.9 (ii BC), UPZ 110.199 (ii BC); dependants, Lxx Ge. 46.5, al.’

GE’s analysis is very odd to begin with; but the main point is that no documentary evidence is mentioned, even though it is important for the history of the word and indispensible to determining the meaning in the LXX. LSJ Suppl. (1996) supplied some, enough to show the new meaning family, dependants that appears in Ptolemaic Greek.28 Not only ought some of this evidence to have been reported in GE but even a brief look at it would have greatly improved GE’s account of the LXX uses.29 27 28 29

See Aitken, No Stone Unturned, 98–102. See further Lee, ‘Ἀποσκευή in the Septuagint’ [= Essay 2]. In Exod 10:24 ‘all members of the house’ is inexact: ‘family, dependants’ is more like it. In Exod 12:37 what GE says is nonsense: the text is οἱ ἄνδρες πλὴν τῆς ἀποσκευῆς, i.e., ‘the men apart from their dependants.’ In Jdt 7:2 it looks like ‘baggage train.’

THE BRILL DICTIONARY OF ANCIENT GREEK

419

5.3. ὀψάριον GE: ὀψάριον -ου, τό [ὄψον] course of a meal, a little dish, esp. of fish ARISTOPH. fr. 45 PLAT.1 102.2 PHERECR. 32 etc.; ἔχει πέντε ἄρτους κριθίνους καὶ δύο ὀψάρια he has five barley loaves and two fishes NT John 6.9. LSJ: ὀψ-άριον, τό, Dim. of ὄψον, Ar.Fr.45, Pl.Com.95, Pherecr.27, Philem.32, Test.Epict.6.11, PPetr. 3 p.327 (iii B.C.), PCair. Zen.440.3 (iii B.C.), etc.; … a jar of pickled fish, BGU1095.17 (i A.D.), cf. PRyl.229.21 (i A.D.), Ev.Jo.6.9, al., OGI484.12 (Pergam., ii A.D.). LSJ Suppl. (1996): ὀψάριον, after ‘dim. of ὄψον’ insert ‘foodstuff, esp. fish’ and add … SEG 26.382 (Athens)’. Rocci: ὀψάριον, ου, τό, dim. di ὄψον, propr. piccolo companatico, ma, in Atene, spec. pesciolino; pesce, vari COM. in AT. 385; P.; δύο ὀψάρια,VOLG. duos pisces NT. Io. 6, 9.

As well as a complete lack of documentary evidence in GE, we find a very unsatisfactory statement of meaning. LSJ had good data to work with (added in LSJ), but committed the sin of not telling us the meaning, instead relying on ‘Dim. of ὄψον’ to do the job. This is worse than useless: the user is given the impression that ὀψάριον is a small ὄψον – whatever that is – and turns to ὄψον only to see a range of meanings, any of which could be relevant. LSJ Suppl. (1996) tried to remedy this but didn’t do it well: ‘foodstuff’ covers all food; ‘esp. fish’ vacillates. GE has made it worse.30 Rocci was good within his limits. What is needed, instead of guesses based on a preceding lexicon, is a fresh assessment of all the data and a clear statement of the meaning so far as it can be known. 5.4. Further examples The three examples above barely touch on the amount of under-reporting of documentary evidence in GE. In regard to the papyri alone, the extent of it is dumbfounding given the claims made. Some are standard and frequent technical terms, such as ἔντευξις (‘petition’). Here are 50 wellattested words in papyri that lack any reference to that evidence in GE: ἀμάω, ἄμφοδον, ἀξίνη, ἀξιόχρεως, βῆμα, διαθήκη, διάπρασις, διαπράσσω, διαστέλλω, διάστημα, διαστολή, διαστολικόν (no entry), διασῴζω, διαταγή, διάταγμα, διάταξις, διατάσσω, ἐγγύη, ἐγγυητής, ἔγγυος, ἐλαιών, ἔντευξις, ἐντυγχάνω, ἐξακολουθέω, ἐπικεφάλαιον, ζωή, ζῷον, ἡμίονος, κάμινος, κοινωνός, κόκκος, λογίζομαι, μέρος, 30

GE’s ‘course of a meal, a little dish’ translates piatto di pietanza, un po’ di pietanza in GI.

420

COLLECTED ESSAYS ON THE GREEK BIBLE

μεταβολή, μοναστήριον, ναύκληρος, νομός, ὀφείλημα, παράδεισος, παραθήκη, πράκτωρ, πρόγραμμα, προσευχή, προστάσσω, πυρός, ῥόα, σιτομέτρης, ὑποζύγιον, φερνή, χωρίον.

Of course this simply reflects the reality, that to try to graft onto a lexicon of Classical Greek adequate coverage of even a part of the vast body of later Greek is impossible without a giant effort.

6. NT AND LXX Coverage of the NT and LXX in GE is difficult to assess fully (and fairly). It is evident that references to these texts are incorporated frequently throughout GI/GE (as they were in Rocci’s Vocabolario).31 Moreover, much of the NT/LXX vocabulary is no different from that of other ancient texts, so most words and meanings are as it were automatically covered. Even so, GE comes nowhere near providing the in-depth treatment available in the specialist lexicons,32 and in some cases there are significant gaps. This is the natural outcome of a lexicon tradition in which the NT and LXX were treated as an adjunct to a lexicon of Classical Greek. The coverage was never really satisfactory in LSJ or any of its predecessors, even though these texts received some notice from the first edition onwards. GE’s material has come into existence in a similar way and the results are similar. 6.1. θυμός The old word θυμός has a range of meanings rather difficult to analyse, but one, ‘anger,’ begins to predominate in Classical Greek and continues into Modern Greek.33 In the LXX there are 300+ examples of this sense. A careful search in GE’s long and packed entry finds one citation: ‘2Kgs. 5.12,’ under ‘D. … || extens.’.34 Why this one was chosen (in GI) is somewhat mystifying when so many others, e.g., from the Pentateuch and Psalms, were available. (The reference is not taken from Rocci or LSJ.) The LXX also has a few examples of θυμός in the sense of ‘mental 31

32 33 34

Cf. GE, vi: ‘Significant consideration is also given to later forms of Greek, in particular Greek of the imperial age … (Old and New Testament, Patristics, etc.), up to the VIth century.’ BDAG for the NT; Muraoka, Lexicon for the LXX. Chadwick, Lexicographica Graeca, 143, 149; full analysis of θυμός, 143–50. = 4 Kgdms 5:12. This happens to be in the ‘Kaige recension’ of Kingdoms, not exactly a good representative of LXX Greek.

THE BRILL DICTIONARY OF ANCIENT GREEK

421

disposition, ‘temper’,’ as Muraoka defines it, but there is no mention in GE. As for the NT, where ‘anger’ is the predominant meaning too (16 or so examples), GE cites none at all. 6.2. νουθετέω Again, GE manages one example in the LXX (Job 36:12), under ‘2. pass. to be admonished, reproved, corrected,’ rightly enough. But it gives none from the NT, where the word occurs eight times. The NT evidence would have been a useful – one might think indispensible – addition to sense 1. act. ‘advise, admonish’ etc., where currently all the citations are Classical except two from Philo and Lucian. 6.3. λογικός This is a difficult word, found some 9,000 times in Greek literature to VI AD. It is commonest in philosophers and theologians (which is warning enough). The two NT examples present special challenges, especially Rom 12:1, which is the subject of ongoing debate: παραστῆσαι τὰ σώματα ὑμῶν θυσίαν ζῶσαν ἁγίαν εὐάρεστον τῷ θεῷ, τὴν λογικὴν λατρείαν ὑμῶν. to present your bodies as a sacrifice living, holy, pleasing to God, your λογική service /(act of) worship.

LSJ steered clear of this one. GE has plunged in and come up with ‘|| later spiritual ORIG. … etc.; ἡ λ. λατρεία the spiritual cult NT Rom. 12.1’ (GI ‘il culto spirituale’).35 The other example, 1 Pet 2:2 τὸ λογικὸν ἄδολον γάλα, escapes notice, perhaps fortunately. 6.4. ὑπομονή ‘Endurance, patience’ in the face of adversity, ὑπομονή, is an important Christian concept, with parallels in Jewish and ‘secular’ thought. GE has improved on LSJ by at least including one NT reference (Rom 2:7), out of 30 or so. The LXX examples, however, do not rate a mention.36 35

36

Rocci did not include any NT example; but under λατρεία ‘2) culto; adorazione’ he notes exx. in Plato and adds ‘Set.; NT.’ In Lampe, PGL, λογικός fills two columns. Some renderings of Rom 12:1: KJV reasonable (< Tyndale < Vg. rationabile); RSV, NRSV spiritual; NEB fn. for such is the worship which you, as rational creatures, should offer. LSJ had none either, but LSJ Suppl. (1996) makes some rather complicated LXX additions, including a new meaning hope (which may not be for the better).

422

COLLECTED ESSAYS ON THE GREEK BIBLE

More noteworthy, in the light of the statement already quoted (n. 31), is the absence of any subsequent evidence like that seen in Lampe, or that can be found by TLG searching (2000+ exx. to VI AD). 6.5. Other observations In contrast with the above examples, GE’s treatment of the LXX does sometimes show advances. An instance is seen in καθιζάνω, which has a transitive use (‘cause to sit’) that LSJ missed. GI/GE includes it, with Prov 18:16 as the example (also found in Job 12:18).37 The attestation of the word given in GE, however, could have been fleshed out significantly.38 To conclude this section, I have taken a sample of 50 NT words from αὐτόπτης to ἄψυχος. In GE, 19 have a NT reference; 31 do not. Why the variation I cannot discern: it appears that the NT material in GI/GE was collected unsystematically. In the same sample the LXX is slightly more often referred to, that is, in those words occurring in the LXX (31 of the 50): 14 have a LXX reference; 17 do not. So the hit rate for both NT and LXX is less than half.

7. A PARTICLE Should we expect lexicons to help with particles? Whether we should or not, we certainly do. We hope to get a quick indication to help us out. Few of us are ready to face the ordeal of reading Denniston’s famous book or some other specialist study (inaccessible to us anyway). The lexicons do try to cover particles, usually in some depth. In GE μέν, for instance, is given a very extensive treatment filling three columns; the material in LSJ is of similar length; Rocci was shorter but still thorough. We have no reason not to trust most of this. Rather than try to examine μέν, let us take a less frequent particle, καίτοι, to see (a) whether GE is an advance on LSJ; and (b) how well both lexicons cover the data and usage of καίτοι. 37

38

I owe this example to Claude Cox. For his discussion of GE, focused especially on LXX Job, see his page at . Cox has noted an important failing of GE, the use of the LXX text of Rahlfs (1935), not the standard Göttingen edition where available. TLG searching produces many more exx. of καθιζάνω (c.60) than the 7 in GE, in a much greater range of authors. There is also one in UPZ 1.78.20 (159 BC, record of a dream).

THE BRILL DICTIONARY OF ANCIENT GREEK

423

καίτοι GE: καίτοι or καί τοι adv. certainly, truly, indeed: κ. ἐμοί yes, for me also IL. 13.267 cf. EUR. Med. 344; καίτοι σοφοῦ παρὰ φωτὸς εἰρημένον indeed said by a wise man SIM. 37.12 || advers. however, yet, nonetheless: κ. τί φημι and yet what am I saying AESCHL. Pr. 101; καίτοι γε nevertheless XEN. Mem. 3.12.7 | concess. although, even if: καίτοι περ although HDT. 8.53. Rocci: καίτοι, certo; a dire il vero; veramente, κ. ἐμοί, certo, anche a me, IL.13, 267; v. 1, 426; EU. Med. 344. – b) tuttavia; pure; per altro; nondimeno, ATT.; κ. τί φημι; ma che dico? ESCHL. Pr. 101; καίτοι γε, SEN. Mem. 3, 12, 7. – c) quantunque; sebbene, ATT.: κ. περ, quantunque, ER. 8, 53. Bailly: καίτοι, adv. 1 et certes, et en vérite [sic], Il. 13,267; ISOCR. Pan. 67, etc. || 2 quoi qu’il en soit, cependant, toutefois, ATT.; … κ. τί φημι; ESCHL. Pr. 101; … || 3 quoique, devant un part. … renforcé par περ . καίτοι περ, HDT. 8,53. LSJ: καί τοι, and indeed, and further, freq. in Hom. with one or more words between, Il. 1.426, al.; καὶ σύ τοι E.Med.344; καὶ τἆλλά τοι X.Cyr.7.3.10: once in Hom. as one word, Il.13.267. II. after Hom. usu., and yet, to mark an objection introduced by the speaker himself, freq. in Rhetorical questions, καίτοι τί φημι; A.Pr.101 …: without a question …: strengthd., καίτοι γ᾽ … etc.: mostly separated, καίτοι...γε … X.Mem.3.12.17, etc. …; so καίτοι περ v.l. in Hdt. 8.53. III. with a participle, much like καίπερ, Simon. 5.9. …

It will give us some perspective to start from outside the lexicons. Denniston’s analysis finds four main uses of καίτοι (adversative; continuative; logical; combined with other particles), with subgroups in some, a total of 14 sections. Each is illustrated, in Denniston’s usual fashion, by a wealth of examples, all from the Classical period except one. At a rough count he gives 250 examples.39 Blomqvist covers the same ground for his selection of Hellenistic authors (330–30 BC). He includes a table of frequency in his Hellenistic texts and some Classical authors. It shows that while there are fewer in later texts (70+), the numbers in Classical authors add up to hundreds; these include high numbers in prose writers such as Demosthenes (190+), Plato (120+), and Aristotle (200+).40 It is obvious that our lexicons offer an exceedingly limited selection of evidence for καίτοι. GE cites only six examples. Only two of these are from prose (Xen., Hdt.) and none are post-Classical. LSJ has more, a total 39

40

Denniston, Greek Particles, 555–64. At 561 he includes some totals for Isocrates (× 125) and Lysias (× 106). Blomqvist, Greek Particles, 35–45; table at 36. There are c.50 in papyri; c.15 in inscrr. The TLG tally just to II AD is 5,000+.

424

COLLECTED ESSAYS ON THE GREEK BIBLE

of 27 (with ‘etc.’ sometimes), and they are slightly more representative, including more from prose (D., Pl.) and even a couple of late authors (Plb., Philo). But the coverage is still very inadequate in LSJ, and in GE entirely unsatisfactory. LSJ appears to recognise three meanings or uses of καίτοι. The second and third are clearly described, but the first is rather strange and unclear: it seems as if and indeed, and further is frequent in Homer, but only if καί τοι is split by another word (except once), and that this use occurs only a couple of times later. The fit to the noted examples does not seem right either: for instance, the first, Iliad 1.426 καὶ τότ᾽ ἔπειτά τοι εἶμι, (‘and then after that I will go’), can hardly be an example of καί + τοι. Close scrutiny of the tradition suggests the first meaning was never intended to be a separate sense back in Passow and LS1 where it originated, but was an ambient meaning to cover the word as a whole; treatment of Homer was simply the next item in Passow. Proof of this is that sense I. was originally and yet, yet in LS1 (< Passow doch, und doch, doch auch, dennoch). It was changed in the 5th edition (1861) to and indeed, and further, the meaning we have now, at the same time as a new sense II. and yet was introduced. (LSJ’s sense III., the concessive use, was separated from the beginning.) These changes caused confusion in sense I. The match between these apparent three meanings in LSJ and Denniston’s analysis is a worry too. Sense II. matches D.’s (1) ‘adversative’ use, and III. matches (1.vii) the ‘concessive’ use; but where are D.’s (2) ‘continuative’ and (3) ‘logical’ uses? Various other problems could be pointed out but need not be gone into here. GE’s entry is no more than an attenuated remnant of its predecessors. The first sense or use (there are no numbers, only || and |) has been freshened up with input from Bailly via Rocci, but the illustrations doubtfully support it (one, SIM. 37.12, has migrated from sense [3]).41 The second sense, again from Bailly via Rocci, is an inexact catch-all, and the translation of Xen. Mem. 3.12.7 (included in LSJ only to illustrate καίτοι separated from γε) is simply wrong. The third is real but the example is a dubious v.l. (many others were available).42 41

42

GE begins with Iliad 13.267, but Denniston (Greek Particles, 555) says ‘This compound is not found in Homer …’; in Il. 13.267 (he says) ‘καί goes closely with ἐμοί (‘Know that for me too’)’. In E., Med. 344, GE’s next ex., Denniston (541) takes τοι as attached to σύ (i.e., ‘you are their father, you know’). [Simonides 37.12 (= earlier 5.9) was in LSJ’s III., moved up in GI, hence in GE.] The OCT (Hude) at Hdt. 8.53.1 prints καίπερ. On concessive καίτοι see Denniston, Greek Particles, 559; Blomqvist, Greek Particles, 39–43.

THE BRILL DICTIONARY OF ANCIENT GREEK

425

This look into καίτοι has enabled us to see not just what is wrong with the current entries but how their history has controlled the outcomes. The major underlying problem is reliance on an existing lexicon entry as the basis for the next, however much ‘revised.’ Unless the old entry is completely laid bare and all its data are reassessed, its problems will be passed on or made worse. Hence our conclusion that (a) GE’s entry for καίτοι is not an advance on LSJ’s; and (b) the coverage of data and usage is inadequate in both lexicons, but much worse in GE than in LSJ.

8. CONCLUSIONS The foregoing samples may be a small selection but they are representative of widespread phenomena in the new lexicon. They justify an assessment as follows. (a) Is GE a ‘better’ lexicon than LSJ? No. (b) Is it a competitor to LSJ? Yes; but only because it is a major lexicon aimed at covering the same territory. (c) Is it the equal of LSJ? No. (d) Can it replace LSJ? No. (e) Will it serve the purposes of the ordinary student beyond elementary level? Yes, up to a point. (f) Can it serve the purposes of scholars? Again, up to a point. (g) Does it apply an improved method of defining meaning? No. (h) Is it better at lexical analysis than LSJ? No. (i) Does it offer generally better coverage of meanings and attestation than LSJ? No. (j) Does it actually provide better coverage of post-Classical Greek, as it claims to do? No, very seldom. (k) Does it omit information that LSJ offers? Yes, often. (l) Is its presentation easier to follow than LSJ’s? To a slight extent only. (m) Is its numbering system easy to follow? No. (n) What does it offer that LSJ does not? Limited additions to the attestation of words and some meanings. (o) Is it basically a translation of its predecessor GI? Yes. (p) Is GI basically a revision of a predecessor? Yes. GE is a useful lexicon, but it is not the great advance that we long for. The shape of the lexicography of Ancient Greek has not been changed by GE, but remains where it was. GE simply supplies another lexicon of similar size and coverage to LSJ, doing the job no better and often rather worse. The principal reason is that GE is a descendant of LSJ itself, with some revisions, but with most of the faults as well as merits of LSJ preserved. The major, far-reaching overhaul that LSJ’s material needs has not been undertaken, either by the editors of GE or by anyone. Montanari rightly describes the situation in his preface to GE:

426

COLLECTED ESSAYS ON THE GREEK BIBLE

The Liddell-Scott-Jones dictionary is used by scholars of antiquity as the reference dictionary for Greek, although the need for a profound revision of LSJ itself or for a completely new dictionary of Ancient Greek has been put forward many times and is indeed increasingly felt, in order to take into account more organically the advances of knowledge achieved over the years.43

I read these words as a true statement of where we are, not a claim that GI/GE is itself the ‘profound revision’ that is needed. Montanari wisely refrains from making such a claim. If after all he intended to imply it, or others take his words in that way, it needs to be stated that there is no justification for it.44 What, then, is to be done? Lexicons like GE are a legacy of the past. They deal primarily with Classical Greek, and contain both too much and too little. They retain old material and have difficulty adding new. They also preserve the mistakes of their predecessors and have difficulty removing them. The coverage of post-Classical Greek is out of control because there is too much material. Trying to graft small portions of it onto existing lexicons is unworkable. Lexicons like GE are now a holding operation. Rather than continue to produce new lexicons based on old ones, the next generation of Greek lexicographers must rethink how to gather and present their material. Gathering of data by electronic means is now the only option; and a reappraisal of the meanings is essential. The whole corpus of Ancient Greek is too vast to deal with at once, but a systematic approach could move us forward. If the corpus were divided into manageable segments, the data could be gradually combined into one major electronic resource. Printed lexicons would still be needed, but could be slimmer and more selective, and based on better data.

43 44

GE, p. v. The second preface by Nagy, Muellner, and Goh (p. vii) apparently sees GE as the lexicon that incorporates ‘recent advances made in scholarship on the ancient Greek world.’

27 GREEK IDIOM IN THE LXX-PENTATEUCH: THE PREPOSITION PARA 2019

Abstract This paper examines the preposition παρά in the LXX-Pentateuch, where it occurs 223 times. A full analysis of the Hebrew matches shows that παρά renders a great variety of equivalents. A lexical analysis, with parallels in non-LXX Greek, is presented, showing that all the senses of παρά are in accordance with Greek usage. The conclusion is drawn that the choice of παρά is not due to consistent matching with any Hebrew equivalent but is determined by context and Greek usage alone. The hypothesis is put foward that the same may be true of all prepositions in the LXX-Pentateuch. The value of the LXX as evidence for contemporary Koine Greek is also noted.

1. INTRODUCTION Greek and Hebrew each have their own prepositions, and in each language they are part of a web or system of relationships proper to that language. While there are many ready matches between prepositions in the two languages, the two systems do not exactly align. Even when a preposition in one language is the natural equivalent of one in the other, they will almost certainly differ in their range of usage. It is also likely that there will be some spare capacity on the Greek side. The number of Greek prepositions used in the Pentateuch is 17 (that is, ‘proper’ prepositions). The Hebrew ‘simple’ prepositions are about the same in number, but most of the Greek ones can be used with more than one case: about half are used with two or three cases. The second most common preposition in the Pentateuch, ἐπί, appears with three cases, making it in effect three prepositions, not one. It also has a great variety of uses with each case.1 One 1

There are 39 senses (10 + 17 + 12) in Muraoka, Lexicon, s.v. ἐπί. The most frequent preposition is ἐν (2,200 +), then ἐπί (1,400 +), followed by εἰς (1,200 +). Hebrew of course expands its repertoire by ‘complex’ prepositions such as ‫לפני‬.

428

COLLECTED ESSAYS ON THE GREEK BIBLE

might wonder what Hebrew match or matches call for ἐπί to be used over 1,400 times. Only systematic enquiry can find out. The subject of how the LXX translators use the Greek prepositions and match them to the Hebrew has received limited attention.2 Investigation of this question is likely to produce new insights into the translators’ methods and their familiarity with Greek. An examination of the preposition παρά in the Pentateuch is the subject of this paper.3 παρά occurs 223 times, with all three cases. This is rather a lot for a preposition whose immediate Hebrew match is not obvious. One first wonders if some of the frequency could be due to a match to Hebrew ‫אצל‬, which has the meaning ‘beside’ like παρά + acc. In fact ‫ אצל‬occurs only 10 times in the Pentateuch, only 7 of these are translated by παρά, and only 4 are παρά + acc., ‘beside’ (the other 3 are παρά + dat., ‘with, near’). That explanation does not take us far. What then are the matches to παρά? The aim of this paper is to answer that question, and to find out how παρά is actually used and why so frequently. Also to be addressed is the question whether the usage of παρά in the Pentateuch is in accordance with natural Greek usage. Some general conclusions from the evidence will complete the paper.

2. THE HEBREW MATCHES OF ΠΑΡΑ

IN THE

LXX-PENTATEUCH

Let us first see a summary of all the Greek–Hebrew matches of παρά, that is, what παρά corresponds to in the Hebrew text (of MT) at each occurrence. The data are based on a full list I have compiled, which need not be presented in its entirety here.4 The matches for the three cases are as follows:

2

3

4

Some questions are explored in five papers by Soisalon-Soininen in Studien zur Septuaginta-Syntax: ‘Wiedergabe hebräischer Zeitangaben mit ‫‘ ;’ב‬Comparative Expressions with ‫‘ ;’מן‬ἐν für εἰς in der Septuaginta’; ‘Wiedergabe des ‫ ב‬instrumenti’; and ‘Wiedergabe des partitiven ‫מן‬.’ Cf. also Sollamo’s classic Semiprepositions, and ‘Repetitions of Prepositions.’ Johannessohn, Präpositionen has good observations but is only a beginning. Revised version of a paper presented in the Greek Bible Section at the SBL Annual Meeting, Denver 2018. My thanks to Jim Aitken for comments at the session and to Trevor Evans for reading the final version; also to Siegfried Kreuzer for retyping the Hebrew and an anonymous reviewer for helpful suggestions. Prepared with the aid of ‘Accordance’ and checked against the printed Göttingen text. The possibility of a different Heb. original of course arises, but affects very few cases.

GREEK IDIOM IN THE LXX-PENTATEUCH: THE PREPOSITION PARA

429

2.1. παρά + acc. Key: + = added without match in MT. >MT = text lacking in MT. ~ = corresponds to. ‫אל‬ Gen

1

Exod

1

Lev

6

Num

2

‫ב אצל‬ 1

‫מן מן‬# ‫ †על־ על‬other*

+

2

3

2

1

2 1

3

2

3

6

1 1

5

1

4

5

9

16

3 2

1

2

3

10

1

2

4

9

4

free >MT 2

1

4

2

Deut Totals

‫לפני‬

8 1

2

1

13 3

13

4

2 2

3

# In ‫מבל)ת(י‬ [Total 90] † ‫על־יד‬, ‫על־שׂפת‬, (‫על־דבר )אשׁר‬ * Gen: ‫ל‬, ‫עם‬. Exod: ‫אחר‬, ‫בקצה‬, ‫מדי‬. Num: ‫עלי‬, ‫מעבר ל‬, ‫ ~( ימה‬παρὰ θάλασσαν)

2.2. παρά + gen. ‫ על מעם מן מלפני מיד מאת מאת ל את‬other* + free >MT ‫פניו‬ Gen

1

6

Exod

4

9

Lev

1

8

Num

1

1

1

44

4

2

2

2

1

5

1 1

3

1

1

1

1

1

5 1

2

16

Deut Totals

1

1

1

1

4

4

9

8

1

6

1

1 2

1

12

2

* ‫ ~( מלבי‬παρ᾽ ἐμαυτοῦ)

1

[Total 91]

2.3. παρά + dat. Gen

‫אצל‬

‫את‬

‫ביד‬

3

5

2

Exod

4

Lev

1

‫בין‬

‫בית‬

‫בעיני‬

3 1

10

1

2

1

1

2

Deut 3

+

1+2

free

>MT 3

1

5

Num Totals

‫עם‬/‫עמדי‬

2

1

3

3

2+1

1

14

2

1

3

[Total 42]

430

COLLECTED ESSAYS ON THE GREEK BIBLE

2.4. Comments The striking thing is the great variety of the Hebrew matches. For all three cases of παρά there are multiple Hebrew equivalents. The total, if one unpacks the ‘other’ category, is over 30. We note also that παρά is quite often added without a match in the Hebrew (30 times), and there are a few ‘free’ renderings as well. Are there any constant matches? The only one that stands out is παρά + gen. ~ ‫ ×( מאת‬44). But what is its significance? It hardly constitutes a case of stereotyping, when the total is no more than 69% of the renderings of ‫( מאת‬total 64), and there are significant and deliberate variations from παρά.5 It is not in any case an exact match or calque representing the two elements of ‫( מאת‬lit. ‘from with’): παρά equates only to the ‘from’ element.6 Can we detect interference, that is, use of παρά in an unGreek way caused by the match to ‫ ?מאת‬Hardly: παρά + gen. (receive) ‘from’ (a person) is an entirely natural Greek use and is exactly the right expression for the contexts in which it appears. None of the other fairly frequent matches comes to a different result. For instance, παρά renders 22 ‫ מן‬times (with two cases). But παρά is not the obvious equivalent: that would be ἀπό or ἐκ. So interference cannot be a factor. On examination it turns out that (a) παρά + acc., which renders 31 ‫ מן‬times, mostly renders ‫ מן‬meaning ‘more than, beyond’, for which it is a natural Greek expression;7 and (b) παρά + gen., which renders ‫ מן‬9 times, is mostly used for the meaning ‘from’ (a person), again natural Greek,8 the same as the rendering of ‫מאת‬. Similarly, the match of παρά + dat. to ‫עם‬/‫עמדי‬, which is found 14 times, comes about because certain uses of παρά are more suitable in the contexts than the predictable equivalent μετά.9 In short, the choice of παρά is not governed by the 5

6 7

8

9

Summary of other renderings of ‫מאת‬: ἀπό × 12; ἐκ × 4; ὀπίσω × 1 (not = MT); omitted × 3; total 20. Specimens: Exod 5:20 ἐκπορευομένων αὐτῶν ἀπὸ Φαραώ (‫;)מאת פרעה‬ Num 31:2 ἐκδίκει τὴν ἐκδίκησιν … ἐκ τῶν Μαδιανιτῶν (‫)מאת המדינים‬. These are contextual choices. Cf. Lee, Greek of the Pentateuch, 244 for further comments on ‫ מעם‬and ‫מאת‬. Ten times: Gen 37:3; 43:34b; Exod 18:11; 33:16b; Num 12:3; Deut 7:6, 7a, b, 14; 10:15. The other three are: Deut 2:8 (I.2 ‘beside’); 7:8 (I.4 ‘on account of’); 9:28b (I.4 ‘on account of’). Cf. Soisalon-Soininen, ‘Comparative Expressions with ‫מן‬,’ discussed in Lee, Greek of the Pentateuch, 30–1. Seven times ‘from’: Gen 18:14; 23:13; 24:50; Exod 3:22; 12:35; Lev 25:33; Num 3:12. The other two are: Gen 49:24(25) (unclear: II.2 ‘from’?) and Deut 22:3 (II.4 ‘by’). παρά meaning ‘with, near, in the hands of’ (III.1): Gen 24:25; 31:32c; Exod 22:25(24); Lev 47a, b; Deut 29:17(16); 32:34; meaning ‘in the household of’ (III.2): Gen 29:27; Lev 25:35, 39, 40; Num 22:8, 9; Deut 15:16. This is not to say μετά is not used for ‫עם‬: see Lee, Greek of the Pentateuch, 317–8 for full lists of ‫ ~ עם‬μετά + gen. and ‫ עם‬with other renderings.

GREEK IDIOM IN THE LXX-PENTATEUCH: THE PREPOSITION PARA

431

Hebrew word in the original but by other factors, namely context and Greek usage (see further below, §5).

3. SAMPLES

OF THE USAGE OF ΠΑΡΑ

For a closer look at how παρά is used, let us see the occurrences with each case in three different books. The second-last column shows the Hebrew match; the number in the last column refers to the lexical analysis of παρά in §4.1 below. 3.1. παρά + acc. in Genesis ‫וישׁב באלני ממרא‬

‫ ב‬I.2

‫ישׁב בשׁער־סדם‬

‫ ב‬I.2

ὡς τὴν ἄμμον τὴν παρὰ τὸ χεῖλος τῆς θαλάσσης

‫וכחול אשׁר על־שׂפת הים‬

‫ על‬I.2

24:11

ἐκοίμισεν τοὺς καμήλους... παρὰ τὸ φρέαρ τοῦ ὕδατος

‫אל־באר המים‬

‫ אל‬I.2

25:11

κατῴκησεν... παρὰ τὸ φρέαρ τῆς ὁράσεως

‫עם־באר לחי ראי‬

‫ עם‬I.2

29:20

παρὰ τὸ ἀγαπᾶν αὐτὸν αὐτήν

‫באהבתו אתה‬

ℶ I.4

36:37

ἐκ Ῥοωβὼθ τῆς παρὰ ποταμόν

‫מרחבות הנהר‬

+ I.2

37:3

ἠγάπα τὸν Ἰωσὴφ παρὰ πάντας τοὺς υἱοὺς αὐτοῦ

‫אהב את־יוסף מכל־בניו‬

‫ מן‬I.3

41:3a

καὶ ἐνέμοντο παρὰ τὰς βόας

41:3b

παρὰ τὸ χεῖλος τοῦ ποταμοῦ

41:17

ᾤμην ἑστάναι παρὰ τὸ χεῖλος τοῦ ποταμοῦ

13:18

κατῴκησεν παρὰ τὴν δρῦν τὴν Μαμβρή

19:1

ἐκάθητο παρὰ τὴν πύλην Σοδόμων

22:17

‫ אצל ותעמדנה אצל הפרות‬I.2 ‫על־שׂפת היאר‬

‫ על‬I.2

‫הנני עמד על־שׂפת היאר‬

‫ על‬I.2

43:34b ἐμεγαλύνθη δὲ ἡ μερὶς Βενιαμὶν παρὰ τὰς μερίδας πάντων

‫ותרב משׂאת בנימן ממשׂאת‬ ‫כלם‬

‫ מן‬I.3

Ζαβουλὼν παράλιος κατοικήσει, καὶ αὐτὸς παρ᾽ ὅρμον πλοίων

‫והוא לחוף אניות‬

‫ ל‬I.2

49:13

In most of these παρά has the one meaning, ‘beside’ (I.2), yet it corresponds to six different Hebrew prepositions: ‫ב‬, ‫על‬, ‫עם‬, ‫אל‬, ‫עצל‬, ‫ל‬. In 49:13 there is some interpretation by the translator: the Hebrew has ‘and he (will be) as/for a shore/haven(?) of ships,’ but the translator has made it

432

COLLECTED ESSAYS ON THE GREEK BIBLE

‘and he (will be) beside an anchorage of ships.’ There is interpretation similarly in 36:37, which is a good example of added παρά: the Hebrew says ‘Rehoboth of the river,’ but the translator clarifies with ‘R. beside the river.’10 In all of the examples the translator says what he wants to say using παρά to do it, instead of representing the Hebrew by a stock equivalent. In 29:20 there is an example of παρά + articular infinitive meaning ‘on account of,’ a use that may be unfamiliar to some but is well attested in documents (see sense I.4 below). As can be seen, the translator has declined to use ἐν as the match to ‫ ב‬and also converted the noun (‫)אהבתו‬ into an infinitive. The match παρά + acc. ~ ‫‘ מן‬more than,’ discussed above, is exemplified in 37:3 and 43:34b. 3.2. παρά + gen. in Deuteronomy 2:6a

βρώματα ἀγοράσατε παρ᾽ αὐτῶν

2:6b

καὶ ὕδωρ μέτρῳ λήμψεσθε παρ᾽ αὐτῶν ἀργυρίου

3:4

οὐκ ἦν πόλις, ἣν οὐκ ἐλάβομεν παρ᾽ αὐτῶν

10:12

τί κύριος ὁ θεός σου αἰτεῖται παρὰ σοῦ ἀλλ᾽ ἢ φοβεῖσθαι...;

17:18

καὶ γράψει ἑαυτῷ τὸ δευτερονόμιον τοῦτο εἰς βιβλίον παρὰ τῶν ἱερέων

18:3a, b καὶ αὕτη ἡ κρίσις τῶν ἱερέων, τὰ παρὰ τοῦ λαοῦ, παρὰ τῶν θυόντων τὰς θυσίας...·

‫אכל תשׁברו מאתם בכסף‬

‫ מאת‬II.2

‫וגם־מים תכרו מאתם בכסף‬

‫ מאת‬II.2

‫לא היתה קריה אשׁר לא־לקחנו‬ ‫מאתם‬

‫ מאת‬II.2

‫מה יהוה אלהיך שׁאל מעמך‬ ‫כי אם־ליראה‬

‫ מעם‬II.2

‫ מלפני וכתב לו את־משׁנה התורה הזאת‬II.2 ‫על־ספר מלפני הכהנים‬ ‫וזה יהיה משׁפט הכהנים מאת‬ ‫העם מאת זבחי הזבח‬

‫ מאת‬II.2, ‫ מאת‬II.2

‫ככל אשׁר־שׁאלת מעם יהוה‬

‫ מעם‬II.2

18:16

κατὰ πάντα, ὅσα ᾐτήσω παρὰ κυρίου τοῦ θεοῦ σου

22:3

ὅσα ἂν ἀπόληται παρ᾽ αὐτοῦ καὶ εὕρῃς

‫אשׁר־תאבד ממנו ומצאתה‬

‫ מן‬II.4

23:15 (16)

οὐ παραδώσεις παῖδα τῷ κυρίῳ, ὃς προστέθειταί σοι παρὰ τοῦ κυρίου αὐτοῦ

‫לא־תסגיר עבד אל־אדניו‬ ‫אשׁר־ינצל אליך מעם אדניו‬

‫ מעם‬II.2

10

παρὰ ποταμόν is a set phrase that does not require the article, as we might say ‘Roobothon-river’: cf. Henley-on-Thames, Southend on Sea (and many others). Cf. Wevers, Genesis, 607, dubiously: ‘One might have expected του ποταμου [sic], but the word is here rationalized as meaning ‘beside a river’.’

GREEK IDIOM IN THE LXX-PENTATEUCH: THE PREPOSITION PARA

23:21 (22)

ὅτι ἐκζητῶν ἐκζητήσει κύριος ὁ θεός σου παρὰ σοῦ

33:23

καὶ ἐμπλησθήτω εὐλογίαν παρὰ κυρίου

433

‫כי־דרשׁ ידרשׁנו יהוה אלהיך‬ ‫מעמך‬

‫ מעם‬II.2

‫ומלא ברכת יהוה‬

+ II.2

Here παρά ‘from’ (a person) is frequent (× 11) because it is, as was said above (§2.4), the best fit in the contexts. In this sample it renders ‫ מאת‬often (× 5), but note that it renders ‫ מעם‬nearly as often (× 4). If we try out ἀπό or ἐκ as an alternative to παρά in these places, we can see that they are not well-suited to the contexts. In 10:12, for example, ἀπὸ σοῦ might just be possible, but ἐκ σοῦ does not seem right at all; a check confirms that αἰτῶ/-οῦμαι ‘ask for’ (from a person) is never followed in the Pentateuch by ἀπό or ἐκ, only παρά, regardless of what the original has.11 The ‘from’ words used with λαμβάνω, like παρά in 2:6b and 3:4, are a longer story, but when ἀπό and ἐκ are used with this verb, which they are, they have meanings different from παρά.12 Similar data could be gathered for the other verbs in this sample. Two other examples here are worth comment. In 18:3 the translator has twice translated ‫ מאת‬by παρά, but in the first instance has inserted τά before it. Why? To make the meaning clear as he sees it: ‘This is the decision for the priests as regards the things from the people (τὰ παρὰ τοῦ λαοῦ), from those sacrificing (παρὰ τῶν θυόντων).’ The first phrase gives the subject-matter, the second specifies. The last occurrence in the table, 33:23, is another case of added παρά reflecting an interpretation by the translator: ‘blessing from the Lord,’ instead of ‘blessing of the Lord.’13 3.3. παρά + dat. in Exodus 2:21

κατῳκίσθη δὲ Μωυσῆς παρὰ τῷ ἀνθρώπῳ

16:18

ἕκαστος εἰς τοὺς καθήκοντας παρ᾽ ἑαυτῷ συνέλεξαν

22:25 (24)

ἐὰν δὲ ἀργύριον ἐκδανείσῃς τῷ ἀδελφῷ τῷ πενιχρῷ παρὰ σοί

11

12

13

‫ויואל משׁה לשׁבת את־האישׁ‬

‫ את‬III.2

‫ אישׁ לפי־אכלו לקטו‬free ‫אם־כסף תלוה את־עמי‬ ‫את־העני עמך‬

III.2

‫ עם‬III.1

αἰτῶ/-οῦμαι + παρά: Exod 3:22 (‫ ;)מן‬11:2 (‫ ;)מאת‬12:35 (‫ ;)מן‬22:14(13) (‫;)מעם‬ Deut 10:12 (‫ ;)מעם‬18:16 (‫)מעם‬. λαμβάνω + ἀπό or ἐκ in Deut: 1:15 (ἐξ ὑμῶν, ‘from among’ > MT), 23 (ἐξ ὑμῶν, ‘from among’ ~ ‫)מכם‬, 52 (ἀπὸ τοῦ καρποῦ, ‘some of’ ~ ‫ ;)מפרי‬2:62 (ἀπὸ τῆς ἀπαρχῆς, ‘some of’ ~ ‫)מראשׁית‬, 4 (ἐκ τῶν χειρῶν σου, ‘out of’ ~ ‫)מידך‬. It is not evident why παρά is inserted here in particular, when the same phrase ‫ברכת‬ ‫ יהוה‬is translated by a simple genitive in Gen 39:5; Deut 12:15; 16:17; 33:13 (‫)מב ֶֹר ֶכת‬. ְ

434 31:13

COLLECTED ESSAYS ON THE GREEK BIBLE

ἔστιν γὰρ σημεῖον παρ᾽ ἐμοὶ καὶ ἐν ὑμῖν

33:12b οἶδά σε παρὰ πάντας, καὶ χάριν ἔχεις παρ᾽ ἐμοί 33:16a ὅτι εὕρηκα χάριν παρὰ σοί

‫כי אות הוא ביני וביניכם‬

‫ בין‬III.1

‫ בעיני ידעתיך בשׁם וגם־מצאת חן‬III.3 ‫בעיני‬ ‫ בעיני כי־מצאתי חן בעיניך‬III.3 ‫הנה מקום אתי ונצבת‬ ‫על־הצור‬

‫ את‬III.1

καὶ παρ᾽ ᾧ εὑρέθη βύσσος

‫וכל־אישׁ אשׁר־נמצא אתו‬ ‫תכלת‬

‫ את‬III.1

καὶ παρ᾽ οἷς εὑρέθη ξύλα ἄσηπτα

‫וכל אשׁר נמצא אתו עצי‬ ‫שׁטים‬

‫ את‬III.1

33:21

ἰδοὺ τόπος παρ᾽ ἐμοί, στήσῃ ἐπὶ τῆς πέτρας

35:23 35:24

To render ‫ עם‬and ‫את‬, both roughly equivalent to English ‘with,’ the translators of course frequently use μετά + gen.; but there are many other renderings, and μετά is used for only 64% of the occurrences of ‫עם‬, and 47% of ‫את‬.14 For the meaning ‘with, near, in the hands of’ (III.1 below), παρά + dat. is more appropriate. How often this is the case in the Pentateuch overall cannot be pursued here, but in this sample we see it several times. Thus in 33:21, ‘Here is a place near me, you shall stand on the rock,’ παρ᾽ ἐμοί is more suitable than μετ᾽ ἐμοῦ (= ‘in company with me’), which would hardly give the right sense. The sample above also shows two instances of an interesting use of παρά + dat. that emerged from the evidence of the documentary papyri, namely, ‘in the household of’ (sense III.2 below). This is identifiable in 2:21 and 16:18 above, and several times elsewhere, including two notable instances in Genesis (see below, §4.2). A case of interpretative rendering appears in 31:13, where the Exodus translator changes ‘[my Sabbaths] are a sign between me and between you’ (‫ )אות ביני וביניכם‬into ‘a sign with me and among you’ (σημεῖον παρ᾽ ἐμοὶ καὶ ἐν ὑμῖν).15 4. LEXICAL ANALYSIS

OF ΠΑΡΑ, WITH PARALLELS IN NON-LXX

GREEK

I next offer a lexical analysis of the occurrences of παρά in the Pentateuch, with parallels in non-LXX Greek. Parallels were readily found in III BC papyri (one II BC) for all the meanings except the first, one of the oldest and most basic; but it was easily found in older sources from Homer 14 15

Lee, Greek of the Pentateuch, 317–20. The standard rendering of ‫ בין‬is ἀνὰ μέσον. The Exodus translator again avoids it in 31:17 with reference to the Sabbath: διαθήκη αἰώνιος (17) ἐν ἐμοὶ (‫ )ביני‬καὶ τοῖς υἱοῖς (‫ )ובין בני‬Ἰσραήλ· σημεῖόν ἐστιν αἰώνιον. There has to be an exegetical reason for this.

GREEK IDIOM IN THE LXX-PENTATEUCH: THE PREPOSITION PARA

435

onwards. There can be no doubt about its currency in the language of the third century BC.16 In this analysis the method of indicating meaning is by gloss (in italics) and collocation (in brackets, plain text). An improved method could be achievable.17 For each of the senses one example in the Pentateuch is offered (sometimes two), with the Hebrew match to παρά. 4.1. Lexical analysis I. παρά + acc. 1. (movement) alongside (a place) Exod 2:5 αἱ ἅβραι αὐτῆς παρεπορεύοντο παρὰ τὸν ποταμόν. (‫)על־יד‬ Homer, Iliad 1.34 βῆ δ᾽ ἀκέων παρὰ θῖνα πολυφλοίσβοιο θαλάσσης. And he went in silence along the shore of the loud-sounding sea. Plato, Gorgias 511e.5 ταῦτα διαπραξάμενος ἐκβὰς παρὰ τὴν θάλατταν καὶ τὴν ναῦν περιπατεῖ. After carrying out (this service), he goes ashore and walks back and forth beside the sea and the ship. 2. (to or at a position) beside, at (a place) Exod 15:27 παρενέβαλον δὲ ἐκεῖ παρὰ τὰ ὕδατα. (‫)על‬ Lev 19:21 προσάξει... τῷ κυρίῳ παρὰ τὴν θύραν τῆς σκηνῆς... κριὸν πλημμελείας. (‫)אל־פתח‬ PCairZen 1.119.6 εἰς βο|τανισμὸν τῆς παρ᾽ αἰ|γιαλὸν ἀρ(ουρῶν) ιε. For weeding of the 15 arouras of land beside the shore. PHeid 8.421.10 (II BC) ἐν τῷ τρ[ί]τῳ καταβαίνουσι | ἀπὸ τῆς . . υφης π[α]ρὰ τὴν θύραν τοῦ | ἀνδρῶνος. In the third (room) they come down from the ceiling(?) beside the door of the men’s apartment. 3. more than, beyond Num 12:3 Μωυσῆς πραῢς σφόδρα παρὰ πάντας τοὺς ἀνθρώπους. (‫)מן‬ PSI 4.422.34 ὁ πατήρ | μου προετέραι (= -ᾳ) παρὰ πάντας | τοὺς ἐκεῖ. My father was superior to all the (farmers) there. 4. on account of (+ artic. inf.) Exod 14:11 παρὰ τὸ μὴ ὑπάρχειν μνήματα ἐν γῇ Αἰγύπτῳ. (‫)מבלי‬ PCairZen 3.377.5 παρὰ τὸ μή σε χορηγεῖν ἡμῖν | τὰ κατ∖ὰ∕ τὴν συνγραφήν. Because you have not supplied us with what was stated in the contract.

II. παρά + gen. 1. (movement) from (a place, person) Lev 9:24 καὶ ἐξῆλθεν πῦρ παρὰ κυρίου. (‫)מלפני‬ 16 17

A later example spans the gap: Mark 1:16 παράγων παρὰ τὴν θάλασσαν τῆς Γαλιλαίας. As is being explored in the project ‘A Greek-English Lexicon of the Zenon Archive’ conducted by Trevor Evans and this author. The Pent. data might seem to require many senses of παρά, but that corpus produces even more.

436

COLLECTED ESSAYS ON THE GREEK BIBLE

PMich 1.34.16 ἀπόλωλέμ μοι ὄνος θήλεια | λευκὴ ἐκπηδήσασα παρὰ | Νικίου νυκτός. I have lost one white female donkey which escaped from Nikias at night. 2. (obtain, receive) from (a person) Lev 25:36 οὐ λήμψῃ παρ᾽ αὐτοῦ τόκον οὐδὲ ἐπὶ πλήθει. (‫)מאתו‬ PCairZen 1.120.2 διὰ τὸ προλαβεῖν παρ᾽ αὐτοῦ κερμάτιον εἰς ἐφόδια. Because we received in advance from him money for travel expenses. Idiom: παρ᾽ ἐμαυτοῦ (‘from my own resources’/ ‘on my own account’) Gen 31:39 ἐγὼ ἀπετίννυον παρ᾽ ἐμαυτοῦ κλέμματα. (‫)מידי‬ Num 24:13 οὐ δυνήσομαι παραβῆναι τὸ ῥῆμα κυρίου ποιῆσαι αὐτὸ πονηρὸν ἢ καλὸν παρ᾽ ἐμαυτοῦ. (‫)מלבי‬ PMich 1.58.27 προστέ|θεικα δὲ καὶ παρ᾽ ἐ|μουτοῦ (= ἐμαυ-) ἐρίων μν(ᾶν) α. I have also added from my own resources 1 mina of wool. 3. (a person) attached to (a person) Num 31:49 οἱ παῖδές σου εἰλήφασιν τὸ κεφάλαιον τῶν ἀνδρῶν τῶν πολεμιστῶν τῶν παρ᾽ ἡμῶν. (‫)אשׁר בידנו‬ PCairZen 1.33.7 ἠξίουν κατα|σταλῆναί τινα τῶν παρὰ σοῦ πρὸς Λυσίμαχον. I asked that one of your people be sent down to Lysimachos. PCairZen 4.532.24 τῶι παρ᾽ Ἀπολλωνίου Ζήνωνι. To Zenon who is Apollonios’s agent. 4. (action) by (a person) Deut 22:3 ὅσα ἂν ἀπόληται παρ᾽ αὐτοῦ καὶ εὕρῃς. (‫)ממנו‬ PCol 3.46.3 ...τὴν πα|ρὰ σοῦ δοθῖσαν γρα|φὴν τῶν ἱερέων. [Contrary to] the list of pigs given by you.

III. παρά + dat. 1. with, near, in the hands of (a person) Gen 39:15 καταλιπὼν τὰ ἱμάτια αὐτοῦ παρ᾽ ἐμοὶ ἔφυγεν. (‫)אצלי‬ Lev 19:13 οὐ κοιμηθήσεται ὁ μισθὸς τοῦ μισθωτοῦ παρὰ σοὶ ἕως πρωί. (‫)אתך‬ PCairZen 1.120.1 παρὰ Διονυσοδώρωι τῶι Διονυσίου υἱῶι κατελίπομεν | βατιάκια β ἐν θήκηι. With Dionysodoros son of Dionysios we left 2 dishes as security. 2. in the household of (a person) Gen 40:3 καὶ ἔθετο αὐτοὺς ἐν φυλακῇ παρὰ τῷ ἀρχιμαγείρῳ εἰς τὸ δεσμωτήριον. (‫)בית שׂר בטבחים‬ Gen 40:7 ...τοὺς εὐνούχους Φαραώ, οἳ ἦσαν μετ᾽ αὐτοῦ ἐν τῇ φυλακῇ παρὰ τῷ κυρίῳ αὐτοῦ. (‫)בית אדניו‬ PCairZen 2.192.5 ...γράψαι | [σο]ι περ[ὶ αὐ]τ[οῦ], ἵνα κατατάξῃς που αὐτὸν παρὰ σοί. [They asked me] to write to you about him, to give him some position with you. 3. in the estimation of (a person) Exod 33:16a ὅτι εὕρηκα χάριν παρὰ σοί. (‫)בעיניך‬ PCairZen 3.481.18 γίνωσκε δὲ καὶ | παρὰ τοῖς κεραμεῦσιν | διαβολὴν ἔχοντά με· Know also that I am subject to malicious accusation among the potters.

GREEK IDIOM IN THE LXX-PENTATEUCH: THE PREPOSITION PARA

437

4.2. Remarks Brief comments may be made on some meanings and examples here. The approach to idioms (i.e., fixed expressions that have a meaning beyond that of the parts) is to place them under the meaning of the preposition itself, as one can best understand it. The idiom παρ᾽ ἐμαυτοῦ turns up in the Pentateuch and is able to be paralleled in a contemporary text. It is placed here under sense II.2, as the closest to what seems to be the meaning of παρά itself.18 A further point related to the lexical analysis. The lexicographicallyinclined reader may notice that sense I.2 wraps together two meanings, one implying movement to an end-point, i.e., ‘to’ a position beside something, the other without movement, ‘at’ a position beside something. This is about ‘lumping’ and ‘splitting.’ The lexicographer has to decide if the difference is significant enough to justify splitting, that is, setting up separate senses, or not. In this case many of the occurrences are not clearly one sense or the other, or may imply a bit of both, so separate senses are difficult to maintain. Sense II.3 ‘attached to’ is a somewhat counter-intuitive meaning: one expects παρά + gen. to mean ‘from.’ But this sense is well-attested in documents, and the transition is illustrated by PCairZen 1.33.7 (in II.3 above), which could possibly be taken as ‘from.’ Num 31:49 is clear and represents the meaning of the original: ‘Your servants have taken the total of the fighting men who belong/-ed to us’ (τῶν παρ᾽ ἡμῶν ~ ‫אשׁר בידנו‬, ‘who are/were under our control’).19 Sense III.2 ‘in the household of’ has been noticed above in Exod (§3.3). Most of the examples correspond to ‫עם‬, and some can be hard to distinguish from III.1 ‘with.’20 But in the two quoted (Gen 40:3, 7), the original has ‫בית‬, ‘in the house’ (of), leaving little doubt how the Greek was intended.21 Knowing that παρά has this meaning makes the narrative 18

19 20

21

The parallel is exact for the first, literal use (Gen 31:39); the other (Num 24:13) is metaphorical. The next exx. in documents are BGU 8.1882.7 (I BC), SEG 25.501.56 (c.85 BC), both literal; there are others later. I have not searched literary Greek. LSJ, s.v. παρά A.II.2, has exx. with 3rd pers. reflexive ἑαυτοῦ. Other III BC examples of sense II.3: PCairZen 1.4.46; 2.247.2; 2.217.5. Full list of definite and possible exx. of ‘in the household of ’: Gen 24:23 (‫)בית‬, 25 (‫;)עם‬ 29:27 (‫ ;)עמדי‬39:2 (+); 40:3 (‫)בית‬, 7 (‫ ;)בית‬Exod 2:21 (‫ ;)את‬16:18 (free); Lev 25:35 (‫)עם‬, 39 (‫)עם‬, 40 (‫)עם‬, 47a (‫)עם‬, 47b (‫ ;)עם‬Num 22:8 (‫)עם‬, 9 (‫ ;)עם‬Deut 15:16 (‫)עם‬ [total 16]. Muraoka (Lexicon, s.v. παρά) includes four of these under II.1.a (I give M.’s translations): Gen 40:3 (‘at the chief cook’s’), Num 22:9 (‘at your place’), Gen 29:27 (‘you will work with me’), 39:2 (‘he entered … the Egyptian master’s service’); but they are combined with others under one meaning stated as: ‘indicates proximity, ‘beside, at’.’ Other documentary exx. are: PCairZen 1.15.vrp.4; 3.491.19. Also in Gen 24:23, a rather different context but evidently intended in the same sense: εἰ ἔστιν παρὰ τῷ πατρί σου τόπος ἡμῖν καταλῦσαι; (...‫)הישׁ בית־אביך מקום‬.

438

COLLECTED ESSAYS ON THE GREEK BIBLE

clear, and proves that the translator was fully cognizant of what his original meant to say.22

5. CONCLUSIONS Let us begin with a summary of characteristics of the usage of παρά in the LXX-Pentateuch, based on the foregoing data: (a) (b) (c) (d) (e) (f) (g)

Inconsistency in Greek–Hebrew matches Limited consistency in ‫ ~ מאת‬παρά + gen. No evident cases of Hebrew interference Full range of uses of παρά All natural Greek uses Used in free or interpretative renderings Added without any Hebrew match

The evidence for (a) is seen in the summary of the Hebrew matches in §§2.1–3 and in the samples in §§3.1–3. It is hard to detect any consistent, one-to-one matches, except in παρά + gen. ~ ‫ ;מאת‬but this match only reflects the fact that παρά + gen. is the natural Greek way of expressing the idea in the contexts (§2.4). The same is true of other somewhat frequent matches (§2.4). Hebrew interference (c) cannot be identified if the use of παρά is normal Greek – which it is in every case. The evidence for (d) and (e) was found in the analysis of the uses of παρά in the Pentateuch, with parallels in contemporary documentary texts (or earlier) (§4.1). The significance of (f) and (g) is that they show that the translators were able to deploy παρά idiomatically without being constrained by the original. A final point may be added. Though a fine-grained study of each translator would be needed for finality, the impression is strong that there are no differences between the (five) translators in their usage of παρά. That there were five translators I think likely on other grounds, but παρά does not seem to offer help to prove it. The conclusions I draw are the following. (1) The Pentateuch translators do not follow any fixed equivalence in the use of παρά but use it, more or less instinctively, when it best suits the context in accordance with natural Greek. (2) The evidence I think justifies putting this conclusion in a stronger (perhaps surprising) form: it does not matter what the Hebrew preposition is; the translator’s choice of παρά is determined by context and natural 22

Wevers (Genesis, 663) claims that the Hebrew original in Gen 40:3 ‘troubled’ the translator; but his own understanding is troubled by unawareness of this meaning of παρά.

GREEK IDIOM IN THE LXX-PENTATEUCH: THE PREPOSITION PARA

439

Greek usage alone. This is not to say that the Hebrew preposition plays no role: obviously the translator reads (or hears) the Hebrew text with the preposition in it and takes in its meaning before choosing a rendering; but which Greek preposition will be chosen (or whether there will even be one) is not determined by the Hebrew preposition. (3) On the basis of this result for παρά and the similar findings of studies of κατά and the renderings of ‫ עם‬and ‫את‬,23 I here put forward a somewhat bold proposal, namely, that the same conclusion as reached for παρά applies to all prepositions in the LXX-Pentateuch: that is, the choice of Greek preposition is generally if not always determined by the needs of context and Greek usage, not the preposition used in the Hebrew original. This is a hypothesis to be tested by further enquiry. It is intended to challenge the easy assumption that the translators’ usual practice was simply to replicate the Hebrew preposition by a standard Greek equivalent. This is not true of παρά, as we have seen, nor of κατά and the renderings of ‫ עם‬and ‫את‬. (4) The Pentateuch translators’ use of παρά is further proof of their intimate acquaintance with Greek to the point of native-speaker competence. (5) The value of studying LXX usage alongside that of contemporary Koine Greek is again demonstrated. There is a further outcome of that approach: the LXX Pentateuch itself may provide evidence for contemporary Greek.24 Its value in this respect varies. It may do no more than confirm an already well-attested meaning or use, as is the case with most of the uses of παρά, or it may provide useful support, as in its attestation of the idiom παρ᾽ ἐμαυτοῦ (II.2) and its confirmation of the meaning ‘in the household of’ (III.2).

23

24

Lee, Greek of the Pentateuch, 154–9 (κατά written May 2015); 242–7 (‫ עם‬and ‫ את‬July 2014). Cf. Lee, Greek of the Pentateuch, 5.

28 BACK TO THE QUESTION OF GREEK IDIOM 20201

Abstract The paper first discusses Soisalon-Soininen’s recognition of the presence of idiomatic Greek renderings in the LXX, although he takes these to be departures from the translators’ usual strict word-for-word method. The aim of the paper is to identify clear cases of idiomatic Greek and to establish a method for recognising them. An analysis of all renderings into three categories is proposed. Two cases of idiomatic Greek are then presented in detail, the enclitic pronoun/ adjective τις and the particle ἄν. The conclusions drawn are: (a) these features are evidence for the translators’ native-speaker competence in Greek; (b) the LXX-Pent. should be viewed as in essence Greek with Hebrew interference; and (c) its Greek has value as evidence for Koine Greek. The last is illustrated by the ‘iterative’ use of ἄν + indicative.

1. SOISALON-SOININEN AND GREEK IDIOM The title of my paper intentionally echoes that of a 1990 paper by Soisalon-Soininen, ‘Back to the Question of Hebraisms,’ providing both a link to his work and a starting-point for the present discussion.2 That paper was, naturally, about Hebraisms, which he saw as neglected in recent work on literal versus free rendering in the LXX. He did not only discuss Hebraisms, however, but made significant mention of idiomatic renderings: ‘… nearly all the translators occasionally produce very skilful, idiomatic renderings.’ He also said that ‘our work is based on (the question) which 1

2

This paper was presented at the Soisalon-Soininen Symposium on the Septuagint, held in Helsinki in 2017. I very much appreciate the invitation to attend and the opportunity to visit Helsinki for the first time, and most of all the kindness and hospitality of all those involved in organising and conducting the Symposium. Soisalon-Soininen, ‘Zurück zur Hebraismenfrage.’ English translation by Theo van der Louw (‘Back to the Question of Hebraisms’) in Kauhanen and Vanonen, The Legacy, 283–300. Page refs. are to the original; translations are van der Louw’s with occasional revisions.

442

COLLECTED ESSAYS ON THE GREEK BIBLE

renderings are Hebraistic and which are idiomatic.’ By idiomatic renderings he evidently meant those that use natural, normal Greek; and the work to which he referred was ‘to study the differences between individual translators.’3 Soisalon-Soininen thus recognised the presence of idiomatic Greek in the LXX, and his paper gave some illustrations, as well as examples of what he termed Hebraisms. But it seems to me he did not really set out a means of identifying the two, though he did give a definition of a Hebraism.4 As to idiomatic renderings, he only said that they are ‘renderings which involve, to a greater or lesser degree, a break with the customary word for word approach’5 – which is rather limited and I think not true: a ‘word for word’ rendering may still be idiomatic Greek. (I admit to some trouble following his thinking here, as at times elsewhere). In other publications also, Soisalon-Soininen made observations on the presence of idiomatic Greek in the LXX. I take two further papers that show this and reveal more about the significance he attached to these features. A 1973 paper dealt with the genitive absolute.6 This, as SoisalonSoininen of course knew well, is a purely Greek construction without any equivalent construction in Hebrew, and therefore a case of idiomatic Greek. It appears only sporadically in the LXX as a whole, though more frequently in books like Job and Proverbs (as one would expect). When it does appear, it gives us (he says) a glimpse of ‘the true language competence of the translators.’7 Though he does not quite spell it out, by ‘competence’ SoisalonSoininen evidently means competence or skill in Greek, not competence in translation. He adds, moreover, in his summing up, that ‘we should give 3

4

5

6

7

‘Auf der anderen Seite bringen fast alle Übersetzer gelegentlich sehr geschickte idiomatische Wiedergaben’ (Soisalon-Soininen, ‘Zurück,’ 43); ‘[Ob wir es geradeheraus sagen oder nicht,] so gründet sich unsere Arbeit dann darauf, welche Wiedergaben hebraistisch, welche idiomatisch sind’ … ‘[Haben wir denn überhaupt die Möglichkeit,] die Unterschiede zwischen den verschiedenen Übersetzern zu untersuchen?’ (‘Zurück,’ 51). ‘Ein Hebraismus in der Septuaginta ist ein aufgrund des Hebräischen entstandener Ausdruck, der nicht mit dem Sprachgebrauch des gleichzeitigen Griechisch im Einklang steht oder der vom Hebräischen aus erklärliche Gebrauch möglicher Vokabeln oder Ausdrücke im Griechischen in Zusammenhängen, in denen sie nach dem griechischen Sprachgebrauch nicht passend sind’ (Soisalon-Soininen, ‘Zurück,’ 39). ‘… Wiedergaben, die einen größeren oder kleineren Eingriff in die übliche Wiedergabe der einzelnen Teile des Ausdrucks voraussetzen’ (‘Zurück,’ 43). Soisalon-Soininen, ‘Gebrauch des genetivus absolutus.’ English translation by Theo van der Louw (‘The Use of the Genitive Absolute in the Septuagint’) in Kauhanen and Vanonen, The Legacy, 223–8. Page refs. are to Studien; translations are van der Louw’s. ‘[D]ie wirklichen Sprachkenntnisse der Übersetzer’ (Soisalon-Soininen, ‘Gebrauch,’ 180). He also speaks of ‘ihr wirkliches Sprachgefühl and ihr Können’ (176) und ‘die Geschicklichkeit … der Übersetzer’ (180).

BACK TO THE QUESTION OF GREEK IDIOM

443

more weight to these rare cases,’8 a sentiment that is in harmony with the aim of this paper – though I would not agree that they are rare. Clearly, then, Soisalon-Soininen recognised the presence of Greek idiom in the occasional appearances of the genitive absolute. Yet he adds the (to me) curious interpretation that ‘in those cases, the constraining factors stepped back for some reason.’9 What are those ‘constraining factors’? What does he mean? Evidently they are the constraints imposed by attempting to produce a close rendering of the original Hebrew. This is deducible from Soisalon-Soininen’s preceding discussion, where he speaks of the syntax of the LXX as ‘translationese’ and asserts that ‘one could say that it is not Greek but rather Hebrew with Greek words.’10 Free renderings, then, are not expected, they are not supposed to occur; the translators were supposed to render the Hebrew exactly and not depart from it. And why ‘for some reason’? Could the answer be in doubt? Soisalon-Soininen does not explain further or seem to want to explore why a translator might depart from the strict letter of the original. My third sample is Soisalon-Soininen’s 1977 paper on the relative pronoun.11 The main interest here is in what Soisalon-Soininen calls ‘nominal relative clauses,’ that is, those in which a Hebrew relative clause introduced by ‫ אשׁר‬is translated as a noun phrase with the article, as for example: Gen 43:16 ‫ = יאמר לאשׁר על־ביתו‬he said to the (one) who (was) over his house. καὶ εἶπεν τῷ ἐπὶ τῆς οἰκίας αὐτοῦ = he said to the (one) over his house. Gen 19:11 ‫ = את־האנשׁים אשׁר־פתח הבית‬the men who (were) at the door of the house. τοὺς ἄνδρας τοὺς ὄντας ἐπὶ τῆς θύρας τοῦ οἴκου = the men the (ones) being at the door of the house.

The renderings are natural Greek idiom; the literal equivalent with a relative pronoun would have been possible but less natural. Soisalon-Soininen describes such forms of expression as ‘idiomatic, one might even say stylish from the view-point of the Greek language’ and ‘good Greek style’ (57). His appreciation of these as Greek idiom is thus plainly and well stated. 8

9

10

11

‘[M]üssen wir das Gewicht eben auf diese seltenen Fälle legen’ (Soisalon-Soininen, ‘Gebrauch,’ 180). ‘Es sind aus irgendeinem Grunde die hemmenden Faktoren zurückgetreten’ (SoisalonSoininen, ‘Gebrauch,’ 176). ‘Übersetzungssprache’; ‘Man könnte sagen, es ist kein Griechisch, sondern Hebräisch mit griechischen Worten’ (Soisalon-Soininen, ‘Gebrauch,’ 175). Soisalon-Soininen, ‘Hebrew Relative Clause.’ Page refs. are to Studien; the original is in English.

444

COLLECTED ESSAYS ON THE GREEK BIBLE

But he nevertheless seems to want to push – he does push – the Greek into a closer match to the Hebrew. He does this by an argument that equates the article with the relative pronoun: ‘the [Greek] relative pronoun is replaced by the almost identical article’ (56), and ‘in some cases the form is actually identical’ (57). One may respond that to equate the Greek article and relative pronoun is to ignore their difference of function; and it is not really true that the forms are ‘almost identical.’12 But on this basis Soisalon-Soininen goes on to assert that ‘this type of translation is not to be taken as a significant indication of the liberties taken by the Septuagint translators’ (57). We are thus back to the position that the translator is following his Semitic original exactly and not really departing from it, even when he turns a Hebrew relative clause into a natural Greek equivalent with different syntax – and does depart from it. 2. A DIFFERENT VIEW This brief exploration of three of Soisalon-Soininen’s papers suffices to give an indication of his thinking on the subject of Greek idiom in the LXX. Now my aim here is not to criticise everything he said, or to diminish the value of all the fine work he did. Rather, I wish to bring the focus on to the element that he himself identified, the idiomatic Greek renderings. He and I agree that they are present; how we ought to view them is where we differ. He evidently sees them as somehow unexpected or anomalous, even difficult to explain, in translations presumed to follow the original as exactly as possible. To me this is to approach the matter around the wrong way. The LXX translators, at least in the Pentateuch, were not tied to producing a close, exact translation, which they departed from only when forced to – for some strange reason. Rather, natural Greek, of which they had an intimate knowledge, was their starting-point, and they deviated from it because of their generally literal method. They follow this method with great inconsistency: they are just as likely to produce natural Greek as to produce Hebraic Greek. That is why natural Greek idiom keeps appearing. It may be said, of this or that Hebrew feature, that an idiomatic rendering is statistically infrequent compared with a Hebraistic one, and therefore the former is of little or no significance: the translators’ aim was still to make the translation as literal as possible, and deviations are in the nature 12

The only forms that are the same are ὁ, ἡ, οἱ, αἱ, and these are distinguished by the accent (rel. pron. ὅ, ἥ, οἵ, αἵ). Further, ὅ is neuter nom./acc., so cannot function like masc. ὁ.

BACK TO THE QUESTION OF GREEK IDIOM

445

of anomalies. Again I take a contrary view. That a certain feature of Greek idiom appears at all is significant, all the more so if it involves an unrequired or unforced change, as in the case of a shift of word-order. One instance of that kind is just as telling as many. Whether one agrees with all this or not, the question of the presence of idiomatic Greek in the LXX is a very important one, and I suggest is worth pursuing for its own sake. The aim of this paper is to demonstrate the presence of incontestable cases of Greek idiom in the LXX-Pentateuch. The enclitic pronoun/adjective τις and the particle ἄν will be our examples for this purpose. But we need to begin with sound criteria for identifying Greek idiom.

3. IDENTIFYING GREEK IDIOM The question of Hebrew interference in the Greek of the LXX has been the subject of enduring discussion, in various guises, since the days of Thackeray. But the sort of methodological framework suited to our purpose has not yet been proposed (as far as I know). I suggest that it is helpful for the present purpose to divide all LXX renderings into three categories, as set out below. This is not, of course, a complete system for analysing all features of a rendering: it is aimed at finding renderings free of interference. I recognise also that the categories are broad and the terminology imprecise. But further refinements can follow, and some will emerge in the discussion below. The illustrations all involve Hebrew ‫אישׁ‬ in some combination, and all have been chosen from the Leviticus translator, deliberately so, to show that he has no less control of Greek idiom than his fellow-translators. These are the three categories of rendering that I propose: 1. The rendering matches the Hebrew and is not natural Greek: interference is present. Example: (1) Lev 17:3 ἄνθρωπος ἄνθρωπος τῶν υἱῶν Ἰσραήλ, ὃς ἂν σφάξῃ μόσχον … (4) καὶ λογισθήσεται τῷ ἀνθρώπῳ ἐκείνῳ αἷμα· A man a man of the sons of Israel who slaughters a calf … (4) then bloodshed shall be reckoned to that man. ‫( דם יחשׁב לאישׁ ההוא‬4) ...‫אישׁ אישׁ מבית ישׂראל אשׁר ישׁחט שׁור‬

2. The rendering matches the Hebrew and is consistent with natural Greek: interference may or may not be present. Example:

446

COLLECTED ESSAYS ON THE GREEK BIBLE

(2) Lev 25:17 μὴ θλιβέτω ἄνθρωπος τὸν πλησίον· καὶ φοβηθήσῃ κύριον τὸν θεόν σου. Let a man not oppress his neighbour; and you shall fear the Lord your God. ‫ולא תונו אישׁ את־עמיתו ויראת מאלהיך‬

3. The rendering does not match the Hebrew and is natural Greek: interference is not present. Example: (3) Lev 6:40(7:10) καὶ πᾶσα θυσία ἀναπεποιημένη ἐν ἐλαίῳ καὶ μὴ ἀναπεποιημένη πᾶσιν τοῖς υἱοῖς Ἀαρὼν ἔσται, ἑκάστῳ τὸ ἴσον. And every sacrifice made up with oil or not made up shall be for all the sons of Aaron, to each an equal portion. :‫וכל־מנחה בלולה־בשׁמן וחרבה לכל־בני אהרן תהיה אישׁ כאחיו‬

Brief explanations will be useful. Example (1) shows a rendering that is not acceptable as natural Greek, even though the meaning would not be totally opaque to a Greek speaker. Comprehension would be aided by ‘distributive doubling,’ a known though rare phenomenon in Greek, as well as by context. But a natural idiomatic Greek equivalent would be different, and one is produced by the Leviticus translator himself in Lev 20:2 ἐάν τις ἀπὸ τῶν υἱῶν Ἰσραὴλ … (‫)אישׁ אישׁ מבני ישׁראל‬. Renderings in this first category can be labelled ‘Hebraisms,’ if one wants to use the term. In renderings that fall into category 2 – and they would probably be the great majority of all renderings – the question of interference remains open. Although they match the Hebrew, they are not inconsistent with natural Greek, that is, they are capable of being read as natural Greek, and are therefore not certainly Hebraisms. Example (2) is in fact not an entirely natural Greek form of expression: we would be unlikely to find it in a free Greek text. But one cannot say it would be impossible in terms of Greek grammar, syntax, or word-usage. Renderings of type 3 are by definition one thing and not the other. They are undoubted cases of Greek idiom because they do not match the Hebrew. In example (3) the Hebrew literally means ‘a man as/like his brother.’ The translation departs from a literal rendering and gives an idiomatic Greek equivalent that conveys the sense of the original in different words, literally ‘to each the equal (portion).’ The insertion of τό is Greek idiom (where English uses ‘an’). No better proof of the Leviticus translator’s grip on Greek could be found than this rendering. Some further observations arising from the discussion in the Helsinki Symposium and afterwards may be added here.13 As regards the term 13

Discussion of LXX syntax may advance even in an ice-cream shop or at a tram-stop (thank you Theo van der Louw and Jim Aitken).

BACK TO THE QUESTION OF GREEK IDIOM

447

‘Hebraism,’ I would reserve it for individual renderings of type 1 alone, where natural Greek is by definition not present; I would not use it for those of type 2, where there is nothing wrong with the rendering as Greek. This means that a great many renderings are left open; but methodologically that is sounder than classing everything that matches the Hebrew as a ‘Hebraism.’ Other considerations may come into play to help decide the question. Many of the renderings of type 2 seem to me to be a matter of style. That is, they reflect the style of the original, which like the content is carried over into the translation. It would be surprising if this were not so – unless we think that a Hebrew book can be easily converted into a Greek one. Example (2), as we have noted, may not be the sort of thing a Greek writer would say, but it is acceptable Greek. The reason for the oddity is that the form of expression, like the content, comes from the original. Style is also the answer, in my opinion, to the perennial issue of frequency. If a Hebrew expression rendered by a natural Greek equivalent appears more frequently than in a normal Greek text, is it a ‘Hebraism’? My answer is definitely no. Since each individual instance is not a Hebraism, the frequency only reflects that of the original, so it is a feature of style. Soisalon-Soininen himself expressed it well in his discussion of ἰδού: ‘So, even ἰδοὺ φωνὴ μεγάλη is not a Hebraism per se. But the frequency of ἰδού gives the text a Hebraizing flavour.’14 The term ‘Hebraizing flavour’ is of course inexact, but I am not sure of a better alternative. ‘Stylistic interference’ from the Hebrew original might work and is worth a trial. It is only renderings of type 3 that provide us with undoubted idiomatic Greek. We turn now to examples. 4. THE ENCLITIC PRONOUN/ADJECTIVE ΤΙΣ The enclitic pronoun/adjective τις is found 79 times in the LXXPentateuch.15 A full analysis appears in my book and the data presented here are derived from it.16 First let us consider the matches between τις and the Hebrew original. Hebrew has no exact equivalent, since it does not 14

15

16

‘So ist sogar ἰδοὺ φωνὴ μεγάλη an sich kein Hebraismus. Das sich oft wiederholende ἰδού gibt aber dem Text ein hebraisierendes Kolorit’ (Soisalon-Soininen, ‘Zurück,’ 41). Similarly: ‘… eine wiederholte Verwendung … dem Ganzen ein hebräisches Kolorit geben kann’ (39). Cf. Sollamo, Semiprepositions, 299: ‘… exceptionally frequent usage [of the studied expressions] in the LXX is Hebraistic (a stylistic Hebraism).’ Initial search via Accordance©; readings checked in the Göttingen edition. I do not include Exod 22:9(8) ὅ τι οὖν [sic Gött. ed.], which is an instance of the relative pronoun ὅστις + -οὖν. Lee, Greek of the Pentateuch, 128–37.

448

COLLECTED ESSAYS ON THE GREEK BIBLE

employ enclitics like this at all. The Hebrew word for the idea is ‫אישׁ‬, which has the primary meaning ‘man,’ with a subsidiary function to mean ‘someone, a person.’ The Pentateuch translators do sometimes use the ‘literal’ equivalent ἀνήρ or ἄνθρωπος to render ‫ אישׁ‬meaning ‘someone, a person,’ as in the renderings by ἄνθρωπος in examples (1) and (2) above. Mostly, however, they employ τις or ἕκαστος (‘each’), both of which are a move away from a ‘literal’ rendering in favour of more natural Greek.17 Summary data for τις and the corresponding Hebrew in the Pentateuch are as follows: Total of τις in the Pentateuch: 79 In clauses: 54 In noun phrases: 22 In μή τι: 3 With equivalent in Hebrew: 43 ~ ‫אישׁ‬: 40 ~ ‫אחד‬, ‫מאומה‬: 3 Without equivalent in Hebrew, i.e., added: 36 Position of τις: In same position as Hebrew equivalent: 19 Repositioned: 24

The tabulation includes data on the position of τις in its clause – and this is where the story becomes really interesting. Like other enclitics, τις conforms to certain position rules. The main one is that it tends to gravitate to the second position in its clause, or the ‘second slot’ (to use a term of my own). Position in the second slot is a manifestation of Wackernagel’s Law, best known for its effects on the position of the enclitic personal pronoun.18 For τις I have formulated 11 position rules, based on a close analysis of the 79 occurrences in the Pentateuch, supported by parallels in nonLXX Greek. This list of rules is drawn on here and presented in full in my book.19 The question of interest is how much the position of τις matches that of its equivalent ‫אישׁ‬. Some might expect it to match consistently, being changed only where ‫ אישׁ‬is first word and τις has to be moved, or perhaps not even then. In fact, while the position of τις often matches that of ‫אישׁ‬, it is more often changed (24 times), the new position being of course in 17

18 19

ἕκαστος is found 80 times in the LXX-Pent. and corresponds to ‫ )אשׁה( אישׁ‬or some combination thereof in 77 of them. This word ἕκαστος, too, could be used to demonstrate the presence of idiomatic Greek. See, e.g., Horrocks, Greek, 108–9; Sollamo, ‘Enclitic Personal Pronouns.’ Lee, Greek of the Pentateuch, 129–30.

449

BACK TO THE QUESTION OF GREEK IDIOM

comformity with the Greek position rules. Moreover, in a large proportion of the examples (36 out of 79), τις is added without any match to a word in the original. Why would this be? The answer can only be that the addition of τις is natural Greek idiom. Let us now consider τις in relation to the categories of rendering defined above, with selected examples and parallels. There are no instances of τις in category 1, that is, matches to the Hebrew that are also not natural Greek. In all the cases where τις keeps the same position as ‫אישׁ‬, none are unnatural Greek positions. Where keeping the Hebrew order would have led to an unnatural position, as in example (7) below, the translators avoided it by moving τις to a new position. There are instances of τις in category 2, that is, those in which τις keeps the same position as ‫( אישׁ‬or another Hebrew equivalent), and is in a natural Greek position. According to my analysis there are 19 of these. Two examples of τις in category 2: (4) Exod 16:20 ἀλλὰ κατέλιπόν τινες ἀπ᾽ αὐτοῦ ‫ויותרו אנשׁים ממנו‬ BGU 6.1280.17 (210 BC) ἐὰ]ν δὲ μὴ βεβαιοῖ ἀλλʼ ἐγβάλλῃ τις Διοκλῆν Pl. Theaet. 175c.1 ὅταν δέ γέ τινα αὐτός, ὦ φίλε, ἑλκύσῃ ἄνω, καὶ ἐθελήσῃ τις αὐτῷ ἐκβῆναι (5) Exod 22:5(4) ἐὰν δὲ καταβοσκήσῃ τις ἀγρόν ‫כי יבער־אישׁ שׂדה‬ BGU 14.2367.17 (III BC) ἐὰν δʼ ἐπάναγκες δανείζῃ τις μετα[---] Pl. Symp. 221e.1 εἰ γὰρ ἐθέλει τις τῶν Σωκράτους ἀκούειν λόγων

Theoretically these could be due to interference, but other considerations apply that make this unlikely to be the main reason for the placing of τις. In example (4), the translator has good reason not to reposition τινες before the verb, though this is a standard position in the case of ἐὰν δέ at the head of the clause, as in examples (6) and (7) below: τις cannot be placed immediately after a co-ordinating conjunction such as ἀλλά. In example (5), in which τις is ‘postponed’ till after the verb when it could have gone before, there are two possible factors to consider: (a) Exod 22:5(4) occurs in a rather lengthy string of regulations, many of them expressed in the same form as in example (6) below, that is, ἐὰν δέ τις + verb.20 It is possible the translator placed τις in a different position simply for variety; (b) the second factor is a possible link to historical developments in the placement 20

Exod 21:7, 14, 20, 26, 33; 22:1(21:37), 7(6), 10(9). With the order ἐὰν δέ + verb + τις: Exod 22:5(4), 14(13), 16(15).

450

COLLECTED ESSAYS ON THE GREEK BIBLE

of enclitics. Over time Greek shows a tendency towards relocation of enclitics to a position after the verb.21 As can be seen in the parallel above, this was already happening in the time of Plato. It is possible the translator was unconsciously influenced by this undercurrent. Of course it could have been a little of both factors. At any rate, both these examples are definitely natural Greek and this explains them better than a resort to interference from the Hebrew. This is true of τις throughout the Pentateuch. But it is the instances that fall into category 3 that provide incontestable cases of Greek idiom. In these, τις is repositioned or added, contrary to the Hebrew, while being at the same time natural Greek. Examples of τις in category 3: a. τις corresponds to ‫אישׁ‬, but is repositioned (6) Exod 21:7 ἐὰν δέ τις ἀποδῶται τὴν ἑαυτοῦ θυγατέρα ‫וכי־ימכר אישׁ את־בתו‬ (7) Lev 25:29 ἐὰν δέ τις ἀποδῶται οἰκίαν οἰκητήν ‫ואישׁ כי־ימכר בית מושׁב‬ SB 18.13256.9 (III BC) ἐὰν δέ τις παραθῆται, ἀποτίνειν αὐτόν … Pl. Phaedo 60b.6 ἐὰν δέ τις διώκῃ τὸ ἕτερον …

b. τις is added, without correspondence in the Hebrew (8) Gen 6:5 καὶ πᾶς τις (‘everyone’) διανοεῖται ἐν τῇ καρδίᾳ αὐτοῦ ‫וכל־יצר מחשׁבת לבו‬ UPZ 1.60.17 (179–168 BC) ἀλλὰ πᾶς τις πειρᾶται Pl. Theaet. 178a.5 ἔτι τοίνυν ἐνθένδε ἂν μᾶλλον πᾶς τις ὁμολογήσειεν

In (6) the translator moves τις to the standard position in the second slot, where the Hebrew equivalent ‫ אישׁ‬came later, after the verb. There is another word there as well, the particle δέ, and τις is placed after it. This order is not accidental: there are word-order rules within the second slot. (More will be said on these under ἄν.) In example (7) the position of ‫אישׁ‬ could not have been kept without breaking one of the basic rules for τις, that it is not placed first in its clause. Even καί τις (~‫ )ואישׁ‬would not have been idiomatic. In (8) τις is just a natural addition to πᾶς in the manner of idiomatic Greek since Classical times. So τις is a good example of Greek idiom in the LXX-Pentateuch, indubitably in the cases in category 3 and potentially in all cases in category 2. 21

See Horrocks, Greek, 108–9.

BACK TO THE QUESTION OF GREEK IDIOM

451

We may note again the point made earlier that the mere choice of τις as opposed to a ‘literal’ equivalent of ‫ אישׁ‬is a step towards more idiomatic Greek. 5. THE

PARTICLE ἌΝ

The particle ἄν occurs just over 400 times in the Greek Pentateuch.22 It is an entirely Greek feature without any counterpart in Hebrew, and therefore an instance of pure Greek idiom. It occurs in the LXX as a component of various clause types, as required by Greek syntax. These are familiar to us, even if we do not always notice ἄν or realise that it is not there by accident. Most of these clause types are illustrated in the examples below; parallels in contemporary (III BC) papyri are added. The ‘indefinite construction’ in temporal and relative clauses is common, as in examples (10), (13), (14), (15), (18). In examples (9) and (17) we have the stylish addition of ἄν to ὅπως, a feature often seen in Ptolemaic officialese. In (11) and (16) there is the highly idiomatic Greek use of ἄν with a verb in the indicative to convey an ‘unfulfilled’ condition (‘I would have sent you …’). In example (12) we see the rare use of ὡς ἄν + optative in a simile. (9) Gen 12:13 ὅπως ἂν εὖ μοι γένηται BGU 8.1738.32 (72 BC) ὁπηνίκʼ ἂν εὖ συνθῶμεν (10) Gen 21:6 ὃς γὰρ ἂν ἀκούσῃ PCairZen 1.33.7 ἐπόρισεγ γὰρ ἂν αὐτοῖς (11) Gen 31:27 ἐξαπέστειλα ἄν σε μετ᾽ εὐφροσύνης PMich 1.29.3 ἀπέστ[ειλα] | ἄν ∖σοι∕ αὐτήν (12) Gen 33:10 εἶδον τὸ πρόσωπόν σου, ὡς ἄν τις ἴδοι πρόσωπον θεοῦ PCairZen 1.93.18 κέχρηται ἡμῖν ὡς ἂν εἴ τις ἐχθρῶι χρήσαιτο. (13) Exod 32:34 ᾗ δ᾽ ἂν ἡμέρᾳ ἐπισκέπτωμαι PHib 1.29.10 ἧι δʼ ἂν ἡμ[έρ]αι (14) Lev 13:51 κατὰ πάντα ὅσα ἂν ποιηθῇ PCairZen 5.815.4 καὶ ὅσα ἂν εὑρεθῆι (15) Lev 16:17 ἕως ἂν ἐξέλθῃ PPetr 2.3a.1 διὸ ἐπέχω ἕως ἂν ἐξέλθηι (16) Num 22:29 εἰ εἶχον μάχαιραν … ἤδη ἂν ἐξεκέντησά σε PCairZen 4.599.8 καὶ εἰ μὴ … ἐπεβοήθει … πάλαι ἂν ἠσχημόνουν (17) Num 16:40(17:5) ὅπως ἂν μὴ προσέλθῃ μηθείς PLond 7.2033.6 ὅπως ἂν μὴ κατέχηται ὁ Καλλικῶν 22

In the current Göttingen text 394 of ἄν (as counted by Accordance©) + 12 of ἐάν = particle ἄν. [Cf. discussion of ἄν now in Lee, Greek of the Pentateuch, 137–9.]

452

COLLECTED ESSAYS ON THE GREEK BIBLE

(18) Deut 12:26 ἃ ἄν σοι γένηται PCairZen 3.358.3 ὅσα ἄν σοι δόξῃ

These uses of ἄν in the Pentateuch are interesting enough as instances of Greek idiom; but there is more. This particle exhibits word-order behaviour very similar to that of τις, even though it is not an enclitic. Like τις it gravitates to the second slot in its clause, but it has an even stronger tendency to do so than τις; it also precedes some words that τις does not, and τις itself. The rules can be seen played out in our selection: ἄν follows the particles γάρ (10) and δέ (13), but otherwise comes immediately after the headword of its clause, preceding the enclitic pronouns σε (11), σοι (18), the negative μή (17), and τις (12). These word-order rules can be traced back as far as Homer.23 This of course does not mean that the LXX translators learnt them from Homer; it means that the patterns were established in early Greek and continued to operate all the way to the time of the translators.

6. CONCLUSIONS The aim of identifying Greek idiom in the LXX-Pentateuch can be readily achieved. As we have seen, there are undoubted instances according to the criteria we have established. In the case of ἄν, every use falls into this category. In the case of τις, we have found clear instances in which Hebrew interference cannot be involved (60 in all); in the rest the possibility of interference remains open according to our criteria, but other factors apply that can tip the balance against interference. The approach through three defined categories of rendering (§3) provides a method applicable to all cases of potential interference, in effect to all renderings in the LXX. There are three further conclusions that I think are warranted. First, the evidence of idiomatic Greek woven into the fabric of the Greek Pentateuch demonstrates the translators’ complete competence in Greek. If one also takes into account other evidence beyond τις and ἄν – though these are significant enough – it is not too much to assert that the Pentateuch translators had native-speaker competence.24 It is hard to see what other explanation there is for their intimate knowledge of the usage of τις and ἄν, 23 24

See Ruijgh, ‘La place des enclitiques,’ 213–33. Further evidence supporting this conclusion is offered in Lee, Greek of the Pentateuch, esp. ch. 4 (123–72).

BACK TO THE QUESTION OF GREEK IDIOM

453

including the word-order patterns. We can say further that this observation applies to all the translators of the Pentateuch, however many there were. I believe that there were five, but the point is not affected even if this is not true: the spread of τις and ἄν through the Pentateuch shows that they were all equally competent in Greek. This is not the same as competence as translators. Nor do these two features enable us to characterise the techniques of individual translators: as far as I can see, they all use τις and ἄν in the same way. A second conclusion is that the presence of Greek idiom as in τις and ἄν supports the contention that the Greek translation of the Pentateuch should be viewed as in essence Greek with Hebrew interference, rather than as Hebraic Greek into which idiomatic Greek occasionally intrudes, for no evident reason. This is to some extent just a matter of viewpoint, but there is a genuine issue underlying it. What we expect to see will have an impact on what we find, when we try to analyse the Greek of the Pentateuch and the translators’ methods. Soisalon-Soininen’s approach to ‘nominal relative clauses’ might be seen as an illustration (see §1). A third conclusion relates to Koine Greek as a whole. Our usual approach to the Greek of the LXX, founded on a long-standing assumption that there is something anomalous about it and every feature needs to be proved to be acceptable Greek, has been to pursue a quest for parallels outside the LXX to provide the proof. What we have found in τις and ἄν in the Pentateuch, that is, normal usage of characteristically Greek idiom, suggests that the Pentateuch itself has value as evidence for Koine Greek (with of course the usual cautions and provisos). In regard to the uses of τις and ἄν that are already well-attested, as in those presented above, the Pentateuch has little new to offer. But things are different in the case of another use of ἄν, the ‘iterative’ use with indicative in subordinate clauses, as seen in this example: (19) Exod 33:8 ἡνίκα δ᾽ ἂν εἰσεπορεύετο Μωυσῆς εἰς τὴν σκηνήν. … And whenever Moyses went into the tent. … … ‫והיה כצאת משׁה אל־האהל‬ (Lit.) And it happened at the going out of Moses to the tent. …

Though this new Koine Greek construction is well attested later, there are no examples in third-century BC documents and there is a general lack of evidence for it in early Koine Greek.25 The Pentateuch, with some 15 instances, actually provides the (so far) missing proof that it had begun 25

See Mayser, Grammatik, II.1, 295; II.3, 79–80, 94; Muraoka, Syntax, 773.

454

COLLECTED ESSAYS ON THE GREEK BIBLE

in the third century BC. As can be seen from the example, it has nothing to do with the Hebrew original, nor can there be any reason to think the translators were using something that was not current Greek. To conclude. The study of Hebraisms has its place, but rather than go back to the question of Hebraisms I would like to see us go forward to the question of Greek idiom in the LXX. Our evidence has shown that Greek idiom is undoubtedly present in the LXX-Pentateuch, but the pursuit of such evidence has been slow up till now. We could now turn our attention to the many ways in which the translators followed Greek idiom, rather than the ways in which they departed from it.

BIBLIOGRAPHY ABBOTT-SMITH, G., A Manual Greek Lexicon of the New Testament (Edinburgh, 1922; 2nd ed. 1923; 3rd ed. 1937). ADAMS, J. N., Bilingualism and the Latin Language (Cambridge, 2003). ADAMS, S. A., ‘Atticism, Classicism, and Luke–Acts: Discussions with Albert Wifstrand and Loveday Alexander,’ in Porter and Pitts (eds.), The Language of the New Testament (2013), 91–111. ADRADOS, F. R., ‘Sobre los orígenes del vocabulario ático. II,’ Emerita 25 (1957), 81–121. AEJMELAEUS, A., ‘Translation Technique and the Intention of the Translator,’ in C. E. Cox (ed.), VII Congress of the International Organization for Septuagint and Cognate Studies, Leuven 1989 (Atlanta, 1989), 23–36. AITKEN, J. K., ‘Phonological Phenomena in Greek Papyri and Inscriptions and their Significance for the Septuagint,’ in J. Corley and V. Skemp (eds.), Studies in the Greek Bible: Essays in Honor of Francis T. Gignac, SJ (CBQMS 44; Washington DC, 2008), 256–77. — ‘The Literary Attainment of the Translator of Greek Sirach,’ in J.-S. Rey and J. Joosten (eds.), Texts and Versions of the Book of Ben Sira: Transmission and Interpretation (Leiden, 2011), 95–126. — ‘The Characterization of Speech in the Septuagint Pentateuch,’ in D. J. A. Clines and J. C. Exum (eds.), The Reception of the Hebrew Bible in the Septuagint and the New Testament: Essays in Memory of Aileen Guilding (Sheffield, 2013), 9–31. — No Stone Unturned: Greek Inscriptions and Septuagint Vocabulary (Winona Lake, 2014). — ‘The Origins of καί γε,’ in Aitken and Evans (eds.), Biblical Greek in Context (2015), 21–40. — (ed.), The T&T Clark Companion to the Septuagint (London, 2015). AITKEN, J. K., and T. V. EVANS (eds.), Biblical Greek in Context: Essays in Honour of John A. L. Lee (Leuven, 2015). ALAND, K., Vollständige Konkordanz zum griechischen Neuen Testament, II (Berlin, 1978). ALAND, K., and B. ALAND, The Text of the New Testament (trans. Erroll F. Rhodes; Grand Rapids, 1987). ALLEN, W. C., S. Matthew (ICC; Edinburgh, 1912). ANDERSEN, F. I., ‘Advances in the Lexicography of Biblical Languages: A Review Essay,’ ANES 45 (2008), 235–59. ANDERSON, G., The Second Sophistic: A Cultural Phenomenon in the Roman Empire (London, 1993). ANZ, H., Subsidia ad cognoscendum graecorum sermonem vulgarem e Pentateuchi versione alexandrina repetita (Diss. Phil. Halenses, 12.2; Halle, 1894). ARGYLE, S., ‘A New Greek Grammarian,’ CQ 39 (1989), 524–35.

456

COLLECTED ESSAYS ON THE GREEK BIBLE

ARGYRIADES, G. P., Νεοελληνική Γλώσσα· Ιστορικές και Γλωσσολογικές Διαστάσεις (Θεσσαλονίκη, 1990). ARNOLD, B. T., ‘Luke’s Characterizing Use of the Old Testament in the Book of Acts,’ in B. Witherington III (ed.), History, Literature, and Society in the Book of Acts (Cambridge, 1996), 300–23. ARNOLD, W. R., ‘The Meaning of ‫בתרון‬,’ AJSL 28 (1912), 274–83. ARNOTT, W. G., ‘The Confrontation of Sostratos and Gorgias,’ Phoenix 18 (1964), 110–23. — ‘A Note on the Antiatticist (98.17 Bekker),’ Hermes 117 (1989), 374–6. ARRIBAS ARRANZ, F., Paleografía documental hispánica, I. Láminas; II. Transcripciones (Vallodolid, 1965). AUWERS, J.-M., Concordance du Siracide (Grec II et Sacra Parallela) (CahRB, 58; Paris, 2005). BABINIOTES, G. D., Λεξικό της Νέας Ελληνικής Γλώσσας (Athens, 2002). BAER, D. A., When We All Go Home: Translation and Theology in LXX Isaiah 56– 66 (Sheffield, 2001). — ‘What Happens in the End? Evidence for an Early Greek Recension in LXX Isaiah 66,’ in van der Kooij and van der Meer (eds.), Old Greek of Isaiah (2010), 1–31. BAGNALL, R. S. (ed.), The Oxford Handbook of Papyrology (Oxford, 2009). BAILLY, A., Dictionnaire Grec-Français (Paris, 1894; rev. eds. 1950, 1963). BARDY, G., ‘Basilius von Caesarea,’ RAC 1 (1950), 1263. BARKER, N., Aldus Manutius and the Development of Greek Script and Type in the Fifteenth Century (Sandy Hook, Conn., 1985). BARR, J., Comparative Philology and the Text of the Old Testament (Oxford, 1968). — The Typology of Literalism in Ancient Biblical Translations (Göttingen, 1979). BARRETT, C. K., The Gospel According to St John (London, 1955). — The Acts of the Apostles (ICC; Edinburgh 1994, 1998). BARTHÉLEMY, D., Les devanciers d’Aquila (VTSuppl. 10; Leiden, 1963). BATAILLON, M., Érasme et l’Espagne (1st ed., Paris, 1937. Nouvelle édition en trois volumes, ed. D. Devoto and C. Amiel. Genève, 1991). BAUER, H., and P. LEANDER, Historische Grammatik der hebräischen Sprache des Alten Testamentes (Halle, 1922). BEENTJES, P. C., The Book of Ben Sira in Hebrew: A Text Edition of All Extant Hebrew Manuscripts and a Synopsis of All Parallel Hebrew Ben Sira Texts (Leiden, 1997). BEKKER, I., Anecdota Graeca (3 vols.; Berlin, 1814–1821). BENTEIN, K., ‘Perfect Periphrases in Post-Classical and Early Byzantine Greek: An Ecological-Evolutionary Account,’ Journal of Greek Linguistics 12 (2012), 205–75. — ‘Particle-usage in Documentary Papyri (I–IV AD): An Integrated, Sociolinguistically Informed Approach,’ GRBS 55 (2015), 721–53. — ‘Dimensions of Social Meaning in Post-Classical Greek,’ JGL 19 (2019), 119–67. BENTLEY, J. H., Humanists and Holy Writ: New Testament Scholarship in the Renaissance (Princeton, 1983).

BIBLIOGRAPHY

457

BÉRARD, F., et al., Guide de l’épigraphiste: Bibliographie choisie des épigraphies antiques et médiévales (3rd ed.; Paris, 2000). BETHE, E. (ed.), Pollucis Onomasticon (2 vols.; Lexicographi Graeci 9:1–2; Leipzig, 1900, 1931). — ‘Pollux,’ in Pauly-Wissowa, RE 10:773–9. BETZ, H. D., Galatians (Hermeneia; Philadelphia, 1979). BLACK, M., An Aramaic Approach to the Gospels and Acts (3rd ed.; Oxford, 1967). BLOMQVIST, J., Greek Particles in Hellenistic Prose (Lund, 1969). BOISACQ, É., Dictionnaire étymologique de la langue grecque (Paris, 1916). BOLING, R. G., Judges (New York, 1975). BOLINGER, D., Aspects of Language (2nd ed.; New York, 1975). BOWIE, E. L., ‘Greeks and Their Past in the Second Sophistic,’ Past and Present 46 (1970), 3–41. BOYD-TAYLOR, C., ‘A Place in the Sun: The Interpretative Significance of LXXPsalm 18:5c,’ BIOSCS 31 (1998), 71–105. — ‘The Evidentiary Value of Septuagintal Usage for Greek Lexicography: Alice’s Reply to Humpty Dumpty,’ BIOSCS 34 (2001), 47–80. — ‘Lexicography and Interlanguage – Gaining our Bearings,’ BIOSCS 37 (2004), 55–72. — Reading Between the Lines: The Interlinear Paradigm for Septuagint Studies (BTS 8; Leuven, 2011). — ‘The Classification of Literalism in Ancient Hebrew–Greek Translation,’ in LXX.H, 3:139–60. BOYD-TAYLOR, C. (ed.), A Question of Methodology: Albert Pietersma, Collected Essays on the Septuagint (BTS 14; Leuven, 2013). BRANDWOOD, L., A Word Index to Plato (Leeds, 1976). BRATCHER, R. G., and E. A. NIDA, A Translator’s Handbook on the Gospel of Mark (Leiden, 1961). BRAUN, A., ‘Nota sui verbi greci del “volere”,’ AIV 98 (1938/9), 337–55. BREITENSTEIN, U., Beobachtungen zu Sprache, Stil und Gedankengut des Vierten Makkabäerbuchs (Basel, 1978). BROCK, S. P., ‘The Phenomenon of the Septuagint,’ OtSt 17 (1972), 11–36. BROWN, R. E., The Gospel According to John: Introduction, Translation and Notes (AB; 2 vols.; Garden City, 1966–1970). BROWNE, G. M., Sortes Astrampsychi, Volumen I: Ecdosis Prior (Leipzig, 1983). — ‘The Origin and Date of the Sortes Astrampsychi,’ Illinois Classical Studies 1 (1976), 53–8. BROWNING, R., Medieval and Modern Greek (2nd ed.; Cambridge, 1983). BRUCE, F. F., The Epistle of Paul to the Galatians (NIGTC; Exeter, 1982). BÜCHNER, D. L., Review of LEH, Lexicon, BIOSCS 37 (2004), 139–47. BUCK, C. D., ‘Greek ἄμϕοδον, Oscan amvíanud, and the Oscan eítuns-inscriptions,’ CP 17 (1922), 111–8. — A Dictionary of Selected Synonyms in the Principal Indo-European Languages (Chicago, 1949). BUCK, C. D., and W. Petersen, A Reverse Index of Greek Nouns and Adjectives (Hildesheim, 1970). BURGUIÈRE, P., Histoire de l’infinitif en grec (Paris, 1960). BURNEY, C. F., The Aramaic Origin of the Fourth Gospel (Oxford, 1922).

458

COLLECTED ESSAYS ON THE GREEK BIBLE

BURRIDGE, R. A., ‘The Gospels and Acts,’ in Porter (ed.), Handbook (1997), 507– 32. BUTLER, Dom C., The Lausiac History of Palladius (2 vols.; Cambridge, 1898, 1904). CADBURY, H. J., The Style and Literary Method of Luke (Cambridge, Mass., 1920). CADELL, H., ‘Vocabulaire de l’irrigation: la Septante et les papyrus’ in B. Menu (ed.), Les problèmes institutionnels de l’eau en Égypte ancienne et dans l’antiquité méditerranéenne (Cairo, 1994), 103–17. — ‘Vocabulaire de la législation ptolemaïque: problème du sens de dikaiôma dans le Pentateuque,’ in Dorival and Munnich (eds.), Κατὰ τοὺς Οʹ (1995), 207–21. CAIRD, G. B., ‘Towards a Lexicon of the Septuagint. I,’ JTS n.s. 19 (1968), 453– 75. — ‘Towards a Lexicon of the Septuagint. II,’ JTS n.s. 20 (1969), 21–40. CARSON, D. A., Exegetical Fallacies (2nd ed.; Grand Rapids, MI, 1996). CASE, S. J., ‘The Use of ἄμϕοδον in the Septuagint and the New Testament,’ in Buck, ‘Greek ἄμϕοδον,’ 117 n. CAVALLO, G., and H. MAEHLER, Greek Bookhands of the Early Byzantine Period, A.D. 300–800 (ICS Bulletin Suppl. 47; London, 1987). CHADWICK, J., ‘Semantic History and Greek Lexicography,’ in F. Létoublon (ed.), La langue et les textes en grec ancien: Actes du Colloque Pierre Chantraine (Grenoble – 5-8 Septembre 1989) (Amsterdam, 1992), 281–8. — ‘The Case for Replacing Liddell and Scott,’ BICS 39 (1994), 1–11. — Lexicographica Graeca: Contributions to the Lexicography of Ancient Greek (Oxford, 1996). CHANTRAINE, P., La formation des noms en grec ancien (Paris, 1933). — ‘Les mots designant la gauche en grec ancien,’ in H. Kronasser (ed.), Μνήμης χάριν: Gedenkschrift Paul Kretschmer (2 vols.; Vienna, 1956–1957), 1:61– 9. — Grammaire homérique (2 vols.; Paris, 1958, 1963). CHARLES, R. H., The Revelation of St. John (ICC; Edinburgh, 1920). COBET, C. G., Variae lectiones (Leiden, 1873). COHN, L., ‘Griechische Lexikographie,’ in K. Brugmann, Griechische Grammatik, 4th ed., rev. Albert Thumb (München, 1913), 679–730. — ‘Ammonios,’ in Pauly-Wissowa, RE 1:1866–7. COLLARD, C., M. J. CROPP, and K. H. LEE / J. GIBERT, Euripides: Selected Fragmentary Plays, with Introductions, Translations and Commentaries (2 vols.; Warminster, 1995, 2004). CONNOLLY, A. L., ‘Atticism in Non-literary Papyri of the First Seven Centuries AD: A Study in Several Features of Orthography and Syntax’ (BA diss., University of Sydney, 1983). CONTOPOULOS, N., Νέον Λεξικόν Ελληνοαγγλικόν (3rd ed.; London, 1887). COSTACHE, D., and P. KARIATLIS (eds.), Cappadocian Legacy: A Critical Appraisal (Sydney, 2013). COSTAS, P. S., An Outline of the History of the Greek Language: With Particular Emphasis on the Koine and the Subsequent Stages (Chicago, 1936). [CRASTONUS, Joannes,] Dictionarium graecum copiosissimum secu[n]dum ordinem alphabeti cum interpretatione latina (Venetiis, 1497).

BIBLIOGRAPHY

459

CRIBIORE, R., Gymnastics of the Mind: Greek Education in Hellenistic and Roman Egypt (Princeton, 2001). CROSS, F. M., and S. TALMON, Qumran and the History of the Biblical Text (Cambridge, MA, 1975). CUNNINGHAM, I. C. (ed.), Συναγωγὴ Λέξεων Χρησίμων (Berlin, 2003). CURTIUS, G., Grundzüge der griechischen Etymologie (1st ed., Leipzig, 1858–1862; 5th ed., Leipzig, 1879). DAIN, A. (ed.), Le ‘Philétaeros’ attribué à Hérodien, (Paris, 1954). DANIEL, S., Recherches sur le vocabulaire du culte dans le Septante (Paris, 1966). DANKER, F. W., Multipurpose Tools for Bible Study (Minneapolis, 1993). — The Concise Greek-English Lexicon of the New Testament (Chicago, 2009). — Review of Muraoka, Lexicon (2009), RBL 04/2010. DARDANO, M., and P. TRIFONE, La lingua italiana (Bologna, 1985). DARIS, S., ‘I Quartieri di Ossirinco: Materiali e Note,’ ZPE 132 (2000), 211– 21. DARLOW, T. H., and H. F. MOULE, Historical Catalogue of the Printed Editions of Holy Scripture in the Library of the British and Foreign Bible Society (2 vols.; London, 1911). DAVIES, W. D., and D. DAUBE (eds.), The Background of the New Testament and its Eschatology: Studies in Honour of C. H. Dodd (Cambridge, 1956). DE BORRIES, J. (ed.), Phrynichi Sophistae Praeparatio Sophistica (Leipzig, 1911). DEBRUNNER, A., and A. SCHERER, Geschichte der griechischen Sprache (2 vols.; 2nd ed. Berlin, 1969). DECKER, R. J., ‘Markan Idiolect in the Study of the Greek of the New Testament,’ in Porter and Pitts (eds.), Language of the New Testament (2013), 43–66. DEFERRARI, R. J. (trans.), Basil: Letters 1–58 (LCL 190; Cambridge, MA, 1926). — Basil: Letters 249–368 (LCL 270; Cambridge, MA, 1934). DEFERRARI, R. J., and M. R. P. MCGUIRE (trans.), Basil: Address to Young Men on Reading Greek Literature (LCL 270; Cambridge, MA, 1934). DEISSMANN, A., Bible Studies: Contributions Chiefly from Papyri and Inscriptions to the History of the Language, the Literature, and the Religion of Hellenistic Judaism and Primitive Christianity (trans. A. Grieve; 2nd ed.; Edinburgh, 1909). — Light from the Ancient East (trans. L. R. M. Strachan; New York, 1927). DE JONGE, H. J., ‘De nieuwe Nestle: N26,’ NedTT 34 (1980), 307–22. — Opera Omnia Desiderii Erasmi Roterodami, IX.2, Apologia respondens ad ea quae Iacobus Lopis Stunica taxaverat in prima duntaxat novi testamenti aeditione (Amsterdam, 1983). DENNISTON, J. D., The Greek Particles (2nd ed.; Oxford, 1954). DE SOUSA, R. F., Eschatology and Messianism in LXX Isaiah 1–12 (New York, 2010). DE TROYER, K., T. M. LAW, and M. LILJESTRÖM (eds.), In the Footsteps of Sherlock Holmes: Studies in the Biblical Text in Honour of Anneli Aejmelaeus (CBET 72; Leuven, 2014). DHONT, M., ‘Multicausality in Septuagint Studies’ [forthcoming in JSCS]. DICKEY, E., Ancient Greek Scholarship: A Guide to Finding, Reading, and Understanding Scholia, Commentaries, Lexica, and Grammatical Treatises, from Their Beginnings to the Byzantine Period (Oxford, 2007).

460

COLLECTED ESSAYS ON THE GREEK BIBLE

DIETERICH, K., Untersuchungen zur Geschichte der griechischen Sprache: von der hellenistischen Zeit bis zum 10. Jahrhundert n. Chr (Leipzig, 1898. Repr. Hildesheim, 1970). DINES, J. M., The Septuagint (London, 2004). DOGNIEZ, C., and M. HARL, La Bible d’Alexandrie: Le Deutéronome (Paris, 1992). DORIVAL, G., La Bible d’Alexandrie: Les Nombres (Paris, 1994). DRIVER, G. R., ‘Linguistic and Textual Problems: Jeremiah,’ JQR 28 (1937), 97–129. DU BOUCHET, J., ‘Quartiers, Îlots et Rues: Remarques sur ἄμϕοδον,’ RPhil 78 (2004), 43–55. EBELING, H., Griechisch-deutsches Wörterbuch zum Neuen Testamente (Hannover, 1913; 3rd ed. 1929). EDMONDS, J. M., The Fragments of Attic Comedy: After Meineke, Bergk, and Kock (3 vols. in 4; Leiden, 1957–1961). EISSFELDT, O., ‘Ein gescheiterter Versuch der Wiedervereinigung Israels,’ La Nouvelle Clio 3 (1951), 110–27. ELLIOTT, J. K., ‘The Text and Language of the Endings of Mark’s Gospel,’ TZ 27 (1971), 255–62. — ‘The United Bible Societies Greek New Testament: An Evaluation,’ NovT 15 (1973), 278–300. — Essays and Studies in New Testament Textual Criticism (Córdoba, 1992). ELLIS, J., and J. N. URE, ‘Language Varieties: Register,’ in A. R. Meetham (ed.), Encyclopedia of Linguistics, Information and Control (Oxford, 1969). EPP, E. J., and G. D. FEE, Studies in the Theory and Method of New Testament Textual Criticism (Grand Rapids, 1993). ERASMUS, Desiderius, De recta latini graecisque sermonis pronuntiatione (Basle, 1528. Facsimile ed., Menston, 1971). ERBSE, H., Untersuchungen zu den Attizistischen Lexika: Aelii Dionysii et Pausaniae atticistarum fragmenta (AAWB 1949.2:95–221; Berlin, 1950). EVANS, T. V., Verbal Syntax in the Greek Pentateuch: Natural Greek Usage and Hebrew Interference (Oxford, 2001). — ‘The Court Function of the Interpreter in Genesis 42:23 and Early Greek Papyri’ in J. K. Aitken, J. Dines, S. Pearce, and T. Rajak (eds.), Jewish Perspectives on Hellenistic Kingship (Berkeley, 2007), 238–52. — ‘Numbers,’ in Aitken (ed.), T&T Clark Companion (2015), 58–67. EVANS, T. V., and D. D. OBBINK (eds.), The Language of the Papyri (Oxford, 2010). EXLER, F. X. J., The Form of the Ancient Greek Letter (Washington, 1923). FABRICIUS, C., ‘Der sprachliche Klassizismus der griechischen Kirchenväter. Ein philologisches und geistesgeschichtliches Problem,’ JAC 10 (1967), 187– 99. FARMER, W. R., The Last Twelve Verses of Mark (Cambridge, 1974). FEE, G. D., Papyrus Bodmer II (P66): Its Textual Relationships and Scribal Characteristics (Studies and Documents 34; Salt Lake City, 1968). — ‘Rigorous or Reasoned Eclecticism—Which?,’ in J. K. Elliott (ed.), Studies in New Testament Language and Text: Essays in Honour of G. D. Kilpatrick (NT Suppl. 44; Leiden, 1976), 174–97. — The First Epistle to the Corinthians (NICNT; Grand Rapids, 1987).

BIBLIOGRAPHY

461

FERNÁNDEZ-ARMESTO, F., ‘Cardinal Cisneros as a Patron of Printing,’ in D. W. Lomax and D. Mackenzie (eds.), God and Man in Medieval Spain: Essays in Honour of J. R. L. Highfield (Warminster, 1989), 149–68. FESTUGIÈRE, A.-J., Historia Monachorum in Aegypto (Subsidia Hagiographica 34; Brussels, 1961). FEYERABEND, K., Langenscheidt’s Pocket Greek Dictionary: Classical Greek– English, (Berlin, 1910). FISCHER, E. (ed.), Die Ekloge des Phrynichos (SGLG 1; Berlin, 1974). FITZMYER, J. A., ‘The Study of the Aramaic Background of the New Testament’ (1975), in A Wandering Aramean: Collected Aramaic Essays (SBLMS 25; Missoula, MO, 1979), 1–27. — The Gospel According to Luke I-IX (AB; New York, 1981). FLEISCHMAN, S., The Future in Thought and Language: Diachronic Evidence from Romance (Cambridge, 1982). FOX, W., ‘βούλεσθαι und (ἐ)θέλειν,’ Berliner Philologische Wochenschrift 37 (1917), 597–606, 633–9. FRAME, J. E., Thessalonians (ICC; Edinburgh, 1912). FRÄNKEL, H., Einleitung zur kritischen Ausgabe der Argonautika des Apollonios (Göttingen, 1964). FRÖSÉN, J., Prolegomena to a Study of the Greek Language in the First Centuries A.D.: The Problem of Koine and Atticism (Helsinki, 1974). GALLAY, P. (ed.), Saint-Grégoire de Nazianze: Lettres (2 vols.; Paris, 1964). GARCÍA SANTOS, A. A., Diccionario del Griego Bíblico: Setenta y Nuevo Testamento (Estella, Navarra, 2011). GARLAN, Y., ‘Cités, armées et stratégie à l’époque hellénistique d’après l’oeuvre de Philon de Byzance,’ Historia 22 (1973), 16–33. — Recherches de poliorcétique grecque (Athens, 1974). GEANAKOPLOS, D. J., Byzantium and the Renaissance: Greek Scholars in Venice (Harvard, 1962; repr. Hamden, 1973). GEHMAN, H. S., ‘Adventures in Septuagint Lexicography,’ Textus 5 (1966), 125–32. GERLEMAN, G., Studies in the Septuagint I: The Book of Job (Lund, 1946). GIANNAKOPOULOS, I., Ἡ Παλαιὰ Διαθήκη κατὰ τοὺς Ο´ (26 vols.; 4th ed.; Thessaloniki, 1955–68). GIGNAC, F. T., A Grammar of the Greek Papyri of the Roman and Byzantine Periods (2 vols.; Milan, 1976, 1981). GINGRICH, F. W., Shorter Lexicon of the Greek New Testament (Chicago, 1965; 2nd ed. rev. F. W. Danker, 1983). GLARE, P., ‘Starting from the Wrong End,’ in G. Cigman and D. Howlett (eds.), Birthday Celebration for Naky Doniach (Yarnton, Oxford, 1991), 35–41 GOETZ, G., and GUNDERMANN, G., Glossae Latinograecae et Graecolatinae, in G. Loewe and G. Goetz (eds.), Corpus Glossariorum Latinorum (vol. 2; Leipzig, 1888). GOMME, A. W., and F. H. SANDBACH, Menander: A Commentary (Oxford, 1973). GOODSPEED, E. J., Index Patristicus (Leipzig, 1907). GRAY, G. B., Numbers (ICC; Edinburgh, 1912). GREENLEE, J. H., A New Testament Greek Morpheme Lexicon (Grand Rapids, MI, 1983). GREGORY, C. R., Textkritik des Neuen Testamentes (3 vols.; Leipzig, 1900–1909).

462

COLLECTED ESSAYS ON THE GREEK BIBLE

GREGORY, M., ‘Aspects of Varieties Differentiation,’ JLing 3 (1967), 177–98. GREGORY M., and S. CARROLL, Language and Situation: Language Varieties and their Social Contexts (London, 1978). GUY, J.-C., Recherches sur la tradition grecque des Apophthegmata Patrum (Subsidia Hagiographica 36; Brussels, 1962; repr. with suppl. 1984). — Les Apophtegmes des Pères du Désert: Série alphabétique: Traduction française (Paris, 1967–1968). — Les Apophtegmes des Pères: Collection systématique (3 vols.; Paris, 1993– 2005). HAENCHEN, E., The Acts of the Apostles (ET; Oxford, 1971). HAHN, E. A., Naming-Constructions in Some Indo-European Languages (APA Monograph 27; Cleveland, 1969). HALL, B., ‘The Trilingual College of San Ildefonso and the Making of the Complutensian Polyglot Bible,’ in G. J. Cuming (ed.), Studies in Church History (Leiden, 1969), 5:114–46. HANSEN, D. U. (ed.), Das attizistische Lexikon des Moeris: Quellenkritische Untersuchung und Edition (SGLG 9; Berlin, 1998). HARL, M., La Bible d’Alexandrie: La Genèse (Paris, 1986). HARL, M., G. DORIVAL, and O. MUNNICH, La Bible grecque des Septante: du judaïsme hellénistique au christianisme ancien (Paris, 1988). HARRISON, N. V., Saint Basil the Great: On the Human Condition (Crestwood, 2005). HART, J. H. A., Ecclesiasticus: The Greek Text of Codex 248, edited with a Textual Commentary and Prolegomena (Cambridge, 1909). HATZIDAKIS, G. N., Einleitung in die neugriechische Grammatik (Leipzig, 1892). — Μεσαιωνικὰ καὶ νέα Ἑλληνικά (2 vols.; Athens, 1905, 1907. Repr. Amsterdam, 1989, 1990). HAUSPIE, K., ‘The LXX Quotations in the LSJ Supplements of 1968 and 1996,’ in Taylor et al. (eds.), Language and Lexicography (2004), 108–25. HELBING, R., Die Kassusyntax der Verba bei den Septuaginta (Göttingen, 1928). HEMER, C. J., ‘Towards a New Moulton and Milligan,’ NovT 24 (1982), 97–123. HENDERSON, J., ‘A1–Zythum: Domimina Nustio Illumea, or Out with the OLD (1931–82),’ in C. Stray (ed.), Classical Dictionaries: Past, Present and Future (London, 2010), 139–76. HENNIG, D., ‘Staatliche Ansprüche an privaten Immobilienbesitz in der klassischen und hellenistischen Polis,’ Chiron 25 (1995), 235–82. HERKENNE, H., De Veteris Latinae Ecclesiastici capitibus I–XLIII, una cum notis ex eiusdem libri translationibus Aethiopica, Armeniaca, Copticis, Latina altera, Syro-hexaplari depromptis (Leipzig, 1899). HICKIE, W. J., Greek-English Lexicon to the New Testament (New York, 1893). HOCK, H. H., and B. D. JOSEPH, Language History, Language Change, and Language Relationship: An Introduction to Historical and Comparative Linguistics (Berlin, 1996). HOHLWEIN, N., L’Égypte Romaine: Recueil des termes techniques relatifs aux institutions politiques et administratives de l’Égypte romaine, suivi d’un choix de textes papyrologiques (Bruxelles, 1912). HOLLANDER, H. W., and M. DE JONGE, The Testaments of the Twelve Patriarchs: A Commentary (Leiden, 1985). HOLLEAUX, M., Études d’épigraphie et d’histoire grecques, III (Paris, 1942).

BIBLIOGRAPHY

463

HOLTON, D. W., ‘The Formation of the Future in Modern Greek Literary Texts up to the 17th Century,’ in N. M. Panayotakis (ed.), Αρχές της νεοελληνικής λογοτεχνίας / Origini della litteratura neogreca (Venice, 1993), 1:118– 28. HOLTON, D., P. MACKRIDGE, and I. PHILIPPAKI-WARBURTON, Greek: A Comprehensive Grammar (2nd ed., rev. V. Spyropoulos; London, 2012). HOMZA, L. A., Religious Authority in the Spanish Renaissance (Baltimore, 2000). HOPPER, P. J., and E. C. TRAUGOTT, Grammaticalization (2nd ed.; Cambridge, 2003). HORROCKS, G. C., Greek: A History of the Language and its Speakers (1st ed. 1997; 2nd ed.; Chichester, 2014). HORSLEY, G. H. R., and J. A. L. LEE, ‘A Preliminary Checklist of Abbreviations of Greek Epigraphic Volumes,’ Epigraphica 56 (1994), 129–69. — ‘A Lexicon of the New Testament with Documentary Parallels: Some Interim Entries, 1,’ FilNT 10 (1997), 55–84. HUDSON, R. A., Sociolinguistics (Cambridge, 1980). HUGO, P., Review of Troxel, LXX-Isaiah (2008), JHebS 10 (2010) (electronic). HUMBERT, J., Syntaxe grecque (3rd ed.; Paris, 1960). HUNKIN, J. W., ‘Pleonastic ἄρχομαι in the New Testament,’ JTS 25 (1924), 390– 402. IMHOLTZ, A. A., ‘“Liddell and Scott”: Precursors, Nineteenth-Century Editions, and the American Contributions,’ in Stray (ed.), Oxford Classics (2007), 117–34. JACQUES, X., List of New Testament Words Sharing Common Elements (Rome, 1969). JAEGER, W., Early Christianity and Greek Paideia (Cambridge, MA, 1961). JANNARIS, A. N., An Historical Greek Grammar: Chiefly of the Attic Dialect as Written and Spoken from Classical Antiquity down to the Present Time (London, 1897). JANSE, M., ‘Aspects of Bilingualism in the History of the Greek Language,’ in J. N. Adams, M. Janse, and S. Swain (eds.), Bilingualism in Ancient Society: Language Contact and the Written Text (Oxford, 2002), 332–90. JELLICOE, S., The Septuagint and Modern Study (Oxford, 1968). JEREMIAS, J., Unknown Sayings of Jesus (London, 1964). JOBES, K. H., and M. SILVA, Invitation to the Septuagint (Grand Rapids, 2000). JOHANNESSOHN, M., Der Gebrauch der Präpositionen in der Septuaginta (MSU III.3; Berlin, 1926). JOHNSON, A. F., The First Century of Printing at Basle (London, 1926). — The Italian Sixteenth Century (London, 1926). JOLY, R., Le vocabulaire chrétien de l’amour est-il original? Φιλεῖν et Ἀγαπᾶν dans le grec antique (Bruxelles, 1968). JOSEPH, B. D., Morphology and Universals in Syntactic Change: Evidence from Medieval and Modern Greek (New York, 1990). JOSEPH, B. D., and P. A. PAPPAS, ‘On Some Recent Views Concerning the Development of the Greek Future System,’ BMGS 26 (2002), 247–73. JOUGUET, P., La vie municipale dans l’Égypte romaine (Paris, 1911; repr. 1968). JOÜON, P., ‘Θέλειν au sens d’“être sur le point de” dans Jean 1, 43,’ RSR 29 (1939), 620–1.

464

COLLECTED ESSAYS ON THE GREEK BIBLE

— ‘Les verbes βούλομαι et θέλω dans le Nouveau Testament,’ RSR 30 (1940), 227–38. — A Grammar of Biblical Hebrew (trans. and rev. T. Muraoka; 2 vols.; Rome, 2006). JOYAL, M., I. MCDOUGALL, and J. C. YARDLEY, Greek and Roman Education: A Sourcebook (London, 2009). JUDGE, E. A., ‘Paul’s Boasting in Relation to Contemporary Professional Practice,’ in D. M. Scholer (ed.), Social Distinctives of the Christians in the First Century: Pivotal Essays by E. A. Judge (Peabody, MA, 2008), 57–71. KARRER M., and W. KRAUS, with M. MEISER (eds.), Die Septuaginta: Texte, Kontexte, Lebenswelten. [1.] Internationale Fachtagung veranstaltet von Septuaginta Deutsch (LXX.D), Wuppertal 20.–23. Juli 2006 (WUNT 219; Tübingen, 2008). KATZ, P., Philo’s Bible (Cambridge, 1950). — ‘Septuagintal Studies in the Mid-Century: Their Links with the Past and their Present Tendencies,’ in Davies and Daube (eds.), Background of the New Testament (1956), 176–208. KAUHANEN, T., and H. VANONEN (eds.), The Legacy of Soisalon-Soininen: Towards a Syntax of Septuagint Greek (DSI 13; Göttingen, 2020). KAZAZIS, J. N., ‘Atticism,’ in A. F. Christidis (ed.), A History of Ancient Greek: From the Beginnings to Late Antiquity (Cambridge, 2007), 1200–12. KIEFFER, R., Au delà des recensions? L’évolution de la tradition textuelle dans Jean VI, 52-71 (Lund, 1968). KIESSLING, E., ‘Die Aposkeuai und die prozeßrechtliche Stellung der Ehefrauen im ptol. Ägypten,’ APF 8 (1927), 240–9. KILPATRICK, G. D., ‘Atticism and the Text of the New Testament,’ in J. Blinzler et al. (eds.) Neutestamentliche Aufsätze: Festschrift für J. Schmid (Regensburg, 1963), 125–37. — Review of Rydbeck, Fachprosa (1967), JBL 88 (1969), 354–5. — ‘What John Tells Us about John,’ in Studies in John Presented to Dr. J. N. Sevenster (NT Suppl. 24; Leiden, 1970), 75–87. — ‘Eclecticism and Atticism,’ ETL 53 (1977), 107–12. —‘Some Thoughts on Modern Textual Criticism and the Synoptic Gospels,’ NovT 19 (1977), 275–92. — ‘Atticism and the Future of ζῆν,’ NovT 25 (1983), 146–51. — Review of Greenlee, Morpheme Lexicon (1983), NovT 27 (1985), 382–4. KIM, L., ‘The Literary Heritage as Language: Atticism and the Second Sophistic,’ in E. J. Bakker (ed.), A Companion to the Ancient Greek Language (Chichester, 2010), 468–82. KINZIG, W., ‘The Greek Christian Writers,’ in Porter (ed.), Handbook (1997), 633– 70. KLAMP, D., ‘Das Testament des Taharpaesis,’ ZPE 2 (1968), 81–150. KOLITSARAS, I. Th., Ἡ Παλαιὰ Διαθήκη κατὰ τοὺς Ἑβδομήκοντα (5 vols.; Athens, 1989). KOOIJ, A. van der, ‘Isaiah in the Septuagint,’ in C. G. Broyles and C. A. Evans (eds.), Writing and Reading the Scroll of Isaiah (Leiden 1997), 513–29. — The Oracle of Tyre: The Septuagint of Isaiah XXIII as Version and Vision (VTSuppl. 71; Leiden, 1998).

BIBLIOGRAPHY

465

— ‘The Septuagint of Isaiah and the Issue of Coherence: A Twofold Analysis of LXX Isaiah 31:9b–32:8,’ in van der Kooij and van der Meer (eds.), Old Greek of Isaiah (2010), 33–48. KOOIJ, A. van der, and M. N. VAN DER MEER (eds.), The Old Greek of Isaiah: Issues and Perspectives, Papers Read at the Conference on the Septuagint of Isaiah, held in Leiden, 10–11 April 2008 (CBET 55; Leuven, 2010). KOTSEVALOV, A., ‘Koine Syntax of Greek Colonies on the Black Sea,’ in G. E. Mylonas and D. Raymond (eds.), Studies Presented to D. M. Robinson (2 vols.; Saint Louis, MO, 1951, 1953), 2:434–42. LADAS, G. G., and A. CHATZIDIMOS, Προσθῆκες, διορθώσεις καὶ συμπληρώσεις στὴν Ε ̔ λληνικὴ Βιβλιογραϕία τοῦ Émile Legrand γιὰ τοὺς αἰῶνες XV, XVI καὶ XVII (Athens, 1976). LANE, W. L., The Gospel According to Mark: The English Text with Introduction, Exposition and Notes (London, 1974). LAGRANGE, M.-J., Evangile selon Saint Marc (5th ed.; Paris, 1929). LA ROI, E., ‘A Rephilologized Diachronic Analysis of “Post-Classical Greek”: Pitfalls and Principles for Progress,’ JGL 20 (2020), 213–38. — ‘The Insubordination of If- and That-Clauses from Archaic to Post-Classical Greek: A Diachronic Constructional Typology,’ Symbolae Osloenses 95 (2021), 1–63. LE BOULLUEC, A., and P. SANDEVOIR, La Bible d’Alexandrie: L’Exode (Paris, 1989). LEE, J. A. L., ‘A Note on Septuagint Material in the Supplement to Liddell and Scott,’ Glotta 47 (1969), 234–42. — ‘Ἀποσκευή in the Septuagint,’ JTS n.s. 23 (1972), 430–7. — A Lexical Study of the Septuagint Version of the Pentateuch (SCS 14; Chico, Calif., 1983). — ‘The Future of ζῆν in Late Greek,’ NovT 22 (1980), 289–98. — ‘Equivocal and Stereotyped Renderings in the LXX,’ RB 87 (1980), 104–17. — ‘Some Features of the Speech of Jesus in Mark’s Gospel,’ NovT 27 (1985), 1–26. — ‘Συνίστημι: A Sample Lexical Entry,’ in Melbourne Symposium (1990), 1– 16. — ‘A Non-Aramaism in Luke 6.7,’ NovT 33 (1991), 28–34. — ‘The United Bible Societies’ Lexicon and its Analysis of Meanings,’ FilNT 5 (1992), 167–89. — ‘Translations of the Old Testament I. Greek,’ in Porter (ed.), Handbook of Classical Rhetoric (1997), 775–83. — ‘Hebrews 5:14 and Ἕξις: A History of Misunderstanding,’ NovT 39 (1997), 151–76. — A History of New Testament Lexicography (New York, 2003). — ‘The Present State of Lexicography of Ancient Greek,’ in Taylor et al. (eds.), Biblical Greek Language and Lexicography (2004), 66–74. — ‘Ἐξαποστέλλω,’ in J. Joosten and P. J. Tomson (eds.), Voces Biblicae: Septuagint Greek and its Significance for the New Testament (Leuven, 2007), 99–113. — ‘A Lexicographical Database for Greek: Can it be Far Off? The case of amphodon,’ in Karrer et al. (ed.), Die Septuaginta: Texte, Kontexte, Lebenswelten (2008), 214–20.

466

COLLECTED ESSAYS ON THE GREEK BIBLE

— ‘Auxiliary θέλω,’ in Evans and Obbink (eds.), Language of the Papyri (2010), 15–34. — ‘Why Didn’t St Basil Write in New Testament Greek?,’ Phronema 25 (2010), 3–20. — Review of Muraoka, Lexicon (2009), BIOSCS 43 (2010), 115–25. — ‘Releasing LSJ from its Past,’ in C. Stray (ed.), Classical Dictionaries (London, 2010), 119–38. — ‘The Atticist Grammarians,’ in Porter and Pitts (eds.), Language of the New Testament (2013), 283–308. — ‘Accuracy and Idiom: The Renderings of mittahat in the Septuagint Pentateuch,’ in De Troyer, Law, and Liljeström (eds.), In the Footsteps (2014), 79–99. — The Greek of the Pentateuch: Grinfield Lectures on the Septuagint 2011–2012 (Oxford, 2018). — ‘The Use of the Septuagint in the Liturgy and Lectionary of the Greek Orthodox Church,’ in A. G. Salvesen and T. M. Law (eds.), The Oxford Handbook of the Septuagint (Oxford, 2021), 573–87. LEE, J. A. L., and G. H. R. HORSLEY, ‘A Lexicon of the New Testament with Documentary Parallels: Some Interim Entries, 2,’ FilNT 11 (1998), 57–84. LEGG, S. C. E., Nouum Testamentum Graece: Euangelium secundum Marcum (Oxford, 1935). LEGRAND, É., Bibliographie hellénique ou description raisonnée des ouvragres publiés en grec par des Grecs au XVe et XVIe siècles (4 vols.; Paris, 1885– 1906. Repr. Paris, 1962). LEMMELIJN, B., A Plague of Texts? A Text-Critical Study of the So-Called ‘Plagues Narrative’ in Exodus 7:14–11:10 (Leiden, 2009). LE MOIGNE, P., ‘Le livre d’Esaïe dans la Septante: ecdotique, stylistique, linguistique, ou esquisse d’une poétique de la Septante’ (Diss. École Pratique des Hautes Études, Paris, 2001). LIETZMANN, H., Griechische Papyri (Bonn, 1910). LLOYD-JONES, H., and P. PARSONS (eds.), Supplementum Hellenisticum (Berlin, 1983). LOADER, J. A., ‘An Explanation of the Term prosêlutos,’ NovT 15 (1973), 270–7. LOADER, W., The Septuagint, Sexuality, and the New Testament (Grand Rapids, 2004). LOBECK, Chr. A. (ed.), Phrynichi Eclogae nominum et verborum atticorum: cum notis P. J. Nunnesii, D. Hoeschelii, J. Scaligeri et Cornelii de Pauw, partim integris partim contractis (Leipzig, 1820). LOLOS, Y., ‘Greek Roads: A Commentary on the Ancient Terms,’ Glotta 79 (2003), 137–74. LONGENECKER, R. N., Galatians (WBC; Dallas, 1990). LOUW, J. P., Semantics of New Testament Greek (Philadelphia, 1982). — ‘The Analysis of Meaning in Lexicography,’ FilNT 6 (1993), 139–48. — (ed.), Lexicography and Translation (Cape Town, 1985). LOUW, J. P., and E. A. NIDA et al., Greek-English Lexicon of the New Testament: Based on Semantic Domains (2 vols.; 2nd ed., New York, 1989). LOUW, T. A. W. van der, Transformations in the Septuagint: Towards an Interaction of Septuagint Studies and Translation Studies (CBET 47; Leuven, 2007).

BIBLIOGRAPHY

467

LOWRY, M. The World of Aldus Manutius: Business and Scholarship in Renaissance Venice (Oxford, 1979). LYELL, J. P. R., Cardinal Ximenes: Statesman, Ecclesiastic, Soldier and Man of Letters, with an Account of the Complutensian Polyglot Bible (London, 1917). MAAS, P. M., Greek Metre (trans. Hugh Lloyd-Jones; Oxford, 1962). MAHAFFY, J. P., The Flinders Petrie Papyri (3 vols.; Dublin, 1891–1905). MAIDHOF, A., Zur Begriffsbestimmung der Koine: besonders auf Grund des Attizisten Moiris (Würzburg, 1912). MANDILARAS, B. G., The Verb in the Greek Non-Literary Papyri (Athens, 1973). MARICHAL, R. (ed.), Quart Livre de Pantagruel (Geneva, 1947). MARKOPOULOS, T., The Future in Greek: From Ancient to Medieval (Oxford, 2008). MARSHALL, I. H., The Gospel of Luke (NIGTC; Exeter, 1978). MARTINI, C. M., ‘Eclecticism and Atticism in the Textual Criticism of the Greek New Testament, in M. Black and W. A. Smalley (eds.), On Language, Culture and Religion in Honor of E. A. Nida (The Hague/Paris, 1974), 149–56. MASSON, É., Recherches sur les plus anciens emprunts sémitiques en grec (Paris, 1967). MAUERSBERGER, A., Polybios-Lexikon (Berlin, 1956–2006). MAYSER, E., Grammatik der griechischen Papyri aus der Ptolemäerzeit (2 vols. in 6 parts; Berlin and Leipzig, 1906–1938; repr. 1970). MCLEAN, B. H., An Introduction to Greek Epigraphy of the Hellenistic and Roman Periods from Alexander the Great down to the Reign of Constantine (323 B.C.– A.D. 337) (Ann Arbor, MI, 2002). MEECHAM, H. G., The Letter of Aristeas: A Linguistic Study with Special Reference to the Greek Bible (Manchester, 1935). MEER, M. N. van der, Review of NETS (2007), BIOSCS 41 (2008), 114–21. — ‘Bridge over Troubled Waters? The gephura in the Old Greek of Isaiah 37:25 and Contemporary Greek Sources’ in M. K. H. Peters (ed.), XIII Congress of the International Organization for Septuagint and Cognate Studies, Ljubljana 2007 (Atlanta, 2008), 305–24. — ‘Trendy Translations in the Septuagint of Isaiah: A Study of the Vocabulary of the Greek Isaiah 3:18–23 in the Light of Contemporary Sources,’ in Karrer, Kraus, and Meiser (eds.), Die Septuaginta (2008), 581–96. — ‘Papyrological Perspectives on the Septuagint of Isaiah,’ in van der Kooij and van der Meer (eds.), Old Greek of Isaiah (2010), 107–33. MELANCTHON, Philip, Grammatica graeca (Hagenau, 1527. Facsimile ed., Menston, 1969). METZGER, B. M., The Text of the New Testament: Its Transmission, Corruption, and Restoration (Oxford, 1st ed. 1964; 2nd ed. 1968). — A Textual Commentary on the Greek New Testament (London/New York 1975). — Manuscripts of the Greek Bible: An Introduction to Greek Palaeography (New York/Oxford, 1981). MIGLIORINI, B., Storia della lingua italiana (3rd ed.; Firenze, 1961). — The Italian Language (rev. and trans. T. G. Griffith; London, 1966). MIHEVC-GABROVEC, E., Études sur la syntaxe de Ioannes Moschos (Ljubljana, 1960).

468

COLLECTED ESSAYS ON THE GREEK BIBLE

MILLAR, F., A Greek Roman Empire: Power and Belief under Theodosius II (408– 450) (Berkeley, 2006). MILLIGAN, G., Selections from the Greek Papyri (Cambridge, 1912). MIRAMBEL, A., ‘Essai sur l’évolution du verbe en grec byzantin,’ BSL 61 (1966), 167–90. MONTEVECCHI, O., ‘Continuità ed evoluzione della lingua greca nella Settanta e nei papiri,’ Actes du Xe Congrès international de Papyrologues (Warsaw, 1964), 39–49. MORGAN, T., Literate Education in the Hellenistic and Roman Worlds (Cambridge, 1998). MORISON, S., Politics and Script: Aspects of Authority and Freedom in the Development of Graeco-Latin Script from the Sixth Century B.C. to the Twentieth Century A.D (Oxford, 1972). MORRIS, L., Studies in the Fourth Gospel (Grand Rapids, 1969). — The Gospel According to John (Grand Rapids, 1971). MORWOOD, J., and J. TAYLOR, The Pocket Oxford Classical Greek Dictionary (Oxford, 2002). MOULE, C. F. D., An Idiom Book of New Testament Greek (Cambridge, 1963). MURAOKA, T. (ed.), Melbourne Symposium on Septuagint Lexicography (Atlanta, 1990). — A Greek-English Lexicon of the Septuagint (Twelve Prophets), (Louvain, 1993). — A Greek-English Lexicon of the Septuagint: Chiefly of the Pentateuch and the Twelve Prophets (Louvain, 2002). — ‘Recent Discussions on the Septuagint Lexicography,’ in Karrer et al. (eds.), Die Septuaginta (2008), 221–35. — A Greek-English Lexicon of the Septuagint (Louvain, 2009). MUSSIES, G., The Morphology of Koine Greek as used in the Apocalypse of St. John (Leiden, 1971). NABER, S. A. (ed.), Photii Patriarchae Lexicon (2 vols.; Leiden, 1864–1865; repr. Amsterdam, 1965). NEBRIJA, Antonio de, Gramatica castellana. Reproduction phototypique de l’édition princeps (1492). Publiée avec une préface par E. Walberg (Halle a. S., 1909). — Elio Antonio de Nebrija: Diccionario Latino-Español (Salamanca 1492) (ed. G. Colón and A.-J. Soberanas; Barcelona, 1979). — Vocabulario español-latino (Salamanca ¿1495?): sale nuevamento a luz reproducida en facsímile por acuerdo de la Real Academia Española (Madrid, 1951). NESTLE, E., Introduction to the Textual Criticism of the Greek New Testament (trans. W. Edie from 2nd ed.; London, 1901). NEWMAN, B. M., A Concise Greek-English Dictionary of the New Testament (London, 1971; rev. ed.; Stuttgart, 2010). — ‘A Concise Greek-English Dictionary of the New Testament: Reflections and Ruminations,’ in Taylor et al. (eds.), Biblical Greek Language and Lexicography (2004), 91–3. NGUNGA, A. T., Messianism in the Old Greek of Isaiah: An Intertextual Analysis (Göttingen, 2013).

BIBLIOGRAPHY

469

NICKAU, K. (ed.), Ammonii qui dicitur liber De Adfinium Vocabulorum Differentia (Leipzig, 1966). NIDA, E. A., ‘Varieties of Language,’ BT 23 (1972), 316–22. NIDA, E. A., and J. P. LOUW, Lexical Semantics of the Greek New Testament (Atlanta, 1992). NORDEN, E. E., Die antike Kunstprosa: Vom VI. Jahrhundert v. Chr. bis in die Zeit der Renaissance (2 vols.; 5th ed. [= 2nd ed. 1909]; Stuttgart, 1958). NORTON, F. J., Printing in Spain 1501–1520 (Cambridge, 1966). — A Descriptive Catalogue of Printing in Spain and Portugal 1501–1520 (Cambridge, 1978). OATES, J. F., et al., Checklist of Editions of Greek, Latin, Demotic and Coptic Papyri, Ostraca and Tablets (5th ed.; 2001: updates at ). O’CONNELL, S., From Most Ancient Sources: The Nature and Text-Critical Use of the Greek Old Testament Text of the Complutensian Polyglot Bible (OBO 215; Fribourg, 2006). O’CONNOR, M., and J. A. L. LEE, ‘A Problem in Biblical Lexicography: The Case of Hebrew tap and Greek aposkeuê,’ ZAW 119 (2007), 403–9. Ó FEARGHAIL, F., ‘The Imitation of the Septuagint in Luke’s Infancy Narrative,’ PIBA 12 (1989), 58–78. OLIN, J. C., Catholic Reform from Cardinal Ximenes to the Council of Trent, 14951563 (New York, 1990). OLLEY, J. W., ‘Righteousness’ in the Septuagint of Isaiah: A Contextual Study (SCS 8; Missoula, 1979). ORLANDOS, A. K., and I. N. TRAVLOS, Λεξικὸν Ἀρχαίων Ἀρχιτεκτονικῶν Ὅρων (Athens, 1986). O’SULLIVAN, J. N., A Lexicon to Achilles Tatius (Berlin/New York, 1980). OTTLEY, R. R., The Book of Isaiah According to the Septuagint (Codex Alexandrinus) (2 vols.; Cambridge, 1904, 1906). — A Handbook to the Septuagint (London, 1920). PAGE, D. L., Euripides Medea: The Text Edited with Introduction and Commentary (Oxford, 1938). PALMER, L. R., The Latin Language (London, 1954). PAPPAS, P. A., ‘The Microcosm of a Morphological Change: Variation in thelô + infinitive futures and êthela + infinitive counterfactuals in Early Modern Greek,’ Diachronica 18 (2001), 59–92. PARKE, H. W., Greek Mercenary Soldiers (Oxford, 1933). PASCHOS, P. B., Τὸ Γεροντικόν, ἤτοι Ἀποφθέγματα Ἁγίων Γερόντων. Πρόλογος, Κείμενον, Γλωσσάριον, Σχόλια, Εὑρετήριον θεμάτων (3rd ed.; Athens, 1981). PASSOW, F., Handwörterbuch der Griechischen Sprache (4th ed.; Leipzig, 1831). PAUL, S. M., et al. (eds.), Emanuel: Studies in Hebrew Bible, Septuagint, and Dead Sea Scrolls in Honor of Emanuel Tov (Leiden, 2003). PELÁEZ, J., ‘New Testament Lexicography: A Critical Introduction,’ in Peláez and Mateos, New Testament Lexicography (2010), 3–54. PELÁEZ, J., and MATEOS, J., New Testament Lexicography: Introduction, Theory, Method (trans. A. Bowden, ed. D. S. du Toit; FSBP 6; Berlin, 2010). PELLETIER, A., ‘Le vocabulaire du commandement dans le Pentateuque des LXX et dans le Nouveau Testament,’ RSR 41 (1953), 519–24.

470

COLLECTED ESSAYS ON THE GREEK BIBLE

PENTZAS, B., Εἶπε Γέρων· Τὸ Γεροντικὸν σὲ νεοελληνικὴ ἀπόδοση (Athens, 1983). PERKINS, L. J., ‘The Lord is Warrior,’ BIOSCS 40 (2007), 121–38. PERRY, B. E., Aesopica (Urbana, 1952). PIETERSMA, A. (trans.), A New English Translation of the Septuagint: The Psalms (New York/Oxford, 2000). — ‘A Panel Presentation on Ronald Troxel’s LXX-Isaiah,’ in Boyd-Taylor (ed.), A Question of Methodology (2013), 339–58. — ‘A Response to Muraoka’s Critique of Interlinearity,’ in Boyd-Taylor (ed.), A Question of Methodology, 315–37. — ‘Context is King in Septuagint Lexicography—Or Is It?,’ in Aitken and Evans (eds.), Biblical Greek in Context (2015), 165–75. PINKSTER, H., ‘The Development of Future Tense Auxiliaries in Latin,’ Glotta 63 (1985), 186–208. PORTER, S. E. (ed.), Handbook of Classical Rhetoric in the Hellenistic Period 330 B.C.–A.D. 400 (Leiden, 1997). — ‘Paul of Tarsus and His Letters,’ in Porter (ed.), Handbook (1997), 533–85. — ‘Lexical Semantics and New Testament Greek: A Review Article of Some Major Works,’ FilNT (2019), 113–55. PORTER, S. E., and A. W. PITTS (eds.), The Language of the New Testament: Context, History, and Development (Leiden, 2013). PREISIGKE, F., Fachwörter des öffentlichen Verwaltungsdienstes Ägyptens in den griechischen Papyrusurkunden der ptolemäisch-römischen Zeit (Göttingen, 1915). PRELLWITZ, W., Etymologisches Wörterbuch der griechischen Sprache (Göttingen, 1892). PROCTOR, R., The Printing of Greek in the Fifteenth Century (London, 1900). PSICHARI, J., Quelques travaux de linguistique, de philologie et de littérature helléniques (1884–1928) (vol. 1; Paris, 1930). QUASTEN, J., Patrology, Vol. 3 (Utrecht, 1960; repr. Allen, Tex., n.d.). RAJAK, T., Translation and Survival: The Greek Bible and the Ancient Jewish Diaspora (Oxford, 2009). REESE, J. M., Hellenistic Influence on the Book of Wisdom and its Consequences (AnBib 41; Rome, 1970). REEVE, A., and M. A. SCREECH (eds.), Erasmus’ Annotations on the New Testament: Acts–Romans–I and II Corinthians (Leiden, 1990). REGNAULT, L., Les sentences des pères du désert: Collection alphabétique (Solesmes, 1981). REITERER, F. V., ‘Review of Recent Research on the Book of Ben Sira,’ in P. C. Beentjes (ed.), Book of Ben Sira in Modern Research (Berlin, 1997), 23–60. RENGSTORF, K. H., A Complete Concordance to Flavius Josephus (Leiden, 1973– 1983). RENOUARD, A. A., Annales de l’imprimerie des Alde: ou histoire des trois Manuce et de leurs éditions (3rd ed.; Paris, 1834). RIESENFELD, H., Zum Gebrauch von θέλω im Neuen Testament (Uppsala, 1936). RINK, H., Strassen- und Viertelnamen von Oxyrhynchus (Giessen, 1924). RITSCHL, F. (ed.), Thomae Magistri sive Theoduli monachi ecloga vocum atticarum (Halis Saxonum, 1832).

BIBLIOGRAPHY

471

ROBERT, L., À travers l’Asie Mineure: Poètes et prosateurs, monnaies grecques, voyageurs et géographie (Paris, 1980). ROBERTSON, A. T., A Grammar of the Greek New Testament in the Light of Historical Research (4th ed.; Nashville, 1934). ROCCI, L., Vocabolario Greco Italiano (3rd ed.; Rome, 1943). RÖDIGER, R., ‘βούλομαι und ἐθέλω,’ Glotta 8 (1917), 1–24. RÖNNEGÅRD, P., Threads and Images: The Use of Scripture in Apophthegmata Patrum (CB, NT Series 44; Winona Lake, IN, 2010). ROSS, W. A., ‘Lexical Possibilities in LXX Research: Revision and Expansion,’ in W. Kraus et al. (eds.), XV Congress of the International Organization for Septuagint and Cognate Studies Munich, 2013 (Atlanta, 2016), 341– 59. ROUSSEAU, P., Basil of Caesarea (Berkeley, 1994). ROWE, G. O., ‘Style,’ in Porter (ed.), Handbook of Classical Rhetoric (1997), 121–57. RUIJGH, C. J., ‘La place des enclitiques dans l’ordre des mots chez Homère d’après la loi de Wackernagel,’ in H. Eichner and H. Rix (eds.), Sprachwissenschaft und Philologie: Jacob Wackernagel und die Indogermanistik heute (Wiesbaden, 1990), 213–33. RUSSELL, N., The Lives of the Desert Fathers (London, 1981). RUTHERFORD, W. G. (ed.), The New Phrynichus: being a revised text of The Ecloga of the Grammarian Phrynichus (London, 1881). RYDBECK, L., Fachprosa, Vermeintliche Volkssprache und Neues Testament (Uppsala, 1967). SÁENZ-BADILLOS, Á., La filología bíblica en los primeros helenistas de Alcalá (Estella, 1990). SAKALIS, D. T., ‘Ἀνάττικα καὶ Ψευδαττικὰ ἀπὸ τὸν Ἀττικιστὴ Μοίρη,’ Δωδώνη 6 (1977), 441–70. SATTERTHWAITE, P. E., ‘Judges: To the Reader,’ NETS, 195–200. SCHLEUSNER, J. F., Novus Thesaurus Philologico-criticus: sive Lexicon in LXX. et Reliquos Interpretes Graecos ac Scriptores Apocryphos Veteris Testamenti (2nd ed.; Glasgow, 1822). SCHMID, W., Der Atticismus in seinen Hauptvertretern von Dionysius von Halikarnass bis auf den zweiten Philostratus (5 vols.; Stuttgart, 1887–1897). — Review of LSJ Part 4. II, PhilWoch 51 (1931), 705–18. SCHMIDT, J. H. H., Synonymik der griechischen Sprache (4 vols.; Leipzig, 1876– 1886.) SCHOLDERER, V., Greek Printing Types, 1465–1927: Facsimiles from an Exhibition of Books Illustrating the Development of Greek Printing Shown in the British Museum 1927 (London, 1927). SCHRENK, G., ‘θέλω, θέλημα, θέλησις,’ in TDNT, 3:44–62. SCOTT, J. A., ‘The Vocative in Homer and Hesiod,’ AJP 24 (1903), 192–6. — ‘The Vocative in Aeschylus and Sophocles,’ AJP 25 (1904), 81–4. — ‘Additional Notes on the Vocative,’ AJP 26 (1905), 32–43. SCREECH, M. A., Rabelais (London, 1979). SCRIVENER, F. H. A., A Plain Introduction to the Criticism of the New Testament: For the Use of Biblical Students (1st ed., 1861; 4th ed., 2 vols.; ed. E. Miller, London, 1894).

472

COLLECTED ESSAYS ON THE GREEK BIBLE

SEELIGMANN, I. L., The Septuagint Version of Isaiah: A Discussion of its Problems (Leiden, 1948). SHAW, F., Review of Muraoka, Lexicon, BMCR 2010.04.20. SHIPP, G. P., ‘Some Observations on the Distribution of Words in the New Testament,’ in E. C. B. MacLaurin (ed.), Essays in Honour of Griffithes Wheeler Thatcher 1863–1950 (Sydney, 1967), 127–38. SILVA, M., Biblical Words and their Meaning: An Introduction to Lexical Semantics (Grand Rapids, 1983). — ‘Esaias: To the Reader,’ NETS, 823–5. — Review of Muraoka, Lexicon, WTJ 72 (2010), 435–8. SIMPSON, E. K., Words Worth Weighing in the Greek New Testament (London, 1946). SKEHAN, P. W., and A. A. DI LELLA, The Wisdom of Ben Sira: A New Translation with Notes (AB 39; New York, 1987). SKINNER, J., A Critical and Exegetical Commentary on Genesis (ICC; 2nd ed.; Edinburgh, 1930). SMEND, R. (ed.), Die Weisheit des Jesus Sirach: Hebräisch und Deutsch, mit einem hebräischen Glossar (Berlin, 1906). — Griechisch-Syrisch-Hebräischer Index zur Weisheit des Jesus Sirach (Berlin, 1907). SMITH, H. P., The Books of Samuel (ICC; Edinburgh, 1899). SMYTH, H. W., Greek Grammar, rev. G. M. Messing (Cambridge, MA, 1956). SOISALON-SOININEN, ‘Der Gebrauch des genetivus absolutus in der Septuaginta’ (1973), in Studien zur Septuaginta-Syntax (1987), 175–80. — ‘The Rendering of the Hebrew Relative Clause in the Greek Pentateuch,’ in Proceedings of the Sixth World Congress of Jewish Studies Held at the Hebrew University of Jerusalem, 13–19 August, 1973 (vol. 1; Jerusalem, 1977), 401– 6. Repr. in Studien zur Septuaginta-Syntax (1987), 55–61. — ‘Die Wiedergabe einiger hebräischer Zeitangaben mit der Präposition ‫ ב‬in der Septuaginta’ (1978), in Studien zur Septuaginta-Syntax (1987), 107– 15. — ‘Renderings of Hebrew Comparative Expressions with ‫ מן‬in the Greek Pentateuch’ (1979), in Studien zur Septuaginta-Syntax (1987), 141–53. — ‘ἐν für εἰς in der Septuaginta’ (1982), in Studien zur Septuaginta-Syntax (1987), 131–40. — ‘Die Wiedergabe des ‫ ב‬instrumenti im griechischen Pentateuch’ (1983), in Studien zur Septuaginta-Syntax (1987), 116–30. — ‘Die Wiedergabe des partitiven ‫ מן‬im griechischen Pentateuch’ (1985), Studien zur Septuaginta-Syntax (1987), 154–71. — Studien zur Septuaginta-Syntax: Zu seinem 70. Geburtstag am 4. Juli 1987 herausgegeben von Anneli Aejmelaeus und Raija Sollamo (AASF 237; Helsinki, 1987). — ‘Zurück zur Hebraismenfrage,’ in D. Fraenkel, U. Quast and J. W. Wevers (eds.), Studien zur Septuaginta. Robert Hanhart zu Ehren (Göttingen, 1990), 35–51. SOLLAMO, R., ‘Some “Improper” Prepositions, such as ἐνώπιον, ἐναντίον, ἔναντι, etc., in the Septuagint and Early Koine Greek,’ VT 25 (1975), 773–82. — Renderings of Hebrew Semiprepositions in the Septuagint (Helsinki, 1979).

BIBLIOGRAPHY

473

— ‘Repetitions of Prepositions in the Septuagint of Genesis,’ in F. García Martínez and M. Vervenne (eds.), Interpreting Translation: Studies on the LXX and Ezekiel in Honour of Johan Lust (BETL 192; Leuven, 2005), 371–84. — ‘The Place of the Enclitic Personal Pronouns in the Old Greek Psalter,’ in M. K. H. Peters (ed.), XII Congress of the International Organization for Septuagint and Cognate Studies, Leiden 2004 (Atlanta, 2006), 153–60. — ‘The Usage of the Article with Nouns Defined by a Nominal or Pronominal Genitive in LXX Genesis,’ in Kauhanen and Vanonen (eds.), The Legacy of Soisalon-Soininen (2020), 43–53. SOPHOCLES, E. A., Greek Lexicon of the Roman and Byzantine Periods (New York, 1887). SOTIROPOULOS, D., ‘Diglossia and the National Language Question in Modern Greece,’ Linguistics 197 (1977), 5–31. SOUTER, A., A Pocket Lexicon to the Greek New Testament (Oxford, 1916). — ‘Ἀγαπητός,’ JTS 28 (1926–1927), 59–60. SPARKS, H. D. F., ‘The Semitisms of St. Luke’s Gospel,’ JTS 44 (1943), 129–38. — ‘The Semitisms of the Acts,’ JTS n.s. 1 (1950), 16–28. SPEISER, E. A., Genesis (New York, 1964). SPOTTORNO, V., ‘The Relative Pronoun in the New Testament: Some Critical Remarks,’ NTS 28 (1982), 132–41. STAIKOS, K. S., and T. E. SKLAVENITES, Πεντακόσια Χρόνια· Εντύπης Παράδοσης του Νέου Ελληνισμού (1499–1999) (Athens, 2000). STASINOPOULOS, M. D., Μορφὲς ἀπὸ τὸν Τέταρτον Αἰῶνα μ. Χ.: Ἱστορικὴ Εἰσαγωγὴ στὸν Λόγο πρὸς τοὺς Νέους τοῦ Μεγάλου Βασιλείου (Athens, 1972). STEEN, H. A., ‘Les clichés épistolaires dans les lettres sur papyrus grecques,’ Class et Med 1 (1938), 119–76. STEPHANUS, H., Thesaurus Graecae Linguae, ed. C. B. Hase et al. (Paris, 1831– 1865). STOLK, J. V., ‘Dative and Genitive Case Interchange in Greek Papyri from Roman– Byzantine Egypt,’ Glotta 93 (2017), 182–212. STRAY, C., (ed.), Oxford Classics: Teaching and Learning 1800–2000 (Oxford, 2007). — Classical Dictionaries: Past, Present and Future (London, 2010). STRAY, C., et al. (eds.), Liddell and Scott: The History, Methodology, and Languages of the World’s Leading Lexicon of Ancient Greek (Oxford, 2019). STROUT, D., and R. FRENCH, ‘Phrynichos,’ in Pauly-Wissowa, RE 20:920–5. STURZ, F. G., De dialecto macedonica et alexandrina (Lipsiae, 1808). SWAIN, S., Hellenism and Empire: Language, Classicism, and Power in the Greek World, AD 50–250 (Oxford, 1996). SWETE, H. B., An Introduction to the Old Testament in Greek (rev. R. R. Ottley; Cambridge, 1914). SWINN, S. P., ‘Ἀγαπᾶν in the Septuagint,’ in Muraoka (ed.), Melbourne Symposium (1990), 49–81. TABACHOVITZ, D., Die Septuaginta und das Neue Testament: Stilstudien (Lund, 1956). TAGLIAFERRO, E., in ‘Nota linguistica filoniana (Cher. 1–10),’ Helikon 18–19 (1978– 1979), 415–424. TALSHIR, Z., 1 Esdras: A Text Critical Commentary (Atlanta, 2001).

474

COLLECTED ESSAYS ON THE GREEK BIBLE

TARAZI, P. N., Galatians (OBS; Crestwood, 1994). TASKER, R. V. G., ‘The Complutensian Polyglot,’ CQR (1953), 197–210. TAYLOR, B. A. (ed.), IX Congress of the International Organization for Septuagint and Cognate Studies, Cambridge 1995 (Atlanta, 1997). — X Congress of the International Organization for Septuagint and Cognate Studies, Oslo 1998 (Atlanta, 2001). TAYLOR, B. A., et al. (eds.), Biblical Greek Language and Lexicography: Essays in Honor of Frederick W. Danker (Grand Rapids, 2004). TAYLOR, V., The Gospel According to St. Mark (2nd ed.; London, 1966). THACKERAY, H. St. J., ‘The Greek Translators of the Prophetical Books,’ JTS 4 (1903), 578–85. — A Grammar of the Old Testament in Greek According to the Septuagint (vol. 1; Cambridge, 1909). — The Septuagint and Jewish Worship: A Study in Origins (Schweich Lectures 1920; London 1921). THESLEFF, H., Review of Rydbeck, Fachprosa (1967), Gnomon 42 (1970), 551– 5. THIERSCH, H. G. J., De Pentateuchi versione Alexandrina libri tres (Erlangae, 1841). THOMAS, H., Spanish Sixteenth-Century Printing (London, 1926). THOMPSON, A., Review of Lee, History, BIOSCS 36 (2003), 113–27. — ‘Βάπτω: An Illustration of the State of our Ancient Greek Dictionaries,’ in Stray et al. (eds.), Liddell and Scott (2019), 353–94. — ‘Response: Thompson on Goldhill on Stray, Clarke and Katz, Liddell and Scott,’ BMCR 2021.06.35. THOMPSON, S., The Apocalypse and Semitic Syntax (SNTSMS 52; Cambridge, 1985). THRALL, M. E., Greek Particles in the New Testament: Linguistic and Exegetical Studies (Leiden, 1962). THUMB, D., Die griechische Sprache im Zeitalter des Hellenismus (Strassburg, 1901). — Handbook of the Modern Greek Vernacular: Grammar, Texts, Glossary (trans. S. Angus; Edinburgh, 1912). THURÉN, L., ‘The General New Testament Writings,’ in Porter (ed.), Handbook (1997), 587–607. TORREY, C. C., Our Translated Gospels (London, n.d.). TOV, E., The Septuagint Translation of Jeremiah and Baruch (HSM 8; Missoula 1976). — ‘Three Dimensions of LXX Words,’ RB 83 (1976), 529–44. — ‘Compound Words in the LXX Representing Two or More Hebrew Words,’ Biblica 58 (1977), 189–212. — The Text-Critical Use of the Septuagint in Biblical Research (1st ed. 1981; 3rd ed.; Winona Lake, 2015). — ‘Did the Septuagint Translators Always Understand Their Hebrew Text?,’ in A. Pietersma and C. Cox (eds.), De Septuaginta: Studies in Honour of John William Wevers on his Sixty-fifth Birthday (Mississauga, 1984), 53–70. — The Greek and Hebrew Bible: Collected Essays on the Septuagint (Leiden, 1999).

BIBLIOGRAPHY

475

— ‘The Septuagint,’ in M. J. Mulder and H. Sysling (eds.), Mikra: Text, Translation, Reading, and Interpretation of the Hebrew Bible in Ancient Judaism and Early Christianity (Peabody, 2004), 161–88. TOY, C. H., Proverbs (ICC; Edinburgh, 1899). TREGELLES, S. P., An Account of the Printed Text of the Greek New Testament (London, 1854). TRENCHARD, W. C., A Concise Dictionary of New Testament Greek (Cambridge, 2003). TRIANTAPHYLLIDES, M. A., Νεοελληνική Γραμματική, Πρώτος Τόμος· Ιστορική Εισαγωγή (Αθήνα, 1938). — Νεοελληνική Γραμματική· Αναπροσαρμογή της Μικρής Νεοελληνικής Γραμματικής (Αθήνα, 1986). TRONCI, L., ‘Future Forms in Postclassical Greek: Some Remarks on the Septuagint and the New Testament,’ in D. Rifayenko and I. A. Seržant (eds.), Postclassical Greek: Contemporary Approaches to Philology and Linguistics (TiLSM 335; Berlin, 2020), 111–43. TROXEL, R. L., LXX-Isaiah as Translation and Interpretation: The Strategies of the Translator of the Septuagint of Isaiah (Leiden, 2008). — ‘βουλή and βουλεύειν in LXX Isaiah,’ in van der Kooij and van der Meer (eds.), Old Greek of Isaiah (2010), 153–71. TRUDGILL, P., Sociolinguistics (Harmondsworth, 1974). [TURNER, C. H.], ‘Ἀγαπητός,’ JTS 20 (1919), 339–44. TURNER, C. H., ‘Ὁ υἱός μου ὁ ἀγαπητός,’ JTS 27 (1925–1926), 113–29. — ‘Marcan Usage: Notes, Critical and Exegetical, on the Second Gospel,’ JTS 28 (1926–1927), 349–62. — The Early Printed Editions of the Greek Testament (Oxford, 1924). TURNER, E. G., Greek Papyri: An Introduction (Oxford, 1968). — Greek Manuscripts of the Ancient World (2nd ed., rev. P. J. Parsons; ICS Bulletin Suppl. 46; London, 1987). TURNER, G. W., Stylistics (Harmondsworth, 1973). TURNER, N., ‘The Literary Character of New Testament Greek,’ NTS 20 (1973– 1974), 107–14. VALENTE, S., The Antiatticist: Introduction and Critical Edition (Berlin, 2015). VEITCH, W., Greek Verbs Irregular and Defective (Oxford, 1866). VOELZ, J. W., ‘The Language of the New Testament,’ ANRW II 25.2 (1984), 893– 977. VORM-CROUGHS, M. van der, ‘LXX Isaiah and the Use of Rhetorical Figures,’ in van der Kooij and van der Meer (eds.), Old Greek of Isaiah (2010), 173–88. WADDELL, W. G. (ed.), Herodotus Book II (London, 1939). WAGNER, C., Die Septuaginta-Hapaxlegomena in Buch Jesus Sirach: Untersuchungen zu Wortwahl und Wortbildung unter besonderer Berücksichtigung des textkritischen und übersetzungstechnischen Aspekts (Berlin, 1999). WAGNER, J. R., Reading the Sealed Book: Old Greek Isaiah and the Problem of Septuagint Hermeneutics (Tübingen, 2013). WALBANK, F. W., A Historical Commentary on Polybius (3 vols.; Oxford, 1957– 1979). WALTERS, P., The Text of the Septuagint: Its Corruptions and their Emendation (ed. D. W. Gooding; Cambridge, 1973).

476

COLLECTED ESSAYS ON THE GREEK BIBLE

WARD, B., The Sayings of the Desert Fathers: The Alphabetical Collection, Translated, with a Foreword (rev. ed.; London, 1984). WEINFELD, M., Book of Genesis (Tel Aviv, 1975) [in Hebrew]. WEINREICH, U., Languages in Contact (New York, 1953). WELLHAUSEN, J., Einleitung in die drei ersten Evangelien (2nd ed.; Berlin, 1911). WENDEL, C., ‘Moiris,’ in Pauly-Wissowa, RE 15:2501–12. WESSELY, C., Papyrorum Scripturae Graecae Specimina lsagogica (Leipzig, 1900). WEVERS, J. W., Notes on the Greek Text of Genesis (SCS 35; Atlanta, 1993). — Notes on the Greek Text of Exodus (SCS 30; Atlanta, 1990). — Notes on the Greek Text of Numbers (SCS 46; Atlanta, 1998). — Notes on the Greek Text of Deuteronomy (SCS 39; Atlanta, 1995). WHITMARSH, T., The Second Sophistic (G&R: New Surveys in the Classics 35; Oxford, 2005). WIFSTRAND, A., ‘Die griechischen Verba für wollen,’ Eranos 40 (1942), 16–36. — ‘Stylistic Problems in the Epistles of James and Peter,’ Studia Theologica 1 (1948), 170–82. WILCKEN, U., ‘III. Referate: Papyrus-Urkunden,’ AFP 8 (1927), 63–167. WILCOX, M., ‘Semitisms in the New Testament,’ ANRW II.25.2 (1984), 978–1029. WILK, F., ‘Between Scripture and History: Technique and Hermeneutics of Interpreting Biblical Prophets in the Septuagint of Isaiah and the Letters of Paul,’ in van der Kooij and van der Meer (eds.), Old Greek of Isaiah (2010), 189– 209. WILLI-PLEIN, I., Review of Lee, Lexical Study, ThLZ 110 (1985), 94–5. WILSON, N., Mediaeval Greek Bookhands: Examples Selected from Greek Manuscripts in Oxford Libraries (Plates, Text; Cambridge, MA, 1972–1973). WILSON, N. G. (ed.), Saint Basil on the Value of Greek Literature (London, 1975). — ‘Vettor Fausto, Professor of Greek and Naval Architect,’ in A. C. Dionisotti et al. (eds.), The Uses of Greek and Latin: Historical Essays (London, 1988), 89–95. WINSTON, D., The Wisdom of Solomon: A New Translation with Introduction and Commentary (AB; New York, 1979). WITHERINGTON III, B., The Acts of the Apostles: A Socio-Rhetorical Commentary (Grand Rapids, 1998). WOODHEAD, A. G., The Study of Greek Inscriptions (2nd ed.; Cambridge, 1981). WOODY, K. M., ‘A Note on the Greek Fonts of the Complutensian Polyglot,’ Papers of the Bibliographical Society of America 65 (1971), 143–9. WORTLEY, J., ‘How the Desert Fathers “Meditated”,’ GRBS 46 (2006), 315–28. WRIGHT, B. G., No Small Difference: Sirach’s Relationship to its Hebrew Parent Text (SBLSCS 26; Atlanta, 1989). — ‘The Septuagint and Its Modern Translators,’ in Karrer et al. (eds.), Die Septuaginta (2008), 102–14. YOUNG, F., L. AYRES, and A. LOUTH, The Cambridge History of Early Christian Literature (Cambridge, 2004). ZERWICK, M., Biblical Greek: Illustrated by Examples, trans., rev. J. Smith (Rome, 1963). ZGUSTA, L., Manual of Lexicography (The Hague/Paris, 1971). ZIEGLER, J., Untersuchungen zur Septuaginta des Buches Isaias (Münster, 1934).

BIBLIOGRAPHY

477

— ‘Zum Wortschatz des griechischen Sirach,’ in J. Hempel and L. Rost (eds.), Von Ugarit nach Qumran: Beiträge zur Alttestamentlichen und Altorientischen Forschung. Otto Eissfeldt zum 1. September 1957 dargebracht (BZAW 77; 2nd ed.; Berlin, 1961), 274–87. — ‘Die Vokabel-Varianten der O-Rezension im griechischen Sirach,’ in D. Winton Thomas and W. D. McHardy (eds.), Hebrew and Semitic Studies Presented to Godfrey Rolles Driver in celebration of his seventieth birthday, 20 August 1962 (Oxford, 1963), 172–90. — ‘Ursprüngliche Lesarten im griechischen Sirach,’ in Mélanges Eugène Tisserant (Studi e Testi 231; 7 vols.; Città del Vaticano, 1964), 1:461–87. — (ed.), Sapientia Iesu Filii Sirach (Septuaginta Vetus Testamentum Graecum Auctoritate Academiae Scientiarum Gottingensis editum XII.2; 2nd ed.; Göttingen, 1980). ZORELL, F., Novi testamenti lexicon graecum (Paris, 1911; 5th ed. Rome, 1999). — Lexicon Hebraicum Veteris Testamenti (Rome, 1984).

NOTICE OF ORIGINAL PUBLICATION [WITH PERMISSIONS] 1 ‘A Note on Septuagint Material in the Supplement to Liddell and Scott,’ Glotta 47 (1969), 234–42. By permission of Brill Deutschland GmbH, Paderborn.

2 ‘Aposkeuê in the Septuagint,’ JTS 23 (1972), 430–7. By permission of Oxford University Press.

3 ‘Equivocal and Stereotyped Renderings in the LXX,’ RB 87 (1980), 104–17. By permission of Revue Biblique.

4 ‘The Future of zên in Late Greek,’ NovT 22 (1980), 289–98. By permission of Koninklijke Brill NV, Leiden.

5 ‘Some Features of the Speech of Jesus in Mark’s Gospel,’ NovT 27 (1985), 1–26. By permission of Koninklijke Brill NV, Leiden.

6 ‘A Non-Aramaism in Luke 6.7,’ NovT 33 (1991), 28–34. By permission of Koninklijke Brill NV, Leiden.

7 ‘The United Bible Societies’ Lexicon and its Analysis of Meanings,’ FilNT 5 (1992), 167–89. By permission of the editor of Filologia Neotestamentaria and Ediciones El Almendro.

8 ‘A Lexical Study Thirty Years On, with Observations on “Order” Words in the LXX Pentateuch,’ in S. M. Paul et al. (eds.), Emanuel: Studies in Hebrew Bible, Septuagint and Dead Sea Scrolls in Honor of Emanuel Tov (Leiden: Brill, 2003), 513–24. By permission of Koninklijke Brill NV, Leiden.

9 ‘The Present State of Lexicography of Ancient Greek,’ in B. A. Taylor et al. (eds.), Biblical Greek Language and Lexicography: Essays in Honor of Frederick W. Danker (Grand Rapids, MI: Eerdmans, 2004), 66–74. Reprinted by permission of the publisher, Wm. B. Eerdmans Publishing Co.

10 Review of: T. Muraoka, A Greek-English Lexicon of the Septuagint: Chiefly of the Pentateuch and the Twelve Prophets (2002), BIOSCS 37 (2004), 127– 39. By permission of the editor of BIOSCS.

11 ‘Dimitrios Doukas and the Accentuation of the New Testament Text of the Complutensian Polyglot,’ NovT 47 (2005), 250–90. By permission of Koninklijke Brill NV, Leiden. Image of page of the Complutensian Polyglot vol. 5 (1514) from the holdings of the Biblioteca Nacional de España.

12 ‘Exapostellô,’ in J. Joosten and P. J. Tomson (eds.), Voces Biblicae: Septuagint Greek and its Significance for the New Testament (CBET 49; Leuven: Peeters, 2007), 99–113. By permission of Peeters, Leuven.

480

COLLECTED ESSAYS ON THE GREEK BIBLE

13 ‘A Lexicographical Database for Greek: Can it Be Far Off? The Case of amphodon,’ in M. Karrer et al. (eds.), Die Septuaginta – Texte, Kontexte, Lebenswelten: Internationale Fachtagung veranstaltet von Septuaginta Deutsch (LXX.D), Wuppertal 20.–23. Juli 2006 (Tübingen: Mohr Siebeck, 2008), 214–20. By permission of Mohr Siebeck GmbH & Co., Tübingen.

14 ‘The Complutensian Polyglot, the Text of Sirach, and a Lost Greek Word,’ BIOSCS 42 (2009), 95–108. By permission of the editor of BIOSCS.

15 Review of: T. Muraoka, A Greek-English Lexicon of the Septuagint (2009), BIOSCS 43 (2010), 115–25. By permission of the editor of BIOSCS.

16 ‘Auxiliary thelô,’ in T. V. Evans and D. D. Obbink (eds.), The Language of the Papyri (Oxford: OUP, 2010), 15–34. By permission of Oxford University Press.

17 ‘Releasing LSJ From Its Past,’ in C. Stray (ed.), Classical Dictionaries: Past, Present and Future (London: Duckworth, 2010), 119–38. Bristol Classical Press, by permission of Bloomsbury Publishing Plc.

18 ‘Why Didn’t St Basil Write in New Testament Greek?,’ Phronema 25 (2010), 3–20. By permission of Phronema, Journal of St Andrew’s Greek Orthodox Theological College, Sydney.

19 ‘The Atticist Grammarians,’ in S. E. Porter and A. W. Pitts (eds.), The Language of the New Testament: Context, History and Development (Leiden: Brill, 2013), 283–308. By permission of Koninklijke Brill NV, Leiden.

20 ‘Etymological Follies: Three Recent Lexicons of the New Testament,’ NovT 55 (2013), 383–403. By permission of Koninklijke Brill NV, Leiden.

21 ‘The onoma Rule,’ NovT 56 (2014), 411–21. By permission of Koninklijke Brill NV, Leiden.

22 ‘The Literary Greek of Septuagint Isaiah,’ Sem et Class 7 (2014), 135–46. DOI: 10.1484/J.SEC.5.103523 By permission of Brepols Publishers, Belgium.

23 ‘The Sayings of the Desert Fathers: Their Evidence for Late Koine Greek,’ Phronema 30 (2015), 23–42. By permission of Phronema, Journal of St Andrew’s Greek Orthodox Theological College, Sydney.

24 ‘The Vocabulary of the LXX and Documentary Evidence,’ in E. Bons and J. Joosten (eds.), Handbuch zur Septuaginta: Handbook of the Septuagint, vol. 3 Die Sprache der Septuaginta: The Language of the Septuagint (Gütersloh: Gütersloher Verlagshaus, 2016), 98–108. Author’s copyright.

25 ‘The Puzzle of John 21:15–17: A Formality Solution,’ NovT 59 (2017), 27– 30. By permission of Koninklijke Brill NV, Leiden.

26 ‘The Brill Dictionary of Ancient Greek: New Clothes for an Old Emperor,’ NovT 59 (2017), 415–35. By permission of Koninklijke Brill NV, Leiden.

NOTICE OF ORIGINAL PUBLICATION

481

27 ‘Greek Idiom in the LXX-Pentateuch: The Preposition para,’ JSCS 52 (2019), 39–51. By permission of the editor of JSCS.

28 ‘Back to the Question of Greek Idiom,’ in T. Kauhanen and H. Vanonen (eds.), The Legacy of Soisalon-Soininen: Towards a Syntax of Septuagint Greek (DSI 13; Göttingen: V&R, 2020), 13–25. Vandenhoeck & Ruprecht, by permission of Brill Deutschland GmbH.

INDEX OF GREEK WORDS The following are not indexed: illustrations from Muraoka, Lexicon in Essays 10 (pp. 138– 42) and 15 (p. 225); illustrations of accents in Essay 11 (pp. 147–8); lists of examples from the Atticist grammarians in Essay 19 (p. 313); the list of common Koine vocabulary in Essay 23 (p. 383); the list of words in papyri in Essay 26, §5.4 (pp. 419–20). ἀγαπάζω 259 ἀγαπητός 258–65, 272–6 ἀγαπῶ 259, 400–1, 403–5 ἀγαυρίαμα 362 ἄγγελος 335 ἀγρός 331 ἀδόκιμος 6 ἀεί 303, 311, 365 ἀθετῶ 6 αἴνιγμα 325–6, 329 αἰτῶ 83, 433 αἰχμαλωτίζω 300 ἀκμήν 299–300, 311 ἄλλος 85–6 ἄμφοδον 201–6 ἀμῶ 316, 365 ἄν 49, 67, 100, 451–2, 453–4 ἀνάβα 306–7 ἀναβάζω 382 ἀναβαίνω 90–1 ἀνακαινοτομῶ 157 ἀνδρὸς ἀνδρός 231–2 ἄνθρωπος ἄνθρωπος 445 ἀνίσταμαι 310 ἄνισχυς 363 ἀνομία 98 ἀντιπίπτω 6 ἀπαμαυρῶ 363 ἀπατῶ 329 ἀπάτη 313 ἀπιστέω/-ία/-ος 336 ἀπό 380 ἀπόγραφος 158 ἀποκρίνομαι 39 ἀπὸ λιβός 230 n. 23 ἀποσκευή 11–20, 418

ἀποσκορακίζω 363 ἀποσκορακισμός 363 ἀποστέλλω 83, 183–96 ἀπὸ τοῦ νῦν 381 ἀργός 326 ἀριστερός 307–8, 335 ἄριστον 82–3 ἄρκος 308–9, 311 ἄρκος ἀπορουμένη 235–6 ἄρκτος 308–9, 311 ἀροτήρ 365 ἄρουρα 230 n. 23 ἀρραβών 334 ἄρρην 303, 317 ἄρσην 303, 317 ἀρτύματα 39 ἀρχάριος 383 ἄρχω/-ομαι 83, 386 ἄρωμα 213–6 ἀρωματίζω 215–6 ἀσπάλαθος 209–16 ἄτοπος 69 αὐθέντης 301–2 αὐξάνω/-ομαι 84 αὔξω 365 αὐτός 157, 385 ἀφίω/-ιῶ 382 βαθύφωνος 363 βάλλω 331, 386 βάρβαρος 98 βᾶρις 204 βάσανος 5–6 βασίλειον 204 βασίλισσα 316 βελόνη 315

484

INDEX OF GREEK WORDS

-βηθι 306–7 βιάζομαι 413 βίος 415–7 βιώσομαι 40, 42–3 βλέπω 94 βόλβιτον 39 βουλεύομαι βουλήν 368 βούλομαι 365, 381, 413 βοῦς 333 βοῶ 69 βράζω 383 βρέχω 309 γαμῶ 86 n. 13 γε 49 γεμίζω 380 γέμω 38–9 γεοῦχος 383 για 382 γίνομαι 97 γνωμόνιον 157 γνωρίζω 380 γνώσεται 303 γραικοτύπωσις 158 γραΐς 383 γραμματεύς 370 n. 61 γραμματικός/-οί 370 n. 61, 371 δακρύω 69 δάκτυλος 325, 328 δέδοικα 366 δεῖπνον 82–3 δεσπότης 326, 333 δεῦρο 305–6 δέω 333 δή 364 δηλαυγῶς 333 δήπου 315 διά 382 διαθήκη 31 n. 29 διὰ παντός 303 διαστέλλω 136 διαστολή 136 διὰ τί; 95 διάφορος 39 διαφορότης 303 δίδω/-ῶ 382 δίδωμι 5

δικαίωμα 30 δίκην 157 διχόγνωμος 158 δράξ 383 δύναμις 22 n. 4 δύο 332 δυοῖν 298–9 δυσί 298–9 δύσκολος 325, 328 ἑαυτόν 383 ἐβίων 43 n. 29 ἐγείρομαι 310 ἐγκολλῶ 6 ἔδομαι 316, 413 ἔζησα 43 n. 29 εἰ 93, 368 εἰμι 385 εἰρήνη 29 εἷς 332, 385 εἰς 101, 301 εἴσεται 303 εἰς τὰ ἴδια 386 εἴσω 300–1 εἶχον 385 ἕκαστος 448 ἑκάστοτε 303 ἑκατόν 331 ἐκβάλλω 97 ἐκδέχομαι 28–9 ἐκεῖ 301 ἐκεῖσε 301 ἐκκαίω 2–3 ἐκκλησία 337 ἐκκρούω 6 ἐκλογή 320, 323 ἐκπνέω 69 ἔκπρακτος 218 ἐλπίζω 83 ἐλπίς 83 ἐμπρησμός 300 ἐμπυρισμός 6, 300 ἐν 231 n. 25, 301 ἐν ᾅδου 362 ἔνδον 300–1 ἐνδόσιμον 158 ἐν ἐμοί 233–4 ἔνθεν ἑλών 281

485

INDEX OF GREEK WORDS

ἐντέλλομαι 109–16 ἐντυπῶ 158 ἐντύπωσις 158 ἐξαποστέλλω 183–96 ἐξηγοῦμαι 109 n. 12 ἕξις 49 ἔξωθεν 67, 329 ἐπαίρω 383 ἐπάνω 380, 385 ἐπί 427–8 ἐπιθυμῶ 140 ἐπιούσης/-ῃ 308 ἐπίπεμτος 3 ἐπίσταμαι 366 ἐπιτάσσω 109 n. 12 ἐργασία 157 ἐργολαβία 158 ἐρεθίζω 34 ἐρυθρός 366 ἔρχομαι 413 ἐρωτῆσαι εἰς εἰρήνην 34–5 ἐσθίω 84, 311, 413–4 ἐσθίω τὸν ἑαυτοῦ ἄρτον 84, 85 ἔσο 382 ἐσχάτως ἔχω 314 ἔσωθεν 67 ἕτερος 86 εὖ 56–8 εὐδία 326, 329 εὐίλατος 399–400, 418 εὐλογία 83 εὐπορῶ 77 n. 18 εὑρίσκομαι 76, 380 εὑρίσκω 71–8, 94, 380 εὐρυθμία 157 εὐώνυμος 307–8, 311 ἔφαγον 413 ἐφέτος 386 ἔχω 77, 243, 385 ἔχω πρᾶγμα 386 Ζεύς 326 ζῆλος 169–70 ζήλωσις 169–70 ζηλωτής 169–70 ζῆν 37–46 ζήσω/-ομαι 37–46 ζητῶ 91

ζῶ 94 ζωή 416 ἡγοῦμαι 94, 98 ἤθη 305 ἥκω 86 ἥλιος 333 ἡμέραν ἐξ ἡμέρας ἤν 291 ἤρεσε 303 ἦς 317 ἦσθα 317 ἡττῶμαι 367

387

θάλασσα 228, 230, 325, 328 θέλετε 255 θέλησον/-ήσατε 242–3, 253–4 θέλω 91–3, 237–55, 378, 381, 384–5 θεωρία 383 θεωρῶ 380 θησαυρός 335 θλίβω 381 θυμός 420–1 θύραι 67 ἴδιος 334 ἰδού 386, 447 ἱκανός 34 n. 34 ἱλαρός 333 ἱμάς 333 ἵνα 100, 385–6 ἴσασι 315 ἱσπανία 158 ἴστε 315 καθαίρω 329 καθάπαξ 383 καθάπερ 316 καθαρός 329, 335 καθιζάνω 422 καθό 316 καθότι 316 καὶ γάρ 364 καινοτομῶ 157 καίτοι 422–4 κάλαμος 214, 331 καλός 226 καλῶς 56–8

486 κάμινος 410–1 κάνθαρος 6–7 κατάβα 306–7 καταβαίνω 95 κατάκλιτος 363 καταλέγω 320 καταστρώννυμι 25–6 κάτεργον 134–5 κέλευσον 386 κελεύω 109–16 κεραία 156 κερδαίνω 381 κῆνσος 334 κλαίω 69, 412–3 κλεῖν 306 κοινόν 302–3 κολλήγιον 380 κόλος 325 κόλπος 3–4 κοράσιον 311, 314, 325 κορίδιον 311 κράβαττος 286, 311, 334 κράζω 69 κραυγάζω 325 κρούω 383 κῦδος 362 κυνῶ 404 n. 7 κύω 363 κῶλον 5, 230 n. 22 κωλύω 325 κῶμος 325, 329 λαγύνιον 384 λαλητός 219 λαμβάνω 413, 433 λαμπάδιον 135–6 λέγω 87–9, 109 n. 12, 320 ληκύνθιον 382 λογικός 421 λόγιος 320 λόγοι 289 n. 36 λόγος 133–4 λοιπόν 386 λούω/-ομαι 84 λυπῶ/-οῦμαι 84 μακρόβιος 363 μέγιστος 315

INDEX OF GREEK WORDS

μεθεκτέον 291 μελανός 382 μέλλω 242 μέμφομαι 323–4 μέν 47–52, 97 n. 27, 364 μετά 434 μετάβα 306–7 μεταδετέον 410 μετάνοια 381 μετανοῶ 381 μεταξύ 316 μή 95 μηδαμῶς 327 μὴ θέλε 255 μὴ θελήσῃς/-ήσητε 242–3, 254 μὴ οὐ; 386 μήτηρ 331 μήτι 95 μιμητής 169–70 μισθοφορία 384 μῶμος 323–4 νάρδος 334 νερόν 384 νῆστις 326 νίπτω 327 νουθετέω 421 νοῦς 366 νύξ 32 νώ 303 ξέστης 334 ξύν 280, 284 οἶδας 303 οἴδασι 315 οἶμαι 366 οἶσθα 303 ὅλως 386 ὁμιλία 23 n. 5 ὁμιλῶ 23 n. 5 ὁμίχλη 363 ὄνομα 339–48, 387 ὀπίσω 385 ὀπωροφυλάκιον 9 ὁρῶ 67, 326 ὅς, ἥ, ὅ 332, 444 ὅστις 95

INDEX OF GREEK WORDS

ὅτι; 95 οὐαί 67 οὐ μή 63–7, 381–2 οὖν 67–8, 364 ὄφελον 97 ὀψάριον 419 ὄψις 326 παιδίον 135 παλάθη 208 πάλαθος 208–221 παλαιῶ 156 πανταχόθεν 157 παντοκράτωρ 31, 32, 36 πάντοτε 303, 311 παρά 95, 428–39 παρακαλῶ 86, 89–90 παραπικραίνω 33–4, 36 παρατείνω 8 παραρρέω 361 παρρησία 313–4 παροξύνω 34 παρουσία 313–4 πᾶς 332 πάσχω 331 πατήρ 30–1, 331 πείθω 98 πέμπω 183–96 περίκειμαι 99 περιψήχω 219–20 περνῶ 384 πιστός 337 πλατύνω 9 πνοή 6 ποῖ; 301 ποῖον; 384 ποιῶ 97 n. 27 ποιῶ ἔλεος 381 πολλά 95–6 πολύλαλος 219 πολύς 96, 100 πόρρω 366 πόρρωθεν 366 ποταμός 325 ποτίζω 397–8 ποῦ 301, 326 ποῦ ἔνι; 387 πούς 331

πράσσω 366 πριστηροειδής 363 προκοπή 299 προκόπτειν 299 προσήλυτος 30, 235 προσπορεύομαι 24 προστάσσω 109–16 πρόσωπον 326 προφέρω 99 πρωτοποίητος 157 πυργόβαρις 204 πύργος 204 πυρίκαυστος 363 πύρωσις 6 πῶς 327 πῶς ὁ κόσμος; 387 ῥαβδίσκος 156 ῥαφίς 315 ῥιγάω 382 ῥύμη 316 σαλός 384 σελήνη 335 σκελίζω 7–8 σκίμπους 286 n. 25 σουδάριον 335 σπάλαθος 217 σπλαγχνίζομαι 384 σπουδαῖε 156 στέργω 259 συγγηρῶ 220 συγγίνομαι 22–3, 26–7 σύγκρισις 157 συγχωρῶ 381 συλῶ 384 σύν 280 σύνειμι 23 n. 5 συνομιλῶ 158 συνουσία 23 n. 5 συντάσσω 109–16 συντείνω 157 συντελῶ 157 σωθείης 387 σωτήριος 6 τάλας 360 ταχέως 380

487

488 τε 315, 364 τέκνον 320, 323 τελῶ 384 τέμαχος 299 τέρψις 313 τέσσαρες 333 τί; 7 τίθω/-ῶ 382 τίκτω 320, 323 τίποτε 384 τίς 331 τις 447–51 τό 385 τοιγαροῦν 315, 364 τοίνυν 315, 364 τόκος 82 τόπον ἐκ τόπου 387 τρεῖς 332 τρώγω 306, 311, 414–5 τυγχάνω 232–3 τυπῶ 158 ὕβρις 98 ὕδωρ 327 ὑετός 327 υἱός 5 ὕπαγε 315 ὑπάγω 98

INDEX OF GREEK WORDS

ὑπομονή 421–2 ὑποσκελίζω 8 ὕω 309, 327 φαγίον 384 φάγοι 59–60 φάγομαι 316, 413 φέρω 331 φεύγω 368 φιλῶ 403–5 φράσις 158 φύσει 387 χαίρειν 233 χάρτης 325, 329, 336 χείρ 332 χρή 97, 315 χρυσός 335 ψιχίον 384 ψύλλα 296–7 ψύλλος 296–7 ψυχή 1–2 ὦ 60–3 ὤ 61, 62–3 ὡς, ὡς ἄν 367

‫‪INDEX OF HEBREW WORDS‬‬ ‫מ‬ ‫‪ 430, 438‬מאת‬ ‫‪ 430‬מן‬ ‫‪ 232‬מצא‬ ‫‪ 33–4‬מרה‬ ‫‪ 33‬מרר‬ ‫נ‬ ‫‪ 364–5‬נא‬ ‫‪ 230 n. 23‬נגבה‬ ‫‪ 2‬נפשׁ‬ ‫‪ 5‬נתן‬ ‫ס‬ ‫‪ 362‬סבל‬ ‫‪ 370 n. 61‬ספר‬ ‫ע‬ ‫‪ 430, 439‬עם‬ ‫‪ 368‬עצו עצה‬

‫א‬ ‫‪ 30–1‬אב‬ ‫‪ 360‬אוי‬ ‫‪ 232, 445–6, 448‬אישׁ‬ ‫‪ 445‬אישׁ אישׁ‬ ‫‪ 446‬אישׁ כאחיו‬ ‫‪ 428‬אצל‬ ‫‪ 201, 204–5‬ארמון‬ ‫‪ 230 n. 23‬אשׁל‬ ‫‪ 443‬אשׁר‬ ‫‪ 439‬את‬ ‫ב‬ ‫‪ 233‬בי אדוני‬ ‫‪ 5‬בן‬ ‫‪ 2‬בער‬ ‫‪ 31 n. 29, 234 n. 35‬ברית‬ ‫‪ 216‬בשׂם‬ ‫‪ 8‬בתרון‬ ‫ג‬ ‫‪ 364‬גם‬

‫פ‬ ‫‪ 5‬פגר‬ ‫צ‬ ‫‪ 31, 32‬צבאות‬ ‫‪ 109‬צוה‬ ‫‪ 115‬צוה מצוה‬ ‫‪ 3–4‬צלחת‬ ‫ק‬ ‫‪ 7–8‬קלע‬ ‫‪ 24‬קרב‬ ‫‪16, 19‬‬

‫ר‬ ‫רכושׁ‬

‫שׁ‬ ‫‪ 358‬שׁבר‬ ‫‪ 25‬שׁחט‬ ‫‪ (Aram.) 71‬שׁכח‬ ‫‪ 29, 35, 233‬שׁלום‬ ‫‪ 188 n. 15‬שׁלח‬ ‫‪ 339–40, 345‬שׁם‬

‫‪23‬‬ ‫‪356‬‬

‫ה‬ ‫היה‬

‫ז‬ ‫זה אל־זה‬

‫ט‬ ‫‪ 15–7, 19, 20‬טף‬ ‫י‬ ‫‪ 27‬ידע‬ ‫‪ 228‬ים‬ ‫‪ 232‬יסף‬ ‫כ‬ ‫‪ 5–6‬כלמה‬ ‫‪ 33‬כעס‬ ‫‪ 6–7‬כפיס‬ ‫ל‬ ‫‪ 22–3‬להיות‬

INDEX OF BIBLICAL REFERENCES References to main discussions are given. Mentions in footnotes are not noted unless significant. The list of LXX-Pent. examples on p. 451 is not indexed. Genesis 2:6 2:19 3:1 3:11 4:19 6:5 8:10–12 11:10 12:5 12:20 13:14 14:12 15:11 15:14 18:2 19:5 19:11 22:7 22:8 24:23 28:1 29:20 30:28 31:28 32:27(28) 36:37 38:4 39:10 40:3, 7 42:2 43:8 43:16 43:29 44:32 46:5 49:13

398 340, 348 64 110 340 450 232 5 2 110 228, 230 16 5 n. 14 380 26 443 135 135 437 n. 110 432 136 367 341 432 341 22 437 192 14, 20 443 135 28 17 431

Exodus 1:15

341

1:17 2:13 2:14 3:13 3:15 7:6 9:18 10:8–11 10:24 12:35 12:37 15:3 16:4 16:20 18:3 18:19 20:5 20:17 21:7 22:5(4) 22:9(8) 22:20 23:21 24:14 30:16 30:20 30:23 30:34 31:17 33:8 33:21 35:21 36:2 39:22

110 308 241, 251, 255 340, 341 347 110 69 14 15 110 14, 20 339, 340 69 449 340 134 64 140 450 449 447 n. 30 341 27 n. 134 24 214, 216 214 434 453 434 134 24 18

Leviticus 5:4 5:16 6:40(7:10) 13:54

136 3 446 111

492 15:2 17:3 20:2 25:17 25:29

INDEX OF BIBLICAL REFERENCES

232 n. 445 232 n. 446 450

Numbers 2:34 5:12 5:20, 29 11:3 14:16 14:29 16:27 19:2 25:8 25:14, 15 30:7 32:26

111 231 27 n. 341 25 5 17 136 411 341 136 18

Deuteronomy 5:21 10:12 18:3 19:5 20:13–1 22:20 27:1 31:27 32:16

140 433 433 232 16 134 110 34 33

Joshua 1:7

367

Judges 1:23 6:13, 15 13:8 18:29

341 234 234 341

Ruth 2:1, 19

341

1 Kingdoms 1:26 2:1–10

233, 234 191

2 Kingdoms 2:29

8

3 Kingdoms 3:17, 26 20(21):21

234 2

4 Kingdoms 18:32

41

2 Chronicles 20:25

14

Psalms 98:8 137:7 142:11

399 41 41

Proverbs 19:21(24) 24:43(28) 26:15

3 9 3

Song of Songs 4:13

5

Isaiah 1:7+ 1:11 2:16 3:1 3:9+ 3:10 3:23 5:26 6:2 6:3 6:5 8:9 10:3 10:7 11:9 14:25 17:5 17:13 21:1 23:10 29:11 29:18 30:8 30:13–14

363 365 356 364 368 364 363 364 364 355 360 367 366 366 367 362 365 363, 366 367 364 366 363 364 357

493

INDEX OF BIBLICAL REFERENCES

33:19 37:12 37:30 38:18 40:25 40:30 41:7 41:15 42:8 42:14 43:14 44:4 44:18 48:22 49:15 51:13 51:15 53:10 57:6 57:8 57:10 57:21 59:4, 13 60:14 61:5 61:11 62:7 63:2 66:15 66:20 Jeremiah 10:18 17:27 23:12 30:16 39:29

363 364 365 362 364 363 364 363 341 365 368 361 363 233 368 365 341 363 364 366 366 233 363 366 365 365 362 366 363 367

7 201, 204 8 201, 204 33

Habakkuk 2:11

6

——— 1 Esdras 8:53

399

Tobit 3:10 6:15 10:12

250, 263 250 263

Judith 7:2

19

Wisdom 12:6

301

Sirach 6:35 7:13 10:8 11:16 18:33 21:25 24:15 30:7 41:19

255 255 218 220 219 219 209, 212–6 219 363

Baruch 3:14

363

1 Maccabees 10:33

2

2 Maccabees 12:21

19

———

Ezekiel 16:52, 54 32:24, 30

5 5

Hosea 12:12 13:8

231 n. 235

Micah 3:12

9

Matthew 4:3 5:34 6:13 9:14 10:2 13:22 15:16 15:27

88 88 169 96 345 313 300 384

494

INDEX OF BIBLICAL REFERENCES

15:28 17:17 21:19 25:21+ 26:14 26:15 26:53 26:69

62 62 60 317 92 248 90 317

Mark 1:16 1:44+ 2:27 4:4 4:19 5:9+ 5:10 5:14 6:48 7:4 7:11 7:15 7:18 7:20 7:21, 23 7:28 8:12 9:12 9:19 10:32 11:14 12:2 12:5 12:44 13:20 13:29 14:7 14:21 14:38 14:67 15:34 15:37, 39 16:8 16:19

95 67 382 50 313 345 96 68 92, 248 380 67 67, 68 67, 68 68 67 384 93 51 62 68 59 59 n. 51 68 67 67 57 51 51 317 69 69 194 51

Luke 1:1–4 1:26+

283 345

1:27 1:53 1:63 6:7 9:31 9:41 11:7 11:54 12:15 13:24 16:23 20:10, 11 23:22 23:41 23:46 24:25 24:38 24:49

345 191 347 71 88 62 67 71 67 71 67 191 69 69 69 62 90 194

John 1:6+ 1:43 4:33+ 6:21 8:6 11:35 11:54 13:22 16:19 16:32 18:17, 25 21:15–17

345 249 95 249 77 69 314 94 91 386 95 n. 403–5

Acts 1:1 7:12 7:26 8:26 8:40 9:30 9:36 11:22 12:11 13:6 13:26 14:13 15:29 16:11+

62 192 308 95 380 193 88 193 193 345 192 249 57 308

495

INDEX OF BIBLICAL REFERENCES

17:14 17:28 19:2 19:33 21:3 22:21 24:19 26:4 26:7 26:29

193 284 93 249 308 n. 192 59 315 312 59

Romans 8:3 12:1

193 421

1 Corinthians 5:1 9:4 10:5 15:29 15:33

387 386 26 387 284, 305

Galatians 1:6 4:4, 6

380 193

Ephesians 6:3

57

Colossians 4:6

382

2 Thessalonians 2:2 380 1 Timothy 5:22 6:2

380 89

Titus 2:15

89

Hebrews 2:16 12:17

315 315

James 3:3 5:4

315 316

2 Peter 1:4 2:13

315 313

Revelation 6:8 9:11 10:2 13:2 18:13

345 345 308 n. 309 2

INDEX OF MODERN AUTHORS Abbott-Smith, G. 320–1, 334 n., 336–7 Adams, J. N. 169 n., 230 n., 234 n. Adams, S. A. 316 n. Adler, A. 208 n., 211 Adrados, F. R. 239 n.; see also DGE Aejmelaeus, A. V, XV, 207 n., 230 n. Aitken, J. K. V, 35 n., 70, 127, 197 n., 255, 359 n., 399 n., 400 n., 418 n., 428 n., 446 n. Aland, B. 145 n.; see also BRAA Aland, K. 50 n., 145 n.; see also BRAA Alimeni, G. 179 n. Allen, W. C. 69 n. Andersen, F. I. 234 n. Anderson, G. 294 n., 296 n., 312 n. Anz, H. 111 n., 184 n., 390 Argyle, S. 308 nn. Argyriades, G. P. 149 n. Arndt, W. F. see BAG, BAGD, BDAG Arnold, B. T. 190 n., 192 n. Arnott, W. G. 69 n., 304 n. Arnold, W. R. 8 n. Arribas Arranz, F. 179 n. Auwers, J.-M. 209 nn. Babiniotes, G. D. 195 n. Baer, D. A. 351, 361 n., 367 n. BAG IX, 25 n., 30, 72 n. BAGD 40 n., 43 n., 50 n., 51 n., 54 n., 57 n., 61 n., 64 n., 72 n., 75, 84 n., 82–6, 88–95, 99, 102–3 Bagnall, R. S. 393 nn., 394 n., 396, 397 n. Bailly, A. 409, 423 Balz, H. see EDNT Barber, E. A. see LSJ Suppl. Bardy, G. 282 n. Barker, N. 149 n. Barr, J. 7 n., 9, 233 n., 353 n., 354–5 Barrett, C. K. 95 n., 190 n., 403 Barthélemy, D. 41 nn. Bastianini, G. 185 n. Bataillon, M. 165, 166 n., 168 n., 181

Bauer, H. 177 n. Bauer, W. 320 n.; see also BAG, BAGD, BDAG, BRAA Bauer2 (1928) IX, 121 Baumgartner, W. see HALOT; KB, Lexicon BDAG IX, 6 n., 30, 72 n., 75 n., 121, 126 n., 127, 139 n., 156–8, 184, 185 n., 187 n., 202, 240 n., 267 n., 268 n., 306 n., 315 n., 322, 332 n., 381 n., 386 n., 401, 420 n. BDB IX, 7, 8, 15, 23 n., 204 BDF IX, 40 n., 43 n., 50 n., 51 n., 54 n., 56 n., 57 n., 58 nn., 59 nn., 61 n., 65 n., 65 n., 95 n., 284 nn., 307 n., 312 n., 315, 316 n. BDR IX, 59 n., 61 n., 307 n. Beek, L. van IX Beekes, R. IX, 125 n., 321, 335–6, 363 n. Beentjes, P. C. 216 n. Bekker, I. 38 nn., 39 nn., 298 n., 304, 305 nn. Bentein, K. 70, 197 n. Bentley, J. H. 145 n., 165, 166, 167, 170 n. 37, 181 Bérard, F. 395 n. Bethe, E. 311 n. Betz, H. D. 193 n. BHS  26 n. Bihlmeyer, K. 194 n. Black, M. 71 Blass, F. 62 n.; see also BDF, BDR, KB, Grammatik Blomqvist, J. 47 n., 364 nn., 423, 424 n. BM IX, 26 n. Boisacq, É. 321, 333–4 Boling, R. G. 27 n. Bolinger, D. 52 n. Boned Colera, P. see RBLG Bons, E. 235; see also LXX.H, 3 Bowie, E. L. 294 n. Boyd-Taylor, C. 136, 137 n., 228 n., 229 nn., 236, 354 n., 369 n.

498

INDEX OF MODERN AUTHORS

BRAA 72 n. 4, 75 n. 11, 127 Brandwood, L. 42 n. Bratcher, R. G. 92 n. Braun, A. 239 n. Breitenstein, U. 61 n. Brenton, L. C. L. IX, 23 n. Brock, S. P. 26 n., 106 Briggs, C. A., see BDB Brooke, A. E. see BM Brown, F., see BDB Brown, R. E. 403 n., 405 n. Browne, G. M. 75 Browning, R. 40 n., 53 n., 54 n., 58 n., 238, 239, 285 n., 286 n., 290 n., 294 n., 296 n., 313 n., 346 n., 374 n., 377 nn., 378 n., 385 nn., 396 Bruce, F. F. 193 n. Büchner, D. L. 200 n. Buck, C. D. 113 n., 202; see also BuckPetersen Buck-Petersen 157, 217 n., 363 n., 382 n. Burguière, P. 239 n. Burney, C. F. 345 n. Burridge, R. A. 283 n. Butler, Dom C. 374 n. Cadbury, H. J. 50 n., 69 n., 184 n., 192 n., 314 nn., 315 nn. Cadell, H. 108 n., 392, 398 n. Caird, G. B. 135 n., 136, 204 n., 205 n., 411 n. Cañas Reíllo, J. M., see BGS Cannan, C 265 n. Carroll, S. 52 n. Carson, D. A. 337 n. Case, S. J. 205 n. Cavallo, G. 150 n. CGL IX, 126, 278 Chadwick, J. 9, 106, 119–20, 139 n., 187 n., 258–260, 263, 265 n., 267 n., 268, 269 n., 276 n., 277, 278, 410 n., 414, 417 n., 420 n. Chantraine, P. X, 30, 56 n., 125 n., 213 n., 217 n., 275 n., 297 n., 308 n., 321, 323 343 n., 416 n. Charles, R. H. 345 n. Chatzidimos, A. 166 n. CHB 145 n.

Christchev, Th. 169 n. Chryssavgis, Fr M. 387 n. Clackson, J. 343 n. Clarysse, W. 239 n. Clines, D. J. A., see DCH Cobet, C. G. 43 n. Cohn, L. 270 n., 307 n., 310 n. Coles, R. A. 247 n. Collard C. 247 n. Connolly, A. L. 294 n. Contopoulos, N. 156–8 Costache, D. 279 n. 1, 290 Cotelier, J.-B. 375 Cowley, A. E., see GKC Cox, C. E. 422 n. Crastonus, J. 210 n., 211 Cribiore, R. 286 n., 187 nn., 294 n. Cross, F. M. 41 n. Cullen, M. 178 n. Cunningham, I. C. 208 n. Curran, M. 144 n., 189 n., 207 n., 293 n., 319 n., 342 n., 369 n., 407 n. Curtius, G. 321 n. Dain, A. 308 nn. Daniel, S. 6 n. Danker, F. W. 226 n., 234 n., 320 n., 321, 330–7, 391; see also BAGD, BDAG Dardano, M. 179 n. Daris, S. 203 n. Darlow, M. 145 n., 170 n. DCH 20, 32 n., 204 De Borries, J. 298 n. Debrunner, A. 39 n., 58 n.; see also BDF, BDR; Schwyzer, Grammatik Decker, R. J. 66 n. Deferrari, R. J. 281 n., 282 n., 288 n., 289 n., 291 n. Deissmann, A. 4 n., 106, 390–1 De Jonge, H. J. 250 n. Denniston, J. D. 50 n., 51 n., 422–4 Depauw, M. 272 n. De Rossi, F. Canali 401 n. De Sousa, R. F. 351, 354 n., 370 n. De Troyer, K. V DGE X, 120, 124, 136, 156–8, 184, 186 n., 202, 240, 274 n., 278, 388 n., 407, 409 n. DGENT X, 127

INDEX OF MODERN AUTHORS

Dhont, M. 35 n. Dickey, E. 298 n., 302 n., 304 n., 307 nn., 310 nn., 311 n. Diego Sigeo 173 n. Diels, H. 270 Dieterich, K. 239 Diggle, J. IX Dihle, A. 301 Di Lella, A. A. 212 n., 214, 215 n., 220 n. Dines, J. M. 226 n. DMLBS X, 139, 169 n., 408 Dogniez, C. 233 n., 398 n. Dorival, G. 109 n., 136, 187 n., 411 n.; see also HDM, La Bible grecque Doukas, D. 150 n., 153, 158, 164–82 Driver, G. R. 7, 9 Driver, S. R., see BDB Du Bouchet, J. 202, 204 n., 205 n. Ebeling, H. 321 n. Edmonds, J. M. 40 n., 274 n. EDNT X, 193 n. Eissfeldt, O. 7 n. Elliott, J. K. 37 n., 51 n., 315 n., 317 Ellis, J. 52 n. 22 Epp, E. J. 317 n. Erasmus, Desiderius 178 n. Erbse, H. 307 n. Evans, T. V. V, 58 n., 105 n., 144 n., 237 n., 241, 342 n., 367 n., 368 n., 373 n., 378 n., 392, 401 n., 403 n., 411 n., 428 n., 435 n. Exler, F. X. J. 49 n. Eynikel, E., see LEH, Lexicon Fabricius, C. 281 n. Farmer, W. R. 50 n. Fausto, V. 166 Fee, G. D. 37 n., 284 n., 317 n. Fernández-Armesto, F. 170 n. Festugière, A.-J. 378 n. Feyerabend, K. 269 n. Fischer, E. 298 n., 300 n., 304 n. Fitzmyer, J. A. 71, 72, 190 nn., 191 n. Fleischman, S. 239 n. Fountotos, J. 404 n. Fox, W. 239 n. Frame, J. E. 65 n. Fränkel, H. 30

499

Fraser, P. M. X, 297 n. French, R. 298 n. Frisk, H. X, 125 n., 321 Frösén, J. 54 n., 55 n., 295 n. Funk, R. W., see BDF Furguson, R. A. 293 n. Gaisford, T. 208 n., 211, 219 n. Gallay, P. 282 n. García Santos, A. A. 127 Garlan, Y. 203 n. GE X, 126, 202 n., 278, 407–26 Geanakoplos, D. J. 145, 149 n., 150 nn., 160 n., 165, 166 n., 167, 168, 173 n., 181 n. Gehman, H. S. 230 n. Gerleman, G. 367 n. Gerth, B. see KG, Grammatik Ghisalberti, A. M. X GI X, 126, 202, 278, 407–26 Giannakis, G. K., see EAGLL Giannakopoulos, I. 214, 368 n. Gignac, F. T. 60 n., 239, 382 nn. Gingrich, F. W. 320, 396; see also BAG, BAGD, BDAG Glare, P. G. W. 410 n.; see also LSJ Suppl.; OLD Glaser, O. 184 n. Goetz, G. 219 n. Goh, M. 409, 426 n. Gomme, A. W. 274 n. Goodspeed, E. J. 42 n. Gray, G. B. 26 n., 411 n. Greenlee, J. H. 323–7 Greenslade, S. L. X Gregory, C. R. 145 n. Gregory, M. 52 nn. Gundermann, G. 219 n. Guy, J.-C. 374 n., 375–6, 377 n., 384 n. Haenchen, E. 57 n., 65 n., 190 n., 312 Hahn, E. A. 343 n. Hall, B. 145, 159 n. HALOT X, 32 n., 204 Hamer, E. 150 n. Hanhart, R. XII Hansen, D. U. 302, 307 nn., 308 n., 310 n. Harl, M. 112 n., 228 n., 233 n., 236, 398 n.; see also HDM, La Bible grecque

500

INDEX OF MODERN AUTHORS

Harris, G. 50 n. Harrison, N. V. 290 Hart, J. H. A. 210 n. Hatch, E. see HR, HR Suppl. Hatzidakis, G. N. 238 n., 295 n. Hauspie, K. 10; see also LEH, Lexicon HDM, La Bible grecque X, 106 n., 107 n., 108 n., 392 n. Helbing, R. 24 n. Hemer, C. J. 391 Henderson, J. 278 Hennig, D. 203 n. Herkenne, H. 216 Hernitschek, E. V Hickie, W. J. 320 Hill, C. 178 n. 47 Hock, H. H. 238 n. Hohlwein, N. 203 n. Hollander, H. W. 169, 250 n. Holleaux, M. 11 n., 12 Holton, D. 145 n., 147 n., 148 n., 238 n. Homza, L. A. 170 n., 176 n., 181 n. Hopper, P. J. 237 n., 240 n., 243 n. Horrocks, G. C. 157, 187 n., 238 n., 239, 294 n., 295 n., 312 nn., 377 nn., 378 nn., 385 nn., 396, 448 n., 450 n. Horsley, G. H. R. IX, 122, 126, 391, 395 n.; see also New Documents Hort, F. J. A. 347 Howard, W. F. see MH, Grammar, II Howlett, D. R. 169 n.; see also DMLBS HR X, 64 n., 209 n. HR Suppl. X, 209 n. Hudson, R. A. 293 n. Hugo, P. 371 n. Humbert, J. 60 n., 76 n. Hunkin, J. W. 386 n. Imholtz, A. A. 268 n., 410 n. Jacques, X. 324 Jaeger, W. 290 n. James, A. W. 30 James, P. 265 n. Jannaris, A. N. 48 n., 62 n., 238 n. Janse, M. 347 Jellicoe, S. 4 n., 107, 212 n. Jeremias, J. 65 n. Jobes, K. H. 106 nn., 107 n., 236

Johannessohn, M. 61 n., 428 n. Johnson, A. F. 178 nn. Joly, R. 265 n., 400 n., 404, 405 nn. Jonge, M. De 46, 165 n., 170 n., 181 n. Joosten, J. 236, 367 n.; see also LXX.H, 3 Joseph, B. D. 238, 238 n., 239 Jouguet, P. 203 n. Joüon, P. 177 n., 239 Joyal, M. 288 n. Judge, E. A. 169 n., 279 n., 283 n. Kalogerakis, A. 404 n. Karalis, V. 144 n. Kariatlis, P. 279 n. Karrer, M. M. see also LXX.D, LXX.D Komm. Katz, P. 41 n., 278, 352 Kauhanen, T. 441 n. Kautzsch, E., see GKC Kazazis, J. N. 285 n., 294 n., 295 n., 296 n., 307 n., 312 n. KB, Grammatik X, 40 n. KB, Lexicon X, 4 n., 7, 8, 23 n., 204 Kefallinos, E. 149 n. KG, Grammatik X, 50 n., 63 n., 339 n. Kieffer, R. 37 n. Kiessling, E. 11 n. 13 n.; see also Preisigke, Wörterbuch Kilpatrick, G. D. 37–40, 46, 54 n., 54 n., 55 n., 56 n., 59 n., 305 n., 315 n., 317, 318, 324 Kim, L. 312 n. Kinzig, W. 282 n., 284 nn. KJV 95 n., 219 nn., 220 n. Klamp, D. 203 n. Koehler, L. see HALOT, KB, Lexicon Kolitsaras, I. Th. 195 n., 214, 368 n. Kooij, A. van der 351, 352 n., 354 n., 357, 370–1 Kotsevalov, A. 294 n. Kraus, W. see LXX.D, LXX.D Komm. Kreuzer, S. 428 n. Kriaras, E. XI, 123, 195 n., 208 n., 213 n., 215 n., 217 n., 388 Kühner, R. see KB, Grammatik; KG, Grammatik Ladas, G. G. 166 n. Lagrange, M.-J. 65 n., 66 n.

INDEX OF MODERN AUTHORS

Lampe, PGL XI, 30, 75 n., 123, 156–8, 184, 202 n., 218 n., 219, 239, 263 n., 289 n., 377, 383 n., 388, 421 n., 422 Lane, W. L. 60 n. La Roi, E. 197 n., 255 Lascaris, C. 149 n. Latham, R. E. see DMLBS LBG XI, 123 Leander, P. 177 n. Le Boulluec, A. 109 n., 135 Lee, J. A. L. passim Lee, K. H. 247 n. Legg, S. C. E. 50 n., 51 n., 59 n. Legrand, É. 150 n., 153, 154, 157, 165 n., 166 n., 168 n., 169 n., 171 n., 173 nn. LEH, Lexicon XI, 106 n., 108, 109 n., 121–2, 130, 133, 188 n., 201, 202, 218 n., 235, 392 Lemmelijn, B. 354 Le Moigne, P. 357 Lewis and Short XI, 215 Lex. Acad. XI, 213 n., 215 n., 217 n. Lex. Proia XI, 156–8, 195 n., 297 n., 404 n. Liddell, H. G. see LS, LSJ, LSJ Suppl. Lietzmann, H. 74 n. Lindars, B. 106 Llewellyn Davies, A. 411 n. Llewelyn, S. R. see New Documents Lloyd-Jones, H. 302 n. LN XI, 79–103, 112 n., 121, 127, 139 n., 187 n., 190 n., 240 n. Loader, J. A. 30 Loader, W. 140 n. Lobeck, Chr. A. 298 n., 300, 301 n. Lolos, Y. 202 Longenecker, R. N. 193 n. Louth, A. 290 n. Louw and Nida see LN Louw, J. P. 80 n., 81 n., 84 n., 85 n., 103, 240 n., 268 n., 337 n.; see also LN Louw, T. A. W. van der 355n., 357, 441 n., 442 n., 446 n. Lowry, M. 165 n., 173 n. LS1 (1843) 259–60, 324, 424 LS1 Abridged 324 LS2 (1845) 273 n. LS4 (1855) 260–1

501

LS5 (1861) 201 LS8 (1897) 135 n., 261–2 LSJ 6 n., 23 n., 24 n., 25 n., 27 n., 40 n., 43 n., 63 n., 72, 74 n., 76 n., 111 n., 119– 20, 124, 135 n., 136, 156–8, 184, 200, 201, 213 n., 219, 220 n., 239, 257–78, 299 n., 321, 380 n., 388, 396, 407–25 LSJ Suppl. (1968) 1–10, 17 n., 119 LSJ Suppl. (1996) 10, 119, 201, 263 n., 321 n., 398 n., 411 n., 410, 418, 419, 421 n. Lust, J. see LEH, Lexicon LXX.D XI, 214 n., 368 n. LXX.D Komm. XI, 360 n. LXX.H, 3 XI Lyell, J. P. R. 159 Maas, P. M. 281 n. MacAlister, S. 64 n., 73 n. Maehler, H. 150 n. Mahaffy, J. P. 390 Maidhof, A. 302 n. Mandilaras, B. G. 40 nn., 58 n., 158, 239, 294 n., 386 n. Markopoulos, T. 244 n., 254 Marshall, G. 258 n. Marshall, I. H. 191 n., 192 n. Martini, C. M. 37 n., 40 n. Masson, É. 334 n. Mateos, J. see DGENT Matthews, E. see Fraser, LGPN Mauersberger, A. 112 n. Mayser, E. 48 n., 50 n., 56 n., 60 n., 63 n., 241 n., 339 n., 342 n., 396, 453 n. McGuire, M. R. P. 291 n. McLean, B. H. 393 n., 395 n., 396, 397 n. McLean, N. 106 n. 4; see also BM McNeile, A. H. 411 n. Meecham, H. G. 366 n. Meer, M. N. van der 227 n., 351 n., 352, 359, 363 n., 367 n., 391 n., 392 Mega Lexikon X, 120, 156–8, 184, 195 n., 202, 218 n., 220 n., 378 n., 388 Melancthon, P. 178 n. Melandinos, D. 149 Metzger, B. M. 144, 150 n., 174 n., 194 n. MH, Grammar, II XI, 54 n., 306 n., 307 n. MHT, Grammar, III XI, 22 n., 50 nn., 57 n., 58 n., 59 nn., 60 n., 61 nn., 63 n.,

502

INDEX OF MODERN AUTHORS

95 n., 312 n., 314 n., 315 nn., 316 nn., 345 nn. MHT, Style XI, 54 n., 65 n. Migliorini, B. 179 Migne, J.-P., PG XII, 375, 377, 378 nn. Migne, J.-P., PL XII Mihevc-Gabrovec, E. 238 n. Millar, F. 378 n. Milligan, G. 202, 391; see also MM Minnen, P. van 393 n. Mirambel, A. 239 n. MM XII, 30, 56 n., 64 n., 66 n., 72 n., 102–3, 107, 122, 184, 202, 334 n., 391, 396 Moffitt, D. M. 235 Montanari, F. 407–9, 425–6; see also GE, GI Montevecchi, O. 107, 391, 392 Morgan, T. 286 n. Morison, S. 149 n., 150 n. Morris, L. 403 n., 404 n. Morwood, J. 269 n. Moule, C. F. D. 54 n., 65 n. Moule, H. F. 145 n., 170 n. Moulton, J. H. 391; see also Moulton, Grammar, I; MH, Grammar, II; MHT, Grammar, III; MHT, Style; MM Moulton, Grammar, I XI, 40 n., 58 n., 59 n., 60 n., 65 nn., 66 n., 196 n. Mousouros, M. 157, 158, 171 Muellner, L. 409, 426 n. Munnich, O. see HDM, La Bible grecque Muraoka, T. 106 n., 108, 121, 127, 129– 42, 177 n., 187 n., 202 n., 223–6, 356 n., 368 n., 392, 396, 398 n., 401, 420 n., 421, 427 n., 437 n., 453 n. Musaeus 158, 171 n. Muses, C. A. XII, 23 n. Mussies, G. 60 nn. NA27 347 NA28 XII, 77 n. Naber, S. A. 38 n., 304 n. Nagy, G. 409, 426 n. NEB 15 n., 95 n., 214 Nebrija, A. de 167, 170, 178, 180–1 Nestle, E. 145 n.

NETS XII, 23 n., 214, 218 n., 219 n., 220 nn., 226–34, 348, 368 n., 411 n. New Documents XII, 102 n., 203 n. Newman, B. M. 99, 121, 320, 321, 327– 30, 336–7 Ngunga, A. T. 351 n. Nickau, K. 310 n. Nicolaus de Lyra 161 Nida, E. A. 52 n., 80 n., 85 n., 92 n., 268 n.; see also LN NIV 95 n. Norden, E. E. 284, 314 nn. Norton, F. J. 161 n. NRSV 95 n. 24, 219 nn., 220 n., 227 Núñez de Guzmán, H. 181 Oates, J. F. IX, 395 n. Obbink, D. D. 237 n. OCD 302 n., 311 n. O’Connell, S. 210 nn., 216 n. O’Connor, M. P. 20 OED XII, 139, 268, 271, 408 Ó Fearghail, F. 190 n. OGELD XII, 156–8 OLD XI, 139, 268, 408 Olin, J. C. 170 n. Olley, J. W. 359 n. Orlandos, A. K. 202 n. O’Sullivan, J. N. 73 n. Ottley, R. R. 22 n., 352, 356 n., 360 n., 362 n., 368 n. Page, D. L. 76 n. Palme, B. 393 n. Palmer, L. R. 69 n. Pappas, P. A. 238 n. Parke, H. W. 12 n. Parker, B. 23 n. Parsons, P. 302 n. Paschos, P. B. 375 Passoni dell’Acqua, A. 392 Passow, F. 257, 260, 409, 424 Pauly-Wissowa, RE XII, 42 n. Peláez, J. 103 Pelletier, A. 112 n. Pentzas, B. 375 Perkins, L. J. 340 Perry, B. E. 73 n.

INDEX OF MODERN AUTHORS

Petersen, W., see Buck-Petersen Philippaki-Warburton, I. 145 n. Pietersma, A. 5 n., 10, 226–30, 371 n.; see also NETS Pinkster, H. 239 n. Poliziano, A. 162, 180–1 Porter, S. E. 103, 283 n. Preisigke, F. XII, 6 n., 11 n., 74, 122, 135 n., 202, 203 n., 239 n. Prellwitz, W. 321 n. Proctor, R. 149 n., 150 n., 159–61, 162 nn. Psichari, J. 239 n. Quasten, J. 282 n., 290 n. Rabelais 169 Rahlfs, A. XII, 41 n., 111 n., 209, 348 Rajak, T. 226 n. RBLG XII, 206 n. 23 Redpath, H. A. see HR, HR Suppl. Reese, J. M. 302 n. Reeve, A. 145, n. Regnault, L. 375 Reichmann, V. IX Reiterer, F. V. 215 n. 26 Rengstorf, K. H. 42 n. 27 Renner, T. 394 n. Renouard, A. A. 150 n. Richardson, M. E. J., see HALOT Riesenfeld, H. 239 Rink, H. 203 n. Ritschl, F. 309 n. Robert, L. 203 n. Roberts, T. 342 n., 411 n. Robertson, A. T. 58 n., 65 n., 345 n. Rocci, L. 408–24 Rödiger, R. 239 Rönnegård, P. 374 nn. Rosenbach, Z. 129 Ross, W. A. 109 n. Roumanis, E. 387 n. Rousseau, P. 280 n., 282 n., 289 n., 290 n., 291 n. Rowe, G. O. 357 n. RSV 90, 95 n. Ruijgh, C. J. 452 n. Ruhnken, D. 304 n. Russell, N. 378 n.

503

Rutherford, W. G. 298 n. RV 411 Rydbeck, L. 54 nn. Sáenz-Badillos, Á. 165 n., 173 n., 180, 181 nn., 211 n. Sakalis, D. T. 302 n. Sandbach, F. H. 274 n. Sandevoir, P. 109 n., 135 Satterthwaite, P. E. 230 n. SBLHBS 126 n. Scherer, A. 39 n., 58 n. Schleusner, J. F. 7, 8 n., 108, 121, 129, 208 n., 360 n. Schmid, W. 40 n., 56 n., 294 n., 301 n., 312, 315 n. Schmidt, J. H. H. 112 n. Schneider, G. see EDNT Scholderer, V. 149 n., 150 n., 159 nn., 162 nn., 163 n. Schrenk, G. 239 Schubert, P. 397 n. Schwyzer, Grammatik XII, 43 nn., 44 n., 60 n., 63 n., 150 n., 217 n., 307 n., 339 n. Scott, J. A. 60 n. Scott, R., see LS, LSJ, LSJ Suppl. Screech, M. A. 145, 169 n. Scrivener, F. H. A. 144 Seeligmann, I. L. 352, 354 n., 360 n., 362 n. Shaw, F. 234 n. Sheldon, J. 178 n., 343 n. Shipp, G. P. XII, 30, 69 n., 74 n., 106, 300 n., 302 n., 303 n., 306 n., 309 n., 310 n., 313 n., 314 n., 315 n., 316 n., 366, 367 n., 381 n., 383 nn., 404 n. Silva, M. 84 n., 106 nn., 107 n., 227 n., 230 n., 234 n., 236, 337 n., 404 n. Simpson, E. K. 69 n. Sirivianou, M. G. 384 nn. Skehan, P. W. 212 n., 214, 215 n., 220 n. Skinner, J. 15 n. Sklavenites, T. E. 149 n. Smend, R. 209 n., 214, 214 n., 216 n. Smith, H. P. 15 n. Smyth, H. W. 43 n., 76 n., 154, 339 n., 357 n., 413 n. Soisalon-Soininen, I. 428 n., 430 n., 441– 4, 447, 453

504

INDEX OF MODERN AUTHORS

Sollamo, R. 349, 353 n., 392, 428 n., 447 n., 448 n. Sophocles, E. A. 75 n., 123 n., 156–8, 202, 208 n., 218 n., 239, 377, 383 n. Sotiropoulos, D. 54 n. Souter, A. 265 n., 278 n., 320 Sparks, H. D. F. 65 n., 190 n. Speiser, E. A. 27 n. Spottorno, V. 95 n. Staikos, K. S. 149 n. Stamm, J. J., see HALOT Stasinopoulos, M. D. 280 n., 288 n., 291 n. Stavropoulos, D. N. see OGELD Steen, H. A. 49 n. Stephanus, H. 40 n., 252 n. Stolk, J. V. 197 n. Stray, C. 258 n., 265 n., 278 Strout, D. 298 n. Strugnell, J. 30 n. Stuart Jones, H. 270, 411 n.; see also LSJ, LSJ Suppl. Sturz, F. G. 390 Stylianos, Arbp. 279 n. Swain, S. 294 n. Swete, H. B. 4 n., 107, 209, 352, 390 Swinn, S. P. 400 n., 404 n. Tabachovitz, D. 62 n., 65 n. Tagliaferro, E. 184 n., 185 n., 190 n. Talmon, S. 41 n. Talshir, Z. 400 n. Tarazi, P. N. 193 n. Tasker, R. V. G. 144, 159 n. Taylor, B. A. 121 n. Taylor, J. 269 n. Taylor, V. 47, 60 n., 65 n., 66 n. TDNT 184, 403 n. Thackeray, H. St. J. 4 n., 34 n., 40 nn., 48 n., 203 n., 230 n., 302 n., 352–4, 356 n., 359, 364, 367 n., 368 n., 369, 382 n. Thayer, J. H. 267 n. Theophilos, M. 218 n. Thesleff, H. 54 n. Thiele, W. 216 Thiersch, H. G. J. 390 Thomas, H. 150 n., 178 nn., 178 n. Thompson, A. A. 138 n., 257 n., 278, 319 n.; see also LSJ Suppl.

Thompson, S. 345 n. Thomson, C. XII, 23 n. Thrall, M. E. 47 n. Thumb, D. 43 n., 44 n., 196 n., 238 n. Thurén, L. 284 n. TLG XII, 119, 187 n., 270 n., 375 n., 422 n. TLG Canon XII, 126, 187 n., 279 n. TLL XII, 7 nn., 215 Torrey, C. C. 51 n. Tov, E. 21–36, 105 n., 202, 204–5, 229 n., 354 n. Toy, C. H. 4 n. Trapp, E. see LBG Traugott, E. C. 237 n., 240 n., 243 n. Travlos, I. N. 202 n. Tregelles, S. P. 144 n., 145 n., 148 n., 160 n. Trenchard, W. C. 321–9, 336–7 Triantaphyllides, M. A. 145 n., 149 n., 284 n., 313 n. Trifone, P. 179 n. Tronci, L. 254 Troxel, R. L. 351, 352 n., 355 n., 365, 368 n., 370 n., 371 Trudgill, P. 52 nn. Turner, C. H. 159 n., 259–60, 264–5 Turner, E. G. 107 n., 150 n., 396 n., 397 n. Turner, G. W. 52 nn. Turner, N. 55 n., 305 n., 315; see also MHT, Grammar, III; MHT, Style Ure, J. N. 52 n. Valente, S. 38 n., 304 n., 318 Vanonen, H. 441 n. Vattioni, F. 392 Veitch, W. 40 n. Vergara, Juan de 211, 212 n. Vilaras, Y. 149 Voelz, J. W. 316 n. Vorm-Croughs, M. van der 353–4, 357 Wackernagel, J. 448 Waddell, W. G. 415 Wagner, C. 209 n. Wagner, J. R. 351, 355 n., 365 n., 371 Walbank, F. W. 13 n. Walters, P. 33 n., 61 n., 62 n., 106, 135, 352, 360 n., 367 n.

505

INDEX OF MODERN AUTHORS

Walton, S. 319 n., 337 n. Ward, B. 374 n., 375, 381 n., 382 n. Watt, M. 178 n. Weber, R. 215 n. Weinfeld, M. 23 n. Weinreich, U. 5 n. Wellauer, A. 30 Wellhausen, J. 71 Wendel, C. 302 n., 307 n. Westcott, B. F. 347 Wevers, J. W. 20, 109 n., 111 n., 135, 136, 339, 340, 347–8, 398 n., 411 n., 432 n., 438 n. Whitmarsh, T. 286 n., 287 nn., 294 n. Whittaker, M. 194 n. Wifstrand, A. 55 n., 239 Wilcken, U. 11 n. Wilcox, M. 71 Wilk, F. 370 nn. Willi, A. 239 n. Willi-Plein, I. 108 n.

Wilson, N. 150 n. Wilson, N. G. 166, 280 nn., 281 nn., 288 n., 290 n., 291 n. Winston, D. 301 Witherington III, B. 190 n. Woodhead, A. G. 396 n. Woody, K. M. 160, 162, 176 n., 180 Wortley, J. 374 n. Wright, B. G. 214 n., 215 n.; see also NETS Ximénes de Cisneros, F. Young-Evans, G.

170–2

378 n.

Zervos, I. S. see Mega Lexikon Zerwick, M. 61 n., 66 n. Zgusta, L. 135 n. Ziegler, J. 209, 209 nn., 210 nn., 212–3, 215 n., 216, 352, 359, 360 n., 391 Zorell, F. 204, 321 n. Zúñiga, López de, D. 170, 181

Contributions to Biblical Exegesis and Theology 1. J.A. Loader, A Tale of Two Cities, Sodom and Gomorrah in the Old Testament, early Jewish and early Christian Traditions, Kampen, 1990 2. P.W. Van der Horst, Ancient Jewish Epitaphs. An Introductory Survey of a Millennium of Jewish Funerary Epigraphy (300 BCB-700 CE), Kampen, 1991 3. E. Talstra, Solomon’s Prayer. Synchrony and Diachrony in the Composition of 1 Kings 8, 14-61, Kampen, 1993 4. R. Stahl, Von Weltengagement zu Weltüberwindung: Theologische Positionen im Danielbuch, Kampen, 1994 5. J.N. Bremmer, Sacred History and Sacred Texts in early Judaism. A Symposium in Honour of A.S. van der Woude, Kampen, 1992 6. K. Larkin, The Eschatology of Second Zechariah: A Study of the Formation of a Mantological Wisdom Anthology, Kampen, 1994 7. B. Aland, New Testament Textual Criticism, Exegesis and Church History: A Discussion of Methods, Kampen, 1994 8. P.W. Van der Horst, Hellenism-Judaism-Christianity: Essays on their Interaction, Kampen, Second Enlarged Edition, 1998 9. C. Houtman, Der Pentateuch: die Geschichte seiner Erforschung neben einer Auswertung, Kampen, 1994 10. J. Van Seters, The Life of Moses. The Yahwist as Historian in Exodus-Numbers, Kampen, 1994 11. Tj. Baarda, Essays on the Diatessaron, Kampen, 1994 12. Gert J. Steyn, Septuagint Quotations in the Context of the Petrine and Pauline Speeches of the Acta Apostolorum, Kampen, 1995 13. D.V. Edelman, The Triumph of Elohim, From Yahwisms to Judaisms, Kampen, 1995 14. J.E. Revell, The Designation of the Individual. Expressive Usage in Biblical Narrative, Kampen, 1996 15. M. Menken, Old Testament Quotations in the Fourth Gospel, Kampen, 1996 16. V. Koperski, The Knowledge of Christ Jesus my Lord. The High Christology of Philippians 3:7-11, Kampen, 1996 17. M.C. De Boer, Johannine Perspectives on the Death of Jesus, Kampen, 1996 18. R.D. Anderson, Ancient Rhetorical Theory and Paul, Revised edition, Leuven, 1998 19. L.C. Jonker, Exclusivity and Variety, Perspectives on Multi-dimensional Exegesis, Kampen, 1996 20. L.V. Rutgers, The Hidden Heritage of Diaspora Judaism, Leuven, 1998 21. K. van der Toorn (ed.), The Image and the Book, Leuven, 1998 22. L.V. Rutgers, P.W. van der Horst (eds.), The Use of Sacred Books in the Ancient World, Leuven, 1998 23. E.R. Ekblad Jr., Isaiah’s Servant Poems According to the Septuagint. An Exegetical and Theological Study, Leuven, 1999 24. R.D. Anderson Jr., Glossary of Greek Rhetorical Terms, Leuven, 2000 25. T. Stordalen, Echoes of Eden, Leuven, 2000 26. H. Lalleman-de Winkel, Jeremiah in Prophetic Tradition, Leuven, 2000 27. J.F.M. Smit, About the Idol Offerings. Rhetoric, Social Context and Theology of Paul’s Discourse in First Corinthians 8:1-11:1, Leuven, 2000 28. T.J. Horner, Listening to Trypho. Justin Martyr’s Dialogue Reconsidered, Leuven, 2001 29. D.G. Powers, Salvation through Participation. An Examination of the Notion of the Believers’ Corporate Unity with Christ in Early Christian Soteriology, Leuven, 2001 30. J.S. Kloppenborg, P. Hoffmann, J.M. Robinson, M.C. Moreland (eds.), The Sayings Gospel Q in Greek and English with Parallels from the Gospels of Mark and Thomas, Leuven, 2001 31. M.K. Birge, The Language of Belonging. A Rhetorical Analysis of Kinship Language in First Corinthians, Leuven, 2004

32. P.W. van der Horst, Japheth in the Tents of Shem. Studies on Jewish Hellenism in Antiquity, Leuven, 2002 33. P.W. van der Horst, M.J.J. Menken, J.F.M. Smit, G. van Oyen (eds.), Persuasion and Dissuasion in Early Christianity, Ancient Judaism, and Hellenism, Leuven, 2003 34. L.J. Lietaert Peerbolte, Paul the Missionary, Leuven, 2003 35. L.M. Teugels, Bible and midrash. The Story of ‘The Wooing of Rebekah’ (Gen. 24), Leuven, 2004 36. H.W. Shin, Textual Criticism and the Synoptic Problem in Historical Jesus Research. The Search for Valid Criteria, Leuven, 2004 37. A. Volgers, C. Zamagni (eds.), Erotapokriseis. Early Christian Question-andAnswer Literature in Context, Leuven, 2004 38. L.E. Galloway, Freedom in the Gospel. Paul’s Exemplum in 1 Cor 9 in Conversation with the Discourses of Epictetus and Philo, Leuven, 2004 39. C. Houtman, K. Spronk, Ein Held des Glaubens? Rezeptionsgeschichtliche Studien zu den Simson-Erzählungen, Leuven, 2004 40. H. Kahana, Esther. Juxtaposition of the Septuagint Translation with the Hebrew Text, Leuven, 2005 41. V.A. Pizzuto, A Cosmic Leap of Faith. An Authorial, Structural, and Theological Investigation of the Cosmic Christology in Col 1:15-20, Leuven, 2005 42. B.J. Koet, Dreams and Scripture in Luke-Acts. Collected Essays, Leuven, 2006 43. P.C Beentjes. “Happy the One Who Meditates on Wisdom” (SIR. 14,20). Collected Essays on the Book of Ben Sira, Leuven, 2006 44. R. Roukema, L.J. Lietaert Peerbolte, K. Spronk, J.W. Wesselius (eds.), The Interpretation of Exodus. Studies in Honour of Cornelis Houtman, Leuven, 2006 45. G. van Oyen, T. Shepherd (eds.), The Trial and Death of Jesus. Essays on the Passion Narrative in Mark, Leuven, 2006 46. B. Thettayil, In Spirit and Truth. An Exegetical Study of John 4:19-26 and a Theological Investigation of the Replacement Theme in the Fourth Gospel, Leuven, 2007 47. T.A.W. van der Louw, Transformations in the Septuagint. Towards an Interaction of Septuagint Studies and Translation Studies, Leuven, 2007 48. W. Hilbrands, Heilige oder Hure? Die Rezeptionsgeschichte von Juda und Tamar (Genesis 38) von der Antike bis zur Reformationszeit, Leuven, 2007 49. J. Joosten, P.J. Tomson (eds.), Voces Biblicae. Septuagint Greek and its Significance for the New Testament, Leuven, 2007 50. A. Aejmelaeus, On the Trail of the Septuagint Translators. Collected Essays, Leuven, 2007 51. S. Janse, “You are My Son”. The Reception History of Psalm 2 in Early Judaism and the Early Church, Leuven, 2009 52. K. De Troyer, A. Lange, L.L. Schulte (eds.), Prophecy after the Prophets? The Contribution of the Dead Sea Scrolls to the Understanding of Biblical and Extra-Biblical Prophecy, Leuven, 2009 53. C.M. Tuckett (ed.), Feasts and Festivals, Leuven, 2009 54. M. Labahn, O. Lehtipuu (eds.), Anthropology in the New Testament and its Ancient Context, Leuven, 2010 55. A. van der Kooij, M. van der Meer (eds.), The Old Greek of Isaiah: Issues and Perspectives, Leuven, 2010 56. J. Smith, Translated Hallelujehs. A Linguistic and Exegetical Commentary on Select Septuagint Psalms, Leuven, 2011 57. N. Dávid, A. Lange (eds.), Qumran and the Bible. Studying the Jewish and Christian Scriptures in Light of the Dead Sea Scrolls, Leuven, 2010 58. J. Chanikuzhy, Jesus, the Eschatological Temple. An Exegetical Study of Jn 2,13-22 in the Light of the Pre 70 C.E. Eschatological Temple Hopes and the Synoptic Temple Action, Leuven, 2011

59. H. Wenzel, Reading Zechariah with Zechariah 1:1–6 as the Introduction to the Entire Book, Leuven, 2011 60. M. Labahn, O. Lehtipuu (eds.), Imagery in the Booky of Revelation, Leuven, 2011 61. K. De Troyer, A. Lange, J.S. Adcock (eds.), The Qumran Legal Texts between the Hebrew Bible and Its Interpretation, Leuven, 2011 62. B. Lang, Buch der Kriege – Buch des Himmels. Kleine Schriften zur Exegese und Theologie, Leuven, 2011 63. H.-J. Inkelaar, Conflict over Wisdom. The Theme of 1 Corinthians 1-4 Rooted in Scripture, Leuven, 2011 64. K.-J. Lee, The Authority and Authorization of Torah in the Persion Period, Leuven, 2011 65. K.M. Rochester, Prophetic Ministry in Jeremiah and Ezekiel, Leuven, 2012 66. T. Law, A. Salvesen (eds.), Greek Scripture and the Rabbis, Leuven, 2012 67. K. Finsterbusch, A. Lange (eds.), What is Bible?, Leuven, 2012 68. J. Cook, A. van der Kooij, Law, Prophets, and Wisdom. On the Provenance of Translators and their Books in the Septuagint Version, Leuven, 2012 69. P.N. De Andrado, The Akedah Servant Complex. The Soteriological Linkage of Genesis 22 and Isaiah 53 in Ancient Jewish and Early Christian Writings, Leuven, 2013 70. F. Shaw, The Earliest Non-Mystical Jewish Use of Ιαω, Leuven, 2014 71. E. Blachman, The Transformation of Tamar (Genesis 38) in the History of Jewish Interpretation, Leuven, 2013 72. K. De Troyer, T. Law, M. Liljeström (eds.), In the Footsteps of Sherlock Holmes. Studies in the Biblical Text in Honour of Anneli Aejmelaeus, Leuven, 2014 73. T. Do, Re-thinking the Death of Jesus. An Exegetical and Theological Study of Hilasmos and Agape in 1 John 2:1-2 and 4:7-10, Leuven, 2014 74. T. Miller, Three Versions of Esther. Their Relationship to Anti-Semitic and Feminist Critique of the Story, Leuven, 2014 75. E.B. Tracy, See Me! Hear Me! Divine/Human Relational Dialogue in Genesis, Leuven, 2014 76. J.D. Findlay, From Prophet to Priest. The Characterization of Aaron in the Pentateuch, Leuven, forthcoming 77. M.J.J. Menken, Studies in John’s Gospel and Epistles. Collected Essays, Leuven, 2015 78. L.L. Schulte, My Shepherd, though You Do not Know Me. The Persian Royal Propaganda Model in the Nehemiah Memoir, Leuven, 2016 79. S.E. Humble, A Divine Round Trip. The Literary and Christological Function of the Descent/Ascent Leitmotif in the Gospel of John, Leuven, 2016 80. R.D. Miller, Between Israelite Religion and Old Testament Theology. Essays on Archaeology, History, and Hermeneutics, Leuven, 2016 81. L. Dequeker, Studia Hierosolymitana, Leuven, 2016 82. K. Finsterbusch, A. Lange (eds.), Texts and Contexts of Jeremiah. The Exegesis of Jeremiah 1 and 10 in Light of Text and Reception History, Leuven, 2016 83. J.S. Adcock, “Oh God of Battles! Steal My Soldiers’ Hearts!” A Study of the Hebrew and Greek Text Forms of Jeremiah 10:1-18, Leuven, 2017 84. R. Müller, J. Pakkala (eds.), Insights into Editing in the Hebrew Bible and the Ancient Near East. What Does Documented Evidence Tell Us about the Transmission of Authoritative Texts?, Leuven, 2017 85. R. Burnet, D. Luciani, G. van Oyen (eds.), The Epistle to the Hebrews. Writing at the Borders, Leuven, 2016 86. M.K. Korada, The Rationale for Aniconism in the Old Testament. A Study of Select Texts, Leuven, 2017 87. P.C. Beentjes, “With All Your Soul Fear the Lord” (Sir. 7:27). Collected Essays on the Book of Ben Sira II, Leuven, 2017 88. B.J. Koet, A.L.H.M. van Wieringen (eds.), Multiple Teachers in Biblical Texts, Leuven, 2017

89. T. Elgvin, The Literary Growth of the Song of Songs during the Hasmonean and Early-Herodian Periods, Leuven, 2018 90. D.C. Smith, The Role of Mothers in he Genealogical Lists of Jacob’s Sons, Leuven, 2018 91. V.P. Chiraparamban, The Manifestation of God’s Merciful Justice. A Theocentric Reading of Romans 3-21-26, Leuven, 2018 92. P. Paul, Beyond the Breach. An Exegetical Study of John 4:1-42 as a Text of JewishSamaritan Reconciliation, Leuven, 2021 93. I. Fröhlich, David in Cultural Memory, Leuven, 2019 94. M. Langlois, The Samaritan Pentateuch and the Dead Sea Scrolls, Leuven, 2019 95. A. Livneh, Studies on Jewish and Christian Historical Summaries from the Hellenistic and Early Roman Periods, Leuven, 2019 96. M. Rotman, The Call of the Wilderness. The Narrative Significance of John the Baptist’s Whereabouts, Leuven, 2020 97. O. Lukács, Sabbath in the Making. A Study of the Inner-Biblical Interpretation of the Sabbath Commandment, Leuven, 2020 98. J.J. Spoelstra, Life Preservation in Genesis and Exodus. An Exegetical Study of the Tebâ of Noah and Moses, Leuven, 2020 99. T. Havukainen, The Quest for the Memory of Jesus: a Viable Path or a Dead End?, Leuven, 2020 100. K. De Troyer, The Ultimate and the Penultimate Text of the Book of Joshua, Leuven, 2018 103. Angela Kim Harkins and Barbara Schmitz (eds.), Selected Studies on Deuterocanonical Prayers, Leuven, 2021 104. Torben Plitt, Wachstumsgesetz oder Kürzungstendenz? Gedächtnispsychologische Erkenntnisse zu den inhaltlichen Überhängen in den synoptischen Paralleltraditionen, Leuven, 2021 105. Ian Wilson, Praying to the Temple: Divine Presence in Solomon’s Prayer, Leuven, 2022

PRINTED ON PERMANENT PAPER

• IMPRIME

SUR PAPIER PERMANENT

N.V. PEETERS S.A., WAROTSTRAAT

• GEDRUKT

OP DUURZAAM PAPIER

50, B-3020 HERENT

- ISO 9706