Et sapienter et eloquenter: Studies on Rhetorical and Stylistic Features of the Septuagint 9783666532610, 9783525532614, 9783647532615

Citation preview

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Forschungen zur Religion und Literatur des Alten und Neuen Testaments Herausgegeben von Jan Christian Gertz, Dietrich-Alex Koch, Matthias Köckert, Hermut Löhr, Joachim Schaper und Christopher Tuckett

Band 241

Vandenhoeck & Ruprecht

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Et sapienter et eloquenter Studies on Rhetorical and Stylistic Features of the Septuagint

Edited by Eberhard Bons and Thomas J. Kraus

Vandenhoeck & Ruprecht

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Bibliografische Information der Deutschen Nationalbibliothek Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind im Internet über http://dnb.d-nb.de abrufbar. ISBN 978-3-525-53261-4 ISBN 978-3-647-53261-5 (E-Book)

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen. Alle Rechte vorbehalten. Das Werk und seine Teile sind urheberrechtlich geschützt. Jede Verwertung in anderen als den gesetzlich zugelassenen Fällen bedarf der vorherigen schriftlichen Einwilligung des Verlages. Printed in Germany. Druck und Bindung: b Hubert & Co., Göttingen Gedruckt auf alterungsbeständigem Papier.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Table of Contents

Preface Eberhard Bons/Thomas J. Kraus .....................................................

7

Rhetorical ornamentation in the Septuagint: The case of grammatical variation Jan Joosten .......................................................................................

11

Stylistic Invention and Rhetorical Purpose in the Book of the Twelve Jennifer M. Dines .............................................................................

23

Translating the Septuagint Psalms – some ‘Lesefrüchte’ and their value for an analysis of the rhetoric (and style) of the Septuagint (Psalms) Thomas J. Kraus ..............................................................................

49

Rhetorical Devices in the Septuagint Psalter Eberhard Bons .................................................................................

69

Griechisches im Griechisch der LXX Knut Usener ..................................................................................... English abstract .................................................................................

83 98

The Poetics of Wisdom: Language and style in the Wisdom of Solomon Alexis Léonas ...................................................................................

99

Periphrastic Tense Forms withGK OK andIKIPQOCK in the Septuagint of Ezekiel Katrin Hauspie ................................................................................. 127 Index of Ancient Sources ....................................................................... 153 Index of Modern Authors ....................................................................... 162 Index of Important Subjects ................................................................... 163

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Preface

In recent years, Septuagint studies seem to have come into vogue. During the last two decades, not only have several translations of the Septuagint into the most common modern languages been published, but also an important number of monographs dealing with specific biblical books or subjects. To these works might be added numerous articles focussing on a large range of issues, e.g. divergences between the extant Hebrew sources and the Septuagint, crucial passages in the Greek text, text criticism and text history. Judging by the entries in bibliographies, the number of publications on the Septuagint is still increasing. However, it is beyond any doubt that recent Septuagint research is dominated by some fundamental questions that have intrigued scholars for a long time. Strictly speaking, these questions arose as early as the beginning of text critical studies in the epoch of Humanism when scholars were confronted with two different versions of the Bible text. After the discovery of the Qumran manuscripts, the problem is still more faceted: Does the Septuagint present a Bible text that can be considered – at least in specific quotations, sections or even books – more original and more reliable than the extant Hebrew texts, including the Qumran biblical manuscripts? Can one therefore have recourse to the Greek text in order to correct the Hebrew Bible when the latter offers puzzling or enigmatic expressions? As can be expected these questions did not produce unanimous answers. In some cases the Septuagint text might turn out to be more original than the Hebrew text, in other cases the contrary seems to be true. Obviously, whatever the result of such an investigation, it requires a meticulous analysis of the text-critical, semantic and grammatical data of each individual text and its variants. Nevertheless, this approach to the Septuagint texts can be characterized as follows: The Septuagint text is analysed from the point of view of the Hebrew text, above all where it presents different or allegedly better readings. On the other hand, this approach does not pay attention to another series of questions deserving consideration. In fact, it is worthwhile to analyse the Septuagint especially on the level of morphology, vocabulary and syntax on the assumption that its Greek language displays linguistic features other than those of a Semitic language. In particular, how do the translators take advantage of specific Greek features – e.g. verbal forms unknown in He-

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

8

Eberhard Bons – Thomas J. Kraus

brew, more or less synonymous words, technical vocabulary, idiomatic phrases – in order to render their source text? In other words, to what extent does the Greek Bible text present linguistic and stylistic features of its own, and how can they be classified? Furthermore, can we find rhetorical devices the translators have employed, probably with the intention of lending a certain ornatus to the text in the target language? Thus, we are in search of phenomena that cannot be explained by the underlying Hebrew text, the so called Vorlage. On the contrary, the question is whether these phenomena are rather to be explained by the translators’ endeavour to give the Greek text a specific stylistic profile. A closer inspection of Septuagint bibliography leads to the conclusion that the aforementioned questions have been more or less neglected in the past. It is the aim of the present book to draw due attention to these stylistic and rhetorical issues. Needless to say the following articles cannot cover the full range of stylistic and rhetorical features present in the Septuagint. In singling out specific cases, they raise the question of whether analogous instances are to be found elsewhere in the Septuagint. Once we gather more material of this kind a fresh look at other questions debated for a long time should be undertaken: What idea had the translators of their task as well as of their liberty in translating? Can one draw closer conclusions concerning their education and their milieu as well as the original Sitz im Leben of the translated books? The following articles deal with different topics within the broad field of stylistic and rhetorical issues and of vocabulary: Not without underlining the fact that the style of the Septuagint writings is often unpolished, Jan Joosten raises the question of whether their translators cared about the literary quality of their text. However, despite appearances, translators were not unaware of stylistic necessities as shown by the instances of grammatical variation. Jennifer M. Dines investigates the phenomena of variation and new verbal patterns in the Book of the Twelve Prophets. These observations lead to some considerations about the cultural milieu and purpose of the Septuagint version of this collection of prophetical books. The following two papers are dedicated to the Septuagint Psalter: Thomas J. Kraus addresses a series of selected Psalm verses as well as the Greek translation of Psalm 21. He argues for the hypothesis that rhetorical and stylistic features in contrast to the Masoretic Text are deliberate. Consequently, the Greek Psalter is more than simply a literal translation of a Hebrew Vorlage. Presenting cases of hyperbaton, alliteration and paronomasia in the Greek Psalter, Eberhard Bons aims to provide evidence for the hypothesis that the Psalter translation cannot be considered a slavish translation of a

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Preface

9

Hebrew Vorlage. On the contrary, the spread of rhetorical devices proves the degree of liberty translators obviously possessed in diverging from their Hebrew source text. From the point of view of classical studies, Knut Usener presents some Septuagint passages that obviously show the influence of Greek literature. Although these observations are to be corroborated by further investigations, they shed some light on the translators’ cultural horizon as well as their milieu. Alexis Léonas turns to a book that according to scholarly consensus was written in Greek, i.e. the “Wisdom of Solomon”. He seeks to identify the elements of Septuagint language and style used by the author of this late biblical book. Léonas’ detailed study leads to the conclusion that the “Wisdom of Solomon” employs some elements of Septuagint idiom whereas others are dismissed.  To what extent translators allowed themselves latitude between a wordfor-word-rendering and a translation that met Greek grammatical standards, is finally illustrated by Katrin Hauspie. Focusing on periphrasis in the Septuagint of Ezekiel, she points out the manner in which the Greek translation follows its Hebrew source text or diverges from it, e.g. by inserting a copula. Four of the articles collected in the present volume (Bons, Dines, Kraus, Usener) are reworked versions of papers read at a panel on the style of the Septuagint. The panel was included in the program of the 2007 Congress of the “International Society of Septuagint and Cognate Studies”, held at Ljubljana University, Slovenia. In order to complete these approaches to Septuagint language, we deemed it advisable to invite three colleagues (Hauspie, Joosten, Léonas) to contribute articles to this volume. We are grateful to the authors for allowing their papers to be published, despite an unforeseeable delay that occurred between the 2007 congress and publication. We would also like to express our gratitude to the editors of the series “Forschungen zur Religion und Literatur des Alten und Neuen Testaments”, in particular to Christopher M. Tuckett, to Jörg Persch, Christoph Spill and Maike Linne, Vandenhoeck & Ruprecht Verlag. Strasbourg (France) – Neumarkt i.d.OPf. (Germany), 29/11/2010 Eberhard Bons and Thomas J. Kraus

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Jan Joosten

Rhetorical ornamentation in the Septuagint: The case of grammatical variation*

Diametrically opposed views have been expressed on the quality of the Greek Bible as a literary work. Ever since Antiquity, the cultural and religious eminence of the Greek version has led many of its readers to hold that its linguistic form too should, so to speak, be canonical. The Letter of Aristeas recounts: After the books had been read, the priests and the elders of the translators and the Jewish community and the leaders of the people stood up and said, that since so excellent and sacred and accurate a translation had been made, it was only right that it should remain as it was and no alteration should be made in it (Ep. Arist., 310).

Although the Letter purports to be the work of a Greek eyewitness to the events, in reality it expresses the views of Hellenistic Jews of a somewhat later age. The words attributed to the Jews of Ptolemy’s time reflect the attitude of a generation that had grown up with the Septuagint. However, this reverential attitude towards the language of the Greek Bible is not shared by all the ancients. Although few of their writings are extant, we know from the Church Fathers’ polemics against them that Celsus, Porphyry, Hierocles and Julian deplored the “pedestrian” quality of the biblical writings.1 These pagan philosophers found the Septuagint essentially un-Greek. Tellingly, the Fathers do not usually contradict the philosophers on this point, attempting instead to excuse the admitted “vileness and simplicity” of Scripture.2 It appears the pagan readers were on to something. Both positions—that the style of the Greek Bible is perfectly adapted to its purpose, and that it is intolerably unpolished—are still defended today.3 _______________________ * Thanks are due to Jason Dean who corrected the English style of an earlier draft of this article. 1 See A. Léonas, L’aube des traducteurs (Paris: Cerf, 2007), 134–140. 2 See N. Fernández Marcos, Introducción a las versiones griegas de la Biblia (Madrid: CSIC, 21998), 18. 3 For the first position, see, e.g., P. Harlé / D. Pralon, Le Lévitique. Traduction du texte grec de la Septante, Introduction et Notes (La Bible d'Alexandrie 3; Paris: Cerf, 1988), 47–81; Léonas, Aube, passim. For the second position, see, e.g., A. Pietersma, “A New Paradigm for Addressing

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

12

Jan Joosten

The controversy is due to the complexity of the issue as much as to substantial differences of opinion. Before approaching our subject, it will be useful to disentangle the problem of the Septuagint’s style to some extent.

Two alienating factors in the Greek Bible Whether by brilliant design or by clumsy incompetence, the style of the Greek version is unlike that of any comparable text written originally in Greek. The Septuagint clearly fails to conform to Greek norms usually applied to literature on similar subject matter. It is important to realize that the unwonted character of the text is due to two distinct factors. Although these factors cannot always be kept apart entirely, they should be distinguished on a theoretical level. The first factor is the literal translation technique underlying most of the books of the Septuagint. The style of the version is strange to the extent that the Greek slavishly follows the Hebrew. Extensive research has demonstrated that the Septuagint’s literalism exhibits great diversity.4 In practically all its forms, however, it entails a large amount of carry-over from the Semitic source text into the translation. Lexical stereotyping can lead to the unidiomatic use of Greek words, as in the well-known MCK? G RJTY‚VJUGP 'CWKF GK L GK TJ‚PJP VQW„ RQNGOQW, “David inquired … after the peace of the war” (2 Kings 11:7). The close adherence to the Hebrew word order often induces unusual syntax in Greek—one might think of the everrepeated use of MCKž in the connection of main clauses, for instance. The word-for-word rendering of Hebrew idioms and expressions occasionally creates strange collocations. In some cases, Hebrew words are not translated at all but simply transcribed into Greek. These and other types of transfer confer a strange appearance to the target text. The second factor is the use of colloquial Greek. The vocabulary, the grammar and the syntax of the Septuagint are not representative of classical Greek, nor of the literary koine used by Hellenistic authors such as Polybius, but stands closer to the non-literary language of contemporary documentary papyri. Forms like JŸMCOGP (Gen 47:4, instead of JŸMQOGP) and words like WRQ\WIKQP, “donkey” (Gen 36:24, instead of Q¢PQL), are singular ____________________ Old Questions: The Relevance of the Interlinear Model for the Study of the Septuagint”, in J. Cook (ed.), Bible and Computer. The Stellenbosch AIBI-6 Conference (Leiden: Brill, 2002), 337–364. 4 See J. Barr, The Typology of Literalism in Ancient Biblical Translations (MSU 15; Göttingen: Vandenhoeck & Ruprecht, 1979); E. Tov, The Text-Critical Use of the Septuagint in Biblical Research (Jerusalem: Simor, 1981), 50–66.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Rhetorical ornamentation in the Septuagint: Grammatical variation

13

in literary texts, as are such syntactic peculiarities as the recurrent use of a genitive absolute whose subject is also the subject of the main clause (as in Num 3:4 MCK? G VGNGW‚VJUGP 1CFCD MCK? $DKQWF G¢PCPVK MWTKQW RTQU HGTQ‚PVYP CW VY„P RW„T C NNQ‚VTKQP G¢PCPVK MWTKQW, “And Nadab and Abiud died before the Lord when they were bringing strange fire before the Lord”.5 Elements like these must have been shocking to cultivated readers, Jews and non-Jews alike.6 Both factors affect the style of the Septuagint. It must be asked, however, whether they are stylistic in the sense that they proceed from the conscious choice of the translators. The answer to this question is that they probably do not. At least in regard to the second factor, its artless nature is rather self-evident. The use of non-literary Greek hardly reflects a deliberate option on the part of the translators.7 It is true that, in modern literature, books may be written in colloquial language. In English literature, Catcher in the Rye is a good example. Telling the story from the point of view of a teenager, the author, J. D. Salinger, recreated the common speech of youngsters of his day. An ancient analogue of this can be found in the Greek comedians, who have their characters speak colloquial and street language, as the plot requires.8 In these instances, the use of non-literary language in literature reflects the artistry of the author; it is a kind of “style-switching”. However, the use of vernacular Greek in the Septuagint cannot be explained in this way. In the Greek version, non-literary words and forms are not limited to characters’ speech, but equally distributed over the entire text.9 Since the biblical text must have had high prestige for the translators, it makes no sense to suppose that they intentionally cast the translation in

_______________________ 5

See F.C. Conybeare/S.G. Stock, Grammar of Septuagint Greek (Boston: Ginn, 1905), 58. There are no testimonies from Jewish readers on the stylistic inadequacy of the Septuagint. It is striking to note, however, that Jewish Hellenistic writers quoting the Septuagint often try to improve upon its language. See H.B. Swete, An Introduction to the Old Testament in Greek (Cambridge: Cambridge University Press, 1914), 370. 7 This statement may be true only for the translators of the Pentateuch. At least some of the translators of the later books appear to have adopted this level of style under the influence of the Pentateuch even although they would themselves have been able to use a higher literary register. 8 On the affinity of the Septuagint’s vocabulary to that of the Greek comedians, see H.A.A. Kennedy, Sources of New Testament Greek (Edinburgh: T. & T. Clark, 1895). 9 Exceptionally, one finds a somewhat higher stylistic register in speech attributed to God. Thus John Lee has pointed to the two occurrences of classical WŸŸY “to make it rain” in divine speech, Exod 9:18; 16:4, while elsewhere in the Septuagint the Hellenistic DTGZY is used in this meaning. See J.A.L. Lee, A Lexical Study of the Septuagint Version of the Pentateuch (SCS 14; Chico CA: Scholars Press, 1983), 123–124. 6

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

14

Jan Joosten

substandard language.10 If they used colloquial words and forms, they did so unwittingly. They wrote Greek as well as they were able. The case of Hebraizing is more difficult to decide. Strangeness is not necessarily pejorative. As a literal translation, the Septuagint openly admits its foreign origin. Readers are reminded that this is not a Greek text, but a Hebrew one in translation. While Greek readers unfamiliar with the Bible and with Jewish religion may have experienced some confusion in reading the version, Jewish readers, who possessed some familiarity with the Bible, may have appreciated the exotic quality of the Septuagint. Tell-tale signs of its Hebrew origin may have authenticated the writing in their eyes. Thus, the slavish imitation of Hebrew expressions may stem from the conscious choice of the translators. Many scholars think that the literalism of the translators reflects awe for the sacred words.11 Although the resulting Greek text is at times difficult to understand, this may actually have made it more attractive: divine revelation ought to be slightly mysterious.12 Other scholars, however, have explained the literal translation technique of the Septuagint differently. In many passages, the translation appears to have been made by small segments: clauses, and sometimes parts of clauses, were translated one by one according to an “easy technique”.13 This explanation accounts for the numerous cases of anacoluthon and broken constructions in the Greek version. The translators appear to have rendered the Hebrew word for word because this was for them the most expedient way.14 According to this view, the Hebraisms of the Septuagint do not represent a conscious effort at creating a stylized text. At least in regard to the Pentateuch, the latter view would seem to be closer to the mark than the first.15 _______________________ 10

It has recently been argued that the Septuagint originated as an “interlinear” aid facilitating the study of the Hebrew Bible text. If this were true, the prestige of the text would have been rather low for those who created it. For a variety of reasons, however, the “interlinear paradigm” is probably to be rejected, see J. Joosten, “Reflexions on the ‘interlinear Paradigm’ in Septuagintal studies”, in A. Voitila/J. Jokiranta (ed.), Scripture in Transition. Essays on Septuagint, Hebrew Bible, and Dead Sea Scrolls in Honour of Raija Sollamo (JSJ.S 126; Leiden: Brill, 2008), 163– 178. 11 See S.P. Brock, “The Phenomenon of the Septuagint”, OTS 17 (1972) 11–36, in part. 20– 22. 12 See Léonas, Aube, 163–164. 13 See Barr, “Typology”. 14 See T. van der Louw, “Approaches in Translation Studies and Their Use for the Study of the Septuagint”, in M.K.H. Peters (ed.), XII Congress of the International Organization for Septuagint and Cognate Studies Leiden, 2004 (SCS 54; Leiden: Brill, 2006) 17–28. 15 To be sure, in some parts of the Septuagint the literalism is too consistent to be the result of simple expedience. What may have begun as an easy technique, in the Pentateuch and some historical books, at some point turned into a deliberate process. The translators of the kaige sections in Kingdoms, if not the translators of Jeremiah, Ezekiel, the Twelve and Psalms, actively

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Rhetorical ornamentation in the Septuagint: Grammatical variation

15

In sum, the most striking features of the Septuagint may have nothing to do with literary artistry. The Hebraic quality and the non-literary language simply reflect the conditions under which the translators were working: they were engaging upon an unprecedented task, and they had not been schooled in Greek belles lettres.

Stylistic flourishes in the Septuagint So far, then, the results of our explorations are largely negative. Although the Hebraizing and the colloquial language affect the style of the Septuagint from the point of view of the reader, they do not, at least not necessarily, attest any conscious stylizing on the part of the translator. This does not mean, however, that such stylizing is absent. Indeed, after these somewhat lengthy prolegomena, the question still needs to be asked: is there a style of the Septuagint? Did the translators pay attention to the literary quality of the goal text, or did they just put the Hebrew into Greek without worrying how it would look to a Greek reader? To be sure, this question cannot be asked in such a general way. The non-translated books of the Septuagint, such as 2 Maccabees or Wisdom of Solomon, require separate study in this respect. But even if we disregard the books originally written in Greek, the Septuagint is a collection of writings of different ages and origins that cannot all be lumped together. Obviously, free translation—such as we find in Isaiah, Job, Proverbs, Daniel and Esther—leaves much more room for stylistic aspirations than does literal translation. Even among the books translated literally there is great variety, extending from the freewheeling fidelity of Genesis to the almost mechanical transcription in some of the latter prophets—to say nothing of the kaigeTheodotion parts of Kingdoms, or of Second Ezra. Thus the answer may vary from book to book. But the question is still worth asking. Is it possible to speak of style in the Septuagint, particularly in those books, making up the bulk of the corpus, that were translated literally? Did the translators of these books care about the literary quality of their text? It is surprising how few scholars have addressed this question. In an overview published in 1988 and titled “Problèmes stylistiques. La Septante est-elle une œuvre au sens plein du terme ?” (“Problems of style. Is the Septuagint a literary work in the full sense of the word?”), Marguerite Harl ____________________ created a kind of “translationese”. E.g., in the kaige-Theodotion school, the Hebrew pronoun [P  was rendered as G™IY‚ while [MP was systematically rendered G™IY‚ GK™OK— even where a finite verb followed.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

16

Jan Joosten

did not mention a single in-depth study dealing with the style of the more literal books of the Septuagint, although she did refer to one article dealing with Isaiah and one with Job.16 More recently, however, some swallows have been sighted announcing a new Spring. In a pioneer study published in 1985, the late Nechama Leiter drew attention to instances of lexical variation in the Greek story of Bileam.17 Whereas Hebrew authors tend to repeat the same word over and over again, Greek style requires the use of synonyms. The study argues persuasively that variatio, in the rhetorical sense of the term, was an effect consciously pursued by the Numbers translator.18 A short study by John Lee lists several categories of rhetorical ornamentation in the Septuagint and notably signals some interesting cases of alliteration.19 The challenge, of course, is to distinguish cases of conscious artistry from cases of accidental assonance, as Lee himself well realised. James Aitken, building on the work of Françoise Vinel, has treated the question of rhetoric in Greek Ecclesiastes in depth, showing, somewhat surprisingly, that stylistic touches were added even by the most literal of Septuagint translators.20 Two methodological principles can be drawn from these studies. First, the case for stylistic awareness on the part of the translators can be made only on the basis of renderings implying some type of divergence from the Hebrew. Where Hebrew tropes and figures are translated literally the resulting flourishes can hardly be attributed to the translator’s desire to embellish the goal text. Second, changes that simply reflect the grammatical rules of the target language should not be counted as marks of style. Where a Hebrew infinitive absolute is translated with a participle, for instance, one should take into account that the Greek language has no straightforward equivalent of the infinitive absolute. The case for stylistic ornamentation _______________________ 16

See M. Harl/G. Dorival/O. Munnich, La bible grecque des Septante: Du judaïsme hellénistique au christianisme ancien (Paris: Cerf, 1988), 259–266. Harl does not refer to Thackeray’s study on meter in Greek Proverbs. See H.St.J. Thackeray, “The Poetry of the Greek Book of Proverbs”, JThS 13 (1911) 46–66. 17 N. Leiter, “Assimilation and Dissimilation Techniques in the LXX of the Book of Balaam”, Textus 12 (1985) 79–95. 18 In a second article, Leiter inspects a number of apparent transpositions in the Greek version of Genesis 31. Some of these are attributed, again plausibly, to the stylistic sensitivity of the translator. As is admitted by Leiter, it is almost never possible to exclude the possibility that a transposition was found already in the Hebrew source text. See N. Leiter, “The Translator's Hand in Transpositions? Notes on the LXX of Genesis 31”, Textus 14 (1988) 105–130. 19 See J.A.L. Lee, “Translations of the Old Testament”, in S.E. Porter (ed.), Handbook of Classical Rhetoric in the Hellenistic Period 330 B.C. – A.D. 400 (Leiden: Brill, 1997), 775–784. 20 J.K. Aitken, “Rhetoric and Poetry in Greek Ecclesiastes”, BIOSCS 38 (2005) 55–77.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Rhetorical ornamentation in the Septuagint: Grammatical variation

17

can be built only on renderings diverging from the Hebrew source text in a way that is not dictated by the Greek language. The studies of Leiter, Lee, Vinel and Aitken show that instances of ornamentation do occur in the literal translation units. Still, much work remains to be done.

A case of grammatical variatio An instance of grammatical variation was tentatively identified in the course of a close reading of the Greek text of the Decalogue in Exodus 20. The Greek rendering of this central text is rather literal.21 Although the translator allowed himself some leeway in the choice of equivalents, he took pains to match the Greek text to the Hebrew one word for word, following the same order. Only some very slight discrepancies are observed in this respect, such as, notably, the addition of the verb “to be” in nominal clauses.22 A very minute deviation occurs in the following: Exod 20:5 ]FDV NY]JNJYZVV N QW RTQUMWPJžUGKLCW VQK¨LQW FG?OJƒNCVTGW‚UXLCW VQK¨L You shall not do obeisance to them, nor are you to serve them. 23

The sentence typifies the Hebraizing Greek of the Septuagint. In his wordfor-word translation, the translator mechanically rendered the yiqtol form as a future, according to the most frequent function of the Hebrew verbal form.24 The result is slightly awkward since, in non-translated texts, Greek does not commonly use the future indicative in commands.25 For all the literalness, however, there is a small change: while the first instance of negated yiqtol is rendered with QW  + future, the second is rendered with QW  OJž + aorist subjunctive. Why is this so? The most obvious, and the most frequent, rendering of Hebrew N+ yiqtol is QW + future. A literal rendering _______________________ 21

See J. Joosten, “Le Décalogue et ses enjeux théologiques. Texte hébraïque et texte de la Septante”, forthcoming in a volume on the history of interpretation of the Decalogue to be edited by Matthieu Arnold. 22 Thus in Exod 20:2, 5. 23 The same translation is found in Deut 5:9. The English translation of the Septuagint used in the present article is that of NETS. 24 See A. Voitila, “La technique de traduction du yiqtol (l’imparfait hébreu) dans l'histoire de Joseph grecque (Gen 37, 39–50)”, in C. Cox (ed.), VII Congress of the International Organization for Septuagint and Cognate Studies. Leuven 1989 (Atlanta GA: Scholars Press, 1989), 223– 237. 25 See Conybeare/Stock, Grammar, 72; F. Blass/A. Debrunner/F. Rehkopf, Grammatik des neutestamentlichen Griechisch (Göttingen, 171990), 292–293, § 362.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

18

Jan Joosten

of our verse would therefore have been a simple repetition of this structure. Such a rendering is entirely possible in the translation Greek of the Septuagint, and is in fact attested, in a very literal translation unit: 2 Kings 17:35 ]YFDV NY]JNYYZVV N QW™RTQUMWPJžUGVGCW VQK¨LMCK?QW NCVTGWžUGVGCW VQK¨L You shall not do obeisance to them, and you shall not serve them.26

The second Greek syntagm too, QW™ OJž + aorist subjunctive, is regularly used in rendering N + yiqtol, although it is rather less frequent.27 In fact, grammarians hardly detect a semantic difference between QW™ + future and QW  OJž + aorist subjunctive.28 If the translator wanted to use this syntagm he might have repeated it as in the following: Ezek 24:16 JMDV NYFRUV N QW™OJƒMQR„LQW™FG?OJƒMNCWUS„L You shall not mourn nor weep.29

In our verse, however, the translator rendered adjacent and identical prohibitive constructions in two different ways. It is hard to escape the impression that this variation of grammatical construction is due to the stylistic sensitivity of the Exodus translator. Instead of dully transcribing the Hebrew verse word for word according to a preconceived scheme, he rendered the second prohibition in a different way in order to obtain a pleasing effect. In Greek, the second construction is slightly heavier morphologically and perhaps somewhat stronger semantically. Yet it seems unlikely that the translator wanted to stress the second prohibition more than the first. Rather the conjunction of the two constructions intends to achieve stylistic elegance. All of these points are strengthened when it is observed that other passages, where the Hebrew has the same sequence of negated yiqtol forms, show up the same sequence of constructions. The first example in the Pentateuch is the following: _______________________ 26 Repetition of QW  + future is found several times in Leviticus, see Lev 19:15, 26; 25:11. In Lev 25:11, the text of some manuscripts and a quotation in Philo has substituted QW  OJž + aorist subjunctive for the second clause. 27 In the Pentateuch, QW + future occurs about 670 times, QW OJž + aorist subjunctive about 77 times. 28 See, e.g., M. Zerwick, Biblical Greek (Rome: PIB, 1994), 149. 29 See also Deut 31:6; Isa 8:12; Ezek 24:16, 23; Hos 3:3 and several examples in the Book of Job. This collocation is classical, cf. Plato, Gorgias, 494 d: UWƒ FG? QW  OJƒ G MRNCI„L QW FG? OJƒ CK UZWPS„L

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Rhetorical ornamentation in the Septuagint: Grammatical variation

19

Gen 3:3 But of the fruit of the tree that is in the middle of the orchard, God said, QW HCžIGUSGC RQƒCW VQW„QW FG?OJƒCŸŸ[JUSGCW VQW„„ You shall not eat of it nor shall you even touch it.

In this case, one might argue that the second construction expresses a gradation in regard to the first, as is indeed indicated in the NETS translation. In other passages, however, the two interdictions appear to have the same weight, as they do in Exod 20:5: Exod 22:20 MCK?RTQUJžNWVQPQW MCMYžUGVGQW FG?OJƒSNK[JVGCW VQ‚P You shall not harm a guest, nor shall you oppress him.30

The construction is used also in the third person: Deut 22:5 QW MG¢UVCKUMGW‚JC PFTQƒLG RK?IWPCKMKƒQW FG?OJƒG P FW‚UJVCKC PJƒT UVQNJƒPIWPCKMGKCP There shall not be the equipment of a man upon a woman, nor shall a man put on a woman’s garment.31

And in one instance, a third and a second person prohibition are combined: Exod 23:13 Q¢PQOCSGYPGVGTYPQW MC PCOPJUSJ‚UGUSGQW FG?OJƒC MQWUS„G MVQW UVQžOCVQLWOYP You shall not recall a name of other gods; neither shall it be heard from your mouth.

Altogether there are about a dozen examples of this construction, most of them in the Pentateuch, but a few also in other biblical books.32 The construction is not representative of classical Greek. Indeed, it seems to be unattested in non-translated Greek literature except for some writings that may have been influenced by the Septuagint. Thus we find the sequence QW  + future – QW  OJž + aorist subjunctive once in the Book of Enoch: 1 Enoch 22:13 QW VKOYTJSJžUQPVCKG PJOGTCVJLMTKUGYLQW FG?OJƒOGVGIGTSYUKP G PVGWSGP

_______________________ 30

See also Exod 23:18; 23:24; Deut 7:2. See also Deut 17:16; Dan 6:6. 32 Other ostensible cases of the construction could be found, but they do not exhibit the clear binary structure of the examples noted here. 31

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

20

Jan Joosten

They will not be punished in the day of judgement, nor will they be resurrected from here.

It also appears once in the Revelation of John: Rev 7:16 QW RGKPCžUQWUKPG¢VKQW FG?FK[JžUQWUKPG¢VKQW FG?OJƒRGUG R CW VQWƒLQ JŸNKQL They will hunger no more, and thirst no more; the sun will not strike them.

These passages do not exhibit the tight diction of the Septuagintal examples, nor are they prohibitions.33 Nevertheless, their syntax should probably be attributed to the influence of the style of the Septuagint. While the variation of grammatical constructions certainly reflects good Greek style, the precise grammatical construction QW  + future followed by QW  OJž + aorist subjunctive appears to be an innovation of the Septuagint translators. The distribution of examples suggests that the Septuagint translators were not following literary models when they used the sequence. Within the Septuagint, the construction is used a number of times and in different translation units. The translators of Genesis and Exodus appear to have set a fashion.

Other instances of grammatical variation The sequence discussed in the preceding section is not the only case where grammatical forms or constructions seem to alternate merely for stylistic reasons in the Septuagint. Although it will not be possible to present them at length in the present study, some other constructions may be enumerated to illustrate the extent to which the translators use the device of grammatical variatio: a) The third person optative and the third person imperative at times occur in parallel clauses in poetical passages. See, e.g., Gen 9:27; Exod 15:16; 2 Chron 6:41; Ps 34:26; 67:3. Both forms render the Hebrew jussive, and there does not seem to be a strong difference in meaning between them. The main reason for the variation may be stylistic.

_______________________ 33 In the Septuagint, too, the construction occurs occasionally in predictive statements. See, e.g., Deut 29:22; 31:8; Isa 2:4; 5:27; 13:20; 45:17; Jer 27:20; Ezek 18:6; Prov 2:19; 6:35.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Rhetorical ornamentation in the Septuagint: Grammatical variation

21

b) The present and aorist imperatives are generally used to express aspectual nuances. In a few passages, however, the two forms occur side by side in a way that suggests the translator simply sought to vary the grammar for the sake of elegance: Judg A 5:3 C MQWžUCVGDCUKNGK¨LG PYVK\GUSGUCVTCžRCKFWPCVQKž Hear, O kings; give ear, O mighty satraps.34

See, e.g., Exod 1:16, 22; Deut 4:9; Job 34:2; 1 Chron 16:31; Isa 12:4; Jer 31:31; Ezek 21:17. c) The negative counterparts of the imperatives are OJž + imperative present and OJž + second person aorist subjunctive relatively. These constructions, too, appear to alternate freely in some verses, without aspectual implications. See, e.g., Deut 1:21; Prov 22:22; 24:17; Sir 3:4; 5:1; 7:14; 9:1. All these examples show that the Septuagint translators were prepared to exploit the grammatical possibilities of the target language not only for precision, but also for ornamentation. Other examples of grammatical variatio could probably be found.

Conclusions Robert Aldrich’s movie of 1973, “The Emperor of the North Pole”, tells of the life of wanderers in the United States during the great depression. A young man, “Cigaret”, takes to the road. He meets an experienced tramp, called “A Number One”, who teaches him the ways of the trade. From the start, the young man manifests an opportunistic streak and things go badly. The film culminates in a scene on a running train, where, after flogging the conductor in a terrible fight, A Number One declares to Cigaret, who stayed out of the fight all this time: “Kid, you got no class, you’ll never make a good hobo!”—and throws him off into a pond. A Number One’s remark, and indeed the entire movie, show that the question of class, or style, does not arise only in the cultured classes of society. Even among hoboes, it is important to have class. Mutatis mutandis, something similar might be said about the Greek Bible. Compared to contemporary literary works, the Septuagint is remarkably unpolished. In the company of such writings as Polybius’ History or _______________________ 34

Note the reverse use of the tenses in Num 23:18:C¢MQWGG PYVKUCK.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

22

Jan Joosten

Manetho’s Aegyptiaca, which treat comparable subject matter, the Greek version of the Bible looks out of place. It is a vagrant text, circulating in the Greek world, yet not wholly part of it. But the Septuagint is not a hobo without style. It has its own rules, goes its own way. Despite appearances, the translators were not completely oblivious to literary issues. Admittedly, most translation units exhibit a strong tendency to follow the Hebrew source text, formally as well as semantically. Faithfulness to the original is the overriding concern. In small details, however, one observes sensitivity to the genius of the Greek language. Literary taste prevails over strict literalism: a lexical equivalent is altered for variety; words are chosen for the way their sound interacts with other elements in the clause; and grammatical forms are varied for the sake of elegance. The model of the first translators of the Hebrew Bible was not the style of the Greek classics. Their language stands close to the colloquial and to the utilitarian prose of the documentary papyri. With the modest means at their disposition, they nevertheless try to write in a way pleasing to the reader. As in other aspects of the Septuagint, it is worthwhile here to distinguish between the ambition and the actual accomplishment of the translators.35 Their stylistic performance may leave something to be desired: the literal translation technique to which they were committed forcibly interfered with the writing of good Greek; moreover, they may not have had sufficient mastery of literary Greek to write it themselves. Their project, however, was impeccable: while staying as faithful as possible to the source text, they attempted to satisfy the grammatical and stylistic demands of the target language. The style of the version has attracted little attention, but it merits being studied much more intensively.36

_______________________ 35 See J. Joosten, “Exegesis in the Septuagint Version of Hosea”, in J.C. de Moor (ed.), Intertextuality in Ugarit & Israel (OTS 40; Leiden: Brill, 1998), 62–85, in particular p. 85. 36 “Un vaste domaine est ouvert à l’investigation”, Marguerite Harl, in Harl/Dorival/Munnich, La bible grecque, 265.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Jennifer M. Dines

Stylistic Invention and Rhetorical Purpose in the Book of the Twelve

Work on the translation technique of the twelve Minor Prophets (hereafter, “XII”)1 shows that the translator not only made various grammatical and syntactical adjustments to a predominantly literal translation, but also introduced numerous artistic effects, not all occurring in the Hebrew. These include assonance, alliteration, rhyme and other such devices.2 Any examination of the literary style of XII should take account of all such features; for reasons of space, however, the following discussion will be limited to two prominent stylistic features: the use of synonyms to render close recurrences of the same Hebrew word (“variation”) and the arrangement of words or phrases in alternating, chiastic, or circular patterns. Divergences of LXX from MT in these areas are signs that the translator has, apparently deliberately, introduced features not strictly necessary for an adequate rendering of the Hebrew.3 As the source-text is normally followed closely, these phenomena are particularly interesting, prompting questions – to be discussed briefly at the end – about literary models, cultural milieu, and attitudes towards the task and its raison d’être. In what immediately follows, the treatment is largely descriptive.

1. Synonymous Variation Frequently in XII, a recurring Hebrew word is rendered by different, but virtually synonymous, Greek ones. Lee has drawn attention to the widespread occurrence in LXX of this feature, common in Greek literature,

1 Also called the Dodekapropheton. These books circulated as a collection from at least the early-second century BCE (Sir 49:10). Careful scrutiny of translation technique suggests that one translator was responsible for all twelve; for a survey of the evidence see B.A. Jones, The Formation of the Book of the Twelve (Atlanta: Scholars Press, 1995), 80–128. 2 Many are noted in the relevant volumes of La Bible d’Alexandrie (BA). 3 Unless there is reason to think otherwise, the Vorlage is taken to be represented by the (later) Masoretic Text (MT).

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

24

Jennifer M. Dines

whereby repetition is avoided in the interests of euphony and rhythm;4 he gives numerous examples, although none are from XII.5 Muraoka has provided a few examples,6 and Ziegler already listed many more in his demonstration of translational unity.7 It is clear, especially from Ziegler’s lists, that the practice occurs frequently across all twelve books. What I hope to do, by means of selected examples, is demonstrate the range and complexity of the translator’s stylistic choices.8 Not all variations are for purely aesthetic reasons; many are contextually motivated, as Swete observed and as the following examples show.9 1.1 Contextual Variation 1. Hos 2:9 (11); 11(13). Two different understandings of FYO occur here: In 2:9(11),G™PMCKTCW™VQW “in its time”, synonymous with MCS YŸTCP reproduces the temporal sense of YFYOD In 2:11(13), however, FYOcorrectly receives a cultic sense: VC?LRCPJIW TGKLCW™VJL, “her festivals”. The Hebrew play on the two senses of FYO is thus lost, and with it the echo between the two verses. In this example, correct sense has taken precedence over aesthetic concern, although the prepositions G™Pand MCVC(MT D) provide variation.10

4

The technical terms “metabolé”, or “variatio” cover more than the avoidance of repetition. As far as possible, I avoid using these and other rhetorical terms, since ancient definitions were often inconsistent, and little evidence of usage in the second century BCE survives. 5 J.A.L. Lee, “Translations of the Old Testament. I. Greek”, in S.E. Porter (ed.), Handbook of Classical Rhetoric in the Hellenistic Period, 330 B.C.E. – A.D. 400 (Leiden/New York/Köln: Brill, 1997), 775–83, pp. 777–8. 6 T. Muraoka, “Introduction aux douze petits prophètes”, in E. Bons/J. Joosten/S. Kessler, La Bible d’Alexandrie 23.1.Les Douze Prophètes. Osée (Paris : Cerf, 2002), I–XXIII, on pp. XIX– XX. See also J.K. Palmer, “Not Made With Tracing Paper: Studies in the Septuagint of Zechariah”, University of Cambridge dissertation, 2004, on pp. 30–4, for examples from Zechariah. Palmer identifies the practice as “a finger-print” of this translator (on p. 34). 7 J. Ziegler, “Die Einheit der Septuaginta zum Zwölfprophetenbuch”, in Sylloge: Gesammelte Aufsätze zur Septuaginta (Göttingen : Vandenhoeck & Ruprecht, 1971 [first published 1934/5]), 29–42, on pp. 34–6. 8 Passages are cited in LXX order, which differs from MT in the first six books. 9 H.B. Swete, An Introduction to the Old Testament in Greek (Cambridge: Cambridge University Press, 21914), 328–9; Palmer, “Not Made”, 34–5. 10 RCPJIWTKL also renders FYO (in parallel with IZ in Hos 9:5, see BA 23.1, 74. In Hos 12:9 (10), FYO is rendered by G˜QTVJž, “feast”, which may constitute a spaced variation with RCPJIWTKL in 2:11 (13); for other examples of this phenomenon, see below, pp. 30–1.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Stylistic Invention and Rhetorical Purpose in the Book of the Twelve

25

2. Amos 7:9, 13. Here, FSO, “sanctuary”, receives two different applications. In 7:9, CK˜VG NGVCKž, “the rituals”, underscore the unacceptable nature of Israel’s cult, with a term (found only here in LXX) proper to the Greek mysteries. But in 7:13, CIKCUOCDCUKNGYL, “the king’s shrine”, represents the northern priest Amaziah’s positive view of Bethel.11 3. Jon 1:3, 5. Where, in 1:1–5, MT emphasises Jonah’s progressive descent into chaos through repeated use ofFT[Y, the translator chooses contextually apt compounds of DCKPY: 1:3, MCK? MCVGDJGK™L,QRRJP “and he went down to Joppa”, MCK?G PGDJGK L CW VQP “and he went into (i.e. boarded) it”. .CVCDCKPY is the technical term for “going down” to the coast from inland.12 (ODCKPY is the technical term for “boarding” a ship.13 1:5, ,YPCL FG? MCVGDJ GK™L VJƒP MQKNJP VQW RNQKQW, “Jonas however went down into the hold of the ship”. .CVCDCKPY is again aptly chosen, and MQKNJ is the correct term for a ship’s hold.14 These contextual renderings coincidentally provide variation. The next section examines passages where variation is primarily for aesthetic reasons. Examples show its occurrence within the same verse (a), in consecutive verses (b), in near-by verses (c) and at more widely spaced intervals both within and between books (d). The section ends with examples of related stylistic devices (e–g). 1.2 Stylistic Variation a) Within the Same Verse 1. Mic 7:14. Two verbs are used for JT RQKOCKPG NCQP UQW, “shepherd your people” (?OJT, “pasture your people”) and PGOJUQPVCK VJ?P %CUCPK¨VKP, “they 11 This equivalence is frequent in LXX, but occurs in XII only here and Zech 7:3 (the Jerusalem Temple). 12 E.g. Herodotus Hist. 1.94; Plato Resp. 327a; cf WRQMCVCDCKPYP, Jos. C. Ap. 1.180, in a quotation from Clearchus ascribed to Aristotle. 13 E.g. Polybius 30.9.11. The reading of W, C PGDJ, also has this specialist sense. 14 E.g. Herodotus Hist. 8.119. Other examples of contextual variation include: Amos 8:5, 6 (SJUCWTQWL/IGPJOCVC for TD); 9:2, 3 (C™PCURCUGK/NJO[QOCK for ZSN); Joel 1:18 (DQWMQNKC/RQKOPKC for [TF); Obad 14, 18 (HGWIQPVCL/RWTHQTQL for F[T ); Hab 1:2, 3 (C™FKMQWOGPQL/C™UG‚DGKCP for UOJ); Zech 1:3, 4 (G™RKUVTG[CVG/C™RQUVTG[CVG for YDY); Mal 2:17; 3:5, 22 (FKMCKQUWPJ/MTK UKL/FKMCKY‚OCVC for RO). For further examples in Zechariah, see Palmer, “Not Made”, 35.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

26

Jennifer M. Dines

will graze Basanitis” (_DYT[, “they will find pasture [in] Bashan”). In non-biblical Greek, RQKOCKPYand PGOYare interchangeable; here, the passive of RQKOCKPY could have been used for YT[, while the active of PGOY could have represented JT (cf. Hos 4:16; already found in Homer, Il. 11.243 for “roaming the pastures”). In LXX, RQKOCKPY is more common than PGOY, but both render JTin the majority of cases.15 The lack of a preposition before _D permits a double entendre in Greek, absent from Hebrew: PGOY can mean “occupy, possess”, so that “they will graze Basanitis” also means “they will possess Basanitis” (the situation suggested, in any case, by the metaphor). 2. Nah 1:4. Two verbs in Greek represent NO (“wither”): Y NKIYSJ (NNO ) J %CUC PK¨VKL MCK? Q .CTOJNQL MCK? VCƒ G™ECPSQWPVC VQW /KDCPQW G™EGNKRG

(NNO ), “Basanitis and Carmel have been reduced and the blooms of Lebanon have died”.16 BHS suggests that a word beginning with dalet (e.g. NNF, “languish”)originally opened 1:4, fitting the supposed acrostic scheme in 1:1– 8. But the translator (who does not reproduce any alphabetic pattern) probably found the present reading NNO (attested also by 4QpNah) in his Vorlage, since Q NKIQY (a rare verb apparently first attested in LXX)17 never renders NNF, whereas it renders NO pul‘al also in Joel 1:10, 12.18 The equivalence, in fact, is found only in XII. The second occurrence of NO ,at the end of the verse, is rendered by G™MNGKRY “forsake”, “die out” (amongst other possible meanings). This verb translates NO only here in LXX, but it fits the context well. The two renderings suggest aesthetic variation. 3. Hab 3:2. G™PVG IIK\GKPVC? G¢VJ ... G™PVRCTGK¨PCKVQƒPMCKTQ‚P “when the years

approach ... when the time arrives”. This represents a twofold ][PDTSD, “in the midst of years”. LXX has read an infinitive construct clause DQTSK%, rather than, as MT, a prepositional phrase DGTGS% and has varied the verbs.19

15 In XII, RQKOCKPY occurs also in Hos 13:5; Mic 5:4, 6; Zech 11:4, 7, 9, 17, while PGOY occurs in Hos 4:16; Jon 3:7; Zeph 2:7, 14; 3:13. 16 NRSV translates NNO with variation: “wither ... fades”! 17 The classical form, also found in LXX, is Q™NKIYTGY; see M. Harl/C. Dogniez/L. Brottier/M. Casewitz/P. Sandevoir, La Bible d’Alexandrie. 23.4–9, Les Douze Prophètes. Joël– Sophonie (Paris: Cerf, 1999), 200. 18 In Hab 3:12, obelised by Hatch Redpath, the translator probably understood TE, “be insignificant”, for MT’s FE, “tread”, resulting in Q™NKIY‚UGKLIJP; see BA 23.4–9, 295. 19 For a detailed analysis of other stylistic features in this verse, see BA 23.4–9, 285–6.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Stylistic Invention and Rhetorical Purpose in the Book of the Twelve

27

4. Zech 12:7. Two nouns, MCWZJOC and G¢RCTUKL, represent VT RV, “glory”. This simple variation is interesting because G¢RCTUKL, “a raising up”, occurs only here in XII and is very rare in LXX as a whole.20 In non-biblical Greek, it is uncommon, and is first attested of the emotions (“elation”) in Stoic writers (Zeno 1.8; Chrysippus 3.116). .CWZJOC (“boast”) is also very rare (both inside and outside LXX), although the verb MCWZCQOCK (“boast”) is more frequent. The translator may have taken his lead from Ezek 24:25, where both words also occur.21 b) Within Consecutive Verses 1. Amos 5:14, 15. 5:14, G™M\JVJUCVGVQƒ MCNQƒP (DY) MCK? OJ? VQƒ RQPJTQ‚P (T), “seek good and not evil”. 5:15, OGOKUJMCOGP VC? RQPJTC (T) MCK? J ICRJMCOGP VC? MCNC (DY), “we have hated evil things and loved good things”. The chiastic pattern already exists in Hebrew, but the translator further avoids monotony by switching from singular to plural. 2. Joel 1:19, 20. There is double variation here: two “burn” verbs for NM (literally, “eat”), and two nouns for JF , “countryside”: 1:19, RWTC PJNYUG (JNM ) VC?YTCK¨CVJLG™TJOQWRCPVCVC?EWNCVQW C ITQW (JF J), “fire has consumed the harvests of the wilderness ... all the trees of the countryside”. 1:20, MCK? VCƒ MVJPJ VQW RGFKQW (JF ) ... MCK? RWT MCVGHCIGP (JNM ); “and the flocks and herds of the plain ... and fire has devoured ... ”. The two verbs chosen for NM , “consume”, are interchangeable in Greek, although C PCNKUMY (whose first meaning is “spend”, “waste”) is less common in this sense than MCVGUSKY Outside LXX, neither verb seems to be associated with RW¨T In LXX, MCVGUSKY is very common for NM , frequently with RW¨T as subject (e.g. Amos 1:4 etc.; Joel 2:5; Nah 3:15; Zech 11:1; MT  ). $PCNKUMYis less frequent, but the expression RWTC PJNYUGPCW™VJP occurs in Ezek 19:12 and may have suggested the alternative verb (cf. also 20

It occurs in Ezek 24:25, rendering Y O and O: VJƒPG RCTUKPVJLMCWZJUGYLCW™VYP (“the exalting of their boast ... the exalting of their soul”). MT:]RP OV Y... ]VT RV Y O, lit. “the exultation of their glory ... and the uplifting of their being”. 21 See previous note. It is unclear whether Ezekiel was translated before XII, but some evidence points in that direction (Muraoka, “Introduction”, XIII). It is also possible that the same translator was at work. 22 For the sense “consume”, “be eaten”, see Plato, Prot. 321b; Pausanius, Descr. 10.4,10. MCK? VJ?P G RCTUKP [WZJL CW™VYP

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

28

Jennifer M. Dines

Ezek 15:4, 5).23 In Joel 1:19, 20, the pair C PJNYUG/MCVGHCIGP seems a straightforward example of variation.24 It is more difficult to decide whether the pair VQWC ITQW/VQWRGFKQW, both rendering JF , is merely for versatility or whether there is also a contextual awareness that trees and cattle occupy different types of terrain. 3. Hag 2:6, 7 In 2:6, the eschatological “shaking” ([TO) of the cosmos is expressed through the simple verb UGKUY, but in 2:7, “I will shake” ([VTJ) is expressed by the compound UWUUGKUY. Sometimes compound verbs in XII represent derived stems in Hebrew, but in the present case both verb-forms are hiphil, showing that the practice is not automatic.25 The variation matches the difference in syllable count in Hebrew, perhaps deliberately (cf. 2:21 where UGKY again matches [TO with two syllables).26 4. Zech 14:4, 5 In these verses, which describe the splitting of the Mount of Olives, two different words are used for [I, “valley”: ZCQL(14:4) and HCTCIE(14:5): 14:4, MCK? UZKUSJUGVCK Q¢TQL VYP G™NCKYP  ZCQL ( [I) OGIC UHQFTC “and the mountain of olives will be split ... a very great chasm”. 14:5,MCK? G™OHTCZSJUGVCKHCTCIE ( [I) Q TGYPOQWMCK? G†IMQNNJSJUG VCKHCTCIE ([I) Q™TGYPGŸYL,CUQN, “and the valley of my mountains will be stopped up and the valley of the mountains will be welded to Iasol”. There is no obvious contextual reason for using different terms. )CTCIE is widely used for [Iin LXX, although in XII only here and Mic 6:2 (where it is plural, rendering ][PV , “enduring [foundation]”).27 ;CQL is, however, extremely rare, occurring in LXX only here and Mic 1:6. In nonbiblical Greek, it is primarily a scientific or philosophical term for primal matter or the nether abyss. Perhaps the translator wanted to enhance the supernatural element in this eschatological finale?28

23

See note 21 for the relationship between Ezekiel and XII. Noted also in BA 23.4–9, 56; cf. Zeph 3:8, MCVCPCNYSJUGVCK 25 Cf. Muraoka, “Introduction”, VIII (criticising Barthélemy). See also below p. 32 for similar variation in Amos 9:14, 15. 26 The Antiochian witnesses read the present, UGKY in 2:6 also; see M. Casevitz/C. Dogniez/M. Harl, La Bible d’Alexandrie. 23.10-11, Les Douze Prophètes. Aggée–Zacharie (Paris: Cerf, 2007), 82. 27 The reading (often emended to YP[\ J) is presupposed by LXX, which, however, has no reference to mountains, rendering ][TJ as NCQKž, so the first meaning of _V (“watercourse”) may have been intended (cf. Amos 5:24, although HCTCIE is not used); HCTCIE often renders NZP, especially in the Pentateuch. 28 Other examples include: Hos 8:7, 8 (MCVCHCIQPVCK/MCVGRQSJ for ND); Amos 3:4–5 (CT RCUX/VQWUWNNCDGK¨P for FMN); Zech 8:21, 22 (FGJSJPCK/G™M\JVJUCK for VYNJN). 24

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Stylistic Invention and Rhetorical Purpose in the Book of the Twelve

29

c) Within Nearby Verses 1. Amos 1:5, 8. 1:5, MCK?MCVCMQ[YHWNJP, “and I will butcher a tribe”. 1:8, MCK?G™ECTSJUGVCKHWNJ‚ “and a tribe will be taken out”. Both phrases represent D?OYVY, “and the one who holds the sceptre”. The translator has interpreted Dnot as “sceptre” but as “tribe”, with no variation.29 He has understood ?OYV not as a present participle but as a finite verb. Its basic meaning, “grasp” now made poor sense; accordingly, he has chosen two verbs suiting the general context of judgement and punishment. He has also created a variation in the syntax, using first the active (1:5), then the passive (1:8). Although the context has led him to the choice of verbs, the consistency in meaning – destruction of the “tribe” – makes the variation primarily aesthetic.30 2. Jon 1:12, 15. 1:12, C¢TCVG OG ([P ) MCK? G™ODCNGVG OG ([PN[JY) GK™L VJ?P SCNCUUCP “lift me up and throw me into the sea”. 1:15, MCK? G¢NCDQP (Y [Y) VQ?P ,YPCP MCK? G™EGDCNQP CW VQ?P (YJN[Y) GK™L VJ?PSCNCUUCP “and they took Jonas and cast him into the sea”. The variations seem to have no other purpose than to enhance narrative interest. 3. Zech 9:6, 10. 9:6,MCK? MCSGNYWŸDTKPC NNQHWNYP, “and I will remove the arrogance of foreigners” (][VNR_Y I[VTMJY, “I will strike the pride of the Philistines”). 9:10, MCK? G™EQNGSTGWUGKCŸTOCVCG E(HTCKO, “and he will thoroughly destroy chariots from Ephraim” (][TR ODMT[VTMJY, “I will strike the chariotry from Ephraim”). ™(EQNGSTGWY is the normal rendering for VTMhiphil. Possibly the translator chose MCSCKTY in 9:6 (an equivalence occurring only here in LXX)31 because “arrogance” (WŸDTKL) cannot literally be “struck”. But in non-biblical Greek, MCSCKTY can have a non-literal sense; Herodotus, for example, uses it of “reducing” the “power” (FWPCOKP) of the Persians (Hist. 1.71). So the variation is probably aesthetic.32 29

In Mic 5:1 (4:14), two contextual renderings occur: G PTCDF/VC?LHWNCL could also mean “be exalted”, but the wider context prohibits this. 31 The usual equivalences areUTJ, LVP, occasionally FT[ hiphil. 32 Other examples include: Amos 1:3, 13 (VYP G P *CNCCF/VYP *CNCCFK¨VYP for FNIJ); Joel 2:17, 19 (Q¢PGKFQL/Q PGKFKUOQP for JRTZ); Jon 4:3, 8 (J \JPOG/J \JP for [[ZO); Zech 4:5, 13 (QW  IKPYUMGKL/QW M QK»FCL for VF[ NJ); 12:2, 6 (MWMN/MWMNQSGP for D[DU; MWMN here constitutes 30

G ECTSJUGVCK

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

30

Jennifer M. Dines

All the types of variation considered so far are also found in non-biblical literature. The translator seems, however, to go beyond the normal practice of varying words which occur reasonably close together and often chooses different renderings for the same Hebrew word, even at considerable distance.33 d) Widely Spaced Variation 1. Amos 7:8; 8:1(2). Here, VK UWƒ QT„L and VK UWƒ DNGRGKL both render J TJV JO, “what are you seeing?” The contexts are identical, and dramatic enough to create an echo, even with 7:9–17 intervening. The verbs are interchangeable and there is no reason other than stylistic for the variation. The alternation is noteworthy in the light of Lee’s observation that these synonymous verbs are used with equal frequency only from the mid-second century BCE.34 The translator rings the same changes in Hag 2:3: GK»FGP (J T) / DNGRGVG (][ T) and Zech 5:2: DNGRGKLQTY„(JG QT each time). 2. Nah 1:10; 3:1. 1:10, YLMCNCOJEJTCUKCLOGUVJ( NO), “like straw filled with dryness”. 3:1, Y®RQNKL... C FKMKCLRNJTJL(J NO),“oh city ... full of injustice”. As OGUVQL is extremely rare in LXX, occurring elsewhere only in Esther 5:2, Prov 6:34 and Ezek 37:1 (each time rendering NO), the variation is surely deliberate, even though the two verses are so far apart. 3. Hab 3:17; Hag 1:10. Here, VCƒ IGPJOCVC (Hab 3:17) and VC? G MHQTKC (Hag 1:10) both render NYD[. All three words mean “produce”. The first rendering is uncommon, occurring elsewhere only in Lev 26:4, Deut 32:22 and Zech 8:12, although IGP(P)JOC occurs frequently for semantically related words, especially J YDV. (MHQTKQP is, however, extremely rare in LXX, rendering NYD[elsewhere only in Judg6:4 (A). In Lev 25:19, Deut 28:33 it renders different Hebrew words, while in Mal 3:10 (the only other occurrence in XII), it represents T O, “tithe”. Even with the whole of Zephaniah coming ______________________ the central element in a sequence of five verses, spanning the book, all the rest having MWMNQSGP: 2:5(9); 7:7; 12:2; 12:6; 14:14). 33 Authors composing in Greek may of course practice this kind of variation, but it would be less obvious to the reader. The fact of translation makes the phenomenon clear, at least when Greek and Hebrew are compared. 34 I.e. the consensus date for the translation of XII. By the first century BCE, DNGRY becomes more frequent than QTCY; J.A.L. Lee, A Lexical Study of the Pentateuch (Atlanta: SBL Press, 1983), 131–40.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Stylistic Invention and Rhetorical Purpose in the Book of the Twelve

31

between, it is most likely that the variation, involving such recherché vocabulary, is deliberate.35 These soundings have shown a range of ways in which the translator employs artistic variation. The practice must, however, be put in perspective: there are numerous places where the translator could have introduced variation but did not; sometimes variation is replaced by the opposite stylistic device of repetition (anaphora), that is, one Greek word represents two different ones in Hebrew; in some cases, variation and repetition occur within the same verse(s). e) Examples of non-variation: 1. Hos 3:4. A recurring _[ is rendered consistently by a negative and the present participle of GK™OKž: QW M Q¢PVQL / QW™FG? Q¢PVQL / QW™FG? QW¢UJL / QW™FG? Q¢PVQL / QW™ F?G? / QW™FGEven here, the necessary grammatical changes create variety. 2. Hos 7:9. CW™VQ?L FG? QW™M G¢IPY  MCK? CW™VQ?L QW™M G¢IPY Both clauses represent F[ N YJY(“but he did not know”). The deliberate variation between FG and MCK, where there is no modification in sense to require the former,36 reduces monotony.

3. Hos 9:7; 11:10, 11. 9:7, JŸMCUKPCK˜ JOGTCKJŸMCUKPCK˜ JOGTCK, “the days have come ... the days have come” ([O[Y D... [O[Y D). 11:10, 11 MCK? G™MUVJUQPVCK  MCK? G™MUVJUQPVCK “will be astonished” (YFTZ[Y... YFTZ[Y). In these verses, no variation occurs at all. 4. Hag 2:4. MCVKUZWG / MCK¼ MCVKUZWG / MCK? MCVKUZWGVY(S\ZY…S\ZY…S\Z) “be

strong”. The variation in the third verb is grammatically necessary, marking a shift from second to third person imperative; the first two verbs show no variation.

35 36

Another example occurs in Nah 2:6; 3:18 (OGIKUVCPGLand FWPCUVCLfor [T[F ). On MCKand FG‚ see below, n. 62.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

32

Jennifer M. Dines

f) Repetition 1. Amos 9:14, 15. In 9:14, Y , “they will make (gardens)”, becomes HWVGWUQWUK, “they will plant”. In 9:15, ][VP, “I will plant them”, is rendered exactly by MCVCHWVGWUY CW™VQWL. Where MT uses two different verbs, the translator creates repetition to link the verses in a new and striking way. At the same time, he achieves variety by using first the simple, then the compound form of HWVGWY. 2. Mic 7:2. G™MSNKDQWUKP  G™MSNKDX„, “they afflict ... with affliction”, is a contextual guess for ]TZYFYE[, “they hunt ... with a net”, with no attempt to reflect the two unrelated Hebrew words.37

3. Joel 2:1, 15. UCNRKUCVGUCNRKIIK, “trumpet on a trumpet”, uses cognate verb and noun for TRYYSV, “blow the trumpet”, where two unrelated words are used (cf. Hos 5:8; Zech 9:14).

4. Joel 3:2 (2:29). G™RK? VQW?L FQWNQWL MCK? G™RK? VCƒL FQW‚NCL, “upon male slaves and upon female slaves”. Where MT uses unrelated terms for “slave” (][FDJ and VYZRJ), LXX’s cognates create a new and pleasing effect, reinforcing the equality of the recipients of the spirit.

5. Nah 1:6. C RQ? RTQUYRQW Q TIJL CW™VQW (YO\) VKL W˜RQUVJUGVCK (FYO[[O); MCK? VKL C PVKUVJUGVCK (]YS[[OY) G™P Q TIX„ (_YTZD) SWOQW CW™VQW (YR ); Q SW OQƒLCW™VQW (YVOZ) ... , “before his anger who will stand fast? And who will

withstand in the anger of his wrath? His wrath ...”. Variation in Hebrew becomes repetition in Greek, with Q TIJ used for both ]\and _YTZ, and SWOQL for both X and JOZ. The verb K¶UVJOK is chosen for both FO and ]YS, although variation is achieved through the different compounds (W˜RQ and C PVK). Did the translator run out of Greek words for anger? The two nouns, Q™TIJ and SWOQL, are default renderings for all four Hebrew words. Muraoka comments that the translator has difficulty in matching the wide range of Hebrew terms for anger, though Greek does have other possibilities; he cites the present passage as an

37 A different choice for ]TZoccurs in Hab 1:17. The use of cognates is a frequent strategy in the rendering of infinitive absolutes, e.g. Mic 2:12, 13; 5:2; Joel 1:7.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Stylistic Invention and Rhetorical Purpose in the Book of the Twelve

33

example.38 But given the translator’s extensive and rarified vocabulary (Muraoka cites the abundance of locust words in Joel 1:4, for instance), the restriction here to Q™TIJ and SWOQL may be deliberate, especially as the repetition of SWOQL in 1:6b could be an example of anadiplosis.39

g) Variation and Repetition in the same Verse(s) 1. Hos 1:2, 3. DCFK\G, “go” (?N, 1:2), and G™RQTGWSJ, “ he set out” (?N[Y, 1:3) ring the changes on ?NJ, “go”. But NCDG, “take” (ZS, 1:2), and G¢NCDGP, “he took” (ZS[Y, 1:3), render ZSN, “take” each time. The two verses are thus evenly balanced between variation and repetition. 2. Amos 1:3, 13. 1:3, VYP G™P *CNCCF, “of those in Galaad”, and 1:13, VYP *CNCCFKVY„P, “of the Galaadites”, both render FNIJ, “Gilead”, a true variation. Each time, however,VYTJ(“pregnant women of”) is rendered by VC?LG™PICUVTK? G™ZQWUCL.40 As in Hos 1:2, 3, the combination of variation and repetition is apparently for aesthetic reasons. 3. Obad 13. MT’s threefold YF[ ]Y[D … YF[ ]Y[D … ]F[ ]Y[D, “on the day of their distress … on the day of his distress / on the day of his distress”, appears in Greek as: G™PJOGTRQPYPCW™VYP “on the day of their afflictions”,  G™PJOGTQ NGSTQWCW™VYP“on the day of their destruction”, G™PJOGTC RYNGKCLCW™VYP“on the day of their loss”. Between two regularly repeated expressions (G™P JOGT and CW™VYP, the latter smoothing out MT’s change from plural to singular), the translator 38

Muraoka, ‘Introduction’, XX–XXI. I.e. beginning a clause with a word from the end of the previous one; cf. Hos 11:1 (10:15) for another example (C RGTTKHJUCPC RGTTKHJ); G.O. Rowe, “Style”, in Porter, Handbook, 121–57, p. 131; Lee, “Translations”, 779. The device occurs in Hebrew as well as Greek. Further examples of repetition where MT has variation include Amos 2:14; Joel 3(4):4; Hab 1:17; Zech 8:10; Zech 9:8. 40 The evidence of 5QAmos increases the probability that the translator found VYTJ in his Vorlage at 1:3; see G.J. Brooke, ‘The Twelve Minor Prophets and the Dead Sea Scrolls’, in A. Lemaire (ed.), Congress Volume, Leiden 2004 (Leiden/Boston: Brill, 2006), 19–43, p. 25, with, however, a note of caution about Milik’s reconstruction. 41 The plural is the reading of W; most witnesses have the singular RQPQW. The word-choice prompts BHS to suggest the emendation YPY , but in view of the translator’s liking for variation, this is unnecessary. 39

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

34

Jennifer M. Dines

has found three different nouns for F[ . He thus achieves both repetition, creating a sense of the enormity of Idumea’s behaviour, and enough variety to avoid monotony.42 These examples suggest that the translator was trying, however inconsistently, to reproduce stylistic practices in both source and target language. This twofold character to the translation is reinforced by other artistic features.

2. New Verbal Patterns43 2.1 Amos 1:3–2:6. The most spectacular example is afforded by this long pericope, the so called “Oracles against the Nations”. I have already presented a brief analysis,44 but additional features, not covered there, add to its sophistication. Addressee

Suffix/Pronoun Conj/ Prep.

Verbal Form

1:3 S OF

YP

N

infin constr + ] 



CW™VQP

C PS Y¡¡P

'COCUMQW

(fem sg)

1:6 J\ 

*C\JL 

(fem sg)

(masc sg)

YP

CW™VQWL 

(masc pl)

G¢RTK\QP

(3 pl imperf) N

GŸPGMGP 

infin constr + ]

VQWCK™ZOCNYVGWUCK CW™VQWL

(aor infin + masc pl)

1:9 TE

YP

N

infin constr + ]



7WTQW

CW™VJP

C PS Y¡¡P

UWPGMNGKUCP

(fem sg)

(fem sg)

(3 pl aor)

42 Other examples include: Joel 2:17–18 (repetition with HGK¨UCK/G™HGKUCVQ for JUYZNOZ[Y; variation with Q¢PGKFQL/Q PGKFKUOQP for JRTZ); Zech 8:14–15 (different compounds of PQGQOCK for [VOO\/[VOZP. 43 Cf. Muraoka, “Introduction”, XIX–XX. 44 J.M. Dines, The Septuagint (London/New York: T & T Clark, 2004), 55–6.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Stylistic Invention and Rhetorical Purpose in the Book of the Twelve

1:11 ]YF 

YP

VJL,FQWOCKCLCW™VQWL

(fem sg) 

N

infin constr + Y

GŸPGMC

VQWFKYECKCW™VQWL

(masc pl)

1:13 _YO [PD

YP

WK˜YP$OOYP CW™VQP

(masc pl)

35

(aor infin + masc pl) N

infin constr + ]

C PS Y¡¡P

C PGUZK\QP

(masc sg)

(3 pl imperf)

2:1 D YO

YP

N



0YCD

CW™VQP

C PS Y¡¡P

(? masc sg)

(masc sg)

infin constr + Y MCVGMCWUCP

(3 pl aor)

2:4 JFYJ[

YP

N

infin constr + ]



WK˜YP,QWFC

CW™VQP

GŸPGMC

VQWC RYUCUSCK CW™VQWL

(masc pl)

(masc sg)

(aor infin + masc pl)

2:6 N T [

YP

N

Infin Constr + ]



,UTCJN

CW™VQP

C PS Y¡¡P

C RGFQPVQ

(masc sg)

(masc sg)

(3 pl aor)

Comment: 1. Where MT always has YPas pronominal suffix (YPD[  N, “I will not cause it to return”),45 LXX relates the pronoun to the addressee. The pronoun does not, however, always match the addressee’s gender or number. Instead, an ABCB´A´ sequence is created for 1:3–15, consisting of: 1:3, masculine singular (A), 1:6, masculine plural (B), 1:9, feminine singular (C), 1:11, masculine plural (B´), 1:13, masculine singular (A´). This is followed, in 2:1–6, by three masculine singulars. 2. To introduce the nature of each “impiety”, MT has Nfollowed by infinitive construct with third person plural masculine suffix, except for 1:11; 2:1 where the suffix is third person singular. LXX alternates between C PS Y¡¡P with finite verb, and GŸPGMGP(1:6) or GŸPGMC(1:11; 2:4),46 with articular infinitive and pronoun. The alternation is sustained in 1:3–15, ending, in 1:13, with C PS  Y¡¡P. 2:1 breaks the pattern, repeating C PS  Y¡¡P, and thus introducing a final ABA´ sequence. Together with the threefold CW™VQP, this 45 For a survey of interpretations and for the ambiguity of the expression, see J.R. Linville, “What does ‘It’ mean? Interpretation at the Point of No Return in Amos 1–2”, Biblical Interpretation 8 (2000) 400–24. 46 Perhaps itself a variation; usage and meaning are identical.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

36

Jennifer M. Dines

suggests that a new section is indicated. The slight shuffle also enables the whole passage to begin and end with an inclusion (C PS Y¡¡P, 1:3; 2:6). 3. Whereas MT regularly employs an infinitive construct, LXX creates yet another ABCB´A´ arrangement for 1:3–15 with variation in verb tense and form: 1:3, imperfect (A), 1:6, aorist infinitive (B), 1:9, aorist (C), 1:11, aorist infinitive (B´), and 1:13, imperfect (A´). Again, 2:1–6 has its own ABA´ pattern: aorist (A), aorist infinitive (B), aorist (A´), three aorists matching three masculine singulars. 4. Perversely, in the two places where MT shows variation with a masculine singular suffix (1:11; 2:1),47 LXX maintains masculine plurals. If these singular suffixes appeared in the Vorlage, the impression that the translator was creating his own pattern is reinforced. 5. There is no variation in the introduction to each oracle: “for the three impieties of ... and for the four, I will not turn away from ...” This matches the Hebrew exactly and is a more developed example of the mixture of variation and repetition already demonstrated as a feature of the translator’s style.48 Inferences: 1. The translator evidently treated 1:3–15 and 2:1–6 as separate, but connected, sections. This is interesting because the division matches the standard – but from the point of view of content, unnatural – break between chapters one and two, reflected both by the start of a parashah in MT and by divisions in Greek manuscripts.49 The repetition of C PS Y¡¡P in 1:13; 2:1 may be another example of anadiplosis, marking both continuity and development.50 2. A circular structure for 1:3–15 is clearly marked by the central oracle (1:9–10), where the only feminine singular pronoun (CW VJP) and the only finite verb in the aorist (UWPGMNGKUCP) both occur. 3. The matching of the two imperfects (G¢RTK\QP/C PGUZK\QP, 1:3, 13) creates an inclusion. The further inclusion formed by the reference to the Gileadite

47 48 49 50

1:11; 1:13; 2:1 thus form an ABA´ sequence in MT. Above, pp. 33–4. See Swete, “Introduction”, 342–66. See above, note 39.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Stylistic Invention and Rhetorical Purpose in the Book of the Twelve

37

women (VYPG™P*CNCCF and VYP*CNCCFKVY„P) contrasts with MT which only mentions the women in 1:13.51 4. The more homogeneous treatment of 2:1–6, suggests that the translator was bringing the whole sequence to a solemn conclusion. Where MT achieves a cumulative effect by the hammer-blows of repetition, LXX creates an ebb and flow in 1:3–15 through alternations building up expectations which are brought to closure in 2:1–6 by the change in patterning.52 Why, when the Hebrew of 1:3–2:6 is already so stylised, did the translator create this new double pattern? 1:3–2:6 is the most elaborate set piece in XII. Perhaps the translator was aware of this and tried to do it justice. But it was evidently not enough merely to reproduce it step by step. Perhaps the relentless repetitions of the Hebrew struck an ear also attuned to Greek periods as static, or even clumsy, and so needing to be at least partially lightened by variation and alternation. The translator has not changed everything: he balances new and old, source and translation. Significantly, he did not, apparently, think it inappropriate to alter, and even improve on, the poetics of the original. 2.2 Further Patterning in Amos It is surprising, after the tour de force of 1:3–2:6, that none of the other obvious places in Amos is treated in the same way (although admittedly these are shorter): 3:3–6 (a succession of rhetorical questions); 4:6–11 (a list of disasters, with refrain); 7:1–6 (three visions, with stereotyped beginnings). A few places where there are some attempts at artistry are, however, visible, especially 9:11. Here, a rudimentary pattern already present in MT is enhanced by means of four paired verbs all beginning with C PC underlining the hopeful effects of restoration: C PCUVJUY, “I will raise up (][S , “I will raise”) ... C PQKMQFQOJUY“I will build up” ([VTFIY, “I will repair”53) ... C PCUVJUY, “I will raise up” (][S , “I will raise”) ... C PQKMQFQ OJUY“I will build up” ([V[PDY, “I will build”).54

51

See above, p. 33. Other patterning overlaps with the main structure: in the first four oracles, the addressee pronouns are arranged in a singular/plural sequence: 1:3 CW VQP; 1:6 CW VQWL; 1:9 CW VJP; 1:11 CW VQWL. 53 The object in MT is _J[ETR, “their breaches”; LXX interprets as VC? RGRVYMQVC, “fallen parts”. 54 See Dines, Septuagint, 56. Examples of a different type follow in the next section. 52

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

38

Jennifer M. Dines

2.3 Triadic Sequences Although there is nothing to match the artistry of Amos 1:3–2:6, there are several places where patterning on a smaller scale occurs. In particular, there is a noticeable tendency to create short triadic, often circular, sequences. By circular I mean an ABA´ pattern which I consider a form of chiasm.55 1. Amos 2:14–15. A threefold NO[ N, “will not escape”, is rendered QW OJ?UYUX(2:14), QW  OJ?FKCUYSX„(2:15), QW™OJ?UYUX(2:15).56 2:15 contains a further triad: MCK? Q VQEQVJL, “and the bowman” (VSJ RVY, lit. “the one drawing the bow”), MCK? Q Q EW?L VQK¨L RQUKP, “and the swift of foot” (Y[NITDNSY), MCK? Q K˜RRGWL, “and the horseman”(UYUJDMTY, lit. “the one riding the horse”). The translator could, had he wished, have rendered the first and third elements with phrases exactly matching the Hebrew, as in the second element or, conversely, have found a single noun for the second element (cf. Q FTQOGWL, “the runner”, for NSin 2:14). The variation (single noun / phrase / single noun) seems deliberate. 2. Jon 1:9; 1:13; 2:11. JD[J, “the dry land”, is rendered VJ?PEJTCP (1:9), VJ?PIJP (1:13), VJƒP EJTCP (2:11). There seems no reason other than stylistic for the variation. The context in 1:9 perhaps suggests Gen 1:9–10, where EJTC occurs, but an allusion is less obvious in 2:11. 3. Hag 1:9, 10. There is a small triadic pattern not present in MT: FKC?VQWVQ(JO_[), C PS  Y¡¡P (_[), FKC?VQWVQ (_MN).  4. Zech 10:7. ZCTJUGVCK, “will rejoice” (ZO ), GW™HTCPSJUGVCK, “will exult” (YZO ), ZCTJUGVCK, “will rejoice” (NI[). Variation for the two occurrences of

55 Contra R. Meynet, “The Question at the Centre: A Specific Device of Rhetorical Argumentation in Scripture”, in A. Eriksson / T.H. Olbricht / W. Überlacker (ed.), Rhetorical Argumentation in Biblical Texts (Harrisburg PA: Trinity Press International, 2002), 200–214, who thinks that “concentric construction” should be distinguished from chiasm in the classic ABB´A´ form (on p. 200). Watson’s more inclusive approach seems more helpful; W.G.E. Watson, “Chiastic Patterns in Biblical Hebrew Poetry”, in J.W. Welch (ed.), Chiasmus in Antiquity (Hildesheim: Gerstenberg Verlag, 1981), 118–68. 56 WhenNO[ Nrecurs in 9:1, the alternation continues with QW OJ?FKCUYSX„

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Stylistic Invention and Rhetorical Purpose in the Book of the Twelve

39

ZO and repetition, instead of further variation, for N[I, result in a small chiasm.57 2.4 Other Patterning Other forms of variation occur, including chiasms, and, additionally, extended alternating sequences. 1. Hos 4:12–Zech 10:1. An alternation (ABCA´B´) extending across several books is constituted by the renderings of N , “ask”, a verb which occurs only five times in XII: A Hos 4:12: G™PUWODQWNQKLG™RJTYVYP “they consulted with advisors” (N [YED, “[my people] ask a piece of wood”).58 B Mic 7:3:QC¢TZYPCK™VGK¨ “the ruler requests” (N T J, “the official asks”). C Jon 4:8: MCK?C RGNGIGVQ59VJƒP[WZJƒPCW™VQW„, “and he renounced his life” (YRPV N [Y, “and he asked for his life”, i.e. to die). A’ Hag 2:11: G™RJTYVJUQPVQWƒLK˜GTGK¨L, “consult the priests” (][PJMJV  PN , “ask the priests”). B’ Zech 10:1: CK™VGK¨USGRCTCƒMWTKQW, “request from the Lord” (JYJ[OYN , “ask from the Lord”).60 The verb in Jon 4:8 is adapted to the context. But the variation between G™RGTYVCY and CK™VGY seems an open choice, since both verbs frequently render N elsewhere in LXX. A small and widely spaced ABCA’B’ 57

See Zeph 3:14; Zech 2:10(14) for a more dispersed play on ZCKTYand GW HTCKPQOCK: Zeph 3:14 ZCK¨TG (YPT), GW HTCKPQW(YZO ); Zech 2:10 (14) VGTRQW(YPT), GW HTCKPQW(YZO ). Other triadic sequences include: Hos 9:10/Mic 7:1/Nah 3:12 (UMQRQL/ RTYVQIQPYP/ UMQRQL for JTYMD); Amos 1:8/5:15/9:12 (MCVCNQKRQK/RGTKNQKRQK/MCVCNQKRQK for V[T ); Jon 1:3, 5 (MCVCDCKPY G ODCKPY MCVCDCKPY; above, p. 25); cf. Ziegler, “Die Einheit”, 9. 58 G™P UWODQWNQKL is Ziegler’s preferred reading over G P UWODQNQKL, “through tokens” (i.e. omens); J. Ziegler, Septuaginta. Vetus Testamentum Graecum XIII, Duodecim Prophetae (Göttingen: Vandenhoeck & Ruprecht, 31984), 155. For a discussion, see BA 23.1, 90. 59 The verb occurs only here in LXX. As well as “choose”, in Hellenistic Greek (frequently in Polybius) it can mean “refuse” and, in the Middle, “decline something offered”, “renounce”, “lose heart”. For the Greek as a free rendering of the Hebrew idiom, see BA 23.4–9, 161. 60 NETS follows NRSV and translates “ask”, thereby losing the correlation with Mic 7:3; A. Pietersma / B.G. Wright (ed.), A New English Translation of the Septuagint [NETS] (New York / Oxford: Oxford University Press, 2007), 817.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

40

Jennifer M. Dines

pattern can thus be discerned (or at least, alternation between the two verbs in Hos 4:12/Mic 7:3 and Hag 2:11/Zech 10:1). The rarity ofN in XII makes a deliberate arrangement plausible. 2. Hos 5:2–13:4: alternation between G™IYƒFGand MCK?G™IY‚ The first three occurrences of MCK? G™IY‚, in Hos 1:9; 2:4; 3:3, render both [MP Y (1:9; 2:4) and [P ]IY(3:3). But from 5:2 to 13:4, there is unbroken alternation between G™IYƒFGand MCK?G™IY‚61: this does not correlate with [P Y and [MP Y in Hebrew, or with their position in the sentence:62 5:2, G™IYƒFG([P Y) 5:12, MCK?G™IY‚ ([P Y) 7:13, G™IYƒFG([MP Y) 7:15, MCK?G™IY‚ ([MP Y) 10:11, G™IYƒFG([MP Y) 11:1, MCK?G™IY‚ ([MP Y) 12:9(10), G™IYƒFG([MP Y) 12:11(12), MCK?G™IY‚ ([MP Y) 13:4, G IYƒFG([MP Y) In 5:2; 7:13; 10:11; 12:9(10), G™IY‚ FG has an adversative sense, while in 7:15; 11:1; 12:11(12), MCK? G™IY‚ does not, so the alternation looks contextual. But in 5:12, [P Ywould have been more appropriately rendered by the adversative G™IY‚ FG, while in 13:4, G™IYƒ FG does not really point up a contrast, this having been marked already in 13:3. If the translator has mainly chosen the expression appropriate to the context, he seems to have overridden this principle in favour of preserving the alternation. 3. Hab 2:6–Zeph 2:18. Alternation between the interjections QW™CKž (“ah”, “alas”, often expressing disapproval) and the more general Y¤ (“oh”) stretches across the central sections of these books.63 Before Hab 2:6, LXX, whatever the Hebrew interjection, always has QW CKž (thus reducing variation): Hos 7:13; 9:12 ([Y ); 61

In 7:15, Rahlfs adopts MC IY‚, the reading of B, V. If original, this could have been a small additional variation, cf. the one occurrence of GŸPGMGPin Amos 1:6; see above, p. 35. 62 For FG as marking a new section, see F.H. Polak, “Context Sensitive Translation and Parataxis in Biblical Narrative”, in S.M. Paul/R.A. Kraft/L.H. Schiffman/W.W. Fields (ed.), Emanuel: Studies in Hebrew Bible, Septuagint and Dead Sea Scrolls (FS E. Tov; Leiden: Brill, 2003) 525– 39, pp. 530–1, with n. 21. Examples are from Gen and Exod; none involve G™IYƒFG/MCK?G™IY‚. 63 QW CKž, rarely attested in non-biblical Greek, is used in LXX more frequently than Y¤. Classical QK¢OQK (often written QK¢OOQK in LXX manuscripts) occurs in Joel 1:15 as a threefold wail rendering a single occurrence of JJ , and in Mic 7:1 as a twofold rendering of [N[NN (reduplication perhaps for rhetorical effect; Rowe, “Style”, 129–30; cf. Aeschylus, Ag. 1214; Pers. 985; Sophocles, Oed. col. 224). Other classical interjections, such as CK™CK¨K™QWK™Y‚ RQRQKQ VQVQK¨HGW„, do not occur in LXX.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Stylistic Invention and Rhetorical Purpose in the Book of the Twelve

41

Amos 5:16 (YJ), 18; 6:1 ([YJ).64 Y¤ first occurs in Jon 4:2 (JP ).65 It recurs in Nah 3:1 ([YJ), thus binding Nahum and Jonah; similarly, Nahum and Habakkuk, Habakkuk and Zephaniah, end and begin with QW™CK.66 From Hab 2:6 until the last occurrence in Zech 11:17, Hebrew consistently reads [YJ. But in LXX, Hab 2:6–19 has an unbroken alternation between QW CKž and Y¤, ending in 2:19 with QW™CKž (thus forming an inclusion and constituting a chiastic ABAB´A´ pattern, each interjection being followed by Q and present participle). The three occurrences in Zephaniah begin and end with QW CKž (2:15; 3:18, where it is a plus), enclosing Y¤ in 3:2 (ABA´).67 This is another example of a new section (here, a new book) repeating the last element in the previous part of a patterned structure; it provides another small triadic arrangement. The final three occurrences (Zech 2:6 2x; 11:17), all have Y¤, perhaps – as with the triple CW™VQP in Amos 2:1–668 – indicating closure and balancing the exclusive use of QW™CKž in Hosea and Amos.69 The sequence from Jon 4:2 to Zech 11:7 runs: Jon 4:2, Y¤ (JP ); Nah 3:1, Y¤ ([YJ); 3:17, QW™CKž (no equivalent in MT); Hab 2:6, QW™CKž ([YJ); 2:9, Y¤; 2:12, QW CKž; 2:15, Y¤; 2:19, QW™CKžž; Zeph 2:5, QW™CKžž; 3:2, Y¤, 3.18, QW™CKž (no equivalent in MT); Zech 2:6, Y¤ (x2); 11:17, Y¤; 4. Zeph 2:3. MT has three clauses beginning with YSD, “seek”. In the first two, JYJ[V YSD and SFEYSD, LXX reverses the word order to create a chiasm of the ABB´A´ type: \JVJUCVG VQƒP MWTKQP  MCK? FKMCKQUWPJP \JVJUCVG Changes of word order are rare in XII, which suggests the translator was deliberately creating an effect (perhaps equating “seeking the Lord” with “seeking justice”). The classic chiastic form would surely have been instantly recognisable.70 64 The double QW™CKž in Mic 7:4 has no counterpart in MT; Ziegler brackets it as probably secondary (although pre-hexaplaric); Duodecim Prophetae, 136. 65 In Jon 1:14, appropriately for an entreaty, JP (only here in LXX) is rendered OJFCOYL, “on no account”. 66 In Nah 3:17, QW™CK? CW™VQK¨L presupposes ]JN[Y , “woe to them”, for MT’s anomalous ][ , “where are they?”; see BA 23.4–9, 229. 67 The final QW™CKž would have been more appropriately rendered by Y¤, since this is not a condemnation but an exclamation of sorrow, within an oracle of restoration, over those who had harmed Sion. 68 Above, p. 35. 69 Unless it is contextual. BDB, 223, think [YJ here has “a touch of sympathy or pity”. This would work for 2:6, but not for 11:8 where QW™CKž would have fitted better. 70 The third verb, again YSD, is omitted, and the noun JYP, “humility” is read as part of the verb JP, “answer”, resulting in MCK? C RQMTKPGUSGCW™VC‚ For the omission of redundant, grammatically identical items, see J. Joosten, “A Septuagintal Translation Technique in the Minor Prophets”, in F. García Martinez / M. Vervenne (ed.), Interpreting Translation (FS J. Lust; Leuven/Paris/Dudley MA: Leuven University Press/Peeters, 2005), 217–23.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

42

Jennifer M. Dines

5. Hag 1:5, 7; 2:15, 18. A fivefold arrangement is achieved, the Hebrew each time being ]MDDNYO[ , “set your heart”:71 1:5 VCECVGFJ?VCƒLMCTFKCL, 1:7 SGUSGVCƒLMCTFKCL, 2:15 SGUSGFJƒGK™LVCƒLMCTFKCL, 2:18 W˜RQVCECVG FJ? VCƒL MCTFKCL  SGUSG G™P VCK¨L MCTFKCKL In 1:5, 7, the variation VCECVG/ SGUSG is unlikely to be contextual (1:5 looking back and 1:7 forward), for in 2:15, SGUSG, not VCECVG is chosen for another backward look. In 2:18, the clauses introduced by W˜RQVCECVG and SGUSG form an inclusion round the central element, the inaugural date for the rebuilding of the Temple – the pivotal event for the whole book. The alternation is not, therefore, primarily between two different meanings, but is one of emphasis, for the verbs underline the seriousness of the people’s moral commitment. 2:18 provides yet another example of a short triadic run (SGUSG/W˜RQVCECVG/SGUSG). There is an additional small variation between VCECVG (1:5) and W˜RQVCECVG (2:18). The reversal of the sequence in 1:7; 2:15 (marked by the repetition of SGUSG) is reminiscent of the repetition of C PS Y¡¡P in Amos 1:13; 2:1, creating a link between the verses and opening a new sequence.72 As in the case of Amos 1:3–2:6, none of the foregoing arrangements serve any grammatical, syntactical or exegetical purpose; they neither obscure nor elucidate the Hebrew. In the case of Hab 2:6–Zeph 2:18, the regular occurrence of the more idiomatic and stylish G™IYƒ FG may provide a veneer of Greek sophistication, but neither here, nor in Hos 5:2–13:4 (unlike Amos 1:3–2:6) was there monotonous repetition in the Hebrew to be alleviated. Summary There are sufficient examples of variation, of other rhetorical features, and of more elaborate patterning scattered throughout XII to show that these are significant translational practices. Why, though, did the translator go to this kind of trouble in a predominantly exact rendering? To conclude, I offer some reflections – which raise further questions – relating to the cultural milieu from which the translation emerged, and to its possible purpose and function.

71 2:18 MT adds P- after the first YO[ . The variations in Greek, concerning the complement “heart”, reflect differences in the Hebrew; see BA 23.10–11, 75. These verses are among Muraoka’s examples, “Introduction”, XIX–XX. 72 See above, pp. 35-6.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Stylistic Invention and Rhetorical Purpose in the Book of the Twelve

43

3. Cultural Milieu and Purpose of LXX Twelve 3.1 The implications of variation and other stylistic features Frequent use of synonymous variation and occasional use of other features, especially repetition and chiasm, suggest that the translator had absorbed some at least of the elementary rules of Greek compositional style, even if his practice was intermittent.73 What might this suggest about his educational experience? Courses in literary composition, by now much influenced by rhetoric (originally the art of public speaking), were foundational to Hellenistic secondary education.74 Rhetoric itself was “a regular part of the formal education of young men”.75 Unfortunately, as with much other Hellenistic material, most of the relevant sources, including the manuals of rhetoric, have been lost; our information comes mainly from the Roman period, although continuity may be largely assumed.76 Another problem is that little is known about how Jews in the mid-second century BCE were educated and to what degree they attended schools and gymnasia.77 Some Jewish youths may have gone to them, but others may have been educated privately.78 It has been suggested that bilingual Jewish schools were attached to the evolving institution of the synagogue, but there is no firm evidence.79 In any case, even if there was little formal education in rhetoric available for Jews, there was probably public access to courts of law where orators used their skills,80 and public lectures were delivered by travelling sophists.81 By whatever means, it is clear that recognisable – if spasmodic – 73 Palmer thinks variation resulted from the translator’s “linguistic sensibility” rather than “a conscious translation policy” (Palmer, “Not Made”, 36). The primary concern was clearly to render the perceived sense of the Hebrew, but I would call the translation policy with regard to variation secondary rather than unconscious. 74 H.-I. Marrou, Histoire de l’éducation dans l’antiquité (Paris: Seuil, 61965), 151–336; D.E. Orton/R.D. Anderson, A Handbook of Literary Rhetoric (Leiden/Boston/Köln: Brill, 1998), 10–20. 75 G.A. Kennedy, A New History of Classical Rhetoric (Princeton NJ: Princeton University Press, 1994), 3. 76 Ps-Demetrius On Style is now usually dated to the first, not the third, century BCE. Aristotle’s Rhetoric only became known in the first century BCE. 77 See L.H. Feldman, Jew and Gentile in the Ancient World: Attitudes and Interactions from Alexander to Justinian (Princeton NJ: Princeton University Press, 1993), 56–9. For positive conclusions about diaspora communities, see M. Hengel, Judaism and Hellenism (London: SCM Press, 2 vols 1974), 1.66–70. Feldman is more hesitant (Jew and Gentile, 24). 78 F. Siegert, “Homily and Panegyrical Sermon”, in Porter, Handbook, 421–43, p. 437. 79 E.g. by Hengel who suggests that Paul may have acquired his knowledge of philosophy “from the teaching of the Greek-speaking synagogue” (Judaism, 2.115, n. 444). 80 W. Wuellner, “Arrangements”, in Porter, Handbook, 51–87, p. 54; R.W. Smith, The Art of Rhetoric in Alexandria (The Hague: Martinus Nijhoff, 1974), 61. 81 Smith, Art of Rhetoric, 130.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

44

Jennifer M. Dines

features of Greek rhetorical style occur in the translation of XII. Only a more thorough study of the entire collection could show the extent of the translator’s formal education and literary skills. What is already evident is that his use of variation at least goes beyond the occasional, since he sometimes develops alternations between synonyms in quite elaborate ways. The end effect, and most interesting feature, is that he regularly creates new artistic patterns where his source text did not require it. 3.2 Elaborate Patterning Another marked preference is for circular or chiastic structures, both small and large. Although a basic form of chiasm occurs in Greek authors, and more elaborate structures have been detected in, for instance, Homer,82 the scale on which the translator uses it may stem as much from Semitic as from Greek literary practice, since circular and chiastic arrangements are common in Hebrew.83 Alphabetic acrostics are also found in both Hebrew and Greek.84 The translator does not reproduce the partial alphabetic arrangement in Nah 1:2–8, but alternations like those between G™IYƒ FG and MCK? G™IY‚, or between QW™CKž and the more general Y¤, may have come from a similar impulse to create verbal structures and interconnections. Alternating parallelism, for instance, occurs frequently in Psalms and Isaiah.85 These typically Hebraic skills would, of course, blend with the conventions of Greek rhetorical style, in which taxis (“arrangement”) was as important as lexis (“word-choice”).86 The translator seems to have been motivated by both Greek and Semitic poetics, for he provides evidence of both. The result often looks like inexplicable inconsistency, or as if merely occasional lip-service is being done to Greek learning. The reality is surely more complex.

82 See J.W. Welch, “Chiasmus in Ancient Greek and Latin Literatures”, in J.W. Welch (ed.), Chiasmus in Antiquity (Hildesheim: Gerstenberg Verlag, 1981) 250–68, pp. 250–8. 83 Watson, “Chiastic Patterns”; see above, n. 55. 84 In Hebrew see e.g Pss 37; 112; 119; P.C. Craigie, “Excursus III: Acrostic Psalms”, in Psalms 1–50, Word Biblical Commentary (Waco TX: Word Books, 1983), 128–31. 85 J.T. Willis, “Alternating (ABA´B´) Parallelism in the Old Testament Psalms and Prophetic Literature”, in E.R. Follis (ed.), Directions in Biblical Hebrew Poetry (Sheffield: JSOT Press, 1987), 49–76. 86 See e.g. Rowe “Style”, 129, 137; cf. Wuellner, “Arrangement”, 55. It is not clear, however, whether chiasm played a significant part in taxis at this period.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Stylistic Invention and Rhetorical Purpose in the Book of the Twelve

45

3.3 Implications for LXX Twelve If the translator’s intention was to produce a straightforward version of Hebrew XII (a supposition supported by the overall character of the version), why would he have made even occasional unnecessary embellishments, of either a Greek or a Semitic nature? The question is particularly acute if the purpose of the translation was to “bring the reader to the original” and to act as a quasi “interlinear” guide.87 The abundance of stylistic inventions proper to LXX (and the creation of new alliterations and other poetic features not covered in this paper) suggest that, even if it was initially used in conjunction with Hebrew XII, the translation had an independent character from the start, with literary features that other Greek speaking Jews would have recognised. These readers were presumably capable of responding with both recognition and aesthetic satisfaction to the variations and patterns introduced by the translator even where these were different from the arrangements in the Hebrew. Features associated with Greek literature would not have seemed out of place if the claims, found in some Hellenistic Jewish writers, that Greek authors and philosophers “borrowed” from Moses, were widely embraced.88 Such claims only make sense if Jews had been exposed to Greek culture. If aesthetic factors were at work, the puzzle is to understand why stylistic “improvements” are so erratic. For, despite the numerous (and by no means exhaustive) examples I have given, the translator mainly uses default renderings, and mostly copies the stylistic features of the Hebrew. Until a systematic investigation is undertaken, it is difficult to tell the proportion of variation to non-variation, and similarly for the other features.89 But the range and spread of stylistic embellishments, including carefully crafted passages (especially Amos 1:3–2:6), and other poetic effects, make it clear that the translation was produced with at least some pretension to literary merit and not exclusively as an aid to reading the original Hebrew. This could shed light on the milieu in which the translation was made and first used. It reinforces the likelihood that the primary Sitz im Leben was the school, and that Greek XII had a fundamentally pedagogic purpose (which does not, of course, negate other uses including liturgical). Attention 87 See A. Pietersma, “To the reader of NETS”, in NETS, xiv–xv. For evaluations of this approach by A. van der Kooij and N. Fernández Marcos, see B.A. Taylor (ed.), X Congress Volume of the IOSCS, 1998 (Chico CA: SBL Press, 2000), 229–31; 233–8. 88 E.g. Aristobulus in Eusebius, Praep. ev 13.12.4,13. See C. Holladay, Fragments from Hellenistic Jewish Authors, Vol. 3, Aristobulus (Chico CA: University of California Press, 1995) 165, 189; Dines, Septuagint, 35; cf. Jos, Ag. Ap. 1.62–5; 2.168, 257, 281–2. 89 It would also be helpful to see how far stylistic traits discovered in XII occur in LXX Pentateuch and other books which might have acted as models for the translator.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

46

Jennifer M. Dines

to matters of style would then be natural as the mark of an educated translator. Whether he attended a Greek or a Jewish school, or had benefited from Greek or Jewish private tutors, the curriculum and teaching methods would have been largely the same. On the other hand, the translation is not, for the most part, in a literary Greek of a very high order,90 and the frequent interference of Hebrew syntax, idiom, and word order results in a strange hybrid. It is as though the translator is pulled in two directions at once: wanting to produce good Greek (of which he is quite capable) and simultaneously to reproduce traits of the original Hebrew. From the phenomena of the translation itself, I have a growing impression that, rather than bringing Greek-speaking Jews (and perhaps non-Jews) to the Hebrew, the translator is trying to bring the original to his Greek readers in a way in which they can get its flavour, so to speak, within the broad conventions of Greek compositional style. If the translation originated in a Jewish educational context in Alexandria, or elsewhere in Egypt, the marked prevalence of new verbal patterns, alternations, chiastic arrangements and so on (in addition to ones already represented in Hebrew), may be more easily explained. Such devices would not have been merely decorations, or tours de force demonstrating the translator’s skill. In both Hebrew and Greek pedagogy, they could have had a mnemonic function, acting as markers to help students of the scriptures memorize what they were hearing or reading, as well as bringing out the meaning. Memorizing was an established feature of rhetorical training, with mnemonic aids, albeit of a different kind from those apparent here.91 The translator’s invention of new patterns (Hos 5:2–13:4; Amos 1:3–2:6; Hab 2:6–Zech 3:18 and so on) would then have had a practical raison d’être. Connections across two or more books would be particularly useful if – as seems likely – XII were studied as a unified collection. If I am right about this mnemonic function, the translation provides a precious glimpse into methods of biblical teaching and learning in the mid-second century BCE. A further reason for the prominence of artistic embellishments may lie within the nature of rhetorical style itself. Rhetoric was essentially concerned with persuasion, and I wonder whether the translator saw his task as one of transmitting the sacred teachings of the prophetic books in ways most likely to gain success with his implied readership. Hints embedded in Hebrew XII already point in this direction, such as the wisdom exhortation 90

So Siegert, “Homily”, 436. Mnemonics traditionally originated in sixth-century Athens. Practitioners in the second century included Charmadas and Metrodorus (Cicero De Or. 2.360). The most detailed exposition is in the first-century BCE treatise Ad Herennium; see Kennedy, A New History, 123–4; cf. T.H. Olbricht, “Delivery and Memory”, 163. In Hebrew, the alphabetic acrostics may have served a similar purpose. 91

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Stylistic Invention and Rhetorical Purpose in the Book of the Twelve

47

in Hos 14:10 and (perhaps) the aphorism in Amos 5:13. Greek XII creates several new didactic and paraenetic touches, including the introduction of imperatives or first person individualisations, implying direct address and dialogue.92 Baer explains similar traits in LXX Isaiah 56–66 as homiletic or didactic updatings.93 If the translation is marked by attempts to engage the reader/listener in a personal way, the question arises as to what kind of book the translator thought he was transmitting; in contemporary terminology, its genre. This is not anachronistic, for biblical superscriptions, such as those in Cant 1:1; Prov 1:1; 1 Sam 1:1, imply distinct categories, and in Greek there were rules and models for epic, history, philosophy and so on. It is not, then, inappropriate to wonder what the translator thought he was dealing with in translating the words, visions and actions of the prophets, which include a variety of genres. Poetic passages could have suggested the popular genre of didactic poetry.94 The superscriptions to Hosea, Amos, Micah, Zephaniah, Haggai and Zechariah, as well as the narrative form of Jonah, could have made him think of history,95 a genre which, in Greek, often involved the use of rhetorical literary devices.96 Perhaps closest to the features which mark LXX Twelve is the genre of protreptic philosophy. This was concerned particularly with ethics and aimed to effect a change of life in the reader or hearer through the example of a revered teacher.97 It was closely related to the diatribe, “the paraenetic counterpart of the pro-

92 E.g. Hos 10:12 (“seek the Lord”, MT “it is time to seek”); 11:10 (“I will go after the Lord”, MT “they will go after”); Amos 5:15 (“we have hated evil”, MT “hate evil”); Mic 4:9 (10) (“be courageous”, MT “thrust (?)”; LXX has two significant pluses: “and draw near”; “your God”); Hab 2:2 (“the Lord answered me”, MT “the Lord answered”); Zech 6:14 (“those who endure/those who are serviceable/those who discern”, where MT has personal names); Mal 1:1 (cf. Hag 2:15). 93 D.A. Baer, When We All Go Home: Translation and Theology in LXX Isaiah 56–66 (Sheffield: Sheffield Academic Press, 2001), 52–3. 94 “Didactic poetry as a genre shares with rhetoric the need to consider its audience and, often, to persuade”, R. Webb, “Poetry and Rhetoric”, in Porter, Handbook, 339–69, p. 355. 95 Cf. the rabbinic designation of the historical books as “Former Prophets”. 96 See S. Rebenich, “Historical Prose”, in Porter, Handbook, 265–337, pp. 270, 287; R.G. Hall, “Ancient Historical Method and the Training of an Orator”, in S.E. Porter/T.H. Olbricht (ed.), The Rhetorical Analysis of Scripture (Sheffield: Sheffield Academic Press, 1997), 103–18. 97 Protreptic, a form of epideictic rhetoric (by now covering any discourse serving “to encourage belief, group solidarity and acceptance of a system of values”, Kennedy, A New History, 62), “was one of the literary genres recognised as such by ancient philosophers” (D.M. Schenkenveld, “Philosophical Prose”, in Porter, Handbook, 195–264, on p. 204). No Hellenistic protreptic treatises survive, except for Epicurus Letter 3 (preserved in Diogenes Laertius, De philosophorum vitis 10.122–35). With so few sources from the right period, it is difficult to tell the extent to which rhetoric affected this kind of philosophical writing, though Letter 3 has many literary touches.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

48

Jennifer M. Dines

treptic”.98 It would not have been strange for the translator to think in terms of philosophy: Moses’ work of lawgiving (PQOQSGUKžC) is called “philosophical” (HKNQUQHYVGTQP) by the author of The Letter of Aristeas in the second century BCE, a time when “philosophy was all the rage”,99 and later both Philo (Legat. 245; Mos 2.216) and Josephus (Ant. 1.25; Ag. Ap. 1.54; 2.47) call Judaism a “philosophy”; Josephus also refers to the Essenes and other groups as philosophical “schools” (Ant 18.9, 11, 23). Although speculative, it is not unreasonable to suppose that the translator thought along similar lines. It would make sense of the first and second person changes and would fit with an emphasis on the prophetic message as paideia (education, discipline), one of the distinctive features of LXX XII.100 A paraenetic (hortatory) aim might also help explain why the translation is sometimes literal, and sometimes not: according to the (later) rhetorical manuals, literalness is less important than achieving the desired persuasive effect by whatever means are most appropriate.101 If the translator did see his task in this light, it is possible that the use of stylistic variation and other more properly rhetorical ploys is part of a pedagogical and paraenetic understanding of the scriptures.102 More than just occasional and inconsequential features, they could be signs of a serious attempt to bring the urgent messages of God, given through the prophets of old, to his own generation. To further his aim, he has utilised methods of persuasion culled from both his Jewish and his Greek literary heritage.

98

Schenkenveld, “Philosophical Prose”, 231; cf. Kennedy, A New History, 92–3. O. Murray, “Philosophy and Monarchy in the Hellenistic World”, in T. Rajak/S.J. Pearce/J.K. Aitken/J.M. Dines (ed.), Jewish Perspectives on Hellenistic Rulers (Berkeley/Los Angeles/London: University of California Press, 2007) 13–28; on p. 16; cf. Ep. Arist. 31. 100 E.g. Amos 3:7; Hab 1:12; Zech 3:2, 7; cf. Hos 5:2 (RCKFGWVJL); 7:12, 15; 10:10 (RCKFGWY). For this emphasis on teaching/discipline as a translational parti pris, see BA 23.1, 29–30. On Hellenistic Greek attitudes to paideia, see Hengel Judaism, 1.65. Murray remarks that the philosopher as teacher was a commonplace; see previous note. A thorough analysis is now needed to discover whether there is any correlation between exhortative or didactic content and degree of literalness. 101 Rowe, “Style”, 125; cf. Augustine’s application of rhetorical method to Christian homiletics in Book 4 of De Doctrina Christiana (written between 396 and 427 CE). See also G.A. Kennedy, “A Historical Survey of Rhetoric”, in Porter, Handbook, 3–41, p. 37. 102 Cf. Kennedy, “Genres”, 53; Rowe, “Style”, 124. 99

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Thomas J. Kraus

Translating the Septuagint Psalms – some ‘Lesefrüchte’ and their value for an analysis of the rhetoric (and style) of the Septuagint (Psalms) 1. Preliminary remarks The present study has resulted from my work as a collaborator in the translation and commentary project Septuaginta Deutsch – Das griechische Alte Testament in Übersetzung,1 in the course of which I participated in a group of researchers who focused on the linguistic analysis of the Psalms. All in all I worked on thirty-six Psalms, though in the first place I was much less ambitious as far as workload is concerned.2 The volume with the translations of the Septuagint was published in 2009 with even a second edition in the same year,3 while the second with the commentaries will appear in 2011. It is not necessary to stress that the work on such an ambitious project covered several years, due to the complex nature of the Psalms, individually and as a corpus of texts. The Greek text needed to be translated into adequate German, something that involves methodical questions about translation technique (e.g., between literal and free); and, of course, the significant differences between Greek and Hebrew, between Septuagint and the available texts of the Masora and from Qumran had to be indicated in an appropriate way.4 During the years each collaborator of the Psalm group has made diverse observations that could only partly be integrated into the commentary volume. One requirement for establishing a readable and usable linguistic commentary was to limit explanations to the essentials. Consequently, passages had to be shortened, and other observations were regarded as not vital enough for a two-volume commentary. Nonetheless, as with all the other collaborators of the Psalm group, I collected and saved all 1 For a description of the project, see W. Kraus/M. Karrer, “ ‘Septuaginta – deutsch’ ”, BiKi 56 (2001) 104–105. There is also a brief introduction to the project, its collaborators, guidelines, publications, and texts of the project on the Internet (go to http://www.septuagintaforschung.de and then click on ‘LXX.D’; last access 28/02/2011). 2 These are (Septuagint number): Psalms 13; 20; 51–52; 55–56; 59; 62–63; 66; 85–86; 89; 90; 103; 107; 111; 114–115; 117; 119–120; 122–123; 125–129; 131–132; 136–137; 142–144. 3 W. Kraus/M. Karrer (ed.), Septuaginta Deutsch. Das griechische Alte Testament in deutscher Übersetzung (Stuttgart: Deutsche Bibelgesellschaft, 22010). 4 Especially for the psalms other translations became interesting as well, just as were the discussions and translations by the early Christian writers.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

50

Thomas J. Kraus

these additional observations and explanations that did not make it into the two volumes of Septuaginta Deutsch, with the aim of rearranging and systematizing selected passages for publication.5 On the occasion of the International IOSCS conference in Ljubljana, Slovenia, on 13 and 14 July 2007 I presented a selection of peculiarities which I came across during the work on the Psalms, accompanied by further comments about how and why these peculiarities are interesting. Personally, I prefer the German term “Lesefrüchte” (literally “fruits from reading” or, to be more precise, ‘findings in the course of an intensive reading process’) to cumbersome and bulky phrases such as ‘observations and explanations’, ‘interesting phenomena’, or ‘peculiar aspects in the Greek or in contrast to the Hebrew’, to mention only a few. Basically, I do not offer indepth explanations of these ‘Lesefrüchte”. Instead, my intention is to draw attention to them and to initiate further reflections on them. In other words, I hope to promote, if not provoke, further research on the style and rhetoric of the Septuagint. Of course, due to the limited space in this volume I do not intend to be complete and, thus, have simply provided a rather random selection of ‘Lesefrüchte’. My hope is that they are sufficient for highlighting the issue of evaluating the language of the Septuagint Psalms and of coming closer to descriptive and validating study of their style and rhetoric. Although this study is written in English, the German translation published in Septuaginta Deutsch serves as its textual corpus for reference. In order to supply a readable English text, I provide Albert Pietersma’s NETS translation of the Septuagint Psalms6 to accompany the published German version and to facilitate readability. Only here and there is Pietersma’s translation altered in order to produce a more literal agreement between the Greek and its English translation, even if the English that is produced might read more clumsily.7 Readers may compare the German and the English translations; they should not regard them as rivals, but as supplementing each other. Nonetheless, the focus is on the published German version.

5 See, for instance, T.J. Kraus, “ ‘Der Herr wird deinen Eingang und deinen Ausgang bewahren’: Über Herkunft und Fortleben von LXX Psalm cxx 8a”, VT 56 (2006) 58–75. 6 See A. Pietersma, The Psalms (NETS; New York-Oxford: Oxford University Press, 2000). 7 In addition, the following English translations are used for comparison: King James Version (KJV), Americam Revised Version (ARV), Revised Standard Version (RSV), New King James Version (NKJV), and New Revised Standard Version (NRSV).

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Translating the Septuagint Psalms – some ‘Lesefrüchte’

51

2. Clarifying, defining, disambiguating The title of this study already shows quite plainly some of the most crucial problems, which can at best be only touched on here: ‘Translating the Septuagint Psalms – some ‘Lesefrüchte’ and their value for an analysis of the rhetoric (and style) of the Septuagint (Psalms).’ The translation of a text always represents its own interpretation, at least to some degree. Of course, this sounds abstract and striking, almost meddlesome and vacuous. Nonetheless, it is important to be permanently aware of such a matter of course; the process of translating a text from one language into another always involves several decisions a translator has to make; and these decisions involve e.g. semantic, syntactic, or stylistic matters. By preferring one alternative, another – or even others – is discarded. Such a selection from a choice of alternatives marks a sometimes rather problematic but invariably necessary selection. Furthermore, translators have to adapt their translations to the idiom and terms of the target language,8 but also to a specific register or style. Of course, this is a major challenge translators have to face and cope with all the time. In addition, renderings should be accurate and reliable and take into account the socio-cultural context of the original. Translators must focus on the readership they produce their translation for. Thus, they must make up their minds how they want to transfer the actual content of a text to another language, somewhere between the two extremes of a literal, verbatim translation and a loose, dynamic one.9 Although all of this sounds both rather complicated and self-evident, the terms ‘rhetoric’ and ‘style’ raise a problem that needs to be discussed in relation to the context of translating from one language into another (i.e., from one socio-cultural framework into another). When leafing through relevant handbooks both terms are defined in various ways, and the definitions of ‘rhetoric’ and ‘style’ overlap to a certain extent. ‘Style’ can be said 8 See, for instance, T.J. Kraus, “Der Artikel im Griechischen: Nutzen einer systematischen Beschäftigung anhand von ausgewählten Syntagmata (Hab 1,12; Jud 17; Joh 6,32)”, RB 107 (2000) 260–272, with a comparison of Greek, Hebrew, Latin and Syriac syntagmata, and idem, “Grammatisches Problembewusstsein als Regulativ für angemessene Sprachbeurteilung – das Beispiel der griechischen Negation und 2Petr”, FilNeot 14 (2001) 87–99, with a systematic assessment of the Greek negation, especially in 2 Peter. 9 For more details see F. Siegert, Zwischen Hebräischer Bibel und Altem Testament. Eine Einführung in die Septuaginta (Institutum Judaicum Delitzschianum – Münsteraner Judaistische Studien 9; Münster: LIT, 2001), 121–195, who classifies (122) ‘the Greek translation (the Septuagint) of the Hebrew Bible as its oldest commentary’ (“Die griechische Übersetzung der Hebräischen Bibel ist in gewisser Weise ihr ältester Kommentar.”). Further see B.G. Wright III, “Translation as Scripture: The Septuagint in Aristeas and Philo”, 47–61, and W. Kraus, “Contemporary Translations of the Septuagint: Problems and Perspectives”, 63–83, both in W. Kraus/R.G. Wooden (ed.), Septuagint Research: Issues and Challenges in the Study of the Greek Jewish Scriptures (Septuagint and Cognate Studies 53; Atlanta: Society of Biblical Literature, 2006).

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

52

Thomas J. Kraus

to be “any specific way of using language, which is characteristic of an author, school, period, or genre” and different styles “may be defined by their diction, syntax, imagery, rhythm, and use of figures, or by any other linguistic feature”, according to the The Concise Oxford Dictionary of Literary Terms.10 The Metzler Literatur Lexikon, however, contains an entry in which its author describes ‘style’ in relation to a historical standard and/or the preference for genres. A ‘style’ is shaped by specific attitudes and statements of individuals or groups (peoples, guilds, generations, social classes); and linguistic artwork is to be described by grammar, rhetoric, and figurativeness, determined by the choice, the mixture, and the intensity of use of all the possibilities offered by language. This culminates in the distinction between different categories of style: ‘rhetoric’, ‘poetic’, ‘objective’ and ‘subjective’.11 These two definitions randomly taken from two very popular handbooks – the first more pragmatic and the second more academic and complex – show quite plainly (a) that ‘style’ is not defined clearly and (b) that it is closely related to other terms that are ambiguous. To make things worse, ‘rhetoric’ adds to this scenario of imprecision and uncertainty. Is ‘rhetoric’ the ‘skill to advance a view convincingly by means of speaking publicly and so to influence the way of thinking and acting of others and the theory or academic discipline of this art’? Is ‘rhetoric’ then a special form of communication that focuses on persuading and convincing others?12 Do we have to reckon with “the deliberate exploitation of eloquence for the most persuasive effect in public speaking or in writing”?13 In other words, does ‘rhetoric’ only refer to what speakers or authors are fully aware of saying or writing and how they do it? The two handbooks again demonstrate that (a) definitions vary from each other and – when reading the whole entry in each of them – (b) rhetorical devices and figures of speech are regarded as essential for a definition of ‘rhetoric’. Therefore, ‘rhetoric’ is usually understood as having been formed in antiquity and as consisting 10 C. Baldick, The Concise Oxford Dictionary of Literary Terms (Oxford: Oxford University Press, 1990), 214. See further: “Different categories of style have been named after particular authors (e.g. Ciceronian), periods (e.g. Augustan), and professions (e.g. journalistic), while in the Renaissance a scheme of three-stylistic ‘levels’ was adopted …” 11 H. Weidhase, “Art. Stil”, in G. Schweikle/I. Schweikle (ed.), Metzler Literatur Lexikon. Begriffe und Definitionen (Stuttgart: J.B. Metzler, 21990), 443–444. Then follows the distinction between ‘poetic style’ 12 E. Däschler, “Art. Rhetorik”, in Schweikle, Metzler Literatur Lexikon, 389–390: “Fähigkeit, durch öffentl. Rede einen Standpunkt überzeugend zu vertreten und so Denken und Handeln anderer zu beeinflussen und Theorie bzw. Wissenschaft dieser Kunst.” See further: “Von anderen Formen der sprachl. Kommunikation hebt sich die Rh. Durch die Betonung der impressiven bzw. konnotativen Funktion der Sprache ab (Rh. = persuasive, überzeugende Kommunikation).” 13 Baldick, The Concise Oxford Dictionary, 188–189. Also see (190): “Modern critics sometimes refer to the rhetorical dimension of a literary work, meaning those aspects of the work that persuade or otherwise guide the responses of readers.”

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Translating the Septuagint Psalms – some ‘Lesefrüchte’

53

of three forms of speech according to Aristotle, Rhet. 1358b: genus iudicale (the forensic or judicial speech), genus deliberativum (the political or deliberative speech), and genus demonstrativum (the ceremonial or epideictic speech).14 In order to achieve the goal of persuasion successful ancient orators employed rhetorical devices and tropes, something that is still taken as a sign of excellence today.15 Writing about ‘rhetoric’ could go on for pages. However, there is the question what it is relevant for, or whether or not it is pure embroidery work. The aim was to illustrate the invalidity of the cited definitions for the present study.16 ‘Rhetoric’ integrates ‘style’ as linguistic decoration added to the text secondarily so that, due to its peculiarity, it can be removed again. Consequently, one focuses on the peculiar, the extraordinary, or the striking features so that all other linguistic features remain almost unnoticed.17 Similarly obstructive is the uncritical adoption of rhetorical devices and tropes from rhetorical handbooks in order to search for them in the texts one is looking at. Besides, it is often hard to tell whether an author has made a deliberate choice while writing or, in relation to the Septuagint, a translator while translating. Thus, it seems more appropriate to avoid ‘style’ and ‘rhetoric’ when researching the ‘language’ or ‘the Greek’ of the Septuagint.18 Terms as ‘language’, ‘the Greek (of the Septuagint)’ or the like are less controversial and more useful for a pragmatic approach as the one presented here, because aestheticism and artistic value – both often identi14 Of course, this is just a summary of the detailed description Aristotle offers in his Rhetoric. He also addresses ethos (character), pathos (emotion), and logos (reasoning) as means of persuasion. On these three terms see J.D. Ramage/J.C. Bean, Writing Arguments (Needham Heights/MA: Allyn & Bacon, 41998), 81–82. 15 Cf. H. Lausberg, Handbuch der literarischen Rhetorik. Eine Grundlegung der Literaturwissenschaft (Munich: Max Hueber, 1973 = Stuttgart: Steiner, 31990), with an extensive treatment of rhetorical devices and tropes. Lausberg mainly concentrates on the analysis of dispositio and elocutio with a focus on the latter, and his shorter Elemente der literarischen Rhetorik. Eine Einführung für Studierende der klassischen, romanischen, englischen und deutschen Philologie (Munich: Hueber, 102000). For a handy introduction to ancient rhetoric see W. Eisenhut, Einführung in die antike Rhetorik und ihre Geschichte (Die Altertumswissenschaft; Darmstadt: Wissenschaftliche Buchgesellschaft, 31994); M. Fuhrmann, Die antike Rhetorik. Eine Einführung (Düsseldorf: Patmos, 22008). 16 This does not mean that the definitions themselves are wrong and impractical. ‘Rhetoric’ always must be defined in accordance to the text or body of texts it is applied to. See the studies in C.J. Classen/H.-J. Müllenbrock (ed.), Die Macht des Wortes. Aspekte gegenwärtiger Rhetorikforschung (Ars Rhetorica 4; Marburg: Hitzeroth, 1992). 17 Cf. B. Sowinski, Stilistik, Stiltheorien und Stilanalysen (Sammlung Metzler 263; Stuttgart: Metzler, 1991), 14. 18 Cf., for instance, M. Harl, La langue de Japhet. Quinze etudes sur la Septante et le grec des chrétiens (Paris: Cerf, 1994), and the contributions in J. Joosten/P.J. Tomson (ed.), Voces Biblicae. Septuagint Greek and its Significance for the New Testament (Contributions to Biblical Exegesis & Theology 49; Leuven-Paris-Dudley/MA: Peeters, 2007).

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

54

Thomas J. Kraus

fied as the rhetorical capacities of authors (i.e. their capacity to use rhetorical devices and tropes correctly) – need not be an integral part of the collection and discussion of ‘Lesefrüchte’. Hence, the observation of ‘optional’ (= secondary or rhetorical) and ‘obligatory’ (= primary or grammatical) means of expression form the basis for the ‘Lesefrüchte’ presented here, supplemented by lexical and semantic details. These ‘Lesefrüchte’ taken together and expanded by many more such observations may add up to a description of language usage.19 My firm conviction is that if the ‘Lesefrüchte’ attract attention, they will motivate and maybe sometimes provoke scholars into further, more detailed, research. And they may even show the way for an in-depth description of language usage(s) in the Septuagint Psalms.

3. A selection of ‘Lesefrüchte’ from the Septuagint Psalms What appears at first glance to be a rather random selection and compilation of ‘Lesefrüchte’ from the Septuagint Psalms I was working on during the project20 is by no means a product of arbitrary choice. The examples are selected according to their potential contribution to the issue of whether or not the translators of the Hebrew into the Greek were more than pure translators of a ‘Vorlage’. The ‘Lesefrüchte’ of each category – 3.1 and 3.2 – are arranged in ascending order according to their number in the Septuagint Psalter. Although section 3.3 unveils a suspiciously sensational discovery in one of the Psalms, it turns out on closer inspection to be an illusion or misapprehension. This ‘Lesefrucht’ is both a product of tunnel vision or wishful thinking and a warning against not seeing the wood for the trees any more because of years of routine work on the Septuagint Psalms. Another warning is that, without any doubt, the chronological relationship between Masoretic Text and the Septuagint should not be forgotten, above all the fact that the Masora in the form we have it today is not the (exact) ‘Vorlage’ of the LXX.

19 For a distinctive circumscription of ‘style’ and ‘language use’ that is also applied here see T.J. Kraus, Sprache, Stil und historischer Ort des zweiten Petrusbriefes (WUNT 2.136; Tübingen: Mohr Siebeck, 2001), 27–30. Further see J. Anderegg, Literaturwissenschaftliche Stiltheorie (KVR 1429; Göttingen: Vandenhoeck & Ruprecht, 1977), 24 and 27, and M. Landfester, Einführung in die Stilistik der griechischen und lateinischen Literatursprachen (Die Altertumswissenschaft; Darmstadt: Wissenschaftliche Buchgesellschaft, 1977), 3–4, for a dynamic and pragmatic definition of ‘style’ and ‘stylistics’. 20 See note 2 above.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Translating the Septuagint Psalms – some ‘Lesefrüchte’

55

3.1 Diverse and individual ‘Lesefrüchte’ Psalm 51:3–4, 5, 9 [52:3–4, 5, 9MT] Ps 51:3–5 3 Was rühmst du dich in Bosheit, du Mächtiger, der Gesetzlosigkeit21 den ganzen Tag?22 4 Deine Zunge ersann Ungerechtigkeit. Wie ein geschärftes Schermesser begingst du Betrug. 5 Du liebtest Bosheit mehr als Rechtschaffenheit, Ungerechtigkeit mehr als Gerechtigkeit zu reden. [Why do you wickedly boast, O powerful one, of lawlessness all day long? Your tongue devised injustice. You produced treachery like a sharpened razor. You loved evil more than goodness, and injustice more than speaking justice.]23

While the MT has the plural VYYJ (of JYJ; ‘mischief’, ‘wickedness’, ‘destruction’24), which is repeated at the end of V.9, the LXX has the singular C FKMKC (“Ungerechtigkeit”, ‘injustice’), which is also used in V.5 where the MT has TS (‘evil’, ‘falsehood’; OCVCKQVJL in the LXX) instead. Consequently, the keyword linking of verses in the LXX (4 and 5) is different from that in the MT (4 and 9). Also C PQOKC (“Gesetzlosigkeit”, ‘lawlessness’) in V.3 (MT N FUZ; ‘goodness or loving kindness of God’) and the absence of a designation for God25 may link Vv.3–5 more closely than in the Hebrew text. The unit-like character of Vv.3–5 in the LXX is underlined by the rhyme C ICSYUWPJP and FKMCKQUWPJP (“Rechtschaffenheit” — “Gerechtigkeit”, ‘goodness’ — ‘justice’). Maybe the translator regarded the JNU (FKC[CNOC) as signalling a stop after V.2, so that Vv.3–5 should have been linked with each other rather closely.

21

Words and phrases in italics indicate a difference between the Masoretic Text and the Septuagint. For guidelines of how to use the German translation, see Kraus/Karrer, Septuaginta Deutsch, XVII–XXIII. 22 The German is quoted from Septuaginta Deutsch and given with the relevant formatting, signs, and symbols used there. 23 See note 6 above. 24 Cf., for instance, the King James Version, the American Standard Version, and the Revised Standard Version (and the New Revised Standard Version and the New King James Version) for reference. 25 Possibly, the LXX version is caused by a trend toward parallelism (parallelismus membrorum) and C PQOKC by reading UOZ (‘wrongdoing’) for FUZ (‘kindness’). Cf. K. Seybold, Die Psalmen (HAT 1.15; Tübingen: Mohr Siebeck, 1996) 215; F.-L. Hossfeld/E. Zenger, Psalmen 51– 100 (HThKAT; Freiburg-Basel-Vienna: Herder, 2000) 60 (= iidem, Psalms 2 [Hermeneia; transl. by L.M. Maloney; Minneapolis: Fortress Press, 2005] 26).

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

56

Thomas J. Kraus

Psalm 55:9 [56:9MT] Ps 55:9 9 Mein Leben verkünde ich dir, du hast meine Tränen vor dich gestellt wie auch in deiner Verheißung. [My life I proclaimed to you; you put my tears before you, as in fact you promised.]

This verse offers a slight shift of perspective between MT and LXX but also a noticeable difference in form and meaning. In the Hebrew God himself has apparently written down everything in his book (see JV ‘you’ and ?VTRUD NJ ‘Are they not in thy book?’). But the LXX presents the perspective of an active person who is praying to God (VJƒP\YJPOQWG EJIIGKNC UQW) and there is no mention of a book but of G RCIIGNKC (“Verheißung”, ‘promise’). It is only natural that, by means of a prayer, the petitioner’s tears are already before God (VCƒ FCMTWC OQWG PYRKQPUQW) so that there is no need to collect them in a bottle as God is asked to do in the MT.26 The praying person trusts in God’s promise (G RCIIGNKC; see V.5–6) and this promise is in the focus of the enemies’ hatred. But these ‘enemies will retreat in the day when I call upon you’ (V.10) so that God’s promise can be relied on because it will finally be fulfilled. Pss 56:4–5 [57:4–5MT] — 59:4 [60:4MT] — 107 [108MT] Ps 56:4–5 4 Er sandte vom Himmel aus und rettete mich. Er gab zur Beschimpfung die, die mich mit Füßen traten. Zwischenspiel. Gott sandte sein Erbarmen und seine Wahrheit 5 und befreite meine Seele aus der Mitte der Junglöwen. Ich schlief voller Unruhe. Menschenkinder, ihre Zähne sind Waffen und Pfeile und ihre Zunge ein scharfes Schwert. [He sent from heaven and saved me, he gave over to reproach those who were trampling on me. Interlude on strings God sent forth his mercy and his truth, and he rescued my soul from among lion whelps I slept, though troubled. Human children – their teeth are weapons and darts, and their tongue a sharp dagger.]

The MT has the piel perfect of XTZ that implies a more intense active meaning of what is going on in the meaning of ‘defy, reproach, taunt, mock’, and 26

See Hossfeld/Zenger, Psalmen, 115, 117 (= iidem, Psalms 2, 64, 66).

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Translating the Septuagint Psalms – some ‘Lesefrüchte’

57

together with the preceding qal imperfect ZN and the hifil imperfect of [ it shares the same subject, God. It may be strange to imagine God as the subject of the piel perfect of XTZ, with the result that alternatives constructions are discussed among scholars.27 The LXX version seems more direct and uncontroversial: the verbs are given as aorists and the jussive (i.e. voluntative) understanding of V.4–5 in the MT (‘he shall’, ‘let him’, ‘he may’ or the like) is lost in the Greek.28 Where the MT has the piel perfect of XTZ, the Greek reads the prepositional phrase G¢FYMGP GK L Q¢PGKFQL (but see Ps 54:13LXX with an aorist of Q PGKFK\Y for piel imperfect of XTZ so that God is not subject of the ‘reproach’ but the enemies are. In addition, with the help of G¢FYMGP V.4–5 are linked together: the verbs beginning with Gstructure the stichoi, especially G ECRGUVGKNGP (V.4a), G¢¢FYMGP (V.4b), and G ECRGUVGKNGP (V.4c) at the beginning of each stichos; and it obviously does not matter that there is a FKC[CNOC after V.4b. Then V.5ab has G TTWUCVQ and G MQKOJSJP before WKQKƒC PSTYRYP opens the ending of that passage, so that it might be the central message here (V.5cd). Moreover, there is GVQKOJ J MCTFKCOQW twice at the beginning of V.8ab and strikingly a set of four words beginning with G E- as first words of the stichoi in V.9–10a (twice G EGIGTSJVK, then G EGIGTSJUQOCK and G EQOQNQIJUQOCK). No doubt, these stichoi beginnings have a certain effect on the speed of reading/praying, its rhythm, and the task of memorizing the Psalm as a whole; and, although probably small in importance, the G-anaphora (V.8ab and 9ab) and the position of words beginning with G- structure the Greek Psalm in a particular and distinctive manner. Such a trend towards sound and rhythm with the help of equal initial sounds is not uncommon in the LXX: Ps 59:4 (cf. 60:4MT) can be of help to illustrate its effects. Ps 59:4 [60:4MT] UWPGUGKUCLVJƒPIJPMCKƒUWPGVCTCZCLCW VJƒP K¢CUCKVCƒUWPVTKOOCVCCW VJLQŸVKG UCNGWSJ  Du hast das Land erschüttert und es aufgeschreckt. Heile seine Risse, denn es hat gewankt. [You caused the Land to quake and threw it into confusion; heal its fractures, for it was shaken.]

What can be found here are the results of an earthquake. However, the LXX strings together three composita prefixed by UWP- (the two verbs UWP Psalm 57:4–5MT appear rather difficult and controversial. Cf. Hossfeld/Zenger, Psalmen, 119–120, 127 (= iidem, Psalms 2, 67–68, 73–74 28 Of course, the Hebrew imperfect can also be understood in an indicative meaning, which the translator might have accepted then, although the jussive seems to be more appropriate in the MT here. 27

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

58

Thomas J. Kraus

GUGKUCL and UWPGVCTCZCL, and the nounUWPVTKOOCVC), a near rhyme or repetition of word endings (homoioteleuton) in V.4a with VJƒPIJP – CW VJP, and the aural contrast between CW VJP (V.4a) and CW VJL (V.4b). The effect

becomes obvious when reading the verse out loud. Consequently, the LXX offers a stylistically distinctive form of the verse and the translator evidently utilizes the possibilities the Greek language offers for doing so. Hence, the verse is slightly different from its equivalent in the MT. But the differences marked with italics above (additional MCK and aorist passive of UCNGWY ‘it was shaken’ versus qal perfect of YO ‘it totters’29) do not provide the whole effect caused by the peculiarities in the Greek text. Moreover, parts of the Septuagint Psalms 56 and 59 form Psalm 107. Hence, Ps 56:8–12 corresponds to 107:2–6 and 59:7–14 to 107:7–14 (cf. Psalms 57, 60, and 108 of the MT). Even though the differences between the corresponding Hebrew passages seem to be marginal,30 it is notable that, in relation to the LXX version, there is a surprising inconsistency in differences between the MT and LXX texts: for example, where Pss 57:8–12 and 108:2–6 of the MT differ from each other, Pss 56:8–12 and 107:2–6 do not necessarily do so, and vice versa. The same is true for Pss 60 and 108 of the MT and 59 and 107 of the LXX.31 Pss 56:9 [57:9MT] & 90:6 [91:7MT] Ps 56:9 9 Wach auf, meine Herrlichkeit! Wach auf, Harfe und Leier! Ich will in aller Frühe aufwachen. [Awake, my glory! Awake, O harp and lyre! I will awaken at dawn.]

On closer inspection Ps 56:9LXX and 57:9MT have more to offer than just the fact that the Greek offers a nice example of how to employ anaphora and the same prefix for sound and rhythm (G E-). V.9c has the genitive Q¢TSTQW, a genitivus temporis, which denotes the time (‘at the time of dawn’, better ‘at dawn’).32 The MT, however, has TZJT[ ‘I want to wake the dawn’ with

29

See KJV and ASV ‘it shaketh’, and NKJV ‘it is shaking’. So Seybold, Die Psalmen, 431. For a detailed study see R.J. Tournay, “Psaumes 57, 60 et 108: Analyse et interprétation”, RB 96 (1989) 5–26. 31 For more details I refer to my exemplifications in Septuaginta Deutsch. Das griechische Alte Testament in deutscher Übersetzung. Ergänzungsband: Erläuterungen (ed. W. Kraus and M Karrer; Stuttgart: Deutsche Bibelgesellschaft, forthcoming). 32 Cf. F. Blass/A. Debrunner, Grammatik des neutestamentlichen Griechisch (bearb. V. F. Rehkopf; Göttingen: Vandenhoeck & Ruprecht, 171990) § 186.3 and n. 6. 30

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Translating the Septuagint Psalms – some ‘Lesefrüchte’

59

the hifil imperfect of TY ‘to rouse’ or ‘to stir up’33 (cf. Ps 108:3MT) and so personifies ‘dawn’ by means of a transitive verb (‘to wake sb’).34 Other passages in the MT also offer the idea of a personified ‘dawn’ (e.g., Job 3:9; 41:10; Ps 22:1; 139:9; Isa 14:12; Hos 6:3). The translator of the Hebrew might have regarded TY as an intransitive verb and, consequently translated TZ as Q¢TSTQL but then as a genitivus temporis in order to make good sense of it. Another possibility is however that the translator might have taken offence at the personification of ‘dawn’. Perhaps the possible background of TZin V.9 is “the original mythic configuration of the dawn as the power (or goddess) that bears life day after day”,35 which urged the translator to neutralize it by means of Q¢TSTQW.36 In Ps 90:6, however, the mention of a ‘noonday demon’ in the LXX text presents an interpretation or even an individual determination of the destructive power at work there. Ps 90:6 [91:7MT] 6 vor einer Tat, die in Finsternis einhergeht, vor einem Unglück und einem Mittagsdämon. [or the deed that travels in darkness, or mishap or noonday demon.]37

In the Greek UWORVYOC and FCKOQPKQPOGUJODTKPQP are linked with each other by means of MCK, while in the Hebrew DS ‘destruction’ is determined by a verbal phrase. Consequently, the MT offers a nice parallelism between V.6a and V.6b, which is not true for the LXX. Interestingly, 11QPsApa presents FYY ‘and devastation’ where the MT has a qal imperfect of FY[ ‘to devastate’, so that this might ring a bell and remind one of the derivation F ‘demon’ with which those who believed in JYJ[ denounced the idolatrous doing of others who sacrificed ][FN ‘to demons’ (e.g., among others, Deut 32:17; Ps 106:37MT).38 Therefore, the translator might have had F ‘demon’ in V.6b in front of him and thus translated ][TJEFYor the like 33



The verb TY implies the idea of ‘splitting the dark cloud cover’. Cf. E. König, Hebräisches und aramäisches Wörterbuch zum Alten Testament (Leipzig: Dieterisch’sche Verlagsbuchhandlung, 6–71936) s.v. TCZHand TCZC(p. 494). 34 RSV and NRS “I will awake the dawn”, NKJV “I will awaken the dawn”, while KJV “I myself will awake early” and ASV “I myself will awake right early” have rw( as an intransitive verb. 35 Hossfeld/Zenger, Psalms 2, 74. 36 Cf. H.-J. Kraus, Psalmen, 1. Teilband: Psalmen 1–59 (BK 15.1; Neukirchen-Vluyn: Neukirchener Verlag, 51978), 572–573; Hossfeld/Zenger, Psalmen, 129, 130 (= iidem, Psalms 2, 74, 75). 37 The difference in translation between the German (with preposition ‘vor’) and the English (without preposition) originates from the difference in using the verbs ‘fürchten (vor)’ and ‘fear sb’. 38 See König, Hebräisches und Aramäisches Wörterbuch, s.v. *GY( and GYR (p. 484).

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

60

Thomas J. Kraus

into Greek.39 Moreover, the MT possibly implies the responsibility of demons for ‘the pestilence that stalks in darkness’ (V.6a) and ‘the destruction that wastes at noonday’ (V.6b)40 so that we may identify ‘night demon’ (‘pestilence’) for the first and a ‘midday heat’ or ‘fever demon’ (‘destruction’) for the latter.41 Also Aquila (FCKOQPK\QPVCL OGUJODTKCL) and Symmachus (FCKOQPKYFGLOGUJODTKCL) contain the idea of a ‘noonday demon’ (FCKOQPKQPOGUJODTKPQP), though in a slightly different way.42 The adjective OGUJODTKPQL is rather rare in the LXX (only 1Esd 9:41; Job 5:14; Isa 16:3; but there are twenty-six occurrences for the noun OGUJODTKC, e.g. Ps 36:6; 54:18) and Ps 90:6 provides the only occurrence of both words together. There is the assumption that the translator might have read ][TJEFYor the like or might have had another ‘Vorlage’ (see, for instance, the variants in 11QPsApa) and the considerations made above may account for the LXX version of V.6b. But this and the difference between MT and LXX in V.6a, and also the failure in the LXX to personify the darkness as a ‘demon’, and then to stress that in a certain respect, require further explanation. Why is there the rather weak RTCIOC ‘deed’ where the MT has the semantically stronger TDF ‘pestilence’? Why did the translator, if he had a text similar to the MT as ‘Vorlage’, destroy the parallelism whereas the translators usually seemed to be very fond of constructing a parallelism (parallelismus membrorum) when possible? For a reader of the Septuagint version of Ps 90:6 the effect the Greek of V.6b has in its present state is obvious: it stresses the danger from the ‘noonday demon’ by personifying clearly43 the hazardous demonic power that was taken for granted by the people at the time the Septuagint version came into being and was read. The magical papyri from Egypt particularly 39 So B. Rebiger, ‘Die magische Verwendung von Psalmen im Judentum’, in E. Zenger (ed.), Ritual und Poesie. Formen und Orte religiöser Dichtung im alten Orient, im Judentum und im Christentum (HBS 36; Freiburg et al.: Herder, 2003), 265–281, p. 267. 40 Both translations according to RSV and NRSV. See KJV with “the pestilence that walketh in darkness” and “the destruction that wasteth at noonday”. 41 For details, see S. Landersdorfer, “Das daemonium meridianum (Ps 91 [90], 6)”, BZ 18 (1929) 294–300: Landersdorfer points to the Assyrians who believed in two gods who were responsible for attacks of malaria, especially during the midday heat and the cold of the night, and argues that this belief also became known west of Assyria (cf. his reference to the Syrian poet Mar Jakob from the 5th century CE). 42 Further see Theodoret who translates “a thing moving in darkness, a mischance and demon at noon” and then comments: “He made mention of the noonday demon in keeping with popular opinion …” Theodoret of Cyrus, Commentary on the Psalms, 73–150 (transl. by R.C. Hill; The Fathers of the Church 102; Washington D.C.: The Catholic University of America Press, 2001) 104. 43 The evil powers of death/sickness in the MT might have been meant as personified powers, too (see Hossfeld/Zenger, Psalmen, 622 [= iidem, Psalms 2, 430]), but the LXX is by far clearer with FCKOQPKQP OGUJODTKPQP, which leaves no doubt about the personification and the presence of a ‘demon’.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Translating the Septuagint Psalms – some ‘Lesefrüchte’

61

visualize the omnipresence of demons and gods, of powers which the people regarded as real and against or by whom they sought assistance and protection. Jews and Christians alike used the Psalm for protection against all kinds of evil powers, so that Psalm 91MT and 90LXX can be found, for example, on bowls, walls, armbands, pendants, medals, rings, tablets, in lintel inscriptions and papyri. Possibly, it is this established and traditional use in Judaism (cf., e.g. bSchevu 15b; yShab 6.2.8b; yEr 10.11.26c)44 that influenced the way the Psalm is phrased in Greek. In 11QpsApa (= 11Q11) Psalm 91 is set among texts of exorcism and is directed against demonic powers.45 On closer examination Septuagint Psalm 90 offers some more instances that may represent a mystifying and mythologizing trend during its translation process (see, e.g. ‘the basilisk’ DCUKNKUMQL in V.13).46 But this is not the place to go into that here, and I intend to study this trend in more detail in the near future. Psalm 127:2 [128:2MT] Ps 127:2 Die Früchte deiner Mühen47 sollst du essen. Selig bist du, und gut wird es dir gehen. [You shall eat the labours of your crops48; you are happy, and it shall be well with you.]

The LXX version VQWƒL RQPQWL VYP MCTRYP UQW HCIGUCK is literally ‘you shall eat the labours of your crops’49 according to the common mean44 For more information see Rebiger, “Die magische Verwendung”, 268–271, and, with a list of objects, T.J. Kraus, “Septuaginta-Psalm 90 in apotropäischer Verwendung: Vorüberlegungen für eine kritische Edition und (bisheriges) Datenmaterial”, BZ 125 (2004) 39–73. Further see idem, “Fragmente eines Amulett-Armbands im British Meseum (London) mit SeptuagintaPsalm 90 und der Huldigung der Magier”, JAC 48/49 (2005/2006) 114–127 and plates 2–3; idem, “%QWL, %CKPZYYZ und Septuaginta-Psalm 90? Überlegungen zu den sogenannten ‘Bous’Amuletten und dem beliebtesten Bibeltext für apotropäische Zwecke”, ZAC 11 (2008) 479–491. 45 Also see A. Wagner, “Ps 91 – Bekenntnis zu Jahwe”, 73–97 and M.S. Smith, “Primary and Secondary Religion in Psalms 91 and 139. A Response to Andreas Wagner”, 99–104, pp. 101–102, both in A. Wagner (ed.), Primäre und sekundäre Religion als Kategorie der Religionsgeschichte des Alten Testaments (BZAW 364; Berlin-New York: Walter de Gruyter, 2006). 46 Hence, this trend could have served to intensify the theology of Ps 90 that focuses on hope and protection. Hence, the LXX Psalm can be called a ‘psalm of hope sui generis’ according to Hossfeld/Zenger, Psalms 2, 429: “psalm of trust sui generis” for Ps 91MT, 432 (= iidem, Psalmen, 619–620, 625). Also P. Hugger, Jahwe meine Zuflucht. Gestalt und Theologie des 91. Psalms (MüSt 13; Münsterschwarzach: Vier-Türme-Verlag, 1971) 301. For more information see below for Ps 30 [31MT] and 90 [91MT]. 47 Verbatim translation: “Die Mühen deiner Früchte”. 48 See Pietersma, The Psalms, 612, gives “athe labors of your cropsa and explains a” in a note as “[p]ossibly from the labors you expend on your crops”. 49 So LaG, Augustine (labores fructuum tuorum manducabis), Sy et alii Latini.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

62

Thomas J. Kraus

ings of RQPQL and MCTRQL, so that it resembles a metonymy due to the permutation of accusative and genitive.50 The MT offers a more unambiguous text: ‘You shall eat the fruit of the labour of your hands’ (NM V[M?[RM[P[).51 But MCTRQL can also mean ‘wrist’ or, in human anatomy, ‘carpus’, both terms denoting the joint between the arm and hand which itself consists of eight bones and is attested in such a meaning since Homer (cf., for instance, Il. 24.671 and Od. 24.398; see also, e.g. Euripides, Ion 1009 G RKƒMCTRZGTQL and 891 and Hippocrates, Aër. 20 with MCTRQKƒ ZGKTYP).52 With that meaning MCTRQL would be the more uncommon but adequate rendering of #N (see LaR and the Psalterium Gallicanum with labores manuum tuarum), and RQPQL can then refer to the fruit of effort. Symmachus and Theodoret have MQRQPZGKTYPUQW, which is more similar to the MT. The manuscripts S and A and Origen and Cyril of Alexandria tried to ameliorate the difficult phrasing in the LXX and rearranged the text as VQWƒL MCTRQWƒL VYP RQPYP UQW ‘the crops [or fruit] of your labour’.53 Nevertheless, if the meaning ‘wrist’, ‘carpus’ for MCTRQL is accepted, this might be another instance of Homerism in the Septuagint.54 3.2 ‘Lesefrüchte’ that are deliberate interpretation of a ‘Vorlage’ We do not know the exact ‘Vorlage’ the Septuagint translators used for their translation into Greek, though there are many verbatim agreements so that in a considerable number of passages the LXX resembles a literal translation.55 Nonetheless, a translator dared to cut his own path, especially in

50 51

Cf. Lausberg, Handbuch der literarischen Rhetorik, § 568.3. So RSV and NRSV. See also KJV “For thou shalt eat the labour of thine hands” (similarly

ASV).

52 Cf. W. Pape, Griechisch-deutsches Handwörterbuch. Vol. 1: A-K (Braunschweig: Friedrich Vieweg & Sohn, 31914), s.v. MCTRQL 2); H.G. Liddell/R. Scott, A Greek-English Lexicon (rev. and augmented by H.S. Jones; with a revised supplement 1996; Oxford: Clarendon Press, 91940 [1996]), s.v. MCTRQL (B), and Supplement, s.v. MCTRQL. 53 Cf. R. Brucker, “Schritte auf dem Weg zu einer dokumentierten Übersetzung der Septuaginta. Ein Werkstattbericht am Beispiel des Psalters”, in S. Kreuzer/J.P. Lesch (ed.), Im Brennpunkt: Die Septuaginta. Studien zur Entstehung und Bedeutung der Griechischen Bibel 2 (BWANT 161; Stuttgart: Kohlhammer, 2004), 247–258, p. 252 n. 15. Further see the detailed references in A. Rahlfs, Psalmi cum Odis (Septuaginta. Vetus Testamentum Graecum, Auctoritate Academiae Scientiarum Gottingensis 10; Göttingen: Vandenhoeck & Ruprecht, 31979), apparatus ad loc. 54 For more information and a discussion see K. Usener, “Griechisches im Griechisch der LXX”, in the present volume. 55 Cf. E. Bons, “Psalmoi: Das Buch der Psalmen. Einleitung”, in Kraus/Karrer, Septuaginta Deutsch, 751.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Translating the Septuagint Psalms – some ‘Lesefrüchte’

63

relation to word choice; and by leaving the trail of simply translating a text from one language into another a translator could put his own mark on it.56 Psalm 30 [31MT] & 90 [91MT] As argued above for Psalm 90LXX, a translator could essentially influence the theological tendency of a text by his choice of words. Psalm 90 has ‘hope’ (G NRKL) and ‘to hope’ (G NRK\Y) as central terms in V.2, 4, 9, 14, which form the framework of the whole text. Basically, this is nothing dramatically uncommon, because the LXX Psalter has G NRK\Y where the MT has ZD (so Ps 90:2) ‘to trust’, JUZ ‘to seek refuge’ (V.4), NZ[ ‘to await someone/something’ and JYS ‘to wait’, and for Psalm 90 SZ‘to long for’ (V.14). The substantives derived from these Hebrew verbs usually correspond with G NRKL in the Greek.57 Hence, the LXX version of Psalm 90 can be described as a ‘psalm of hope sui generis’ in comparison to Psalm 91MT that is more a ‘psalm of trust sui generis’.58 But on closer examination LXX Psalm 90 is not a unique phenomenon in this respect. The tendency towards ‘hope’ is even more evident in Psalm 30 [31MT]: the verb G NRK\Y is used five times for JUZ (V.2.20), ZD(V.7.15), and NZ[ (V.25).59 Accordingly, the whole Psalm is framed by G NRK\Y and the idea of ‘hope’. In addition, the special ‘hope for life after death’ is a very prominent feature in the Greek, a conception that is unknown in the MT-Psalm.60 It is no surprise, therefore, that the topic ‘hope for life after death’ was integrated in the course of the Christian and Jewish history of reception of Psalm 30 in various ways.61

56

For examples of nonconformities on the level of word choice see Bons, ‘Psalmoi’, 751–

752.

I thank Eberhard Bons for his comments on Psalm 4:6LXX, in Kraus/Karrer, Septuaginta Deutsch. Ergänzungsband: Erläuterungen (forthcoming). Further see C. Spicq, Notes de Lexicographie néo-testamentaire. Vol. I (OBO 22.1; Fribourg: Editions Universitaires Fribourg, Göttingen: Vandenhoeck & ruprecht, 1972), 501; T.F. Williams, “Towards a Date for the Old Greek Psalter”, in: J.V. Hiebert/C.E. Cox/P.J. Gentry (ed.), The Old Greek Psalter. Studies in Honour of Albert Pietersma (JSOTS 332; Sheffield: Sheffield Academic Press, 2001) 248–276, p. 260. 58 Cf. Hugger, Jahwe meine Zuflucht, 301; Hossfeld/Zenger, Psalmen, 619–620, 625 (= Hossfeld/Zenger, Psalms 2, 429, 432). 59 For a detailed survey of the particularities of the LXX version see Cf. E. Bons, Psalm 31. Rettung als Paradigma. Eine synchron-leserorientierte Analyse (FTS 48; Frankfurt a.M.: Knecht, 1995) 265–270. 60 Cf. E. Bons, “ELPIS – L’espérance de l’au-delà dans la littérature juive hellénistique”, in R. Kuntzmann (ed.), Ce Dieu qui vient. Études sur l'Ancien et le Nouveau Testament offertes au Professeur Bernard Renaud à l'occasion de son soixante-cinquième anniversaire (LD 159; Paris: Le Cerf, 1995) 345–370. 61 Cf. Bons, Psalm 31, 272–274. 57

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

64

Thomas J. Kraus

Psalm 86:5 [87:5MT] Ps 86:5 5 Mutter Sion wird ein Mensch sagen,62 und ein Mensch wurde in ihr geboren, und der Höchste selbst hat ihr Fundament gelegt. [Mother Sion, a person will say and a person was born in it, and the Most High himself founded it.63]

It is rather striking that the LXX has OJVJT where the MT simply links the sentence with Y as _Y[ENY ‘and to Sion’. Attempts to dismiss OJVJT as a secondary reading are not plausible so that the opening of V.5 with 0JVJT 6KYP G TGK¨ C¢PSTYRQL has to be explained as given.64 Moreover, G TGK is active (‘will say’) while the nifal imperfect of TO in the MT has a passive meaning (‘will be said’). Even if the MT implies that the place that brings forth human beings is a female one (see OJVJT in the LXX then),65 this does not help to answer the question: who is this anonymous person speaking in the LXX?66 And how does it come about that V.5a and V.5b in the LXX differ from the Hebrew equivalent in the MT? After the nifal imperfect, there follows JDFN[[ Y[ ‘person for person is born in her’, that is to be understood in a distributive sense as ‘every single person is born in her’.67 It appears plausible that the translator did not take [ Y[ as an expression for ‘everybody’ or ‘every single person’, but [ as the subject of TO [, for which an active meaning was accepted, and [ Y as the beginning of a new statement. V.5a of the MT (‘To Sion, a person will say’) then requires an object in order to answer the question: what will a person say? Therefore, the translator adds OJVJT that is in accordance with the context and the Old Testament tradition68 and does not explain ‘Sion’ as the mother of an individual person or of the Messiah but as the “Mutterstadt fremder

62

Or “Ein Mensch wird Sion Mutter nennen” – “a person will call Sion Mother”. Pietersma, The Psalms, 590: “cWith regard to Sion, surely noc person will say | ‘dA persond was born in it?’ | And the Most High himself foundet it.” Notes: cMother Sion, a = Ra; dPerhaps so-and-so.” 64 For reasons against such attempts see C.M. Maier, “ ‘Zion wird man Mutter nennen.’ Die Zionstradition in Psalm 87 und ihre Rezeption in der Septuaginta”, ZAW 118 (2006) 582–596, pp. 592–594. 65 Cf. Siegert, Zwischen Hebräischer Bibel und Altem Testament, 314. 66 Siegert, Zwischen Hebräischer Bibel und Altem Testament, 314, suggests “ein Heilsbringer wie in Num 24,7”. 67 Cf. GK § 123c; Maier, “Zion”, 584. 68 With Maier, “Zion”, 595–596, against J. Schaper, Eschatology in the Greek Psalter (WUNT 2.76; Tübingen: Mohr Siebeck, 1995), 99–101, who regards the LXX version as a new interpretation of its ‘Vorlage’ focussing on a messianic understanding of C¥PSTYRQL. 63

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Translating the Septuagint Psalms – some ‘Lesefrüchte’

65

Völker”69 (‘mother town of foreign peoples’); and this understanding is supported by the correlating demonstrative pronoun QWŸVQK (plural!) in V.4c.70 3.3 ‘Beware of being professionally blinkered!’ – a very special ‘Lesefrucht’ The work on the Psalms for the Septuaginta Deutsch translation project took years, and the various meetings and discussions helped to produce, improve, and refine astounding phenomena while comparing the Masora and the Septuagint with each other. One of these is presented in the following and should serve as a warning against an eagerness, blinded by routine, to accept findings without a second thought. In other words, such an attitude could end up in a methodological pitfall. Psalm 89:10 [90:10MT] Ps 89:10 10 Die Tage unserer Jahre, in ihnen sind siebzig Jahre, wenn (man) aber kraftvoll (ist), achtzig Jahre, und das meiste von ihnen ist Mühsal und Anstrengung; denn Sanftheit ist auf uns gekommen, und wir werden unterwiesen werden. [The days of our years—in them are seventy years, but eighty years, if we remain in control; and the greater part of them is toil and trouble; for meekness came upon us, and we shall become disciplined.]

This verse can be found in Jub 23:13 (V.10c)71 and 23:15 (V.10ab).72 In Sir 18:9–10 there is a similar phrasing, but the maximum of days is given as ‘a hundred years’ (C TKSOQƒLJOGTYPC PSTYRQWRQNNCƒ G¢VJG˜MCVQP). To my 69

Maier, “Zion”, 596. V.5b “und ein Mensch wurde in ihr geboren” (“and a person was born in it”) may also serve as a reason for V.5a in the sense of “und zwar der Mensch, der in ihr geboren wurde” (‘and the person, who was born in it’). So Maier, “Zion”, 595. But such an understanding depends on the epexegetic use of MCK at the beginning of V.5b (cf. Blass/Debrunner/Rehkopf, Grammatik, § 442.6a) and this is not certain in the given context. 71 “And in those days if a man will live a jubilee and a half, they will say about him, ‘He prolonged his life, but the majority of his days were suffering and anxiety and affliction.’” See the Hebrew fragment 3Q5 frg. 1 that starts at “they will say …”. English translation according to O.S. Wintermute, “Jubilees (Second Century B.C.): A New Translation and Introduction”, in J.H. Charlesworth (ed.), The Old Testament Pseudepigrapha. Vol. 2 (The Anchor Bible Reference Library; New York et al.: Doubleday, 1985) 35–142, p. 100. 72 “Then they will say, ‘The days of the ancients were as many as one thousand years and good. But behold, (as for) the days of our lives, if a man should extend his life seventy years or if he is strong (for) eighty years, then these are evil. And there is no peace in the days of this evil generation.’ ” Translation according to Wintermute, “Jubilees”, 100. 70

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

66

Thomas J. Kraus

knowledge the ‘seventy years’ (GDFQOJMQPVC G¢VJ) occurs only once in Greek literature, namely in Herodotus 1.32,73 apart from Christian quotations of this verse. Solon calculates for King Croesus the length of his life, which is limited to ‘seventy years’ (G LICƒTGDFQOJMQPVCG¢VGCQW¤TQPVJL \QJL C PSTYR RTQVKSJOK) and thereafter converted into single days.74 Solon explains to Croesus that the question about a really successful life can only be answered at its very end. Besides, for Solon the desirable length of life is ‘eighty years’ (Herodotus 3.22: Q IFYMQPVCG¢VGC\QJLRNJTYOC C PFTKƒ OCTMQVCVQPRTQMGKUSCK; see Solon himself in frg. 20 – Q IFYMQP VCGVJ OQK¨TC MKZQK SCPCVQW – in his answer to Mimnermos [= Ligiastades] who calculated ‘sixty years’).75 These are fascinating details and, at first glance, the Septuagint has integrated an idea that is, as far as I know, only present in Herodotus. However, so far the Hebrew has not been considered [90:10MT]: JY][PYOVTYDID] YJP][D]JDYP[VYP[O[

The days of our years [are] seventy and if [we are] strong eighty years …76

Hence, both ideas — the length of life as ‘seventy years’ and the maximum of ‘eighty years’ under the circumstance of ‘strength’ — are already present in the MT version of the Psalm. So the translator might just have rendered the Hebrew ‘Vorlage’ into Greek without any deliberate reference to Herodotus. It would be wrong to ascribe automatically the underlying conception in the Greek to Herodotus without any critical reflection on the Hebrew text(s) of this Psalm. Nevertheless, the overlapping of the same conception between MT and LXX on the one hand and Herodotus on the other remains an interesting facet that must be pointed out. V.10c then has VQƒRNGK¨QP ‘the greater part’ (“das meiste”) while the MT has DJT ‘furiousness’, ‘pride’, ‘arrogance’ or the like, possibly due to reading ]DT (instead of ]DJT of the MT). This difference seems to be explicable, but the next ones in V.10d hardly are: LXX and MT differ absolutely from each other there. While the MT has something like ‘for/indeed, it flees away swiftly, and we fly away (like a bird)’,77 the LXX reads ‘for meekness came upon us, and we shall become disciplined’. 73 I thank my colleague in the project Ralph Brucker, Hamburg, for his precise observations and the reference to Herodotus. 74 Cf. Herodotos. Vol. 1.1 (ed. H. Stein; Berlin: Weidmannsche Buchhandlung, 31870) ad loc. (= 1.32.9–10). 75 Cf. Herodotos 1.1 [Stein], pp. 39–40 n. 10 (= 1.32.9–10) and Herodotos. Vol. 1.3–4 (ed. H. Stein; Berlin: Weidmannsche Buchhandlung, 21868) p. 25 n. 18 (= 3.22.18). 76 See KJV: “The days of our years are threescore years and ten; and if by reason of strength they be fourscore years …” (similarly ASV, RSV, NRSV, and NKJV). 77 Translation according to Hossfeld/Zenger, Psalms 2, 416. Further see their notes on this verse on pp. 417–418.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Translating the Septuagint Psalms – some ‘Lesefrüchte’

67

All in all, Psalm 89LXX offers “numerous differences in detail from the Hebrew text” and “the LXX sought to resolve the textually difficult passages through use of its own ideas.”78 Consequently, this Psalm could also serve as further evidence for a deliberate interpretation of a ‘Vorlage’ (see 3.2 above). In detail, it differs rather dramatically (see V.10d) from the MT so that the issue of its ‘Vorlage’ would also be an interesting point of study.

4. Conclusions and prospects The ‘Lesefrüchte’ selected for this study are only a small extract from all the observations my colleagues and I have made during the work on my own Psalms (and those of others) and, of course, there are many more for all the other Psalms. Accordingly, the present study is no more than a report on work in progress, a ‘Werkstattbericht’. In addition, the LXX version by and large agrees with the Hebrew Bible. Unfortunately, we do not have the ‘Vorlage’ any more, so that the Masoretic Text, which conserves the Hebrew tradition, is a very important point of comparison. If we take the few ‘Lesefrüchte’ presented above and compare them with the agreements between LXX and MT and then calculate a percentage of differences and agreements, we may come to the conclusion that the LXX is a close if not verbatim translation of a Hebrew ‘Vorlage’, which only differs here and there from it. This is a challenge and task at the same time: more ‘Lesefrüchte’ must be collected, analysed, and structured in order to paint a more reliable and systematic picture of the Septuagint Psalms between textual fidelity on the one and sensitive assimilation and reshaping on the other. However, the ‘Lesefrüchte’ presented in this study show that (a)

there are striking differences between the Greek LXX and the Hebrew MT. Of course, some of them may be explained very well by pointing to alternative vocalisation and readings of the Hebrew text, others by referring to an unknown and/or lost ‘Vorlage’.

(b)

differences mostly have to do with word choice, rhyme, rhythm, sound and memorizing effects so that at least some of them may be essential stylistic alterations and/or harmonisations in relation to the nature of the Greek language.

(c)

some of the differences in the LXX produce alterations in meaning, message, and theology (above all, see 3.2 and 3.3). Whether these

78

Hossfeld/Zenger, Psalms 2, 425.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

68

Thomas J. Kraus

were meant the way we understand them today cannot be decided without ambiguity. (d)

there are peculiarities in the LXX for which the common explanations – vocalisation, (mis)reading of the given Hebrew text, unknown ‘Vorlage’, and/or adapting the meaning to another cultural context – do not fully explain (see 3.2 and 3.3).

Consequently, even after a critical analysis of the ‘Lesefrüchte’ offered here and a tendency to be hesitant about accepting the instances listed under 3.1 as a translator’s deliberate decisions and interpretations, the examples in 3.2 and 3.3 are plausibly attributed to a translator’s creative and theological will. In addition, I am convinced that a knee-jerk reference to an unknown and/or lost ‘Vorlage’ is not more than an argumentum ex silentio, and it does not adequately account for the various and varied differences between the Hebrew MT and the Greek LXX selected for the present study. Thus, it seems more appropriate to regard the ‘the Greek translation (the Septuagint) of the Hebrew Bible as its oldest commentary’.79 The process of translating a given text from one language into another always involves a large number of decisions the translator has to make. Hence, translators slip into the roles of interpreters who read and render a text against their own social, cultural, and religious background which in turn influences and shapes the translation/interpretation these translators/interpreters produce.

79 Siegert, Zwischen Hebräischer Bibel und Altem Testament, 122: “Die griechische Übersetzung der Hebräischen Bibel ist in gewisser Weise ihr ältester Kommentar.”

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Eberhard Bons

Rhetorical Devices in the Septuagint Psalter* I. Introduction Usually the LXX Psalter is considered to be a very literal translation of a Hebrew Vorlage very close to the Masoretic consonantal text. Indeed, a thorough comparison of the consonantal text preserved in the MT and the LXX shows that the word order and the syntax of these two Psalters versions diverge only slightly.1 It might be deduced from this comparison that the translator as a general rule confined himself to rendering his Hebrew source text word for word.2 This conclusion implies that the traces of the theological and cultural knowledge the translator has left in the Greek text are only of minor importance. However, a more detailed analysis of the LXX shows that the Greek translator was much more creative than is usually supposed.3 A number of rather free translations indicate that he did not shrink from interpreting his Hebrew Vorlage, e.g. when he had to translate difficult or rare words or expressions. Elsewhere the translator was not afraid to correct the Psalter text.4 This might be the case when he found himself confronted with passages that appeared to him theologically unacceptable, e.g. verses that dealt with an alleged plurality of gods (e.g. Ps 8:6; 97:7; 138:1). Nonetheless, this argument cannot explain all the subtleties of * I wish to express my gratitude to my colleagues Jennifer Dines (Cambridge) and Jan Joosten (Strasbourg) for reading and commenting on a first draft of this paper. 1 F. Austermann, Von der Tora zum Nomos. Untersuchungen zur Übersetzungsweise und Interpretation im Septuaginta-Psalter (MSU 27; Göttingen: Vandenhoeck & Ruprecht, 2003), 42– 106, provides plenty of material in order to prove the accuracy of this hypothesis. 2 See e.g. A. Pietersma, “Exegesis in the Septuagint: Possibilities and Limits (The Psalter as a Case in Point)”, in W. Kraus/R. Glenn Wooden (ed.), Septuagint Research. Issues and Challenges in the Study of the Greek Jewish Scriptures (SCSt 53; Atlanta: Society of Biblical Literature, 2006), 33–45, on pp. 38–9. 3 For this opinion, see e.g. F. Siegert, Zwischen Hebräischer Bibel und Altem Testament. Eine Einführung in die Septuaginta (Münster: Lit, 2001), 311: “Der Psalter ist reich an behutsamen ‘Interpretamenten’, die jedoch nie den Charakter von Glossen oder Erweiterungen haben, sondern den einer interpretierenden Wortwahl.” 4 Several examples are quoted in two earlier articles: E. Bons, “Die Rede von Gott in den PsalmenLXX”, in H.-J. Fabry/D. Böhler (ed.), Im Brennpunkt: Die Septuaginta. Band 3: Studien zur Theologie, Anthropologie, Ekklesiologie, Eschatologie und Liturgie der Griechischen Bibel (BWANT 174; Stuttgart: Kohlhammer, 2007), 182–202, on pp. 185–99; id., “Der SeptuagintaPsalter – Übersetzung, Interpretation, Korrektur”, in M. Karrer/W. Kraus (ed.), Die Septuaginta. Texte – Kontexte – Lebenswelten (WUNT 219; Tübingen: Mohr Siebeck, 2008), 450–70, on pp. 464–468.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

70

Eberhard Bons

the Greek text, which is, in my opinion, much more than a mere word for word translation. In this paper, I would like to raise the question of what kind of rhetorical skill can be recognized in the Septuagint Psalter. Obviously, the passages where the Greek translator does not go beyond a literal translation of a Hebrew rhetorical device do not prove anything. Therefore, one has to single out verses or cola whose rhetorical features cannot be explained by the extant Hebrew Psalter text including the Qumran manuscripts. In recent studies, this problem has not attracted much attention, so that careful and detailed studies on this subject are still missing. In order to fill this gap at least partially, in this paper I shall focus on word order as well as the choice of Greek words in the Greek Psalter text.

II. Some examples of hyperbaton, alliteration and paronomasia 1. Hyperbaton In a small number of cases, words belonging together are separated by at least one other word although the extant Hebrew Psalm texts (Qumran fragments, MT) do not reveal an analogous word order. Let me quote two examples: Ps 34(33):13b:

][[ZLRZJ[ J[O VKLG™UVKPC¢PSTYRQLQ˜SGNYP\YJƒP DYVY TN][O[DJ C ICRYPJOGTCLK™FGK¨PC ICSCL “What man is there who desires life, and covets many days, that he may enjoy good?” (NRSV)

“What man is there that desires life, loving to see good days?” (Brenton)

It is obvious that the translator completed the Greek word JOGTCL by the attribute C ICSCL whereas the Hebrew text considers the corresponding word DYas the direct object of the infinitif VY TN. This syntactical structure would lead to the Greek translation *K™FGK¨PC ICSQP (cf. Job 7:7; Eccl 3:13; PsSal 18:6 [C ICSC]). However, this adjectival form nowhere appears in the extant Greek Psalter manuscripts. Conserving the word order instead of the syntactical structure of the second colon, the Greek text displays a hyperbaton. Apparently, this word order has bothered some of the ancient copyists of the Septuagint text who rearranged the line in the following way: C ICRYP K™FGK¨P J˜OGTCL C ICSCL (so e.g. codex B). Just like this textual variant, the NT quotation of the Psalm verse, 1 Pet 3:10, reads K™FGK¨P

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Rhetorical Devices in the Greek Psalter

71

JOGTCL C ICSCL.5 Nevertheless, it should be stressed that the word order JOGTCL K™FGK¨P C ICSCL is in no way unusual, the hyperbaton is even re-

commended when the verb does not finish the clause (Quintilian, Inst. IX, 4, 26).6 After all, one cannot say that the expression JOGTCK C ICSCK does not make sense. On the contrary, it is not completely unknown to biblical wisdom literature. It occurs e.g. in the Book of Ben Sira where the sage warns his disciple not to deprive himself of a happy day (Sir 14:14). Ps 37(36):16:

S[FENODY MTGK¨UUQPQ NKIQPVFKMCKW˜RGƒT ][DT][T_YOJO RNQWVQPCOCTVYNYPRQNWP7 “Better is a little that the righteous person has than the abundance of many wicked” (NRSV)

“A little is better to the righteous than abundant wealth of sinners” (Brenton)

Whether the Hebrew Vorlage of the LXX read ][DT or another word8, it is sure that in the LXX text RQNWP characterizes RNQWVQP. However, the LXX does not change the word order of the Hebrew text. It only translates its last word differently, making it the exact semantic counterpart of Q NKIQP in the first colon of the verse9 insofar as each of the two adjectives respectively refers to the property of the righteous and the sinners. In so doing, the LXX underscores the opposition between the modest possessions of the righteous on the one hand and the abundance of wealth of sinners on the other. In the wake of the LXX text, some modern commentators plead for an emendation 5 In comparison with the Göttingen Psalter text, the quotation shows three further changes: the infinitive C ICRCP as well as the addition of two words: ICT before SGNYP and MCK before K™FGKP. For a detailed study of the quotation, see e.g. S. Woan, “The Psalms in 1 Peter”, in S. Moyise/M.J.J. Menken (ed.), The Psalms in the New Testament (London/New York: T. & T. Clark, 2004), 213–29, on pp. 219–25. By contrast, the quotation of Ps 33:13bLXX in 1Clem 22.2 does not diverge from the text in the Göttingen edition. 6 A detailed study of the genitive hyperbaton is provided by A.M. Devine / L.D. Stephens, Discontinuous Syntax. Hyperbaton in Greek (New York / Oxford: Oxford University Press, 2000), 106–7. 7 In his Göttingen edition of the Psalter, Alfred Rahlfs does not quote any Greek variants. 8 Analogously to the MT, the Targum reads _[ [IU_[T “[of] many wicked”. In the Qumran Psalm manuscripts, the variant *DT seems not to be attested; see P.W. Flint, The Dead Sea Psalm Scrolls and the Book of Psalms (STDJ 17; Leiden: Brill, 1997), chapter IV. D. Barthélemy, Critique textuelle de l’Ancien Testament, Tome 4: Psaumes (OBO 50.4; Fribourg / Göttingen: Academic Press/Vandenhoeck & Ruprecht, 2005), 215, calls attention to 4Q171 (= 4QpPs37a), col. II, 22 where at least the letter yod of *][DT can be identified. In his Psalterium iuxta Hebraeos, Jerome evidently does not translate the Hebrew text ][DT][T_YOJO(= MT), but a text similar to the LXX: quam divitiae impiorum multae. 9 For a similar example, see Tob 12:8bBA ; slightly different is Prov 15:29 (= 16:8MT).

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

72

Eberhard Bons

of the MT (*DT ][T_YOJO)10 though this would result in an awkward word order.11 2. Alliteration created by parallel use of compound words The first two examples reveal that the translator sometimes takes advantage of a characteristic feature of Greek word construction, i.e. the use of alpha privative which has no equivalent in Hebrew.12 The third example, however, can illustrate that alliteration results from the choice of rare words or hapaxlegomena. Ps 49(48):11a:

YFD [TDYN[UMFZ[ G™RK?VQƒCW™VQƒC¢HTYPMCK¼C¢PQWL C RQNQWPVCK “fool and dolt perish together” (NRSV)

“the fool and the senseless one shall perish together” (Brenton)

The Greek language has at least one positive adjective that would have been suitable for the translation of one of the Hebrew adjectives: OYTQL. This word is quite frequent elsewhere in Biblical literature (see e.g. Deut 32:6; Sir 18:18; 19:11; Isa 19:11; 32:6; Matt 25:2,3,8). The LXX Psalter, however, employs it only once, in Ps 93:8LXX where the MT has N[UM. Instead of OYTQL, in Ps 48:11LXX the translator employs two compound adjectives having an alpha privativum and forming a triple alliteration with the following verb C RQNQWPVCK. As for C¢PQWL, this adjective is not only a rare word in the LXX but also a hapax legomenon of the Greek Psalter. Its use can be explained by the fact that TDin the sense “foolish” is quite rare and not always translated adequately (see Ps 72:22LXX: G™EQWFGPYOGPQL), whereas Ps 91:7; 93:8LXX translate the root TDby C¢HTYP. One can therefore infer that in Ps 48:11LXX the choice of the second adjective may have depended on the first one whose meaning was not doubtful. Repeated use of alpha privativum occurs elsewhere in the Greek Psalter (see Ps 91:7LXX where C PJƒT C¢HTYP  C UWPJVQL form an alliteration) though this is not a rule. 10

E.g. H.-J. Kraus, Psalmen 1–59 (BK.AT XV/1; Neukirchen-Vluyn: Neukirchener Verlag, 1989) 438; P. C. Craigie, Psalms 1–50 (WBC 19; Waco, Texas: Word Books, 1983), 296. 11 For this objection, see already W. Bühlmann / K. Scherer, Stilfiguren der Bibel. Ein kleines Nachschlagewerk (Biblische Beiträge 10; Fribourg: Schweizerisches Katholisches Bibelwerk, 1973), 47. 12 See also Siegert, Zwischen Hebräischer Bibel und Neuem Testament, 138. 6

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Rhetorical Devices in the Greek Psalter

73

Indeed the opposite can be shown by Ps 93:8LXX. Nevertheless, parallel use of alpha privativum is not at all an isolated phenomenon which can be illustrated by the following example. Ps 63(62):2d:

][O[NDX[YJ[ELT D G™PIG™TJOMCK¼C DCVMCK¼C PWFT “in a dry and weary land where there is “in a barren and trackless and dry land” (Brenno water” (NRSV) ton)

In this colon again, the translator is confronted with a relatively rare word, LXX X[, “faint, weary”, which he will translate by C¢PWFTQL in Ps 142:6 . In Ps 62:2d, however, ][O[ND demanded a similar Greek equivalent. Without translating literally by QW™+WŸFYT (see for an analogous case Ps 106:40LXX), the LXX has a compound adjective, C¢PWFTQL.13 As for X[, the use of C¢PW FTQL suggested translating it by a morphologically similar word. This word is certainly not an adequate equivalent of X[. Nevertheless, by inserting C¢DCVQL the translator not only creates an alliteration but also a series of three parallel adjectives defining IJ and ending in the vowel . Needless to say, these phenomena are absent from the Hebrew text which is more heterogeneous on both the syntactical and the morphological level. Ps 51(50):12:

[DTSDFZ_YMPZYTY MCK?RPGWOCGW™SG?LG™IMCKPKUQPG™P VQK¨LG™IMCVQKLOQW

“and put a new and right spirit within me” (NRSV).

“and renew a right spirit in my inward parts” (Brenton)

Both the verb G™IMCKPK\Y and the noun VCƒ G¢IMCVC are rare words in the Greek Psalter. The verb is even a Psalter hapax legomenon. Whereas the Hebrew verb FZ is translated by C PCMCKPK\Y “renew, restore” in Ps 102:5; 103:30LXX, the verb preferred by the translator, G™IMCKPK\Y, might have the connotation of restoration and inauguration of a new era.14 It is beyond doubt that this connotation is perfectly fitting in the specific context of Ps 13 Sometimes an adjective with alpha privativum corresponds to a relative clause, see the examples quoted by E. Tov, “Compound Words in the Septuagint Representing two or more Hebrew Words”, in id., The Greek and Hebrew Bible. Collected Essays on the Septuagint (VT.S 72; Leiden: Brill, 1999) 131–52, on pp. 133–146. 14 For further details see C. Spicq, Lexique théologique du Nouveau Testament (Fribourg / Paris: Éditions universitaires/Cerf, 1991), 407–8.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

74

Eberhard Bons

50:12 whose first colon employs the verbMVK\Y in the sense of “create”15: MCTFKžCPMCSCTCƒPMVKžUQPG™PG™OQKQSGQL. As for G™PVQK¨LG™IMCVQKLOQW as Greek equivalent of [DTSD, one should call to attention the fact that VCƒ LXX G¢IMCVC as the rendering of DTS is only attested in Ps 108:18 . Instead, the translator disposes of a series of alternative renderings: G™PVQLOQW (Ps 38:4; 108:22LXX), G PG OQK (Ps 54:5LXX), G™PVXMCTFKOQW(Ps 93:19LXX). However, he had to avoid G PG OQK because this expression already occurs in v. 12a. The same applies for the noun MCTFKžC. Therefore the choice of G™PVQKL G™IMCVQKLOQW is probably best explained by the need for a noun to give to the end of the verse a certain weight, on the semantic level and on the level of sound. 3. Paronomasia In literature dealing with rhetoric, this technical term has various significations of which one is particularly important for the analysis of the LXX Psalter: the repetition of words deriving from the same root and being employed close together16 (so called annominatio consisting in an “organic” change of the word17). In some cases of paronomasia in the Greek Psalter the underlying Hebrew text displays analogous phenomena which the translator seeks to render as literally as possible (e. g. Ps 30:21a: MCVCMTW[GKL CW™VQW?L G™P C RQMTWH VQW RTQUYRQW UQW [MT: ?[PRTVUD]T[VUV]). On the other hand, this type of rendering can be observed as well in several cases where the Hebrew Psalter text offers words of different roots (e.g. Ps 50:8b: VC? C¢FJNC MCK¼ VCƒ MTWHKC VJL UQHKžCL UQW G™FJNYUCL OQK – Ps 83:4b: MCK¼ VTWIYƒP PQUUKC?P G˜CWVX QW¡ SJUGK VCƒ PQUUKžC CW™VJL – Ps 91:7b: C UWPGVQL QW™ UWPJUGK VCWVC). These few examples permit us to differentiate between two categories: paronomasia between two nouns and paronomasia between a noun and a verbal form. Let us illustrate this phenomenon by singling out some other striking examples:

15

In the Biblical literature, this verb undergoes a semantic change, see E. Bons, “Le verbe comme terme technique de la création dans la Septante et dans le Nouveau Testament”, in J. Joosten/P.J. Tomson (ed.), Voces biblicae (Contributions to Biblical Exegesis and Theology 49; Leuven: Peeters, 2007), 1–15. 16 Cf. F. Blass/A. Debrunner, Grammatik des neutestamentlichen Griechisch (Göttingen: Vandenhoeck & Ruprecht, 161984), § 488,1. 17 See H. Lausberg, Elemente der literarischen Rhetorik (Munich: Hueber, 101990) § 279; id., Handbuch der literarischen Rhetorik. Eine Grundlegung (Stuttgart: Franz Steiner, 31990), § 638, 3a. MVK\Y

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Rhetorical Devices in the Greek Psalter

75

a) Paronomasia between two nouns (substantive – substantive or substantive – adjective) Ps 18(17):16b:

JYJ[?VTIO C RQƒG™RKVKOJUGYLUQWMWTKG ?R ZYTVOPO C RQƒG™ORPGWUGYLRPGWOCVQLQ™TIJL UQW

“at your rebuke, O Lord, at the blast of the breath of your nostrils” (NRSV)

“at your rebuke, O Lord, at the blast of the breath of your wrath” (NETS)

Unlike the parallel text 2 Kings 22:16 where the LXX uses the word RPQJ (C RQƒ RPQJL RPGWOCVQL SWOQW CW™VQW), the Psalter translator opts for inserting another noun, G¢ORPGWUKL. This word is not only a LXX hapax legomenon but is also considered a neologism of the LXX. By so doing, the translator achieves two effects: firstly, he creates the repetition of the syllable RPGW in two following words; secondly, he reinforces the parallelism between the two parts of the colon in matching the equivalent of JOP to the parallel noun in the first part: G™RKVKOJUGYL – G™ORPGWUGYL. No need to say that each of the two nouns is a compound, that they almost have the same length and, finally, the same ending. Ps 45(44):2c:

T[JOTRYU[PYN JINYUUCOQWMCNCOQLITCOOCVGYL Q EWITCHQW “my tongue is like the pen of a ready scribe” (NRSV)

“my tongue is the pen of a quick writer” (Brenton)

The expression T[JOTRYU occurs only once more in the Hebrew Bible, in Ezra 7:6 where the LXX reads ITCOOCVGWƒL VCZWL. Avoiding the rather literal equivalent VCZWL or another usual adjective or participle, e.g. URGW FYP (see Isa 16:5), the Greek Psalter translator prefers a more elegant reading. By the choice of Q EWITCHQL he not only creates a paronomasia between the adjective and the preceding noun, but also expresses the writer’s specific dexterity.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

76 b)

Eberhard Bons

Paronomasia between a verb and noun

Ps 18(17):27:

TTDVVTDP] MCK?OGVCƒG™MNGMVQWG™MNGMVQƒLG¢UX NVRVVS] MCK?OGVCƒUVTGDNQWFKCUVTG[GKL [2 Kings 22:27: UVTGDNYSJU] “with the pure you show yourself pure, “and with the select you will be select, and with the crooked you show your- and with the crooked you will pervert” (NETS) self perverse” (NRSV)

The Greek translation of v. 27a might be considered incorrect, especially the choice of G™MNGMVQL for the stem TTD (used here in the niph‘al, “to purify oneself”, as well as in the hitpa‘el “to show oneself pure”). The translator probably read a word like T[ZD(see e.g. Ps 88:4aLXX) or TYZD (see e.g. Ps 77:31bLXX). Nevertheless, it is clear that the first colon of v. 27 continues the pattern of the preceding v. 26 in so far as the adjective in the genitive introduced by the preposition OGVC stems from the same root as the subsequent verb or predicative adjective. As for the second colon of v. 27, it is obvious that the translator sought to adapt it to the first one in matching the adjective and the following verb to one another. It deserves attention that the Hebrew text of v. 27b does not show any similar construction, the corresponding adjective and verb having different roots: the adjective S“crooked” as well as the verb VRV hitpa‘el “to show oneself perverse”. The latter is a Psalter hapax legomenon whereas Soccurs only once more in the Psalter: Ps 101:4 reads SDDN “a false heart”, an expression the LXX renders by MCTFKCUMCODJ‚“a perverse heart”. Ps 17(16):8:

[PT[VUV?[RPMNED G™PUMGRXVYPRVGTWIYPUQW UMGRCUGKLOG “hide me in the shadow of your wings” “you will shelter me with the shelter of your (NRSV) wings” (NETS)

As results from other Psalm passages, the translator is not at all unaware of the ordinary meaning of the verb TVU hiph. which he renders several times by MTWRVY or C RQMTWRVY (Ps 18:7; 37:9; 68:5; 118:19LXX). However, in Ps 16:8 he deviates from this rule opting for the verb UMGRC\QOCK, that is to say the verb fitting perfectly to the adverb of place G PUMGRX. As the next

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Rhetorical Devices in the Greek Psalter

77

example will illustrate, this phenomenon is not limited to only one Psalm quotation. Ps 61(60):5:

?[RPMTVUDJUZ  UMGRCUSJUQOCKG™PUMGRXVYP RVGTWIYPUQW “I will find refuge in the shelter of your wings” (cf. NRSV)

“I will find shelter under the shelter of your wings” (cf. Pietersma)

Although the Greek Psalter equivalent of the Hebrew verb JUZ “to seek refuge” is usually G™NRK\Y (see Ps 5:12; 7:2 etc.), the translator follows the model of Ps 16:8LXX instead of translating the verb “mechanically” by one and the same equivalent. Certainly, as for the image of the Psalmist’s hiding under God’s shelter, the translator does not proceed systematically, as one can deduce from the two quoted texts. In fact, in texts like Ps 26:5b; 35:8bLXX one does not find expressions consisting of a verb and an adverb of place deriving from the same Greek root. Such passages demand a detailed analysis of the respective contexts that cannot be carried out in this paper. Nevertheless, the tendency of creating paronomastic formulations between a verb and prepositional phrase is observable elsewhere in the LXX Psalter as the following examples can show. Ps 149:3b:

YNYTO\[TYPMYXVD G™PVWORCPMCK?[CNVJTK [CNCVYUCPCW™V “Let them sing praises to him with tambourine and lyre” (NRSV)

“Let them sing praises to him with drum and harp”

In the Greek Psalter, the noun TYPMis translated by two different Greek words: MKSCTC (Ps 32:2a etc.) and [CNVJTKQP (Ps 32:2b etc.). Furthermore, the standard Greek translation of the Hebrew verb TO\ is [CNNY, at least in the Psalter (Ps 7:18 etc.). Therefore, it is possible that this verb has influenced the choice of the preceding noun whereas elsewhere this might not be the case (e.g. Ps 70:22bLXX). A particular use of paronomasia can be noticed in comparisons introduced by YUGKž. In the Hebrew Psalter, the noun following the preposition of comparison as well as the corresponding verb might derive from the same root. According to Ps 103:15, man flourishes like a flower of the field (L[E[_MJFJL[EM). As expected, the LXX follows the Hebrew style on this

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

78

Eberhard Bons

point: YUGK? C¢PSQLVQWC ITQWQWŸVYLG™ECPSJUGK. Elsewhere however, the Greek Psalter stresses this kind of relationship against the Hebrew. Let us quote three examples: Ps 72(71):6b:

LT X[\T\][D[DTM YUGK?UVCIQPGLUVC\QWUCK G™RK?VJƒPIJP “like showers that water [?] the earth”

“like drops dripping on the earth” (NETS)

Ps 101(102):4b:

YTZPFSYOM18[VYOEY MCK?VCƒQ™UVCOQWYUGK?HTWIKQP UWPGHTWIJUCP “and my bones burn like a furnace” (NRSV)

“and my bones were burnt up like firewood“ (NETS)

Ps 108(109):19b:

JTIZ[F[OVZ\ONY YUGK¼\YPJJ PFKCƒRCPVQƒL RGTK\YPPWVCK “like a belt that he wears every day” (NRSV)

“and as a girdle with which he girds himself continually” (Brenton)

A closer look at the three texts reveals that each of them contains a rare word. To begin with, in Ps 72:6b the word represents X[\T\one of the numerous difficult words of the Psalter whose sense and function are far from clear.19 However, the translator knows ][D[DTwhich he had rendered by LXX UVCIQPGLin Ps 64:11 . So it seems not too farfetched to fit in a word not only expressing the effect caused by the drops but also deriving from the same Greek root: UVC\Y “drip”. As for GTZP in Ps 102:4, this word is a Psalter hapax legomenon as well. Rendering the verb TTZ“burn” by UWOHTWIY (see also Job 30:30A; Ezek 24:10A) the translator employs a noun deriving from the same root,

18 Against the BHS text (FSYOM), one should follow the evidence of the Hebrew manuscripts reading only one word. See Barthélemy, Critique textuelle, 221–2. 19 W. Gesenius, Hebräisches und Aramäisches Wörterbuch über das Alte Testament (ed. H. Donner) (Heidelberg: Springer, 181995), 312: “Regenguß”.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Rhetorical Devices in the Greek Psalter

79

HTWIKQP, in order to illustrate the manner the bones of the Psalmist are

consumed as by fire. Finally, the noun Z\O“belt” in Ps 109:19b is also a Psalter hapax legomenon. The translator surely knows the verb TIZ “to gird” as well as its figurative sense which he renders by RGTK\Y‚PPWOK (Ps 44:4; 64:13). Since the parallel colon Ps 109:19a mentions the curse the wicked should put on like a garment (K˜OCVKQP) it is possible that the translator inserted in the second line the noun \Y‚PJ “belt, girdle”, i.e. the piece of clothing fitting to the verb RGTK\Y‚PPWOK.

III. Concluding Remarks The purpose of this short article is to give an idea of the existence and diffusion of rhetorical devices in the Greek Psalter that cannot be explained by a slavish translation of a Hebrew Vorlage. Of course, one should be aware of the fact that the study of these examples can be deepened in several ways: Firstly, it is necessary to take into consideration the immediate context of the passages quoted. Secondly, it would be worthwhile to extend this study to the Greek Psalter as a whole in order to show that the examples presented above are not exceptions. As for the method of study, one conclusion can already be drawn: We have to change our perspective. That means to say that instead of wondering which element of the Hebrew text is reproduced in which manner in the Greek text we have to focus on the Greek text itself. Then the question will be: Which rhetorical devices of the Greek text have no model in the extant Hebrew Psalter texts? If we could answer that question we would gain more insight not only into the translation technique underlying the Greek Psalter, but also into the ideas the translator had of his own task.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Knut Usener

Griechisches im Griechisch der LXX 1. Einleitung „Griechisches im Griechisch der LXX“ impliziert, dass es auch „Ungriechisches“ im Griechisch der LXX gibt. Dass dies so ist, lässt sich nicht bestreiten: Die LXX weist, überwiegend als Übersetzung hebräischer Prätexte entstanden, zahlreiche Elemente auf, die dem Hebräischen geschuldet sind und die insofern auch einem griechischen native speaker einiges an Mitdenken abverlangt. Oftmals verfährt die LXX bei der Übersetzung so, dass sie Wort für Wort übersetzt, ohne darauf zu schauen, ob dies zu einer Gestaltung im Griechischen führt, die dem Wesen dieser Sprache entspricht, und ohne darauf zu achten, ob der Sinn in der Zielsprache noch derselbe ist wie in der Ausgangssprache. Doch davon soll hier nicht die Rede sein. In den folgenden Ausführungen wird der Aspekt geradezu umgekehrt: Es sollen Phänomene in der LXX aufgezeigt werden, die die LXX vom Hebräischen abrücken und griechischen Vorstellungen in sprachlicher und kultureller Hinsicht etwas näher rücken lassen. Es sei gleich dazu gesagt, dass damit keinesfalls behauptet werden soll, die LXX sei „griechischer“ als bislang angenommen: Der generelle Sprachduktus weist der LXX nach wie vor eine Sonderstellung zu, so dass man bei ihr von einer Soziolekt- oder Sondersprachlichkeit reden kann.1 Ziel der folgenden Ausführungen ist es vielmehr, einige Erscheinungen zu benennen, die eben gerade wegen dieser Sondersprachlichkeit als Elemente griechischer Sprache und Kultur ins Auge fallen, bislang aber nicht berücksichtigt worden sind.2 Ein paar Vorbemerkungen sind noch zu machen: 1. Ich befasse mich mit diesem Thema nicht als Theologe, sondern als Altphilologe. 2. Mein Zugang erfolgt daher aus der Perspektive des zielsprachlichen griechischen Textes: Es wird um Phänomene gehen, die die Formulierun1 Dies habe ich bereits kenntlich gemacht in meinem Beitrag „Die Septuaginta im Horizont des Hellenismus“, in S. Kreuzer/J.P. Lesch (ed.), Im Brennpunkt: Die Septuaginta. Studien zur Entstehung und Bedeutung der Griechischen Bibel, Band 2 (BWANT 161; Stuttgart: Kohlhammer, 2004), 78–118. 2 Die sprachlichen Besonderheiten sind in zahlreichen Publikationen bereits dargestellt worden – ich verweise insbesondere auf den in Anm. 1 genannten Sammelband sowie auf H.-J. Fabry/U. Offerhaus (ed.), Im Brennpunkt: Die Septuaginta. Studien zur Entstehung und Bedeutung der Griechischen Bibel (BWANT 153; Stuttgart: Kohlhammer, 2001).

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

82

Knut Usener

gen in der LXX betreffen, insofern sie zunächst griechische Texte sind, die als solche, mithin auch ohne den jeweiligen hebräischen Referenz- oder Ausgangstext, verstehbar und rezipierbar sein sollen. Das Hebräische wird allerdings dennoch und gerade auch da eine Rolle spielen, wo die Differenzen zwischen Quell- und Zielsprache allzu deutlich sind. 3. Die Auswahl der Textstellen erfolgt nicht aus systematischen Erwägungen, und sie basiert auch nicht auf einer bereits abgeschlossenen Datenbank, in der alle für das vorliegende Thema relevanten Phänomene versammelt vorliegen. Vielmehr handelt es sich bei dem, was ich vorstelle, zunächst um einzelne, ja um vereinzelte Beobachtungen, die sich bei der Beschäftigung mit einzelnen Schriften oder Passagen der LXX ergeben haben. Es handelt sich daher um Beobachtungen, aus denen man zunächst keinen allgemeinen Schluss für die LXX insgesamt ziehen kann. Bei der Betrachtung der sprachlichen Gestaltung einzelner Texte der LXX wird es weniger um rhetorische Stilistik in dem Sinne gehen, dass ich einzelne Stilelemente wie Alliteration, Homoioteleuton, Anapher, Figura etymologica, Hyperbaton oder Redundanzen vorführe – entsprechende Beobachtungen sind im vorliegenden Band nachzulesen. Wenn es also auch hier nun erneut um Stilfragen geht, dann um Besonderheiten, die über das übliche rhetorische Gestaltungsrepertoire hinausweisen. Hierzu gehören insbesondere versteckte Anspielungen auf Texte außerhalb der LXX durch Bezugnahmen auf die übrige griechische Literatur, Bezugnahmen also, die eine Abweichung von der interlinearen Übersetzung darstellen und nicht mit einem hebräischen Prätext erklärbar sind. Damit ist zugleich ein bekanntes Grundphänomen angesprochen, das im Hintergrund stets zu beachten bleibt: Bei den Texten der LXX handelt es sich zum größten Teil um Übersetzungsliteratur. Somit liegt in diesen Texten zunächst keine wie auch immer gestaltete originäre sprachliche Literaturschöpfung vor, sondern ein seinem Prätext verpflichteter Text, dessen Inhalt ebenso wie seine Form (zumindest im Horizont der Textsorte) im Wesentlichen bereits vorgegeben ist. Da einerseits eine antike Übersetzungstheorie zu diesen Texten nicht vorliegt und man andererseits mit der Anwendung moderner oder neuzeitlicher Reflexionen zur Übersetzungstechnik auf diese Texte vorsichtig umgehen muss, ist bei jeder sprachlich-stilistischen Untersuchung stets zu berücksichtigen, dass es sich bei den LXXTexten um eine vorgeprägte Textur handelt und dass dementsprechend Abweichungen von einem griechischen Sprachduktus daraufhin zu prüfen sind, • ob sie im Zusammenhang mit einer möglicherweise intendierten Wortfür-Wort- bzw. Interlinear-Übersetzung stehen oder von sondersprachlichen Gewohnheiten der LXX-Übersetzer geprägt sein können (das be-

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Griechisches im Griechisch der LXX

83

trifft insbesondere schwer verständliche und ungriechische Formulierungen sowie Hebraismen, die hier aber nicht weiter beachtet werden sollen), • oder ob andere, etwa literarische Gründe benennbar sind (das betrifft Passagen, die mit keinem der oben genannten Gründe in Zusammenhang zu stehen scheinen). In welchem Rahmen allerdings diese durch den Prätext definierten Vorgaben wirksam waren oder auch nicht, muss jeweils individuell geprüft werden. Somit gehört hierher auch die Frage nach solchen sprachlichen Erscheinungen in der LXX, die ganz spezielle Ursachen im Horizont der griechischen Literatur haben können: Sie können insbesondere • aus Gattungsgepflogenheiten ableitbar sein, die z.B. für das Epos, die Lyrik, das Drama, die Historiographie oder die Philosophie charakteristisch sind, • auf Formulierungen zurückgreifen, die nur für eine bestimmte sprachlich-kulturelle Epoche charakteristisch sind: Solche Formulierungen waren dann etwa in der archaischen Zeit verbreitet, scheinen jedoch später unbedeutend oder aus der Mode gekommen und dann erst wieder in der alexandrinischen Zeit zu neuem Leben erwacht zu sein. Viele dieser sprachlichen Phänomene können dabei allgemein als Kennzeichen des hellenistischen Griechisch gedeutet werden, das gerade auf ältere, insbesondere archaische Formulierungen und Formen bewusst zurückgreift. Zugleich aber verweisen diese Phänomene als hellenistische Stilistika auf eine Form der Gelehrten-Sprache oder auf eine Handhabung der Sprache, wie wir sie bei anspruchsvoller Literatur von ausgesprochen gebildeten und belesenen Autoren antreffen. Diese Besonderheiten der LXX greifen dann Tendenzen der die Gegenwart der LXX-Übersetzer prägenden Sprachlichkeit auf und können dabei Indizien für die Entstehung und Verortung einzelner Bücher der LXX liefern, sofern noch weitere Anhaltspunkte hinzukommen: Sprachliche Kriterien allein vermögen keine Sicherheit in Fragen der Verortung in Raum und Zeit zu gewährleisten. Im Folgenden werde ich nun mit dem Stichwort „Vokabular und Sondervokabular“ ein lexikalisches und zugleich stilistisches Thema ansprechen, das allerdings weniger Fragen der korrekten oder fragwürdigen Wortverwendung und Wortbedeutung betrifft, sondern vielmehr die Ebene der mit der Verwendung eines Ausdrucks gegebenen Intertextualität berührt. Dabei ist diese Intertextualität nicht nur als Phänomen der Literaturwissenschaft zu betrachten, sondern zugleich auch als etwas, das den kulturellen

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

84

Knut Usener

Horizont der LXX-Übersetzer und möglicherweise auch den ihres Publikums beschreibt.

2. Vokabular und Sondervokabular Im Gegensatz zu „Vokabular“ wird unter Sondervokabular der Sammelbegriff für Lexeme verstanden, deren Verwendung im jeweiligen Kontext aus bestimmten Gründen auffällt. Diese Gründe sind zum einen gegeben in der Abweichung des zu erwartenden Ausdrucks – wenn man im Vergleich zum hebräischen Text feststellt, dass gegenüber der Tendenz der LXX, möglichst geradlinig und in Wortgleichungen, wenn nötig auch weitestgehend konkordant zu übersetzen, hier und da eine deutliche, nicht ohne Weiteres aus dem Hebräischen erklärbare Abweichung vorliegt. Zum Zweiten sind die Gründe für Auffälligkeiten gegeben in der Einbeziehung oder im Wechsel zu einer Sprachebene, die dem Vorlagentext nicht unmittelbar entspricht – also etwa, wenn in einem darstellenden Abschnitt eines prophetischen Prosa-Textes lyrisches oder episches Vokabular eingestreut wird, das so den sprachlichen Duktus in ein anderes Licht taucht. Zum Dritten zählt zu den erwähnenswerten Auffälligkeiten das Hinzufügen von Zusätzen zu der in sich sprachlich, semantisch und als Übersetzung einer bereits verständlichen und korrekten Wiedergabe des hebräischen Textsinnes.

3. Methodik und Grenzen der Erfassung Neben der Lektüre der LXX-Texte und der Kenntnis der griechischen Literatur sind Konkordanzen und Lexika insbesondere in elektronischer Form hilfreich – sie können wenigstens zum Teil das Überlesen einer entsprechenden Wendung kompensieren. Doch eine Datenbank basierte und somit elektronische oder „kalte“ Auflistung der Schnittmenge des Vokabulars, das sowohl in der LXX als auch in den Gattungen etwa der Lyrik bzw. der Epik, nicht oder selten aber in der Prosa vor der LXX Verwendung findet, birgt die Gefahr der bezugs- und kontextfremden Schematisierung: In jedem Einzelfall muss der Kontext sowohl in der LXX als auch in der beispielsweise lyrischen oder epischen Quelle genau daraufhin untersucht werden, wie aussagekräftig das jeweilige Lexem oder die Wortfügung für die Frage der Sprachebene ist. Auch eine nicht kommentierte Auflistung, wie man sie etwa bei Zieglers Iob-Arbeit (1985) findet,3 ist zunächst wenig hilfreich. Hier bedarf es einer genauen Differenzierung. 3 J. Ziegler, Beiträge zum griechischen Iob (MSU 18 = AAWG 147; Göttingen: Akademie der Wissenschaften, 1985).

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Griechisches im Griechisch der LXX

85

So ist es, um ein Beispiel aus einem anderen, nicht von Ziegler behandelten Kontext zu nennen, eher unergiebig zu erwähnen, dass der griechische Begriff FWUMCVCRCWUVQL in der Bedeutung „schwer zu stillen“, „schwer zu beruhigen“ bei Aischylos von jammervollem Schmerz (C¢NIQL, Choephoren 470, in lyrischem Versmaß) und Euripides (von der leidgeprüften Seele, Medea 108, in anapästischem Metrum) belegt ist, dann aber erst wieder als LXX-Hapax in 3 Macc 5:7 (von einem Schrei, DQJ) sowie außerhalb der LXX etwa zur gleichen Zeit bei Theophrast und später dann wieder bei Plutarch vorkommt:4 Unergiebig ist dies dann, wenn man nicht den jeweiligen Hintergrund ausleuchtet und aufzeigt, was durch die Verwendung einer seltenen Ausdrucksweise über den semantischen Wert hinaus literarisch oder kulturell zum Ausdruck kommt. Über nützliches Wissen aus dem Bereich der Lexemverwendung werde ich gleich berichten.

4. Besondere Erscheinungen der Stilistik: Zwei Beispiele Das erste Beispiel, das ich präsentiere, ist Ezekiel 25:4 entnommen. Im Zusammenhang der Worte gegen die Ammoniter, gegen Moab, Idumäa und die Andersstämmigen [25:1–17] heißt es in 25:3–4: 3

MCKƒG™TGK¨LVQK¨LWK˜QK¨L$OOYPC MQWUCVGNQIQPMWTKQWVCFGNGIGK MWTKQLC PS Y¡PG RGZCTJVGG™RKƒVCƒC˜IKCOQWQŸVKG™DGDJNYSJMCKƒG™RKƒ VJƒPIJPVQW,UTCJNQŸVKJ HCPKžUSJMCKƒG™RKƒVQƒPQK»MQPVQW,QWFCQŸVK G™RQTGWSJUCPG™PCK™ZOCNYUK

4

FKCƒVQWVQK™FQWƒG™IYƒRCTCFKFYOKW˜OCLVQK¨LWKQK¨L.GFGOGK™LMNJTQ PQOKžCPMCKƒMCVCUMJPYUQWUKPG™PVC RCTVKCW™VYPG™PUQKƒMCKƒFY UQWUKPG™PUQKƒVCƒUMJPYOCVCCW™VYPCW™VQKƒHCIQPVCKVQWƒLMCTRQWL UQWMCKƒCW VQKƒRKQPVCKVJƒPRKQVJVCUQW

3

Dies sagt der Herr: Dafür, dass ihr euch über mein Heiligtum gefreut habt, dass es entweiht wurde, und über das Land Israels, dass es verwüstet wurde, und über das Haus Juda, dass sie in Gefangenschaft gegangen sind, 4 deswegen, siehe, übergebe ich euch den Söhnen von Kedem zum Erbe, und sie werden in dir mit ihren Habseligkeiten ihre Zelte aufschlagen, und sie werden in dir ihre Zelte aufstellen; sie werden deine Früchte verzehren, und sie werden dein Fett5 trinken.6

4

Theophrast, Vent. 35 (abgeleitete Bildung); Plut., Alex. 31. V.4: Fett: wörtl.: „Fettigkeit“; Sinn: Fruchtbarkeit, Reichtum. 6 Die verwendeten deutschen Übersetzungen der LXX sind hier und später der Neuausgabe W. Kraus / M. Karrer (ed.), Septuaginta Deutsch. Das griechische Alte Testament in deutscher Übersetzung (Stuttgart: Deutsche Bibelgesellschaft, 2009, 2. Aufl. 2010) entnommen. 5

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

86

Knut Usener

„Sie werden deine Früchte verzehren, und sie werden dein Fett trinken“ (25:4) ist die für uns hier relevante Passage. In EzekLXX heißt sie: HCIQPVCKVQWƒLMCTRQWLUQW MCKƒCW VQKƒRKQPVCKVJƒPRKQVJVCUQW

Wenn es in EzekLXX heißt, dass die Söhne von Kedem „deine [der Ammoniter] Früchte verzehren und deine RKQVJL trinken werden“, so soll dies wohl den masoretischen Text abbilden, der übersetzt heißt: „Sie werden diene Früchte essen und deine Milch (?DNZ) trinken.“ Statt RKQVJVCschreiben Theodotion und andere Zeugen ICNC– Milch: Erst mit dieser Lesart ist eine genaue Angleichung von griechischem und masoretischem Text gegeben. Der von Theodotion und anderen Übersetzern gewählte Begriff stellt die exakte und darüber hinaus naheliegende Wiedergabe des hebräischen Prätextes dar. Der LXX-Text scheint hingegen diese naheliegende Übersetzung zu Gunsten einer abgelegenen Formulierung geradezu zu vermeiden. Was bewirkt er dadurch? Das griechische Wort RKQVJL, das zunächst „Fettigkeit“, „Fett“ bedeutet, bezeichnet, ausgehend von dieser Grundbedeutung, generell den sehr guten Ernährungszustand (vom Vieh), die Üppigkeit (vom Boden) oder auch den reichen Ertrag, somit also etwas wie Reichtum an Naturalien. „Fett“ bedeutet im Griechischen in vergleichbarem Kontext soviel wie „Reichtum“: Die Menschen werden also „von deinem Reichtum“ sozusagen etwas abschöpfen. Im Griechischen kommt zum Ausdruck für RKQPVCK„trinken“ durch die Verwendung des Begriffes für „Fett“ RKQVJL die Assonanz durch die Buchstabenfolge RKQ- hinzu – auch hierin weicht die LXX vom MT ab, wo von der Figura etymologica diesmal kein Gebrauch gemacht wird. Wie kommt es nun zur Verschiebung von Milch zu Fett (RKQVJL)? Hinter dem Substantiv RKQVJL steckt das Adjektiv RKYP, das „fett, feist, wohlgenährt“ bedeutet. „Fett“ sind aber in einem ganz bestimmten Sprachgebrauch insbesondere Schafe und Ziegen: So bezeichnet Homer (Odyssee 9.464 etc.) diese Nutztiere als RKQPC OJNC bzw. OJNC RKQPC. Schafe und Ziegen (beide können als OJNC bezeichnet werden) dienten den Griechen seit jeher jedoch insbesondere auch als Milch-Lieferanten.7 Aus der bei Homer vorgeprägten Wortfügung oder Formel entsteht so unter Auslassung des Bezugswortes OJNC (Schafe oder Ziegen) und mit Substantivierung des Attributs RKQPC eine semantische Verschiebung, so dass man nun bei Ezekiel 25:4 „Fettigkeit trinkt“ – und damit ist dann wohl Schafs- oder Ziegenmilch gemeint. 7

Der Ausdruck RKQPCOJNC ist sonst nur noch bei Theokrit, Id. 25.86, belegt.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Griechisches im Griechisch der LXX

87

Ob es sich bei dieser hier vorgestellten Formulierung RKQPVCKVJƒP RKQ VJVC um einen Rückgriff auf die dichterische Sprache oder sogar speziell auf Homers epische Dichtung handelt, ist wegen der Abänderung und Auslassung noch nicht sicher: Andere Gründe könnten zur vorliegenden Formulierung geführt haben, wenngleich ich es für sehr wahrscheinlich halte, dass im vorliegenden Falle das homerische Sprachmuster durch Verkürzung abgewandelt wird.8 Durch die Begriffswahl setzt der LXX-Übersetzer allerdings ein – zunächst zumindest – diffuses Signal: Die Vermeidung eines durch den Prätext empfohlenen Wortes und die Wahl eines anderen, besonders auffälligen Begriffs generiert einen zusätzlichen neuen semantischen Kontext. Auf der Sachebene wird der Aspekt von Reichtum und Wohlleben angedeutet – die „Söhne von Kedem“ sind die Nutznießer der strafenden Handlung des Herrn. Auf der literarischen Ebene erfährt der Text eine gewisse Erhabenheit in der Diktion: „Milch trinken“ ist reine Nahrungsaufnahme, „Fett trinken“ hingegen verweist auf Wohlstand. Erkennbar ist zunächst nur, dass die Wiedergabe des naheliegenden Wortes durch einen eher überraschenden Begriff zwar zu keiner inhaltlichen Verschiebung der Aussage führt, dass aber die Formulierung stilistisch, ja durch die Bildhaftigkeit geradezu poetisch aufgeladen wird. Woher die hier gewählte Ausdrucksweise ursprünglich bezogen wird, bleibt zunächst offen, und insofern ist das Signal zunächst diffus. Doch was ist, wenn in nicht allzu weiter Ferne im gleichen Buch ein weiterer, vielleicht noch deutlicherer Hinweis auf die Verwendung oder Kenntnis dichterischer Diktion vorliegt? Wird dann nicht die Berechtigung für die Annahme größer werden, dass der LXX-Übersetzer Sprachgut verwendet hat, das er aus einem besonderen und benennbaren sprachlichen Kontext einer vielleicht sogar konkret benennbaren literarischen Quelle, nicht aber aus einer interlinearen Übersetzung der hebräischen Vorlage ableiten konnte? Ich komme zum zweiten Beispiel: Im Klagelied über Sor (Tyros) zählt der Prophet Ezekiel zwei Kapitel weiter, also in Kap. 27, die Reichtümer und Schätze auf Grund umfangreichster Außenhandelsbeziehungen auf. Das ist sicherlich auch historisch bedeutsam, aber nicht nur dies. Der griechische Text, der in einigen Details von der uns vorliegenden hebräischen Fassung abweicht und vielleicht auf einen Text zurückgeht, der älter ist als der Masoretentext, formuliert in Ez 8 Ein Blick ins NT zeigt, dass auch dort Wendungen vorhanden sind, deren Prägung dichterisch vorgeformt ist; vgl. etwa GK™LVQƒ WRGTQPC PCDCKPY in Apg 1:13; 9:37; 20:8 – das Muster der Wendung ist eine homerische Formel: vgl. Ilias 2.514; 16.184; Odyssee 1.362. Vgl. auch zu weiteren Beobachtungen D.R. MacDonald, Does the New Testament Imitate Homer? Four Cases from the Acts of the Apostles (New Haven/London: Yale University Press, 2003).

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

88

Knut Usener

27:5 einen Sachverhalt, der mit seiner hebräischen Vorlage nicht mehr erklärt werden kann. In EzekLXX heißt es: VCKPKžCKUCPKžFYPMWRCTKUUQWG™MVQW/KDCPQWG™NJOHSJUCPVQW RQKJUCKžUQKK˜UVQWƒLG™NCVKPQWL

…Bretter für Schiffsplanken aus Zypressenholz wurden vom Libanon geholt, um dir Mastbäume aus Tannen zu fertigen.

Dass hier Zypressenholz zu „Mastbäumen aus Tannen“ verarbeitet wird, grenzt in der Welt der Realien an Zauberei – und in der Welt ihrer Worte nennt man das dann allerdings eher Intertextualität: Den faktisch sinnlosen, literarisch aber umso wichtigeren Zusatz „aus Tannen“ (oder „aus Tannenholz“) bietet nur die LXX, nicht aber die Vorlage im hebräischen Text.9 Dieser somit rein griechische Zusatz lässt sich ganz einfach erklären, wenn man eine Formulierung aus der Odyssee (2.424 und 15.289) als Muster annimmt. Dort wird mit jeweils identischen Formulierungen dargestellt, wie ein Schiff zum Ablegen bzw. zum FahrtAufnehmen vorbereitet wird – jeweils im Zusammenhang mit Telemachos, der sich auf die Suche nach seinem Vater Odysseus begibt bzw. wieder zurück fährt: K˜UVQƒPF GK™NCVKPQPUVJUCP

und richteten einen Mast aus Tanne auf …

Der LXX-Übersetzer versteht die Formulierung „Mast aus Tanne“ offensichtlich als gleichbedeutend mit „Schiffsmast“ oder „Mastbaum“, wobei er die präzisierende Materialbezeichnung von Homer übernimmt (nur hier begegnet die Wendung),10 ohne sie im neuen Kontext („Zypressenholz“) als Störung zu empfinden. So kommt es zu einer Formulierung, die zwar inhaltlich dem hebräischen Vorlagentext verpflichtet ist, aber doch markant über diesen hinausweist, indem sie sich als formal von Homers epischer Diktion beeinflusst erweist. Was ist daraus zu schließen? Da es bei Ezekiel, aber auch bei anderen Texten der LXX Anleihen an die Sprache der Epik, speziell an Homer gibt,11 beginnt nun vielleicht ein Kapitel alexandrinischer Lebenskultur anschaulich 9 In der Vulgata (cedrum de Libano tulerunt, ut facerent tibi malum) wird keine weitere Holzart genannt – malus, i, m. kann „Mastbaum“ bedeuten, als fem. Substaniv (malus, i, f.) aber auch „Apfel-“ oder „Obstbaum“. Hier ist jedoch kontextbedingt der Mastbaum gemeint. Damit schließt sich die Vulgta klar dem MT an. 10 Die griechische Formulierung K˜UVQƒLG™(K)NCVKPQL für „Mastbaum aus Tanne“ begegnet nur an den hier genannten Stellen sowie in der hierauf bezogenen antiken Sekundärliteratur (also insbesondere in Homerkommentaren und Glossen sowie bei Theodoret zu Ezek 27:5). Sonst gibt es für die Wendung keine antiken Belege. 11 Diese sprachlichen Anleihen müssen noch eigens gesammelt und ausgewertet werden.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Griechisches im Griechisch der LXX

89

zu werden: Wenn man nicht davon auszugehen hat, dass die Übersetzer bereits in ihrer Schulausbildung mit Homer intensiv in Kontakt gekommen sind (was durchaus möglich ist, aber ebenso durchaus auch bezweifelt werden kann: Das ist eine Frage, die mit dem Phänomen des G˜NNJPKUOQL zusammenhängt), so kann man fragen, ob das Leben der Mitarbeiter des Museion und dasjenige der Diaspora-Juden in Alexandria vielleicht enger miteinander verwoben war, als wir uns das so gemeinhin vorstellen. Jedenfalls muss das geistige Leben recht vielfältig und der kulturelle Austausch anregend gewesen sein. Wenn nun in einigermaßen engem Abstand zwischen einzelnen Passagen sprachliche Phänomene begegnen, bei denen zumindest eines einen sicheren Rückgriff auf epische Formulierungstechniken darstellt, so wird man auch in anderen Fällen damit rechnen dürfen, dass eine sprachliche Fernwirkung vorliegt. Das zuletzt besprochene Beispiel lässt sich als „Homerismus“ bezeichnen – ein Begriff, der zum einen beschreiben soll, woher der Übersetzer zu seiner Formulierung angeregt worden sein kann, und der zum anderen ein Licht auf die Funktion oder Bedeutung einer solchen sprachlichen Gestaltung werfen kann: Der Rückgriff auf spezielle Wendungen, die nur in einmaligem Kontext belegt sind, mag ein Indiz dafür sein, dass der Übersetzer sei es Kenntnisse dichterischer Literatur der Griechen hatte, sei es über Kontakte zu alexandrinischen Philologen der damaligen Zeit verfügte: Dass der oder die Übersetzer jedenfalls nicht ohne Kontakt zur griechischen Kultur gelebt haben und dass dieser Kontakt das Umfeld der LXX-Übersetzer beeinflusst haben dürfte, kann durch derlei Beobachtungen an Wahrscheinlichkeit gewinnen. Wenn nun der Übersetzer eine spezielle und nur bei Homer belegte Wortfügung wählt, obwohl er den kontextuell erforderlichen Sachverhalt nicht nur ohne den Rückgriff auf Homer, sondern auch dadurch einfacher hätte formulieren können, so muss man nach den Gründen fragen. Handelt es sich bei dem Significans um eine erstarrte Wortfügung, die als ein Begriff, der aus zwei Worten besteht, das Significatum abbildet? Dann müsste das Significans auch an anderen Stellen in der Literatur belegt sein. Dies aber ist nicht der Fall. Handelt es sich bei der Wortfügung um eine Formulierung, die so in ihrem ursprünglichen Text sinnvoll und „sachlich richtig“ ist? Dann hätte diese Wortfügung auf ihrem Weg in ihren „zweiten Kontext“ beim LXXÜbersetzer an Sinnhaftigkeit verloren. Dies ist in der Tat der Fall. Wenn der Übersetzer diesen Sinnverlust oder – in diesem Falle sogar – diese in der Formulierung liegende Unstimmigkeit „Mastbäume aus Tannen aus Zypressenholz“ nicht bemerkt oder sogar in Kauf nimmt, so gewiss aus Gründen, die nicht auf der Ebene der Sachinformation liegen. Vielmehr wird es ihm um die sprachliche „Anleihe“ gegangen sein. Dadurch aber verleiht er sei-

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

90

Knut Usener

nem LXX-Text ein poetisches Kolorit: Erhabenheit, Würde und Autorität sind auf der poetologischen Seite, von der klanglichen Wirkung abgesehen, die Resultate für den LXX-Text. Auf der kulturell-soziologischen Seite kommt hinzu, dass man einen Einblick in die Gegenwartssituation des Übersetzers bekommt: Mag der Übersetzer die Quelle der Formulierung aus eigener Homer-Lektüre kennen, oder wird er vielleicht eher Kontakt zur Arbeit alexandrinischer Homerphilologen gehabt haben? Es lassen sich weitere Beobachtungen zu Formulierungen anschließen, bei denen möglicherweise ebenso Wendungen aus der griechischen Literatur im Hintergrund stehen: Im Falle der Ezek-Übersetzung sind dies etwa Passagen in 31:15–17 – dort wird eine Formulierung verwendet, die auch (aber natürlich nicht nur) aus dem homerischen Epos bekannt ist: „In (das Haus) des Hades“ – GK™L ŸFQW. Die Wendung begegnet in der LXX mehr als 20mal. In der Dichte der Phänomene liegt ein Indiz dafür, dass der Übersetzer mit der homerischen oder allgemein archaischen und poetischen Diktion, aber auch mit gemeinhin griechischem Gedankengut vertraut gewesen sein kann. Im besonderen Falle der Bezeichnungen für die Welt der Toten kommt neben dem poetischen Kolorit noch etwas hinzu: Wenn mitNY  die alttestamentarische Vorstellung der Totenwelt zum Ausdruck gebracht wird, so verwenden die LXX-Übersetzer überwiegend den in der griechischen (und somit heidnischen) Religion üblichen Ausdruck $KFJL.12 Dieser Begriff bezeichnet eigentlich den Totengott, und daher lautet die vollständige Formulierung für „in die Unterwelt hinabgehen“ zunächst „ins Haus des Hades hinabgehen“ – MCVCDCKPGKP GK™L FQOQP $KFQW (oder $KFCQ). „Haus“ (FQOQP) wird allerdings meist weggelassen, die zu diesem Wort gehörige Präposition GK™L hingegen bleibt stehen und wird dann mit dem possessiven Genitiv $KFQWverbunden.13 Der Genitiv dieses Namens erklärt sich somit aus der Ellipse des erforderlichen Akkusativs, den ursprünglich FQOQP liefert. Analog wäre etwa die in der englischen Sprache gebräuchliche Wendung „at the butcher’s“ (sc. house / shop). An der Verwendung der Wortfügung GK™L ŸFQW fallen besonders zwei Aspekte auf: Zum einen verwenden die Übersetzer der LXX eine heidnisch 12 An drei Stellen (Job 40:20; 41:24; Prov 30:16) begegnet VCTVCTQL: Die Griechen verbinden mit dem Tartaros ähnliche Schreckensvorstellungen, wie sie hebr. NY  transportiert: Bei Homer noch nicht mit Hades gleichzusetzen, kann in späterer Zeit Tartaros für Hades zur Bezeichnung der Unterwelt oder des Totenreiches verwendet werden. Ezek 26:20 etc. verwendet DQSTQL „Grube“ für die Stätte der Toten in der Unterwelt. Bemerkenswert ist, dass an sehr prominenter Stelle in der Odyssee (10.517) vom Ausheben einer Grube (DQSTQL) im Rahmen der Beschwörung der Totenseelen die Rede ist. Prov 23:14 verwendet für NY  nicht ŸFJL sondern SCPCVQL. Vgl. auch Hab 2:5. 13 Statt $KFQW wird auch die klein geschriebene und aspirierte Form ŸFQW mit der nichtepischen Genitiv-Endung –QWstatt –CQverwendet.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Griechisches im Griechisch der LXX

91

geprägte Wortfügung und treffen damit eine Wortwahl, die in anderen Fällen eher vermieden wird, wenn ein kultischer Begriff in der griechischen Sprache bereits geprägt und somit religiös „aufgeladen“ ist. Dies kann man etwa an den Begriffen für den Altar erkennen: Statt des üblichen Begriffs DYOQL wird eher der seit Homer geläufige Begriff G™UZCTC „Feuerstelle“ (G UZCTJ: Homer, Ilias 10.418; Odyssee 5.59 und öfter; G™UZCTC:Exod 25:4 etc.) oder – weit häufiger – der Neologismus SWUKCUVJTKQP „Opferstätte“ (Gen 8:20 etc.) verwendet. %YOQL wird hingegen verwendet, wenn es um den Gott nicht gefälligen und somit gleichsam heidnischen Altar geht (Exod 34:13 etc.). Zum Zweiten ist die Wahl des Ausdrucks insofern auffällig, als es im Griechischen andere Ausdrucksmöglichkeiten für die Unterwelt gibt, die religiös gleichsam neutral wirken: Statt „in das Haus des Totengottes“ kann man auch einfacher formulieren, dass die Seelen der Toten „nach unten“ (MCVY) zu „denen da unten“ (QK˜ MCVY) gehen. Oder man kann im Griechischen (seit Homer, Ilias 14.274) von G¢PGTSG oder PGTSG sprechen. Mit diesem Adverb werden etwa QK˜ G¢PGTSGSGQK „die Götter da unten“ bezeichnet. Auch MCVC ZSQPKQL „unterirdisch“ kann im Zusammenhang mit dem Totenreich gesagt werden: Mit =GWƒLMCVCZSQPKQLwird in Ilias 9.457 Hades als „Zeus der Unterwelt“ bezeichnet. Doch alle hier genannten Formulierungen kommen in der LXX nicht vor: Lediglich in Deut 32:22; Oden 2:22; Sir 51:6 und Psalmen Salomos 15:10 findet sich eine Verbindung von MCVY mit der Unterweltsvorstellung in der Formulierung GŸYL (fehlt bei Sir) ŸFQW MCVY „bis ins Totenreich hinab“. An den genannten Stellen ist MCVYallerdings ein substantivisches Attribut. Aus der Formulierung GK™L ŸFQW lassen sich keine zwingenden Rückschlüsse auf eine zugrunde liegende Quelle ziehen: Bei Homer wird noch nicht die attische Genitiv-Form ŸFQWverwendet, sondern stets $KžFCQ(Ilias 5.646 etc.); im attischen Drama (Sophokles, Aias 865 etc.) sowie bei Prosaautoren begegnet dann die seither und somit auch in der LXX belegte Form mit Anlaut-Aspiration und Genitiv auf -QW. Mit diesem Wort wird die gemeingriechische Vorstellung vom Totenreich ausgedrückt. Ihr Ursprung ist allerdings auf das Engste hörbar mit Hades als dem Totengott verbunden, der über das Totenreich herrscht. Ebenso durchaus nicht direkt auf eine bestimmt Quelle zu beziehen, in der Formulierung aber dennoch poetischer als der Prätext ist der griechische Text von Ezek 31:15: Dies sagt der Herr: ‚An dem Tag, an dem es in das (Haus) des Hades hinabstieg, trauerte die Tiefe um es, und ich hielt ihre Flüsse an und staute die Fülle des Wassers,

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

92

Knut Usener

und (aus Trauer) um es14 verfinsterte sich der Libanon, alle Bäume der Ebene wurden (aus Trauer) um es gelöst15.’

„Wurden gelöst”: /6 schreibt JRN(„sanken ohnmächtig hin“). Der LXX-Übersetzer greift dafür auf ein episches Bild (denkbar ist auch hier ein Homerismus) zurück und überträgt MT sinngemäß: RCPVC VCƒ EWNC VQW RGFKQW G™R CW™V G EGNWSJUCP. Bei Homer kommt die Wendung … NWVQ IQWPCVC „die Knie lösten sich“ in Ilias 21.114 etc. (nebst einer formalen Abwandlung in 13.360) bzw. IQWPCV  G¢NWUGP (Ilias 5.176 etc.) oder NWUG FGƒ IWKC (Ilias 16.312 etc.) im Zusammenhang einer Sterbeszene vor, in der Odyssee findet sich diese Formel an zahlreichen Stellen jeweils bei der Darstellung eines „Schwächeanfalls“.16 Die Bedeutung der Passagen bei Homer, der diese bildhafte Wendung mit ihrem formelhaftem Charakter literarisch erstmals prägt, ist also stets Ohnmacht oder Tod: Ein Held, dessen Knie „sich lösten“, stirbt oder gerät nahe an den Rand des Todes in den einen Zustand der Kraft- oder Bewusstlosigkeit. Mit der Verwendung von G™MNWY bei Ezek 31:15 kommt es zu einer ähnlichen Formulierung wie bei Homer; so lässt der LXX-Übersetzer im Hintergrund eine gewisse Kriegs- oder Gewaltmetaphorik entstehen, wenn er die „Größe und den Fall der Zypresse“ darstellt. Zugleich wird aber insbesondere der Aspekt der Ohnmacht, den der MT bietet, dann (allerdings nur für einen kompetenten Leser) transparent ins Griechische mit übertragen, wenn man die Odyssee-Passagen im Hintergrund kennt: Die meisten Stellen dort sind mit Odysseus verknüpft, und zwar im Zusammenhang mit Monolog-Szenen, in denen der Held seine jeweilige Situation wahrnimmt und reflektiert. Die bei Ezek verwendete Wiedergabe „alle Bäume … wurden … gelöst“ wird im Sinne des MT nur vor dem Hintergrund der bildhaften Ausdrucksweise in Homers Odyssee verständlich.17 Eine interessante Besonderheit bietet erneut Ezechiel. In 37:7 heißt es: MCKƒG™RTQHJVGWUCMCSYƒLG™PGVGKNCVQOQKMCKƒG™IGPGVQG™PVG™OGƒ RTQHJVGWUCKMCKƒK™FQWƒUGKUOQLMCKƒRTQUJICIGVCƒQ UVCG˜MCVGTQP RTQƒLVJƒPCTOQPKCPCW™VQW

Und ich verkündete, wie er mir aufgetragen hatte. Und als ich verkündete, siehe, da kam es zu einem Beben, und die Knochen näherten sich, ein jeder zu seiner Fügung. 14 V.15: Um es: Gemeint ist Assur, während „ihre“ zuvor wieder das Bild der Zypresse aufgreift. 15 V.15: Wurden gelöst: Sinn: wurden schwach, welk, ohnmächtig. 16 Odyssee 4.703 etc. nebst einer formalen Abwandlung etwa in Odyssee 18.238 etc. 17 Vgl. zu den Monologszenen K. Usener, Beobachtungen zum Verhältnis der Odyssee zur Ilias (Script-Oralia A, Altertumswissenschaftliche Reihe 5; Tübingen: Gunter Narr, 1990), 81–94. Die Wendung NWVQ IQWPCVCfür „sterben“ lebt in der griechischen Literatur stets nur als HomerReminiszenz in ihrer archaischen Sprachgestaltung fort (vgl. etwa Chariton 1.1.14.5; Oppian, Halieutika 2.71).

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Griechisches im Griechisch der LXX

93

Der Begriff CTOQPKC („Fügung“) kommt in LXX nur bei Ezek vor (23:42 und 37:7) und hat im MT keine Entsprechung: Im MT lautet die Elberfelder Übersetzung zu Ezek 37:7: Und ich weissagte, wie mir befohlen war. Da entstand ein Geräusch, als ich weissagte, und siehe, ein Getöse: und die Gebeine rückten zusammen, Gebein an Gebein.

Die auffällige und durch den Prätext nicht eben naheliegende Formulierung verleitet vorsichtig zur Frage, ob hier (wenn auch in gleichsam verdünnter Konzentration) ein philosophischer Einschlag vorliegt: Heraklit18 ist hier durchaus als Bezugspunkt denkbar, da bei ihm der Begriff der „Fügung“ oder CTOQPKC wichtig ist.19 Bei Heraklit allerdings spielt diese „Fügung“ eine besondere Rolle im Zusammenhang mit der Verbindung der Gegensätze, die sich in einer höheren Einheit aufheben: Davon ist bei Ezek 37:7 zumindest an der Textoberfläche nichts zu spüren. Hier geht es vielmehr um die Fügung oder Zusammenfügung dessen, was der Tod auseinandergerissen hat: Die Knochen, die nicht mehr in ihrer einstigen funktionalen Verbindung und Einheit existieren, werden wieder aneinandergefügt. Dass hier von CTOQPKC gesprochen wird, überrascht etwas, zumal der Übersetzer die Möglichkeit gehabt hätte, die im hebräischen Prätext stehende Wendung ins Griechische zu übernehmen. Doch wenn man den Gedanken verfolgt, eine Aussage, wie sie Heraklit treffen könnte, würde auch hier getätigt werden, so würde dies folgenden Gedankengang ermöglichen: „Gebeine rückten zusammen, Gebein fügte sich wieder zu Gebein – und dies, obwohl es doch eigentlich nicht möglich ist, da die nicht mehr zusammengefügten Gebeine durch diese Auflösung ihrer Verbindung tot sind; wären sie hingegen zusammengefügt, hätten sie die Qualität des Lebendigen.“ So stünde dann das Tot-Sein in einem geradezu widersinnigen und doch wiederum verständlichen Spannungszusammenhang mit dem „Lebendig-Sein“. In einer Vision – und um die geht es im Kontext bei Ezek 37:7 – lässt sich eine solche Spannung aushalten. Insofern es sich tatsächlich auch um eine inhaltliche Spannung handelt, ist die Ausdrucksweise allerdings durchaus in gewissem Grade gedanklich Heraklit bzw. seinem Denken verpflichtet. An dieser Stelle lassen sich weitere Beobachtungen anschließen, die nun auch andere Bücher der LXX betreffen.

18

Vgl. Heraklit Frg. 22 B 8 DK; 22 B 51 DK; 22 B 54 DK; vgl. Plat., Symp. 187a. Auffallend häufig begegnet philosophisches Vokabular: NQIKUOQL (Esth 1:1 etc.; in 4 Macc über 100 Belegstellen) HKNQUQHKC (fünf Belege in LXX, alle in 4 Macc). Besonders fällt dies in 4 Macc auf – ein Text, der wohl ohne hebr. Quelltext entstanden ist. Der Sprachduktus unterscheidet sich so auch von anderen Texten der LXX, da er keine Übersetzungsliteratur bietet. 19

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

94

Knut Usener

Eine auffällige Formulierung, die nicht vom hebräischen Prätext nahegelegt wird, findet sich in Isa 3:22-23.20 Dort ist von u.a. „lakonischen Kleidern“ die Rede: VCƒG™RKDNJOCVCVCƒMCVCƒVJƒPQK™MKCPMCKƒVCƒFKCHCPJ/CMYPKMCƒMCKƒ VCƒDWUUKPCMCKƒVCƒW˜CMKPSKPCMCKƒVCƒMQMMKPCMCKƒVJƒPDWUUQP 'KCHCPJL bedeutet „durchsichtig“: Man denkt an einen Stoff wie Tüll

oder Organza – es geht also um Kleidungsstücke, die wie ein Negligé aussehen. Die griechische Formulierung ist z.T. im MT nicht nachvollziehbar, und speziell die FKCHCPJ /CMYPKMC stehen ohne Vorbild da.21 Das Adjektiv FKCHCPJL wird im Zusammenhang mit Kleidungsstücken (so der Kontext bei Jesaja) verwendet. In der griechischen Literatur wird FKCHCPJL auch – aber nicht nur – mit Kleidungsstücken verbunden: So bezeichnet der Komödiendichter Aristophanes „leichte (sozusagen nicht blickdichte) Frauenkleider“ als FKCHCPJZKVYPKC(Aristophanes, Lysistrate 48). Ein weiterer Komödienschriftsteller, Menander, bezeichnet diese Kleider als FKCHCPGƒL ZKVYPCTKQP(Frg. 727,1f.). Als „lakonisch“ werden aber nicht Kleider, sondern Schuhe bezeichnet (wiederum Aristophanes, Vespae 1157f.: G ODCFCL /CMYPKMCL). Diese Schuhe aus Filz sind eher einfach, also nicht besonders kostbar. Bei Jesaja liegt also eine gewisse Spannung in der Wortfügung, die allerdings erklärlich ist, wenn man im Hintergrund die moralisch nicht für jedermann akzeptable Polyandrie der Spartaner (vgl. etwa Polybios 12.6b.8) sieht, die als Ausschweifung gebrandmarkt werden konnte: Dann spricht Jesaja von einem ordinären Kleidungsstil in spartanischer (lakonischer) Manier. Die Formulierung ist dabei recht kompakt: Sie setzt das kulturelle (Klischee-)Wissen beim Rezipienten voraus. Die Formulierung bietet mit gleichsam spartanisch-karg gehaltenen, nicht aber dichterischen sprachlichen Mitteln eine Anspielung und Kritik im Horizont der Auseinandersetzung mit einer fremden Kultur. Dass dabei keine unmittelbare Quelle benannt werden kann, aus der der Übersetzer schöpfen konnte, spricht für eine bewusst eigenständige Formulierung.

5. Weitere Textpassagen Eine weitere poetische Gestaltung einer Formulierung begegnet in Neh 5,14f. (zweimal): Dort wird in der LXX (= 2Esdr 15,14f) für hebr. JZR bzw. ]ZR (Statthalter) statt der Person oder statt seines Amtes der Begriff DKC „Gewalt, Kraft, Macht“ eingesetzt, was einer speziell auch bei Homer 20

Zu Isa 3:22 vgl. Cant 5:7. Vgl. zu diesem Text auch A. Passoni Dell’Acqua, “Colori e trasparenze nella haute couture dell’Egitto greco-romano”, Semitica et Classica 1 (2008) 113–138, bes. 130. 21

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Griechisches im Griechisch der LXX

95

anzutreffenden Formel entspricht: Homer spricht etwa von der DKJ ˜+TC MNGKJ;22 auch Iphikleia wird in Odyssee 11.296 durch den Zusatz DKJ zu einer angesehenen Persönlichkeit.23 Die Bezeichnung einer adligen Person erfolgt bei Homer in archaischer Weise durch die Betonung ihrer durchsetzungsstarken und überlegenen Macht und Kraft gegenüber anderen Personen der Gemeinschaft. Der LXX-Übersetzer verfährt nur an den hier genannten Stellen so – ohne vom hebräischen Text dazu „eingeladen“ zu werden. Dadurch bekommt der Text eine besondere Erhabenheit. Man kann von einer gewissen Episierung sprechen – der Begriff des Homerismus würde das Phänomen etwas einengen. Neben Formulierungen, die (wie oben dargestellt) auf in der Philosophie verwendete Begrifft verweisen, finden sich – wie zu erwarten natürlich eher selten – auch Anspielungen auf Gestalten des griechischen Mythos: Eine weitere, zumindest verbale „Anleihe“ an den griechischen Kulturraum und Mythos findet man bei Job (30:29): C FGNHQƒLIGIQPC UGKTJPYP G˜VCK¨TQL FGƒ UVTQWSYP „ein Bruder der Sirenen bin ich geworden und ein Gefährte der Strauße.“ Zu den Sirenen und deren Belegen in der LXX schreibt Lust:24 „siren, demon of the dead living in the desert (used to translate Hebr. words meaning ostrich, desert owl and jackal).” Die Belege finden sich neben Job 30:29 bei Isa 13:21; 34:13; 4:20; Jer 27:39; Mic 1:8 und 4 Macc 15:21. In der griechischen Mythologie gibt es wohl zunächst nur zwei Sirenen (Homer, Odyssee 12.39ff.), später drei oder vier. Die Vorstellung ihres Aussehens wechselt: Man sah in ihnen hübsche junge Frauen mit verlockenden und zugleich Verderben bringenden Stimmen oder auch dämonische Mischgestalten aus Vogel und Mensch mit Wohnsitz bald im Himmel, bald im Hades.25 In der LXX dient der Name der Sirene nur als loser Anknüpfungspunkt ohne mythisches Korrelat. Besonders „griechisch“ wirkt die Sirene hier, obwohl sie zwar dekontextualisiert ist, aber doch als Begriff für die Vorstellung von einem fabulösen Wesen herangezogen wird, das dem Hebräischen nicht in genauer Entsprechung bekannt ist. In der LXX wird an die Vorstellung der Sirenen bald als Frauengestalt, bald als Fabelwesen angeknüpft, und Job erfindet noch einen Bruder für die Sirenen hinzu, da der hebräische Text vom „Bruder des Schakals“ spricht: Damit aber wird für diese Passage 22

Ilias 2.658; 11.690; Odyssee 11.601. Vgl. auch Ilias 13.770; 781: RQW VQK (781: QK¢Y) 'JKžHQDQL VG DKJ S  (NGPQKQ C¢PCMVQL; Ilias 23.859 … DKJ7GWMTQKQC¢PCMVQL– „die Gewalt des Gebieters Teukros“. 24 J. Lust / E. Eynikel / K. Hauspie, Greek-English Lexicon of the Septuagint (2nd edn; Stuttgart: Deutsche Bibelgesellschaft, 2003), 550. 25 Vgl. hierzu A. Heubeck/A. Hoekstra, A Commentary on Homer's Odyssey. Volume II: Books IX–XVI (Oxford: Oxford University Press, 1989), 118ff. 23

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

96

Knut Usener

deutlich, dass der LXX-Übersetzer nicht an die Fabelwesen der griechischen Dichtung in ihrem mythologischen Kontext dachte, sondern eine Entsprechung zum Schakal gesucht hat, der griechisch Q˜ oder J˜ SY‚L heißt (nicht in der LXX belegt). Der Symbolwert dieses Tieres schien wohl am ehesten dem der Sirenen zu entsprechen. Dieser geradezu dämonische Symbolwert des Untergangs ist bei Isa 13:21 besonders deutlich hörbar. Die Einsetzung der Sirenen kann in der griechischen Übersetzung von Job (30:29) und Isa (34:13; 43:20) damit zusammenhängen, dass der MT im unmittelbaren Kontext der Schakale auch von Straußen spricht: Das ebnet die Verwendung der Sirenen als fabulöser Mischwesen mit Vogelgestalt, die die Schakale gleichsam verdrängen. Dass bei den Sirenen aber auch an Frauen gedacht wird, geht aus der Tatsache hervor, dass bei Jeremia (27:39; vgl. Mic 1:8) von „Töchtern“ die Rede ist. In 4 Macc 21 klingt der Sagenhintergrund, wie er seit der Odyssee bekannt ist, andeutungsweise an (UGKTJPKQK OGNFKCK, „sirenenartige Gesänge“ oder „melodische Gesänge von Sirenen“). Somit ergibt sich bei der Verwendung des Wortes UGKTJP in der LXX ein breites Spektrum: Die Vorstellungen sind bei den Übersetzern bzw. Verfassern nicht einheitlich. Job bezieht noch einmal griechische Mythologie ein, wenn er in 42:14 eine seiner drei spätgeborenen Töchter $OCNSGKCL.GTCL nennt, worunter man auch allegorisch das „Füllhorn“ versteht. Amaltheia, sei es eine Nymphe, sei es eine Ziege, hatte den griechischen Gott Zeus in dessen Kindheit mit Ziegenmilch ernährt. Das abgebrochene Horn der Ziege wird später vom dankbaren Göttervater zum unerschöpflich reichen Füllhorn verwandelt. Diese Sage ist bei den Dichtern Anakreon (Frg. 8 D) und Phokylides (Frg. 7 D) im 6. Jh. v. Chr. erstmals greifbar, wird dann aber insbesondere auch mit dem am Ptolemäerhof tätigen Wissenschaftler und Dichter Eratosthenes verbunden:26 Unter seinem Namen ist ein (wohl aber nicht von ihm stammendes) Gedicht über Gestirn-Konstellationen (Katasterismoi) in Fragmenten erhalten, in dem u.a. die Sage der Amaltheia behandelt wird (Katasterismos 13). Vielleicht ist der LXX-Übersetzer von Job mit diesem Lehrgedicht vertraut gewesen. Er ist jedenfalls der einzige Übersetzer der LXX, der den Namen Amaltheia verwendet. Dabei hat er eine möglichst angemessene Wiedergabe von ?YRJ_TS gesucht: In der Übertragung durch $OCN SGKCL.GTCL wird der Klang eine nicht unerhebliche Rolle (_TSund MGTCL) gespielt haben. Wie schon öfter besonders bei der LXX-Übersetzung von Ezekiel zu beobachten war, so begegnet auch in unmittelbarer kontextueller Nähe bei Job 26

Eratosthenes wurde von Ptolemaios III Euergetes als Erzieher für seinen Sohn Ptolemaios Philopator nach Alexandria berufen. Später leitete Eratosthenes als Nachfolger des Wissenschaftlers und Dichters Kallimachos die Bibliothek in Alexandria.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Griechisches im Griechisch der LXX

97

ein weiteres Wort des griechischen Kulturraumes, das nicht aus dem Hebräischen abgeleitet werden kann: In 42:11 bekommt Job neben einem Lamm auch „ein ungeprägtes Goldstück (im Wert) einer Tetradrachme“, VGVTCFTCZOQP ZTWUQWP C¢UJOQP. Es ist dies der einzige Beleg für die Tetradrachme in der LXX. „Ungeprägt“, C¢UJOQL, wird in der LXX insgesamt dreimal verwendet, und zwar neben Job 42:11 in Gen 30:42 (Lämmer „ohne Zeichnung“ oder „bedeutungs-, wertlos“) und 3 Macc 1:3 „unbedeutend“: Die Bedeutung „ungeprägt“, die auch etwa bei Herodot (9.41), bei Thukydides (2.13 etc.), bei der Alexander-Rhetorik aus dem 2. Jh. v. Chr. (Kap. 69) oder auch in zahlreichen Inschriften anzusetzen ist, trägt das Wort in der LXX nur bei Job.

6. Zusammenfassung und Ausblick Es gibt deutliche und weniger deutliche Hinweise darauf, dass zumindest einige Übersetzer der LXX über den Prätext hinaus auf Formulierungen und Inhalte der griechisch-heidnischen Literatur zurückgreifen. Mythologische (Horn der Amaltheia, Sirenen) und philosophische (CTOQPKC), wirtschaftlich-kulturelle (Tetradrachme, lakonische Kleidchen, Hades, Tartaros) und poetologisch-literarische (DKC K˜UVQL G™NCVKPQL RKQVJLOhnmachtsanfall und Todesnähe) Aspekte werden in der LXX an verschiedenen Stellen aus dem griechisch-heidnischen Kulturkreis übernommen und (ohne dass der hebräische Prätext dies vorgibt) der Übersetzung gleichsam als Zugabe mitgegeben: Dies führt allerdings nicht dazu, dass der jeweilige Sprachduktus insgesamt in höherem Maße griechisch ist als der anderer Übersetzer. EzekielLXX etwa – ein Text, der in überdurchschnittlich vielen Passagen Gedanken und Formulierungen aus der griechischen Literatur bezieht – ist ein gutes Beispiel dafür, wie sich offensichtliche Bezüge auf griechische Literatur in eine Sprache integrieren lassen, die stellenweise sehr linear am hebräischen Text entlang formuliert und nicht selten das Sinnverständnis des griechischen Textes verstellt oder zumindest erschwert.27 Die Bezugnahmen auf griechische Sprachmuster oder literarische Textpassagen sind Inseln in einem Meer der Übersetzungs-Sprache, die das Koine-Griechisch zur Basis hat und darüber hinaus ihren eigenen Regeln folgt: Sie spiegeln das geistige Interesse an einer Kultur wider, in deren Horizont die Übersetzer leben und mit deren exponierten Vertretern sie in 27 Beispiele hierzu habe ich in dem in Anm. 1 genannten Beitrag vorgelegt. Verwiesen sei darüber hinaus auch auf die deutsche Übersetzung der LXX und auf den entsprechenden Erläuterungsband. Vgl. W. Kraus/M. Karrer (ed.), Septuaginta Deutsch. Das griechische Alte Testament in deutscher Übersetzung (Stuttgart: Deutsche Bibelgesellschaft, 2009, 2. Aufl. 2010). Die Publikation des Erläuterungsbands ist für 2011 vorgesehen.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

98

Knut Usener

gewissem Kontakt gestanden haben dürften: Es ist nicht undenkbar, dass in Alexandria, an dem Ort, an dem zumindest eine große Zahl an LXX-Texten entstanden sein wird, ein Austausch mit den Gelehrten des Museion stattgefunden hat. Um das geistige Leben der Diaspora-Juden und ihrer griechischen Umwelt noch klarer erkennen zu können, bedarf es noch weiterer Untersuchungen – die vorgelegten Beobachtungen können nur ein bescheidener Anfang gewesen sein. English abstract: This essay deals with some linguistic and stylistic features of the LXX, which cannot simply be explained on the assumption of a word-for-wordtranslation from the underlying Hebrew text. Diverse translational features are found in many of the texts that are included in the LXX. Alongside phrases that sound non-Greek, there are others that are commonly found in rhetorically elaborated Greek prose, as well as further examples that are mainly suitable for Greek poetry. This suggests that a complex approach has been adopted in undertaking the translation of the Hebrew text. Consequently, it is helpful to ask about the kind of rhetorical features found in the LXX. Significant questions include how best to describe the rhetorical skills of the LXX-translators, and what are the key differences between these and those employed by Greek native speakers? Furthermore, it is useful to glean any information concerning the cultural background of the LXX-translators, and to provide possible reasons for specific formulations that have been chosen, especially in cases where the Hebrew text does not support these renderings? Other pertinent questions include: What can the epic storyteller Homer contribute to contemporary LXX-studies? Why do certain LXX-translators employ phrases that sound philosophical? Did some LXX-translators have knowledge of the philological employment concerning the Museion in Alexandria? Therefore, this essay discusses those conspicuous passages that feature mythological, philosophical, and other literary allusions in the LXX that are beyond the Hebrew pattern.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Alexis Léonas

The Poetics of Wisdom: Language and Style in the Wisdom of Solomon 1. Preliminary questions The scholarly consensus on the Book of Wisdom today favours both its compositional unity and its Greek linguistic origins.1 Nonetheless, for the purposes of our study, the foundations of this consensus merit a brief overview. The unity of the book reveals itself not only through the narrative coherence of its content, but also through multiple formal details, which a hypothetical editor would have found difficult to assemble in a composite document. On the level of the content, the presentation of the practical workings of wisdom, evident in the everyday confrontation of the wise with the unwise in the first part of the book (Ch. 1–5) is followed by an invocation and definition of Sophia by an anonymous royal sage2 (Ch. 6–9) and develops into a demonstration of her works through history (biblical, but presented in the most general way possible), culminating in the story of Exodus, seen as her greatest manifestation and feat (Ch. 10–19). With all due allowance to criticism,3 this scheme seems coherent enough to hold together the different parts of the text. From the more formal point of view, the distribution of rare or artful words throughout the book seems to support the same conclusion.4 Compound Greek words with no easily imaginable Hebrew Vorlage are equally widespread,5 These factors, as well as frequently recurring alliteration and

1 Cf. a thoroughgoing presentation of this view in D. Winston, The Wisdom of Solomon (The Anchor Bible; New York: Doubleday, 1979), 12–20. 2 His identity becomes somewhat more explicit in Wis 9:7–8: he was chosen by God to be the king of His people and to build a Temple on His Holy Mountain – sapientis sat! 3 The early modern criticism held the content in the earlier and later chapters as radically different, cf. J.G. Eichhorn, Einleitung in die apokryphischen Schriften des Alten Testaments (Leipzig: Weidmann, 1795), 88ff, 144ff. 4 Cf. C. Larcher, Etudes sur le livre de la Sagesse (Paris: Gabalda, 1969), 181ff; J.M. Reese, Hellenistic Influence on the Book of Wisdom and its Consequences (Analecta Biblica 41; Rome: Pontificio Istituto Biblico, 1970), 3ff; M. Kepper, Hellenistische Bildung im Buch der Weisheit: Studien zur Sprachgestalt und Theologie der Sapientia Salomonis (BZAW, 280; BerlinNew York: Walter de Gruyter, 1999), 71ff. 5 Cf . Winston, The Wisdom of Solomon, 15 n.5.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

100

Alexis Léonas

wordplay6 – unimaginable on such a scale in a translation7 – argue both for the compositional unity and the genuinely Greek authorship of the text.8 However, the uneven distribution of the parallelismus membrorum (consistently present in Ch. 1–5, often present in 6:1 – 12:18 and nearly absent in 12:19 – 19:22)9 ultimately remains a stumbling block in affirming the book’s unity. The fluctuation in the recourse to the use of parallelism cannot be explained away by mere poetic license, even if we refuse to see in it enough ground for dissecting our text and analysing it as several different sources.10 Thus, a study of the poetics of the Wisdom of Solomon necessarily has a bearing on the book’s unity and origin. In the present investigation I will attempt to address, if not resolve, some of the difficulties involved.

2. Literary artifice in the Book of Wisdom and the questions it raises Richness of vocabulary is probably the most basic indication of a book’s literary ambitions. Indeed, of the 7000 words that make up the Book of Wisdom,11 its actual vocabulary is no less than 1700 words, 20% of which are not attested elsewhere in the Septuagint12. Furthermore, approximately 6 Winston, The Wisdom of Solomon, 16 n. 9–12. Cf. the succinct presentation of data in support of unity in S. Holmes, “The Wisdom of Solomon. Introduction”, in R.H. Charles (ed.), The Apocrypha and Pseudepigrapha of the Old Testament. Vol. I (Oxford: Clarendon, 1913), 521–522. 7 F. Zimmermann is right to point out the frequent wordplay in the Targums, but this evidence is not conclusive because of the linguistic proximity between the Biblical Hebrew and the Targum Aramaic, which makes wordplay easy (F. Zimmermann, “The Book of Wisdom: its Language and Character”, JQR 57 [1966] 4 and n.4; the examples mainly draw on Onkelos). It can be generally observed that the Targums not only use wordplay, but can have surplus literary design absent in the parent text (cf. I. Himbaza, “Le poème acrostiche sur Exode XX, 1–5 dans le Targum Fragmentaire [MSG]”, VT 52 [2002] 42–50). Nevertheless, these texts are much too late to be plausibly used as models for Wisdom. 8 Latest serious attempt to claim the Hebrew background for at least one Wisdom chapter was made by N. Peters, “Ein hebräischer alphabetischer Psalm in der Weisheits Salomons, Kap. 9“, BZ 14 (1916) 1–14. Aramaic origin was advocated by F. Zimmermann (“The Book of Wisdom”, 1–27, 101–35). Despite their ingenuity, these tours de force of retroversion fail to convince. 9 Cf. F. Focke, Die Entstehung der Weisheit Salomos (FRLANT 22; Göttingen: Vandenhoeck & Ruprecht, 1913), 53. 10 S. Holmes has pointed out another fluctuation – probably related to the first – in the distribution of particles in the first and second part of the book (Ch. 1–10 and 11–19). Thus, OGP occurs 3 times in Wis 1–10 and 27 times in the rest of the book. Likewise, FG occurs 52 times in the first part against 82 times in the second, KžPC 7 times vs. 21, C NNC 4 vs. 17, ICT 52 vs. 102 and MCK?ICT 2 vs. 12. Cf. S. Holmes, Introduction, 522. 11 Cf. Kepper, Hellenistische Bildung, 51–52. 12 Cf. the list of some 50 compound words used in Wisdom in H.B. Swete, An Introduction to the Old Testament in Greek (Cambridge: Cambridge University Press, 1900), 311; Kepper, Hellenistische Bildung, 51–52.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

The Poetics of Wisdom: Language and Style in the Wisdom of Solomon

101

300 words of this record appear to be extremely rare, if not wholesale hapax legomena.13 The recent study by M. Kepper has demonstrated other occurrences for most of those 300 words, leaving nevertheless 19 cases, which the author qualifies as hapax totius graecitatis.14 Lexical richness finds characteristic expression in the wide-ranging use of the composite adjectives: MCMQVGZPQL (Wis 1:4; 15:4), PJRKQMVQPQL (11:7), Q NKIQ ZTQPKQL (9:5) etc.15 An interesting trait of the Wisdom’s writing style is the avoidance of personal names. Although the author’s awareness of the Classical literature as well as of the Biblical books is reflected in numerous indirect quotes,16 this background never becomes explicit. The attitude shows some similarity with the oracular style, by definition enigmatic.17 Hellenistic Judaism provides an example of this style in some of the Sibylline oracles (books III–V, nd st d II –I c. BC)18, while Lycophron’s Alexandra (first half of the III c. BC)19 demonstrates the readiness of early Alexandrian poetry to integrate this style in a properly literary context:20 as a literary product, the Wisdom of Solomon lies somewhere between those two. Convoluted periphrastic constructions enhance the enigmatic aspect of Wisdom’s writing (Wis 1:12: RNCPJ\YJL = deviant way of life; Wis 7:1: IJIGPQWL C RQIQPQL RTYVQRNCUVQW = descendant of Adam; Wis 11:7: PJRKQMVQPQL FKCVCIOC = Pharaoh’s decree to kill the infants; Wis 17:2: HWICFGLVJLCK™YPKQWRTQPQKCL = “fugitives from eternal providence” => the godless Egyptians in the darkness; Wis 18:12: G P GPK? Q PQOCVK SCPC VQW = “through a common death”). Likewise, metonymy (EWNQP = “boat”21 in Wis 10:4; 14:5, 7; RQPQL = “fruit of labour” in Wis 8:7; 10:10; ZGTUCK¨C//PJMVC? = “dry-land creatures” [i.e. “Egyptians”] // “aquatic creatures” [i.e. “frogs”] in Wis 19:19) and frequent litotes contribute to the impression of the intricacy of the text’s 13

J.M. Reese catalogues 335 cases of hapax (Hellenistic Influence, 3), while Larcher acknowledges only 315 (Etudes, 181). 14 Kepper, Hellenistische Bildung, 71–72. 15 Cf. the list and further bibliography in D. Winston, The Wisdom of Solomon (The Anchor Bible; New York: Doubleday, 1979), 15 n.5. 16 For Classical authors, cf. P. Heinisch, Die griechische Philosophie im Buch der Weisheit (Münster: Breslau, 1908); Reese, Hellenistic Influence; for biblical quotes cf. J. Fichtner, “Der AT-Text der Sapientia Salomonis”, ZAW 57 (1939) 155–192. 17 Cf. the tale of Croesus in Herodotus, I.53–55. 18 Cf. III Syb. 75ff., 218ff., 381ff., 573ff.; V Syb. 93ff., 137ff. 19 Cf. A.W. Mair/G.R. Mair (ed.), Callimachus, Hymns and Epigrams, Lycophron, Aratus with an English Translation and Notes (The Loeb Classical Library; Cambridge-London: Heinemann, 21955). 20 M. Hadas, A History of Greek Literature (New York: Columbia University Press, 1950), 192–193. 21 For literary background cf. Winston, The Wisdom of Solomon, 266–267.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

102

Alexis Léonas

meaning (Wis 1:2 G OHCPK\GVCK VQK¨LOJƒC RKUVQWUKPCW™V = “reveals itself to those who do not mistrust [i.e. trust] it”, cf Wis 3:11; 11:17; 12:9, 10; 19:22). The author’s fascination with words further manifests itself in the numerous word plays or paronomasia in the language of the rhetoric schools.22 Wis 2:23: GK™MQPCVJLK™FKCLK™FKQVJVQL>ms.C KFKQVJVQL@G RQKJUGPCW VQP made him image of his proper being [or eternity] Wis 6:10: QK˜HWNCECPVGLQUKYLVC?QŸUKCQUKYSJUQPVCK those who kept the holy observances in a holy way will be made holy Wis 14:5: SGNGKLOJ?C TIC?GK»PCKVC?VJLUQHKCLUQWG TIC you wish that the works of your wisdom would not lie idle

Hyperbata, frequently recurring throughout the book, serve as further evidence of the same creative approach to wording. Wis 14:1: C ITKCOGNNYPFKQFGWGKPMWOCVC litt. rowdy – planning to fare through – waves

J.M. Reese counts 240 such examples – an impressive number for a book of that size. This forms a sharp contrast with the rest of the Septuagint, where occurrences of hyperbaton are extremely rare.23 The repetition of similarly sounding words or endings at the end of each line – a variation of homoioteleuton – is yet another literary artifice often used in the book (e.g. Wis 3:11–12; 7:17–21; 8:19–20; 10:13–15; 14:11– 12). Furthermore, H.St.J. Thackeray has noticed the assimilation in scansion of the verse endings. According to him, “The attempt to assimilate the endings of the UVKZQK runs through the whole book, but is much more evident towards the close, where the writer abandons the more Hebraic manner of the early chapters and gives free play to his own genius.”24 The most frequent ending for couplets is found to be hexametric (~ ~ – –)25 and the same compositional technique finds at least one other exponent in

22 Cf. Herodian’s and Alexandre of Tarsus’ treatment of paronomasia in C. Walz (ed.), Rhetores Graeci. Vol. 8 (Osnabrück: Otto Zeller, 1968 [11832–1836]), 595.13 and 477.3–14. 23 Reese, Hellenistic Influence, 26–27. Cf. Kepper, Hellenistische Bildung, 87. 24 H.St.J. Thackeray, “Rhythm in the Book of Wisdom”, JTS 6 (1905) 233. 25 Thackeray, “Rhythm”, 235.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

The Poetics of Wisdom: Language and Style in the Wisdom of Solomon

103

the Letter to the Hebrews.26 Not only the endings, but also the beginnings of the line get repeated frequently: one is tempted to define this proceeding as anaphora (13:17–19; 15:15; 17:16–18) if it were not also a clear imitation of the Septuagint model (cf. MCK at the start of practically every line in the first 9 chapters, e.g. 4:19; 9:9–12). Besides those larger features of composition, rhetoric ornaments such as chiasmus (1:1, 4, 8; 3:15–16), enumeratio (7:22–23; 14:25–26; 17:17–19), sorites (6:7–20) adorn the text, bearing eloquent witness to the author’s education and art. Bypassing further proofs one can agree with Focke, who declared the Book of Wisdom’s author nothing short of a literary genius.27

2.1 The Septuagint heritage in Wisdom’s writing Yet some aspects of Wisdom’s poetic seem to escape definition in terms of the Greek literary craft. We come now to the type of writing, which finds obvious parallels in the Septuagint Bible, and whose origin and function we must yet try to understand. Parallelism is the most salient “Hebraic” feature of the Book of Wisdom.28 Even if not consistently observed, it is present throughout the whole book, including the later chapters (cf. 17:19–20; 18:3, 10–11). Recourse to this literary device leads to renouncing writing long periods dear to the Greek (which the author of Wisdom shows himself quite capable of producing, cf. 12:3–8; 12:27; 13:11–15), in favour of stringing together of short kola. Short sentences are laid out in a quasi unconnected sequence. Independent kola function as isolated parts within an asyndetic composition. Their inner structure is minimalist, often reduced to subject, object, predicate. The verb is preponderant and makes one think of Hebrew or at least the Septuagint imprint. Recurrent use of MCK at the beginning of each verse also finds immediate correspondence in Septuagint Greek. Another similarly “exogenous” trait is the use of words and expressions typical of Biblical Greek, such as GW NQIGK¨P – “to bless” (14:7; cf. 15:19) or C TGUVQƒP G P Q HSCNOQK¨L (“ac26 Thackeray, “Rhythm”, 236. Thackeray’s conclusions were not unanimously accepted, cf. Focke, Die Entstehung, 54 n.1. 27 Focke, Die Entstehung, 62: „Es ist klar, dass wir hier eine hervorragende, ganz individuelle Begabung vor uns haben, die man wegen der sicheren und gewandten Beherrschung des sprachlichen Materials versucht ist genial zu nennen“. 28 Cf E. Norden’s comparison of parallelisms Hebrew and Greek, Agnostos Theos. Untersuchungen zur Formengeschichte religiöser Rede (Leipzig-Berlin: Teubner, 1913), 254–263 and 355–364.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

104

Alexis Léonas

ceptable” – 9:9) etc. Discussion of these uses is necessarily linked to what has already been said about the richness and innovative aspect of the book’s vocabulary: we will now postpone this analysis and proceed to the questions of more general purport.

2.2 The puzzle of Wisdom’s language and style Assessed even in a summary way, the literary features of the Book of Wisdom raise multiple questions. In the first place, why would a writer possessed of such panoply of Greek rhetoric and literature adopt the parallelising style, typical of the Hebrew Bible, and circulated in Greek through the Septuagint?29 We know quite a few Hellenistic Jewish authors who, in their writings, could easily handle the Greek literary mould. Ezekiel the Tragedian wrote in iambic trimeters, Philo the Epic Poet and the author of the Sentences of Phocylides used hexameters as did the authors of the Sibyllines. Last but not least, Philo of Alexandria produced convoluted philosophical prose: if the author of Wisdom did not take any of these options – as well he could – the choice needs be explained. A minor, but equally weighty question is whether the book of Wisdom should be considered poetry or prose. It is often called poetry, even in the recent scholarship.30 Both Ziegler’s and Rahlfs’ editions present the text visually the way poetry usually appears: the lines are delimited by parallelism, inasmuch as one can discern it through the book. However, despite all the advantages of such presentation, and with all respect for the editors’ work,31 one is tempted to think that within the Greek literary system, the Book of Wisdom would normally be regarded as prose. This conclusion inevitably follows from the absence of meter, which the rigid norm of Antiquity regarded as the exclusive sign of poetry. Determining the nature of the text – in terms of other writing contemporary with it – gives us an insight into the author’s intentions while writing. He was certainly not writing verse, whatever else he was trying to put together. 29 It seems fairly clear that the LXX is the source of the Biblical references in Wisdom (cf. Fichtner), although the author may have had some knowledge of the exegetical traditions based on the MT, cf. P. Enns, Exodus Retold. Ancient Exegesis of the Departure from from Egypt in Wis 10:15–21 and 19:1–9 (HSM 57; Atlanta: Scholars Press, 1997), 75ff and 104ff. 30 E.g. P. Beauchamp speaks of the book being “en vers”: “Epouser la Sagesse ou n’épouser qu’elle? Une énigme du livre de la Sagesse” in M. Gilbert (ed.) La Sagesse de l’Ancien Testament. Journées bibliques 1977 (BETL 51; Leuven: Peeters, 21990 [11979]), 358. 31 Already Epiphanius has put forward the poetic (UVKZJTJL) arrangement of the book, De mens. et pond., 4 (162).

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

The Poetics of Wisdom: Language and Style in the Wisdom of Solomon

105

An additional argument to support this view can be deduced from the development of the biblical parallelism in late Antiquity. The fact is well known that, despite its massive presence in the Bible, parallelism fell into disuse and relative oblivion during the period that elapsed between Ben Sira and the early rabbinic scholarship.32 Already the Septuagint translators are not fully consistent in the use of parallelism, which falters in numerous passages of the Greek version.33 Kugel describes this development as outright “forgetting” of parallelism,34 for rabbinic exegesis normally attributed an independent meaning to both halves of the parallel structure. Thus, the verse of Deut 33:10 “They shall teach Jacob thy ordinances, and Israel thy law” comes to be explained in the Sifre as referring to the two Laws – the written and the oral.35 Having escaped formal description and fixation in Antiquity, parallelism had to wait until the Bishop Lowth gave a definition of its principles in the mid–XVIIIth century.36 Therefore, attributing a poetic intention to the Wisdom’s use of parallelism seems anachronistic and risky. Although the writer of Wisdom obviously had an appreciation for parallelism, we cannot describe it in terms of poetry perception characteristic of his epoch. The author’s recognition of the poetic virtue of parallelism many centuries in advance of Lowth – though not altogether unimaginable, given his literary genius – should perhaps be described in some other terms, which are yet to be found. Thus we come to another crucial question: how do the “Hebraic” elements of Wisdom’s composition fit into the book’s aesthetic system? The issue takes us back to our initial query, to the question why would the whole richness of typically Hellenistic literary device be poured into a text largely divergent from the Greek models? To answer these questions, we will first have to consider what models Hellenistic literature had to offer.

32 I am here repeating the conclusions formulated by J.L. Kugel, The Idea of Biblical Poetry. Parallelism and its History (New Haven-London: Yale University Press, 1981). 33 Even the advocate of Hebrew parallelism in Wisdom, E.D. Reymond recognizes that the semantic parallelism is harder to determine in the LXX than in the Hebrew Bible: “because the Septuagint does not always translate the Hebrew word-pairs literally or consistently” (“The Poetry of the Wisdom of Solomon Reconsidered”, VT 52 [2002] 393 n.42). In other words, that simply means that the translators’s work was not guided by the principle of parallelism. 34 Kugel, Biblical Poetry, 94–95. 35 Cf. Kugel, Biblical Poetry, 98. 36 R. Lowth, De sacra Poesi Hebraeorum praelectiones (London: Oxoniensis Ditata, 1753).

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

106

Alexis Léonas

3. Wisdom of Solomon in the Hellenistic literary context Leaving aside the difficult question of genre, let us look at those Hellenistic writings that provide some basic resemblance with Wisdom. Several models have been considered as providing such literary analogy. While obvious parallels exists in the Septuagint Bible, an important similarity is known to exist between Wisdom and the Isis aretalogies. Less often studied but no less interesting in this context is the comparison with the prose nd hymns of the II c. AD rhetor Aelius Aristides. 3.1 The Septuagint: a source but not a model Given that the relation between Wisdom and the style of the Septuagint will be studied later in this article, I now limit myself to a few general remarks. The impact of the Bible, beyond the paraphrase of Genesis and Exodus in Wis 10–19, is evident from numerous quotes from the Septuagint version. If the influence of the Pentateuch37 and the wisdom writings38 seems rather natural, borrowings from the prophetic books, mostly from Isaiah39 but also from Jeremiah, testify to an extensive biblical culture of the author.40 On the other hand, while the influence of the Bible is undeniable on the level of the content, the formal level presents a strange mix of elements typically biblical or “Hebraic” with the traits characteristically rhetorical and “Greek”. The tendency towards greater Hellenisation can of course be interpreted as a continuation of the literary trends already present within the LXX translation. The Septuagint version of the book of Proverbs is an example of Hellenistic writing with a great degree of refinement and autonomy in its relation to the parent text.41 The Wisdom of Solomon shares many aspects of the Greek Proverbs’ poetics. Extensive use of neologisms (some thirty 37

E.g. Wis 11:4 Ù Deut 8:15; Wis 12:8 Ù Ex 23:28, Deut. 7:20; Wis 16:22; 19:21 Ù Num 11:7. Cf. Focke, Die Entstehung, 66–67; Fichtner, “Der AT-Text“, 155–192. 38 E.g. Wis 5:10–13 Ù Prov. 30:19. Cf. Holmes, Introduction, 520–521, 524, P. Skehan, „Wisdom and the Earlier Wisdom Writings“, in idem, Studies in Israelite Poetry and Wisdom (CBQMS, 1; Worcester: Heffernan Press, 1971), 173–191; R.J. Clifford, “Proverbs as a source for Wisdom of Solomon”, in N. Calduch-Benages/J. Vermeylen (ed.), Treasures of Wisdom. Studies in Ben Sira and the Book of Wisdom. Festschrift Maurice Gilbert (BETL 143; Leuven: Peeters, 1999), 255–263. 39 Wis 2:11 Ù Isa 3:10; Wis 14:3 Ù Isa 43:6; Wis 15:10 Ù Isa 44:20; cf. Fichtner, “Der AT-Text”, 191. 40 Cf. Wis 4:20 Ù Jer 2:19. Even Philo of Alexandria, whose Bible knowledge is beyond dispute, admits having encountered the Book of Jeremiah later in his career (de Cherubim, 48–49). 41 Cf. a comprehensive study of the language and style in D.-M. d’Hamonville, La Bible d’Alexandrie 17: Les Proverbes (Paris: Cerf, 2000), 72ff.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

The Poetics of Wisdom: Language and Style in the Wisdom of Solomon

107

hapaxes) and love of compound words (OPJUKMCMQL, Prov 12:28; MC MQHTQUWPJ, 16:18; HKNCOCTVJOYP 17:19)42 bear witness to kindred literary ambition. Furthermore, interest in the phonetic effects, alliteration and assonance is another feature present in the two books.43 Rhythmic patterns were also discovered in Proverbs, where numerous verses show a tendency to iambic and to hexametric arrangement.44 These resemblances show that the Hellenistic literary aspirations were not unheard of in the circles of the Septuagint translators. However, this analogy does not seem quite sufficient to explain the language and style of a book written originally in Greek. The gap between the Wisdom and the traditional wisdom writing is also important on the narrative level. Unlike Proverbs and most other Wisdom books, Wisdom of Solomon is not a collection of wise sayings, however skilfully arranged. Its speculative scope is overwhelmingly wider: Wisdom attempts to define (rather than just to evoke) Sophia and to portray its manifestations. Moreover, this portrayal amounts to the exegesis of a crucial portion of the Torah. Although wisdom literature knows analogous developments, this philosophical and exegetical stance is on the whole unprecedented. A comparison with other genuinely Greek compositions of Hellenistic Jewish literature raises similar problems in a reverse way, leaving us with the question why did the author not adopt one of the many Greek literary forms, as did Ezekiel the Tragedian, Philo and the authors of the Sentences of Phocylides, III Sibyllines, or the Letter of Aristeas? 3.2 The Isis aretalogies Let us now turn to the Isis aretalogies, which can in fact shed new light on Wisdom’s writing craft.45 Isiac texts were widely spread across the Eastern Mediterranean: inscriptions and papyri have been discovered in Egypt (P.Oxy. 1380,46 hymns of Isidorus in Fayum,47 an invocation of Isis in the

42

Cf. d’Hamonville, Proverbes, 80–81. For Proverbs see d’Hamonville, Proverbes, 99–100. 44 Cf. H. St.J. Thackeray, “The Poetry of the Greek Book of Proverbs” in JTS 13 (1911) 46– 66; d’Hamonville, Proverbes, 92–99. 45 Cf. S. Reinach, “Les arétalogues dans l’Antiquité”, BCH 9 (1885) 257–265; A. Kiefer, Aretalogische Studien (Leipzig: Noske, 1929). For a succinct presentation of extant sources cf. Y. Grandjean, Une nouvelle arétalogie d’Isis à Maronée (EPRO, 49; Leiden: Brill, 1975), 8–11; the most extensive anthology of texts remains W. Peek, Der Isishymnus von Andros und verwandte Texte (Berlin: Weidmann, 1930). 46 B.P. Grenfell/A.S. Hunt, The Oxyrhynchos Papyri XI (London: Egypt Exploration Society, 1915), 190–220. 43

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

108

Alexis Léonas

magical papyri48), on Greek islands (Andros),49 in Thrace (Maronea)50 and in Asia Minor (Kyme).51 Apart from the inscriptions, occasional quotes from the aretalogies emerge in literary works: Diodorus of Sicily cites an inscription in praise of Isis in Bibliotheca historica (I.27,4)52 and more famously Apuleius in Metamorphoses (XI.5,1–5 and XI.25,1–6) pays tribute of praise st and prayer to the goddess.53 Most of these texts date from approximately I – nd st nd II c. AD, although some inscriptions go back to the II –I c. BC and may draw on earlier sources (the text in Diodorus, the Andros inscription, Isidorus’ hymns from Fayum, the Maronea inscription). Leaving aside the theological aspects of relation between Isis and Sophia, we must note a significant difference in purpose of the texts we are trying to compare. Isiac inscriptions form a corpus with a great variety of writing styles. This diversity probably reflects a variety of uses, from cultic to literary.54 While the Isiac texts on a whole appear to have been votive, this is not the case of the Book of Wisdom, which is more of a literary product. Keeping in mind this important distinction, we can concentrate on the similarities existing between these texts. As a matter of fact, many of the issues raised by the study of the corpus of aretalogies would seem strangely familiar to a Wisdom scholar: 1. There is a long-standing debate about the genre and purpose of the Isiac texts: whether they were meant as a kind of religious propaganda55 or chiefly for internal use by Isis worshipers. As in the case of Wisdom, the content of the Isiac texts can be alternatively comprehended as hymn, eulogy, invocation, a theological sum or even – as Werner Peek has playfully suggested – an Evangelion.56 47

E. Bernand (ed.), Inscriptions métriques de l’Egypte gréco-romain. Recherches sur la poésie épigrammatique des Grecs en Égypte (Paris: Les Belles Lettres, 1969), 631–652. 48 PGM II, P.VII, pp. 22–23. 49 Peek, Der Isishymnus. 50 Grandjean, Une nouvelle arétalogie. 51 Reproduced in Peek, Der Isishymnus. Besides Isiac texts, one might add here inscriptions dedicated to other Egyptian deities close to the Isis cycle, cf. V. Longo, Aretalogie nel Mondo Greco: I, Epigrafi e Papiri (Genoa: Istituto di Filologia Classica e Medioevale, 1969). 52 Cf. C.H. Oldfather (ed.), Diodorus of Sicily in twelve volumes. Vol. I, Books I and II, 1–34 (Loeb Classical Library; Cambridge-London: Heinemann, 1960 [11933]), 86–89. 53 Cf. J. Gwyn Griffiths (ed.) Apuleius of Madauros. The Isis Book (Metamorphoses, Book XI) (EPRO 39; Leiden: Brill, 1975). 54 The text of P.Oxy. 1380 is titled “invocation” (G RKMNJUKL), whereas the Maronea inscription presents itself as an enkomion (lines 5, 8). 55 Cf. R. Harder, Karpokrates von Chalcis und die memphitische Isispropaganda (Abhandlungen der Preussische Akademie der Wissenschaften. Phil.-hist. Klasse 14; Berlin: Walter de Gruyter, 1944); J. Bergman, Ich bin Isis: Studien zum memphitischen Hintergrund der griechischen Isisaretalogien (Acta Universitatis Upsaliensis, Historia Religionum 3; Lund: Berlingoka Boktryckeriet, 1968); Longo, Aretalogie, 29–34. 56 Peek, Der Isishymnus, 26 n.1.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

The Poetics of Wisdom: Language and Style in the Wisdom of Solomon

109

2.

The possibility of an Egyptian Vorlage for the Isis texts has also been a subject of heated debate. The thesis of Harder,57 who attempted to establish a common Egyptian ancestor to the aretalogies from Kyme, Ios, Andros and Thessalonike, has found strong opponents in A.D. Nock58 and A.-J. Festugière.59 They stressed the Greek nature of the aretalogies, to the exclusion of any Egyptian original. The Isis texts must have been written directly in Greek by Hellenised Egyptian clerics or commissioned to a Greek scribe for the edification of a Greek speaking audience60. This conclusion was reinforced by the meticulous researches of D. Müller61 who had catalogued the formulae common to various aretalogies in order to contrast them with the genuinely Egyptian material. The outcome turned out to be in favour of independent Greek composition: only about one third of the components thus analysed can claim Egyptian origin, while others are more or less purely Greek,62 Although this view did not long remain uncontested,63 we see that just as with the Wisdom of Solomon, Isis aretalogies offer a “harmonious blend”64 of heterogeneous elements that resists easy reduction to a single source.

3.

Compositional features of the aretalogies also have an interesting bearing on the study of Wisdom. The literary form of the aretalogies varies from overtly poetic, put across by hexameters (the hymns of Isidorus, the Andros inscription) and iambic trimeters (Cyrene inscription), to rhetorically elaborate prose (Maronea inscription – a “prose hymn”65) and beyond that to what one is tempted to describe as “translation Greek” (Kyme, Ios inscriptions). This diversity certainly reflects various degrees of hellenisation of a given religious discourse, even if we refuse to see here a progression from a literal translation to free compositions. From the poetics’ point of view, the advantage of treating these texts synchronically is the possibility of comparing dif57

Harder, Karpokrates von Chalcis. A.D. Nock, in Gnomon, 21 (1949) 221–228. 59 A.J. Festugière, “A propos des arétalogies d’Isis”, HTR 42 (1949) 209–234 (repr. in A.J. Festugière, Etudes de religion grecque et hellénistique [Paris: J. Vrin, 1972], 138–163). 60 Festugière, “A propos des arétalogies”, 232: “Tout conduit à penser que les prêtres égyptiens … ont fait rédiger par un Grec (ou par un des leurs qui savait le grec), pour les Grecs, en conformité avec des idées grecques et un genre littéraire grec, une eulogie d’Isis …”. 61 D. Müller, Ägypten und die griechischen Isis-Aretalogien (Abhandlungen der Sächsischen Akademie der Wissenschaften. Phil.-hist. Klasse 53,1; Berlin: Akademie Verlag, 1961). 62 Cf. Müller, Ägypten, 91. 63 Cf. Bergman, Ich bin Isis. 64 Cf. L.V. Zabkar, Hymns to Isis in Her Temple at Philae (Hanover-London: University Press of New England, 1988), 159. 65 Grandjean, Une nouvelle arétalogie, 49 and 106–110. 58

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

110

Alexis Léonas

ferent literary attempts to handle the sacred. In this perspective, prose composition appears to be no less appropriate a medium than hexametric poetry sanctified by the Greek epic tradition. 4.

A highly interesting trait of many of these texts is the recourse to rare and compound words as well as to periphrastic constructions. This is st particularly evident in the hexametric inscription from Andros (I c. BC). The study of W. Peek has been able to find there 40 cases of neologisms (in a text of 150 lines)66, wide-ranging metonymy (DNCUVQP = “child”, VTQRKL = “ship” etc.)67 and the use of periphrastic constructions, bordering on obscurity (MCTRQVQMQL OCVGT = “the Earth”, C VTCRKVQƒL RNCIMVGKTC = “the Zodiac”).68 Peek sums up the literary impact of this writing style as decorative and pathetic.69 One can conjecture that it was deemed appropriate to express religious content. The predilection for composite adjectives in the Andros inscription (C UCNGWVQL RQNWECNYVQL HKNQSTGUMQL)70 is paralleled in many st other Isiac sources: the Fayum hymns of Isidoros (I c. BC),71 the Kios st st hymn (I c. AD),72 and more significantly in P.Oxy. 1380 (I c. AD),73 which bears a much stronger Egyptian imprint. This wording obviously reflects preoccupation with the titles of Isis, ultimately dependant on native Egyptian models.74 One can even surmise that those models could have triggered the proliferation of compound adjectives in more literally elaborate texts.

5.

A closer look at the Isiac prose reveals further similarities with the Wisdom style. The repetition at the beginning of each new line of the formula G IY‚ GK™OK – otherwise well known to biblical scholars – is the nd most salient feature of the texts from Kyme and Ios (AD I–II c. and d III c.). Besides this recurrence, which is akin to Wisdom’s predilection for anaphora, the composition of the inscriptions is based on short verbal sentences, loosely connected. The asyndetic style is very simi66

Peek, Der Isishymnus, 89. Peek, Isishymnus, 94. 68 Peek, Isishymnus, 95. 69 Peek, Isishymnus, 95. 70 Peek, Isishymnus, 93–94. 71 I.26 – RQNWQPWOQL, II.1 – OGICNYPWOQL, III.34 – OGNCPJHQTQL, see Bernand, Inscriptions métriques, 633–634. 72 Line 2: RQNWUGOPQLZTWUQUVGHCPQL; l. 5: RQNWYPWOQL, cf. Peek, Isishymnus, 139. 73 Lines 9 and 70 – RQNWOQTHQL; l. 54 – MCNNKOQTHQL; l. 55 – C PFTQUYVGKTC; l. 66 – IWRQOQTHQL, cf. Grenfell/Hunt, The Oxyrhynchos Papyri XI, 196–197. 74 Cf. J. Gwynn Griffiths, Plutarch’s de Iside et Osiride (Cardiff: University of Wales Press, 1970), 502–503. Lists of divine names and titles are also well attested in the Greek religious culture, cf. Norden, Agnostos Theos, 144–147. 67

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

The Poetics of Wisdom: Language and Style in the Wisdom of Solomon

111

lar to what we find in large portions of the Wisdom of Solomon. There is yet another resemblance: the Isis texts bear an imprint of the parallelismus membrorum: Aret. Cym. 4: G™IYƒPQOQWLC PSTYRQKLG SGOGP MCK?G PQOQSGVJUCC?QW SGK?LFWPCVCKOGVCSGK¨PCK I gave laws to men and decreed things that none can alter.75 Aret. Cym. 41–42: G™IYƒGK™OKRQNGOQWMWTKC G™IYƒMGTCWPQWMWTKCGK™OK I am ruler of war and ruler of thunderbolt am I.76

Its presence can be easily explained: parallelism is not at all limited to the Bible, it is a hallmark of a large variety of Oriental Poetry, including Egyptian. Whether or not the author of the inscription had in mind any specific Egyptian model, it is important for us that he considered this compositional pattern worthy to be reproduced. In his study of the Kyme inscription,77 P. Roussel concludes that the author had had distinct literary ambitions: he tried to attain solemnity, to use choice terms and generally to give a poetic colouring to his prose.78 Furthermore, P. Roussel estimates that the author attempted to produce a specimen of the hieratic style in imitation of the Egyptian models. Even though he could not have known those models firsthand, the indigenous priests would have been enough to provide inspiration.79 nd

6.

st

The inscription from Maronea (II –I c. BC) offers a different literary approach to what appears to be a similar task. The word G™IMYOKQP occurs twice in its preamble (lines 5 and 8)80 and defines the genre and literary programme81. Prose then becomes the natural medium, highly rhetorical and strongly marked by the embellishments as chiasmus, opposition, parallelism and skilful variation of short and long periods.82 The koinè influence is hardly felt.83 Hiatus is skilfully avoided.84 75

Cf. Peek, Isishymnus, 122. Cf. Peek, Isishymnus, 124, see also lines 51–52. 77 P. Roussel, “Un nouvel hymne à Isis”, REG 42 (1929) 137–168. 78 Roussel, “Un nouvel hymne”, 148. 79 “L’auteur a prétendu donner un spécimen de style hiératique à l’imitation de modèles égyptiens, qu’il ne connaissait pas directement, mais dont quelque exégète indigène avait fourni une interprétation”, Roussel, “Un nouvel hymne”, 148. 80 Cf. also l. 12 and l. 21 in Grandjean, Une nouvelle arétalogie, 17. 81 Cf. Grandjean, Grandjean, Une nouvelle arétalogie, 106–110. 82 Grandjean, Une nouvelle arétalogie, 108. 83 Grandjean, Une nouvelle arétalogie, 109. 84 Grandjean, Une nouvelle arétalogie, 109. 76

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

112

Alexis Léonas

The rhythmic pattern is enforced by introducing often similar metric organisation within each period and at its ending85. Parallelism here is obviously different in origin from the Septuagint samples, but in a purely Greek context its functioning appears remarkably similar. As in Wisdom, it can involve semantic apposition as well as metric and phonetic assimilation: 11. GK™IC?TW˜RG?TVJLG OJLMCNQWOGPJUYVJTKCLJ¤NSGL RYLW˜RG?TVJLK¢FKCLVKOJLQW MC P ? G NSQKL If you came when invoked for the sake of my deliverance, how could you not come for the sake of your own praise? 14–15. MCK?RTYVQPG™RK?VQ?IGPQLJEY VYPG™IMYOKYPRQKJUCOGPQLC TZJ?P VJ?PRTYVJPUQWVQWIGPQWLC TZJP First I am going to deal with your origin, making the beginning of your race the beginning of my praise. 35. UQK?RTQ?LMCVQKMJUKP$K¹IWRVQLG™UVGTZSG UW?OCNKUVCVJL(NNCFQLG™VKOJUCLVC?L $SJPCL To you dwelling in Egypt was pleasing, You cherished Athens most of all Greece.86

The rhetorical accomplishment of the Maronea inscription brings us close to another Hellenistic analogy to Wisdom – the prose hymns of Aelius Aristides. 7.

I will now try to assess the results of comparison between the Isis literature and the Book of Wisdom. The diversity of Isis texts is instructive: it shows how the same content finds expression both in poetry and prose. Instead of a text being translated, we have here a message getting expressed. Inscriptions that at first glance appear as standing closer to translation, actually have the same relation to the core-message as the texts showing more literary acumen. Interestingly, the “Egyptianising” inscriptions are actually of a later date st nd d (Kyme – I –II c. AD, Ios – III c. AD), though this may be due to historical chance. The Egyptian roots become a very interesting factor in the genesis of these texts. Although no textual derivation is possible – at least as long as no new evidence is discovered – we find the idea of a hieratic form of expression actively present. Whether the Egyptian 85 86

Grandjean, Une nouvelle arétalogie, 108, 115–117. Cf. Grandjean, Une nouvelle arétalogie, 17–18 and 115–117.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

The Poetics of Wisdom: Language and Style in the Wisdom of Solomon

113

prototype was real or imaginary,87 the strategies of writing that it had inspired are often comparable. Periphrastic constructions, unusual wording, strange composition, obscurity appear as different means to convey the Egyptian couleur locale. Pompous, somewhat mysterious, ornate style could be achieved with equal success through metric poetry, rhetorical prose or even more exotic “orientalised” prose. Plurality of literal forms allows us to derive the literary specifics of typically Greek texts from an Egyptian archetype, and vice versa to perceive the “exogenous” compositional traits of less standard writings as an extension or development of purely Greek writing craft.88 3.3 Aelius Aristides’ prose hymns nd

Aelius Aristides, a II c. AD rhetor from Smyrna,89 was active both in his native town and across the Eastern Mediterranean (Alexandria, Athens, Rome)90. Of his entire literary heritage we are mainly concerned with the prose hymns he dedicated to various gods of the Hellenistic pantheon – Athena, Heracles, Zeus, Dionysus, Sarapis, Asclepius.91 Although these discourses generally stick to the genre of encomium, Aristides himself often uses the word “hymn” to outline their scope.92 In the preamble of the praise to Sarapis, Aristides goes to great length to justify writing in prose: prose had existed before poetry (Or. 45.8), oracles do not disdain prose (45.7), prose writers also know meter (45.10–12). This captatio benevolentiae is apparently needed to introduce Aristides’ special genre, which he sums up in the end of the discourse as simultaneously a “hymn”, a “thanksgiving” 87 The question arises then, how much actual translation might have been necessary to initiate this development? I think one can agree with Roussel and Festugière that in reality no texttranslation may be required at all. Awareness of the key formulae like G IY G KOK and of the cultic titles of Isis would have been enough to provide ferment for this kind of creative writing. Interestingly, another example of “translation without original” in Antiquity would be the Gospels. 88 That is to say, the binary composition structures in the Maronea inscription can be tentatively explained as echoing the parallelism of the genuinely Egyptian texts. On the other hand, one can just as well consider the parallelism occurring in the inscriptions with more Egyptian colouring (Kyme and Ios) as an outgrowth of the Greek rhetoric techniques, pushed to the extremes. 89 His dates are usually given as AD 117–180-ies, cf. C.A. Behr, “Studies on the biography of Aristides”, in ANRW 2.34.2 (1994) 1140–1233. 90 Cf. A. Boulanger, Aelius Aristide et la sophistique dans la province d’Asie au IIe s. de notre ère (Bibliothèque des Ecoles Françaises d’Athènes et de Rome, 126; Paris: E. de Boccard, 1923 [reprinted 1968]); I. Rutherford, Canons of Style in the Antonine Age. Idea-Theory in its Literary Context (Oxford Classical Monographs; Oxford: Oxford University Press, 1998). 91 Or. 37–46 in B. Keil (ed.), Aelii Aristidis Smyrnaei quae supersunt omnia, Vol. II, Orationes XVII–LIII (Berlin: Weidmann, 1898). 92 Cf. D.A. Russel, “Aristides and the Prose Hymns”, in idem (ed.), Antonine Literature (Oxford: Clarendon Press, 1990), 199–219.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

114

Alexis Léonas

and a “prayer” (45.34). We can guess that Aristides was aware of the peculiar nature of his endeavour from the repetition of similar remarks in other discourses. Thus, in the discourse to Zeus, the literary task is formulated as WŸOPQPG™TGKP'KQLMCK? VCWVCC¢PGWOGVTQW (43.2) – a daring enterprise, to judge by the accompanying scruples (43.2–5).93 The preamble to discourse on Heracles (40.1) mentions those who glorify the hero by “chanting praise in prose” – MCVCNQICFJP C¢FQPVGL. Chanting may be just the right word, as the rhythmic patterns are strongly felt through Aristides’ hymns. These claims suggest that Aristides’ use of eulogy – favourite exercise of the rhetoric schools – combines a peculiar expressive form and a distinctly religious aim.94 His hymn to Zeus is a strong expression of piety in many ways approaching Monotheism. The writing has a religious and/or wider expressive dimension, which served to press the very boundaries of prose style. Such literary approach could in fact have been influenced – as A. Boulanger rightly noticed – by the Isis aretalogies then at large in the Hellenistic world.95 The Book of Wisdom provides an interesting background to this literary development. Although it has never been described as a prose hymn – which it could well be – it exhibits the same tendency to “energised” prose, carefully manipulated to become the vehicle of a religious message. The possibility to include Wisdom in the context of this literary development, somewhere between the Isis aretalogies and Aelius Aristides’ prose hymns, indicates the way forward for the study of its language and style.

4. The position of “Hebraisms” in the poetic system of Wisdom Bearing in mind this literary context, we can now turn to the distinctive characteristics of the book of Wisdom. We will examine the exogenous features of its expression – in brief the “Hebraisms” – and attempt to define their purpose and role within the Wisdom’s poetics. The notion of Hebraism must be redefined in this case. Given the book’s Greek authorship, Hebraisms can no longer be a product of the literal trans93

Cf. Or. 37.1; 44.2, 18. The same particularities reappear in the work of the AD IIId c. rhetor Menander, whose treatise on the epideictic genre includes a model discourse in praise of Apollo (D.A. Russel/N.G. Wilson [ed.], Menander Rhetor. A Commentary [Oxford: Oxford University Press, 1981], the Smintiac oration cf. 207–225). It bears a formal resemblance to the religious texts of Aristides, cf. Norden, Agnostos Theos, 165 n.1. 95 Cf. Boulanger, Aelius Aristide, 309. Boulanger suggests a rapprochement between the prose hymns and P.Oxy. 1380. The Maronea inscription, which would be a much better match, was only discovered in 1969. 94

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

The Poetics of Wisdom: Language and Style in the Wisdom of Solomon

115

lation technique. What we identify as Hebraisms in the Book of Wisdom are words, expressions and elements of composition borrowed from the Septuagint, which we perceive as exotic within the Wisdom’s literary texture. This perception was obviously not shared by the book’s author, who had no scruples about including them in his writing. Nonetheless, the question of their position and their functioning in the Wisdom’s poetics is quite legitimate. Our study of the “Hebraisms” will concentrate on two facets of this phenomenon. The first task will consist in investigating the Septuagint and even Hebrew background of a given usage. Having established its position in the Septuagint linguistic system, I will proceed to analyse its place within the Book of Wisdom proper. The combination of these two investigations will hopefully give us further insight into Wisdom’s literary anatomy. Let us now look at some of the most salient “Hebraic” turns of Wisdom:96 a) the expression WK˜QK?C PSTYRYP – “sons of men” occurring in Wis 9.6 is well known to the biblical student. It obviously corresponds to the Hebrew ]F [PDand to the Aramaic P [PD.97 Despite its familiar Hebrew ring, the expression is relatively rare in the Septuagint (Gen 11:5; 1 Sam 26:19; 2 Sam 7:14; 1 Kings 8:39). Another variation of it is WK˜QƒL C PSTYRQW (Num 23:19; in Jer 2:6 it translates ]F ). The Greekness of this expression is vouchsafed by the formula WK¸GL $Z CKYP, current in Homer (Iliad A.237, 276, 368, 392) and which finds obvious parallels in the Septuagint designations of various groups: WKQK?$OOYP =][P] (1 Kings 11:5), WK˜QK?.YTJ =  TS[PD (1 Chron 26:1) and of course WK˜QK? ,UTCJN (Lev 17:3, 10, 13; Num 9:7; 16:2; 17:2, 6). If we now try to position WK˜QK? C PSTYRYP within Wisdom, we notice first of all that it is preceded by WK˜Q?LVJLRCKFKUMJLUQW – which introduces the theme of common humanness: Wis 9:5–6: QŸVKG™IYƒFQWNQLUQ?LMCK?WK˜Q?LVJLRCKFKUMJLUQW C¢PSTYRQLC USGPJ?LMCK?Q NKIQZTQPKQL MC£PICTVKL¤VGNGKQLG™PWK˜QKLC PSTYRYP GK™LQW FG?PNQIKUSJUGVCK For I am your servant and the son of your maidservant Weak and short-lived human …

96 Most of the “Hebraisms” have received a thorough study and are conveniently repertoried in C.E. Purinton, “Translation Greek in the Wisdom of Solomon”, JBL 47 (1928) 276–304; Zimmermann, “The Book of Wisdom”, 1–27, 101–35. 97 Cf. Purinton, “Translation Greek”, 280; Zimmermann, “The Book of Wisdom”, 16 n.15.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

116

Alexis Léonas And in fact even if one would be perfect among the sons of men, … he will not matter anything.

The semantic range of WK˜QL in Wisdom is practically limited to the expression of divine sonship of the just, be it the generalised righteous person of the first chapters (WK˜Q?LSGQW in Wis 2:18; cf. Wis 5:5; 9:7) or Israel as the people of God in the second half of the book (12:19, 21; 16:10; 16:26; 18:4, 13).98 This probably indicates that WK˜QK? VYPC PSTYRYP is a readymade formula and not a spontaneous use of language. It can be compared with some other expressions Wisdom uses to express common humanity – SPJVQ?L C¢PSTYRQL (7:1), [WZCK¨L C PSTYRYP (14:11). ¹$PSTYRQL is consistently used for denoting certain activities as typically commonplace: G™P Q¢[GK C PSTYRYP (3:4), G¢TIC ZGKTYP C PSTYRYP (13:10), C PSTYRYP G RKPQKC (15:4). WK˜QK? C PSTYRYP fits well with this semantic range. The rarity of this formula in the Septuagint precludes considering it as an unconscious imprint of the former’s language. It can be either a deliberate borrowing, or a reinvention following an existing model, motivated by the author’s own poetic needs. b) The use of the word VTKDQL (always in plural VTKDQK, except Wis 14:3) has been frequently pointed out as a Hebraism.99 The word is used with both spatial (6:16; 14:3) and ethical connotations (“way of doing things” 2:15; 5:7; 6:18; 9:18; 10:10). Both usages are attested in Classical Greek.100 One is actually surprised that VTKDQK would be at all brought up in this context: the word is hardly used in the Septuagint. It occurs in the Pentateuch only once (Gen 49:17). Besides Wisdom, the LXX Psalms (16 occurrences), Proverbs and Isaiah share the most occurrences. Only on three occasions does VTKDQL translate ?TF(Prov 2:20; 30:19; Isa 3:12), which can be contrasted with the massive use of QFQL for ?TF and its cognates in the Septuagint. In fact QFQL is also frequently used in the Book of Wisdom (2:16; 5:6–7; 5:12; 10:17; 12:24; 14:3; 18:23; 19:7). Often it appears together with VTKDQL as its synonym (2:15–16; 5:6–7; 14:3). The meaning of QFQL also carried an ethical connotation already in classical usage: Plato used the formula QFQLDKQW (“way of life”) in the Republic (10.600A)101. Compared to its scarcity in the Septuagint, 98 This formula can be compared to C PSTQRQLVQWSGQW frequently used of the prophets in the historical books, cf. 1 Sam 9:7, 8, 10; 1 Kings 13:4–8, 11, 12, 14; 2 Kings 1:9–13. 99 Cf. Zimmermann, “The Book of Wisdom”, 16 n.15. 100 Cf. LSJ s.v. VTKDQL4. 101 Cf. other examples in A. Léonas, Recherches sur le langage de la Sepante (OBO 211; Fribourg-Göttingen: Vandenhoeck & Ruprecht, 2005), 167.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

The Poetics of Wisdom: Language and Style in the Wisdom of Solomon

117

frequent use of VTKDQK in Wisdom can only illustrate the author’s effort to diversify the standard vocabulary of philosophical parenesis. c) Wis 9:3:

G™PGW SWVJVK[WZJLMTKUKPMTKP

[man should] judge the judgement in soul’s honesty The expression MTKUKP MTKPGKP easily suggests a literal translation of the Hebrew infinitive absolute with a finite verb. Hebrew YRR [ or Aramaic P[F_F could have resulted in this expression.102 It occurs in the Septuagint fairly often and in slightly different forms (MTKUKP MTKPGKP Gen 19:9;103 MTKPQWUKP  MTKUKP Deut 16:18; MTKPCKMTKPYP Judg 11:27).104 If the source of this expression is thus clear, its context in Wisdom is somewhat of a puzzle. Unlike most Septuagint books105 Wisdom uses the “etymological figure” only once on this particular occasion. Searching for similarly construed expressions in Wisdom we discover a whole range of formulae that involve play on the same root: Wis 5:17: NJO[GVCK RCPQRNKCP VQƒP \JNQP CW VQWMCK? QRNQRQKJUGKVJƒPMVKUKP (“he will take the armour of his indignation and will arm the creation); Wis 12:24: VYPRNCPJLQFYPOCMTQVGTQPG™RNCPJSJUCP (“blundered far beyond the ordinary paths of blunder”); Wis 18:3: QFJIQƒP C IPYUVQWQFQKRQTKCL (“guide to the unknown paths”). These expressions appear as attempts to imitate the etymological figure by a writer with no Hebrew knowledge. In other words, the expression MTKžUKP MTKžPGKP, although undoubtedly taken from the Septuagint, actually is a part of a completely different poetic system. The Wisdom author’s perception of Septuagint language can be to some degree illustrated from patristic remarks on the Biblical idiom. Hadrian th – a 5 c. AD representative of the Antiochean school who dedicated 102

Cf. F. Zimmermann, “The Book of Wisdom”, 16 n.15. One is tempted to think that the expressive phrase in Gen 19:9: GK¸LJ»NSGLRCTQKMGKP OJ? MCK? MTKUKP MTKPGKP; could have suggested this formula to our author. He shows keen awareness of the story of Lot, to which he turns twice (Wis 10:6–8 and 19:17). The parallel between the blindness of Sodom and of Egypt in 19:7 is both theologically and hermeneutically apt. 104 Cf. 1 Sam 24:16; 25:39; Job 39:32; Isa 11:4; 16:5; Jer 21:12; 37[30]:13. 105 Cf. H.St.J. Thackeray, “Renderings of the Infinitive Absolute in the LXX”, JTS 9 (1908) 597–601; R. Sollamo, “The LXX Renderings of the Infinitive Absolute used with a paronymous finite verb in the Pentateuch”, in N. Fernández-Marcos (ed.), La Septuaginta en la investigación contemporánea (V Congreso de la IOSCS, Textos y Estudios “Cardenal Cisneros” 34; Madrid: Instituto ‘Arias Montano’ Consejo Superior de Investigaciones Científicas, 1985), 101–113; E. Tov, “Renderings of Combinations of the Infinitive Absolute and Finite Verbs in the Septuagint – their Nature and Distribution”, in D. Fraenkel et al. (ed.), Studien zur Septuaginta – Robert Hanhart zu Ehren (MSU 20; Göttingen: Vandenhoeck & Ruprecht, 1990), 64–73, repr. in E. Tov, The Greek and Hebrew Bible. Collected Essays on the Septuagint (VT.S 72; Leiden-Cologne: Brill, 1999), 257–269. 103

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

118

Alexis Léonas

an entire treatise to the peculiarities of the Septuagint idiom106 – perceived the literal translations of etymological figure as a means of intensity. Introducing a series of biblical quotations featuring this construction, Hadrian explains: MCV  G™RKVCUKP VQƒ CW VQƒ NGIGK – “[the Scripture] repeats the same thing for the sake of intensity”.107 We find here two phenomena: the repetition of the same (VQ? CW VQ? NGIGK), describing the procedure and the statement of its purpose and/or effect (MCV  G RKVCUKP). Going back to Wisdom we can conjecture that the author’s strategy was probably based on a similar observation: expressiveness was somehow linked to creative tautology. In the absence of exact knowledge, multiplying words of the same root within a sentence was thought to be the way and thus contributed to the specific “Hebraising” beauty of the book. d) The expression WRQUVGNGK¨VCKRTQUYRQP in Wis 6:7: QW IC?TWRQUVGNGK¨VCKRTQUYRQPQRCPVYPFGURQVJL

For the ruler of all will not recoil before a person. The passage probably reflects the phrasing of Deut 1:17: QW  OJ? WRQUVGKN RTQUYRQP C PSTYRQW QŸVK J MTKUKL VQW SGQW G UVKP

(“do not retreat before a person’s countenance, for the judgement belongs to God”).108 The probability of a direct quote is supported by the rare use of WRQUVGNNGKP in the Septuagint:109 an unwitting recourse to a typical expression seems unlikely in this case. The use of RTQUYRQP in Wisdom does not allow much guesswork. On one occasion the word is clearly used in a literal sense (17:4, HCUOCVC C OGKFJVQKLRTQUYRQKL), while the only other occurrence is again a construction: 7QVGUVJUGVCKQ FKMCKQLMCVC? RTQUYRQPVYP SNK[CPVYPCW VQP (5:1, “Then the just will stand against [over? by?] the face of his oppressors”). Whether RTQUYRQP has here a clear 106 Cf. F. Goessling, Adrians (K UCIYIJGK LVC?LSGKCLITCHCL aus neu aufgefundenen Handschriften herausgegeben, übersetzt und untersucht (Leipzig-Berlin: Reuther, 1887). Obviously Hadrian reflects the Bible understanding, which has already evolved during the Christian centuries. He was also aware of the profane literary criticism. The fact that he still felt the need to explain some of the Septuagint formulae means that they still had to be reconciled with the standard usage. 107 Isagogè § 104, cf. Goessling, Adrians, 116–118. Similar comments are scattered through the patristic commentaries, cf. Léonas, Recherches, 194–196. (RKVCUKL is a technical terms used by the grammarians to describe the effect of words like OCNNQP, cf Apollonius Dyscolus, On Conjunctions, 223.4, in C. Dalimier (ed.) Apollonius Dyscole, Traité des conjonctions (Histoire des doctrines de l’Antiquité classique 25; Paris: Vrin, 2001), 190. 108 Cf. C. Dogniez/M. Harl, La Bible d’Alexandrie 5: Le Deutéronome (Paris: Cerf, 1992), 115; See also the commentary on Ex 23.21, in A. Le Boulluec/P. Sandevoir, La Bible d’Alexandrie 2 : L’Exode (Paris: Cerf, 1989), 239. 109 The other occurrences are Ex 23:21; Job 13:8; Hab 2:4; Hag 1:10; 3 Macc 5:20.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

The Poetics of Wisdom: Language and Style in the Wisdom of Solomon

119

connotation of “a person” and/or whether it was used precisely for its figurative strength cannot be decided on the basis of this evidence only. 8RQUVGNGK¨VCKRTQUYRQP is closely connected to the formula NCODCPGKP RTQUYRQP, which occurs three times in Ben Sira (4:22, 27; 42:1) and is further related to the Pauline use of RTQUYRQNJ[KC (Rom 2:11; Eph 6:9; Col 3:25, cf. Jas 2:1–9).110 Persistence of this usage is perhaps a further indication of its linguistic vigour. e) Wis 9:9: … G™RKUVCOGPJVKC TGUVQ?PG™PQ HSCNOQK¨LUQW 

MCK?VKGW SG?LG™PG PVQNCKLUQW

Knowing what is good in your eyes And what is right in your commandments.

$TGUVQ?PG™PQ HSCNOQK¨LVKPQL (9:9) is recognised as a clear Hebraism by C.E. Purinton and as an Aramaism by F. Zimmermann.111 However, as it is, this expression is extremely rare in the Septuagint (one isolated occurrence in Jdt 12:14), where the standard expression would be C ICSQP112 or GW SG?L G™P Q HSCNOQK¨L113 (not to mention the most amply represented RQPJTQP114). The passage in Gen 16:6 (UQK C TGUVQ?P ¤), which Purinton, following Grimm,115 brings up as evidence, in fact witnesses quite the contrary. It is an example of the Genesis translator’s avoidance of a Hebraism: this could hardly have incited the author of Wisdom to introduce one into his text. While the possibility of a direct imprint coming from the Hebrew or from the Septuagint is implausible, the synonymic range of C TGUVQ?P G P Q HSCNOQK¨L within Wisdom is worth looking at. We have seen that the context of the verse 9:9 allows immediate understanding of the expression, even if it involved some obscurity. The only two other occurrences of C TGUVQL in Wisdom yield expressions with similar semantic array: C TGUVJ?  J¤P MWTK J [WZJ? CW VQW (4:14); VC? C TGUVC UQW G FKFCZSJUCP C¢PSTYRQK (9:18). The formula G™P Q HSCNOQKL also recurs in Wisdom: G¢FQECPG™PQ HSCNOQK¨LC HTQPYP VGSPCPCK (3:2: “They were deemed dead in the eyes of the fools”). A similar expression, which Wisdom often uses to say the same thing is G™P Q¢[GK (e.g. in close connection with G™P Q HSCNOQK¨L in 3:2, Wis 110 Cf. a succinct study of this expression in P. Harlé/D. Pralon, La Bible d’Alexandrie 3: Le Lévitique (Paris: Cerf, 1988), 166–167 (comm. on Lev 19:15). 111 Equivalent of Hebrew [P[DDYJ, Purinton, “Translation Greek”, 293–294; or Aramaic [P[DDTM, Zimmermann, “The Book of Wisdom, 16 n.15. 112 Cf. Judg 17:6(A); 1 Sam 1:23; 24:5. 113 Cf. Judg 17:6(B), 2 Sam 17:4; 2 Kings 18:3. 114 Cf. 2 Kings 17:2; 21:2, 6, 9, 15, 20; 23:32, 37. 115 Cf. Purinton, “Translation Greek”, 293.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

120

Alexis Léonas

3:4 – G™PQ¢[GKC PSTYRYPMQNCUSYUKP – “if they have been punished in the sight of men”).116 The turn is particularly interesting because G™PQ¢[GK is completely absent from the Septuagint. Meanwhile the word G™PYRKQP, which the Septuagint frequently uses, never occurs in Wisdom.117 Thus we see that the expression C TGUVQ?P G P Q HSCNOQK¨L in Wisdom is a part of a linguistic or stylistic pattern which draws on the Septuagint, but is not fully dependant on it. It appears within the range of expressive options beyond the LXX scope (as the use of C TGUVQL or G™PQ¢[GK). The expressive range gets extended towards more exotic formulae eventually found in the Septuagint. Their free imitation explains the making of at least some of the Wisdom’s Hebraisms. f) The expression G™P CRNQVJVK MCTFKCL \JVJUCVG CW VQP – “seek Him in the singleness of heart” – in Wis 1:1 has been habitually cited as a Hebraism.118 This verdict is surprising as this formula practically never occurs in the Septuagint (except once in 1 Chron 29:17).119 It may in fact be inspired by the commandment to love God G E QŸNJL VJLMCTFKCLUQW in Deut 6:5 and by the injunction to worship Him with the same attitude in Deut 10:12. Deuteronomy also explains that God can only be found when sought sincerely – QŸVCP G M\JVJUJVG CW VQ?P G™E QŸNJL VJL MCTFKCL Deut 4:29120 (see also Deut 26:16; 30:2, 6, 10). Wisdom’s awareness of this background becomes equally apparent in the prayer of Solomon (Wis 9:1ff), which he utters G™EQŸNJLVJLMCTFKCLOQW (8:21). Otherwise, Wisdom’s use of MCTFKC varies between the literal (the reason’s bodily residence, 2:2) and the wider figurative use: MCTFKC appears in a parallelism with G™NRKL and DKQL in Wis 15:10. The expression G™PCRNQVJVKMCTFKCL certainly aims at imitating Septuagint idiom, but the formulation itself can be credited to the creativity of the author of Wisdom. g) The expression G™P ZGKTK is a phrase frequent in the Septuagint and has been amply dealt with in the scholarship. The widespread LXX occurences would be impossible to discuss in this note, and I will limit myself to an overview of the use of ZGKT in Wisdom. The word is extensively used as part of various constructions, which makes as116

Other occurrences with the same connotation are Wis 8:11; 14:7; 15:19. Although Wisdom does use G™PCPVKQP (7:9; 11:22). 118 Purinton, “Translation Greek”, 281; Zimmermann, “The Book of Wisdom”, 16 n.15. 119 Cf. J. Amstutz, Aplotes. Eine begriffsgeschichtliche Studie zum jüdisch-christlichen Griechisch (Bonn: P. Hanstein, 1968). 120 The passage in Deut 4.29 is reworded as G PQŸNJ"MCTFKC in Jer 36(29):13. 117

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

The Poetics of Wisdom: Language and Style in the Wisdom of Solomon

121

sessing its proper sense difficult.121 Expressions like G¢TIC ZGKTYP (1:12; 13:10 bis; cf. 13:19) or RQPQK ZGKTYP (8:18) probably presuppose a literal sense for ZGKT. The same can be said for constructions as TWUGVCK  G™M ZGKTQ?L C PSGUVJMQVYP (2:18 cf. 5:16) or RNQWVQL G™P ZGTUK?P CW™VJL (7:11 cf. 7:16, 19:3). However already their analogies in 5.16 (FKCFJOCG MZGKTQ?LMWTKQW) and 7:16 (G P ZGKTK? CW VQWJOGKL cf. same in 3:1) have a strong figurative aspect. Metonymy is clear in 1:16 (VCK¨L ZGTUK?P MCK? VQK¨L NQIQKL – “[summon death] by deed and by word”) and in a different construction in 9:16 – VC?G™PZGTUK?PGWTKUMQOGPOGVC?RQPQW (“we discover with difficulty what is at hand” i.e. the evident things). The degree of figurativeness becomes even more difficult to grasp when we come to ZGKT qua attribute of God (J RCPVQFWPCOQLUQWZGKT, 11:17, cf. 10:20, 16:15). Phrases like “it is impossible to escape your hand” (16:15), offer the reader a free choice between anthropomorphic and metaphorical understanding. Instrumental connotation of ZGKT is amply present in Wisdom where it gets expressed in several different ways: FKC? ZGKTYP (12:6), V ZGKTK (14:6; 19:6) and G™P ZGKTK (11:1).122 The latter formula is most often associated with the literalism in the Septuagint. It often appears there as an etymologising rendering of the Hebrew instrumental preposition F[D. However, Wisdom makes the meaning of ZGKT so blurred as to exclude any exact replication of the Septuagint calque. At the most one can say that the author was aware of the prepositional use of ZGKT in the LXX texts and tried to imitate some of its features. Wis 3:14 is a good illustration of his willingness to use G™PZGKTK in a context where it does not contribute much sense: GW™PQWZQL Q OJ? G TICUCOGPQL G P ZGKTK? C PQOJOC. Does it mean “eunuch who has not done unlawfulness by his hands” or “in his hands?”123 Another indication of Wisdom’s creative approach to the Septuagint heritage is the use of epithets for ZGKT when it comes in a construction: ZGKTQ?L G¢TIQP C TZCKCL (“product of ancient handwork” 13:10) or G™TIC\GVCK ZGTUK?P C PQOQKL (“makes with lawless hands” 15:17, cf. 16:15). Prepositional use of ZGKT in the LXX does not know such embellishments. Wisdom’s author was aware of the peculiar use of expressions with ZGKT in the Septuagint, which he tried to reproduce to his best knowledge and in accordance with his hellenistic taste. 121

Only one occurrence in Wis 8:12 yields the fully literal meaning. Wis 11:1: (W™QFYUGPVC?G¢TICCW™VYPG™PZGKTK?RTQHJVQWCIKQW. 123 Or perhaps even “against the Hand [of God]” because the parallel line goes OJFG? G™P SWOJSGK?LMCVC?VQWMWTKQWRQPJTC. 122

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

122

Alexis Léonas

I abandon here the case by case analysis of Hebraisms in order to proceed to more general remarks on style. I am fully aware that there is much more to be said on the matter. Expressions like G¢MVKUGP GK™L VQ? GK»PCK (1:14), NQIK\GUSCK GK™L VK (2:16), RNJTQWP ZTQPQWL (4:13), GW™NQIGKP (14:7, cf. 15:19) and many others deserve careful study and will reveal more about the Wisdom’s writing skill. The use of the particle G™P in Wisdom offers an interesting insight into the mechanics of the Septuagint influence. We have already encountered it in several constructions, like C TGUVQ?PG™PQ HSCNOQKLG™PQ¢[GKG™PZGKTKG™P CRNQVJVK MCTFKCL. It is noteworthy, that Wisdom never replicates the instrumental use of G™P, characteristic of Septuagint language. In fact, we know that this use was a stumbling block to ancient readers124. Even after several centuries of Bible reading in the Church, Hadrian’s Isagogè is obliged to comment: Q OCMCTKQL'CWKFVJ" G™PC PVK? VJLUW?PMGZTJVCK (“the blessed David uses ‘in’ instead of ‘with’”).125 While this connotation would be understood with difficulty, the idea of inclusion would be extended beyond spatial/temporal meanings to denote condition, state, circumstance. In this perspective it is interesting to observe that Wisdom multiplies cases where the use of G™P would be appear as untypically extensive126 without becoming altogether instrumental. Besides the examples already quoted we encounter expressions as 1:12: G PRNCP\YJL, 4:6: G P G EGVCUO 8:18: G™P RQPQKL ZGKTYP 9:6: G™P WK˜QKL C PSTYRYP, 10:10: GW RQTJUGPCW VQ?PG™POQZSQKL, 10:16: G™PVGTCUKMCK? UJOGKQKL, 14:30: G P FQN, 16:22: RWT HNGIQOGPQP G™P VJ" ZCNC\, 18:9: G™PQOQPQK, 18:12: G™PGPKƒ Q PQOCVKSCPCVQWMVN. The rich number of similar cases probably stems from the attempt at imitating the sacred tongue without outstepping any Greek linguistic norm. We see generally that the Wisdom’s treatment of the Septuagint idiom is far from being indiscriminate. The author picked only those features of style that matched his aesthetic norm. Several characteristic traits of Septuagint language – no matter how prominent in the Greek Bible – were left out. Thus neologisms and extension (sometimes blurring) of the words’ meaning were regarded as acceptable, while Hebrew personal names were uniformly rejected. Among the divine names, only those were retained that 124 Cf. A. Jannaris, A Historical Greek Grammar (London: Macmillan & Co., 1897), §1562– 3; F.M. Abel, Grammaire du grec biblique suivie d’un choix de papyrus (Paris: Gabalda, 1927), 212; H. Rosén, L’Hébreu et ses rapports avec le monde classique. Essai d’évaluation culturelle (Comptes rendus du Groupe linguistique d’études chamito-sémitiques, Suppl. 7; Paris: Geuthner, 1979), 63. 125 Isagogè, § 105, cf. Goessling, Adrians, 118. 126 Cf. the combination of abstract and concrete in Wis 7.4: G P URCTICPQKL C PGVTCHJP MCK?HTQPVKUKP– “I was nursed among swaddlig clothes and care”.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

The Poetics of Wisdom: Language and Style in the Wisdom of Solomon

123

met Hellenistic standards.127 This attitude to barbarian names is otherwise well attested in the Hellenistic world: the model could be found in Plato, whose exposition of the originally Egyptian story of Atlantis in Critias is accompanied by a remark that all the names are hellenised (Critias 113A– B).128 Various forms of figura etymologica – prevalent throughout the Septuagint – never occur in Wisdom. Instead, there is frequent word play involving the same or similarly sounding roots: evidence of an attempt to create the same effect by other means. In several cases one can feel that the author was trying to imitate the Septuagint style, only to come up with what one can call Ersatz-Hebraisms. Formulae like G™P G˜PKƒ Q PQOCVK SCPCVQW (18:12), expressions using the word ZGKT, the turn G™P Q¢[GK and more generally the extensive use of the particle G P seem to exemplify this tendency. The ultimate source of the “Hebraising” expressions is obviously the Septuagint, providing obscure words (although in a text where neologisms abound their strangeness would be attenuated129), quotations (as Wis 6:7 Ù Deut 1:17), inspiration for strange formulae (as variety of expressions with ZGKT) and even a specific compositional structure. We come thus to the composition using parallelism. The recent study of parallelism in Wis 1:1–15 by E.D. Reymond offers an interesting perspective on this subject matter. One can share Reymond’s admiration for the variety within parallelism (semantic, phonetic, grammatical), but not his recognition of this writing style as the first-hand biblical parallelismus membrorum.130 It appears rather as an attempt to imitate the biblical style by an author who had no articulate idea of this style’s actual workings. The variety of types of parallelism in Wisdom indicates the uncertainty in the author’s approach. Having noticed this particular trait of the Septuagint diction, the author of Wisdom reproduced it indiscriminately on different levels – semantic, phonetic, grammatical. No wonder that among the 27 semantic parallels Reymond studied in Wisdom 1:1–15, more than half never occur in the Septuagint.131 He also admits that the phonetic parallels – common stock in Wisdom – are on the whole less frequent in the Hebrew 127 Cf. P.-M. Bogaert, “.WTKQL, le nom incommunicable (Sg 14.21)”, in N. CalduchBenages/J. Vermeylen (ed.), Treasures of Wisdom. Studies in Ben Sira and the Book of Wisdom. Festschrift Maurice Gilbert (BETL 143; Leuven: Peeters, 1999), 387–394; Winston, The Wisdom of Solomon, 249. 128 Cf. Strabo, Geography, 3.3.7; Josephus, AJ, 1.129; more reference in Léonas, Recherches, 104–111. 129 E.g. G™RKUMQRJ3:6, UMCPFCNQP 14:11, GW™NQIKC15:19. 130 E.D. Reymond, “The Poetry of the Wisdom of Solomon Reconsidered”, VT 52 (2002) 385–399. 131 14 out of 27, Reymond, “The Poetry”, 394.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

124

Alexis Léonas

Bible.132 Furthermore, the grammatical and semantic patterns, instead of reinforcing each other as in the Hebrew poetry, in Wisdom tend to “go their own ways”.133 Wisdom’s “remake” of the Biblical poetry thus appears as largely creative and experimental. Vagueness of this poetic method134 could contribute to explaining the occasional disappearances of parallelism towards the end of the book. There is also a certain similarity between the literary effort in Wisdom and the attempts of some of the Isis aretalogies to reproduce an Egyptian style. There the parallelism also makes occasional appearance without becoming prevalent.

5. Conclusions All in all, we find in Wisdom a conscious attempt to imitate the Septuagint idiom as it could have been seen from within the Greek linguistic and literary system. In a somewhat anachronistic way one could call Wisdom a pastiche of the Septuagint. Pastiche presupposes a deliberate literary imitation with specifically aesthetic aims. In a limited sense, this is true of Wisdom, which offers an artful remake of the Septuagint’s language and style.135 One can go even further than that, as the imitation also operates on the level of the content: after all, the book gives a paraphrase of the most essential portion of the Pentateuch. One of the main implications of this conclusion is that our view of the attractiveness of the LXX language and style for Greek readers must stand corrected. Instead of being more or less a literary outcast, the Septuagint should be regarded as having a literary appeal of its own. This fact could already be guessed from Ps. Longinus’ treatment of the first lines of Genesis in the treatise On the Sublime.136 Ps. Longinus included a quotation from 132

Reymond, “The Poetry”, 397. Reymond, “The Poetry”, 396. 134 A reader with no systematic knowledge of Hebrew would normally perceive translation in a way quite different from the scholarly study. For instance, Hadrian’s Isagogè, already quoted here, describes the parallelism in much the same terms as the renderings of etymological figure: both involve repetition and thus both become labeled VCWVQNQIKC. Cf. Isagogè §100 and §104, in Goessling, Adrians, 116–118. 135 Wisdom’s attitude to the Septuagint idiom must be further explored in relation to the main currents within the Hellenistic literacy: atticism and asianism. Unsurprisingly, M. Kepper considers Wisdom to be closer to asianism (Hellenistische Bildung, 88–95). 136 De sublimitate 9.9. Cf. D.A. Russel (ed.), Libellus de sublimitate Dionisio Longino fere adscriptus (Scriptorum classicorum bibliotheca oxoniensis; Oxford: Clarendon Press, 1968); the Septuagint connection has been studied by G. Dorival, “La Bible des Septante chez les auteurs païens (jusqu’au Pseudo-Longin), in Cahiers de Biblia Patristica, 1 (Paris-Strasbourg: Centre d’analyse et de documentation patristiques, 1987), 9–26. 133

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

The Poetics of Wisdom: Language and Style in the Wisdom of Solomon

125

LXX Genesis into his discussion of literary passages – mainly Homeric – that inspire religious awe. Although long known to scholarship, this judgest ment by an I c. aesthete stood alone and used to be taken as eccentricity, if not a token of the author’s Jewish origin. Now the Book of Wisdom can substantiate the evidence of a cultivated reader’s fascination with the Greek Bible. Wisdom’s attempt to copy the LXX language is coming from thoroughly Hellenised circles and constitutes yet another proof of this idiom’s literary success. Another implication of this conclusion is the possibility of using Wisdom as a touchstone to evaluate the “Greekness” of various Septuagint texts. Obviously reflecting the literary taste of a given epoch, Wisdom provides a useful criterion for the study of Septuagint Greek. Its acceptance or rejection of the Septuagint wording is an indication of the latter’s relative difficulty or conformity with the Greek literary use. In this perspective, Wisdom’s parallels with Symmachus, discovered in great number by J. Fichtner,137 merit especial attention. Little as is known of Symmachus’ literary programme,138 it certainly involved reshaping Septuagint language. By no means a total rejection, but a remaking not dissimilar to the literary project of Wisdom. In both cases the aim can be broadly described as “Hellenisation”, although it remains to be seen whether that meant reducing obscurity or cultivating it. In any case, treatment of Septuagint language by the two writers certainly deserves comparison and further study. Another exponent of the same literary approach is the author of the Gospel of Luke. Leaving aside the intricacies of his relation to the other Synoptics, the purely linguistic aspect of Luke’s effort very often consists in smoothing, i.e. Hellenising the formulations found in Mark and Matthew. Preliminary comparisons done by E. Norden139 are of extreme interest for the literary history of the language of the Septuagint.140 Study of Wisdom also has interesting implications for Septuagint study in general. Much of Wisdom’s literary specificity is resulted through the imitation of the Septuagint. Between the “hieratic” Septuagint and the standard Greek literary forms, Wisdom discovered its own way of sacred writing. The same intermediate position between the archaic and the modern is 137

Fichtner, “Der AT-Text”, 192. Cf. A. Salvesen, Symmachus in the Pentateuch (JSSM 15; Manchester: University of Manchester, 1991). 139 Norden, Agnostos Theos, 357–360; idem, Die Antike Kunstprosa vom VI. Jahrhundert v. Chr. bis in die Zeit der Renaissance, I–II (Leipzig-Berlin: Teubner, 1918), 485–492. 140 The Book of Judith is another witness of this literary tendency, cf. J. Joosten, “The Original language and Historical Milieu of the Book of Judith”, in M. Bar-Asher/E. Tov (ed.) Megillot 5–6: A Festschrift for Devorah Dimant (Jerusalem: Bialik Institute and Haifa University Press, 2007), 159–176. I thank Prof Joosten for indicating to me this study, whose conclusions suggest several interesting parallels with Wisdom. 138

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

126

Alexis Léonas

shared by the Isis aretalogies, loosely dependant on their Egyptian sources. The Isis corpus demonstrates how little original text is needed to result imitation. Imitation is a creative work and does not actually require for the original to exist: imitations generate further imitations. If we compare the attitude of Wisdom’s author to the Septuagint and the relation of an Isis aretalogy writer to his/her Egyptian models, the literary aspects of their approach would be much the same. The primary difference is that the literary model of Wisdom is well preserved. With Wisdom’s author we can be sure: s/he imitated Septuagint style. A further question can then be asked: what style did the Septuagint imitate? In other words, what models oriented the Seventy’s choice of style? A simplistic, self-evident answer would be to say that the LXX idiom is the reflection of the Hebrew Vorlage. However, would it be right to regard the Septuagint as the natural outcome of a minimal effort at translation? Perhaps the literary form of Septuagint itself should not be regarded as all that natural. We must not forget that any translation’s poetics are not fully determined by the original. One is tempted to think that the early translators, in their choice of form, were guided by the same concerns as we find in Wisdom. Oppositions as archaic – novel, hieratic – profane, poetic – prosaic, pathetic – commonplace could have influenced the Septuagint’s expression from the very beginning, shaping its language and style to a much greater extent than we now admit. The impact of the Hebrew text-source on this choice of form – in itself undeniable – needs to be carefully measured. The example of Wisdom gives us an insight on the interaction of the expressive forms and of the content inspiring them. Thus, Wisdom’ poetics helps us to put in context the attitude of the LXX translators to their work. It allows to reassess the degree of creativity involved in it. We come to see that translation work entails not merely reproducing a given text, but finding expression for a given content.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Katrin Hauspie

Periphrastic Tense Forms withGK™OK andIKIPQOCK in the Septuagint of Ezekiel1 1. Definition Periphrasis is the use of separate words to express a relationship that is otherwise expressed by one single word, such as a verb, preposition, adverb or noun. The present study focuses on verbal periphrasis, i.e. circumlocution of a verb form. In classical Greek grammars verbal periphrasis is mentioned briefly alongside the use of the perfect tenses, e.g. for constructions like DGDQW NGWOGPQL Y®.2 The passive perfect and pluperfect indicative of mute and liquid verbs, as well as of verbs whose perfect stem ends in a consonant (as far as the third person plural is concerned), are usually formed by the use of a participle withGK™OK. Likewise the active and passive subjunctive and optative of the whole paradigm, as well as the future perfect of some of these verbs, act the same way. This use of a participle with GK™OK produces a periphrasis (“Umschreibung”). Kühner & Gerth also talk of “Umschreibung” when a present, perfect or aorist participle is used with the copula GK™OK, as opposed to a simple verb form, e.g. QW¡VQKOCKPQOGPQKGK™UKP.3 The latter periphrasis disconnects the verb from its complements, thus emphasizing the verb. For both types, the periphrasis is formed by the use of the participle with the verb GK™OK.4 In the latter case GK™OK functions as copula and the participle as predicate (like an adjective or noun do); in the 1 This contribution is based on a detailed study of my doctoral dissertation La version de la Septante d’Ézéchiel: Traduction annotée d’Ez 1–24 et étude du grec d’Ézéchiel par une sélection de particularités lexicales et grammaticales (Leuven, 2002; promotor: Prof. Dr. Willy Clarysse, co-promotor: Prof. Dr. Johan Lust), entitled “La périphrase avec GK™OK et IKIPQOCK”, pp. 364–392. I thank Antoon Ternier for his most useful remarks, and Sergio Scatolini for his kindness to read and correct the English, both of which contributed to the improvement of this article. 2 R. Kühner / F. Blass, Ausführliche Grammatik der griechischen Sprache. I: Elementarund Formenlehre 2 (Hannover: Hahnsche Buchhandlung, 31992), 98–100; 111. This phenomenon is denoted by the term “Umschreibung”. See also R. Kühner / B. Gerth, Ausführliche Grammatik der griechischen Sprache. II: Satzlehre 1 (Hannover: Hahnsche Buchhandlung, 31992 (=11898), 39. 3 Kühner / Gerth, Satzlehre 1, 38–39. See also E. Mayser, Grammatik der griechischen Papyri aus der Ptolemäerzeit mit Einschluss der gleichzeitigen Ostraka und der in Ägypten verfassten Inschriften. II: Satzlehre 1 (Berlin/Leipzig: Walter de Gruyter, 1926), 223. 4 Porter resumes the research conducted on periphrasis (S. E. Porter, Verbal Aspect in the Greek of the New Testament, with Reference to Tense and Mood [Studies in Biblical Greek 1; New York: Peter Lang, 21993], 447–8).

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

128

Katrin Hauspie

former case GK™OK functions as auxiliary. The combination of GK™OK with a participle replaces in both cases a synthetic, simple verb form (QW¡VQKOCK PQOGPQKž GK™UKP = QW¡VQK OCKžPQPVCK). They represent however two distinct kinds of periphrasis, as a result of the different function of the verb GK™OK. As ‘periphrasis’ is a combination of words equivalent to a single word, verbal periphrasis can accordingly be replaced by a simple verb form. However, some – consonantal – verbs supplement their missing forms by means of a periphrastic verb form, e.g. RGRNGIOGPQK GK™UK(P) and J IIGNOGPQK GK™UK(P).5 Fanning describes “periphrasis with the participle” as the combination of a participle with another verb form, so that they together function as a unit, equivalent or nearly equivalent to a simple (or monolectic) verb form.6 He examines periphrases in the New Testament, formed by a present or perfect participle and the verb GK™OK.7 Evans defines verbal periphrasis as the combination of auxiliary and participle or infinitive, equivalent to a synthetic verb form.8 As a result of the formal feature of two separate words and in opposition to a synthetic tense form, he calls them analytic tense forms, after the terminology of Conybeare & Stock ‘analytic tenses’.9 Evans also sees the phrases made up ofOGNNY andSGNY followed by an infinitive as examples of verbal periphrasis expressing future viz. purpose.10 In these cases OGNNY and SGNY act as auxiliaries.11 The present study seeks to undertake the treatment of verbal periphrasis defined as a group of words formed by an 5 Kühner / Blass, Elementar- und Formenlehre 2, 154–5 (mute verbs) and 169–70 (liquid verbs). In some dialects and in some authors the simple form of the third person plural ending in -CVCK, -CVQ is utilized (Kühner / Blass, Elementar- und Formenlehre 2, 78); given the simple form ending in -CVCK, -CVQ, the forms of the type RGRNGIOGPQKGK™UK(P) and J IIGNOGPQKGK™UK(P) are in fact periphrases equivalent to a simple verbal form, out of use or rarely used. See also note 2. 6 B.M. Fanning, Verbal Aspect in New Testament Greek (Oxford: Clarendon Press, 1990), 310. 7 He briefly mentions the few periphrases formed by the verb IKIPQOCK (Fanning, Verbal Aspect, 310 n. 255). 8 T. Evans, Verbal Syntax in the Greek Pentateuch (Oxford: Oxford University Press, 2001), 221 and 256. 9 F.C. Conybeare / St. G. Stock, A Grammar of Septuagint Greek (Grand Rapids: Zondervan Publishing House, 21981 [= reprint from Selections from the Septuagint, according to the texte of Swete, published in 1905 by Ginn and Company, Boston (Grand Rapids: Zondervan, 1980)]), §72. 10 Verbal Syntax, 227–9. He admits thatSGNY is not fully understood as an auxiliary in the Pentateuch. The examples rather point to a verbum volendi meaning “to want”, not yet to an auxiliary meaning “to intend to”. 4GNY as auxiliary features in later Greek (on p. 229). We classify all cases of the verbSGNY with infinitive in the Septuagint of Ezekiel under the completive clauses after a verbum volendi. 11 For a detailed study ofOGNNY as modal verb, see L. Basset, Les emplois périphrastiques du verbe OGNNY. Étude de linguistique grecque et essai de linguistique générale (Collection de la Maison de l’Orient méditerranéen ancien 7, Série philologique 1; Lyon: Maison de l’Orient, 1979), especially on pp. 139–230.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Periphrastic Tense Forms withGK™OK andIKžIPQOCK in Ezekiel

129

auxiliary and a participle;12 this is a basic decision that is furthermore grounded on the absence ofOGNNY andSGNY followed by an infinitive in the Septuagint of Ezekiel.13 In a way our definition comes close to the point of view of Aerts. He calls the combination of GK™OK and G¹ZY with participle, expressing an elementary verbal conception, a periphrasis. Some examples may illustrate this: J¤PFKFCUMYP= G™FKFCUMGP, IGITCOOGPQPG™UVK= IGITCRVCK, MJTWECL G¹ZGK =MGMJTWZGP.14 As a result such a periphrasis consists essentially of a participle, and is equivalent to a simple verb form.15 Porter defines periphrasis in a different way. For him, the auxiliary figuring in the periphrastic construction is “an aspectually vague auxiliary” and the participle “not only must be grammatically in suitable agreement with the auxiliary but must be adjacent to it”.16 Consequently, auxiliary and participle must be close to each other. Equivalence with a simple verb form is not required. Moreover, Porter criticizes this equivalence, what he calls synonymity. Only denotative synonymy can be meant, not absolute synonymy: the periphrasis and the simple verb form refer to the same reality (‘denotative synonymy’), even to a similar context (‘cognitive synonymy’) but they are not semantically identical. When a language distinguishes between these two verbal forms, there must be a semantic difference as well. Porter also emphasizes the juxtaposition of auxiliary and participle, in 12

This opinion is also shared by Fanning, Verbal Aspect, 309 n. 253. The verbOGNNY does not occur in the Septuagint of Ezekiel. 14 W. Aerts, Periphrastica. An Investigation into the Use ofGK™OK and G™ZY as Auxiliaries or Pseudo-auxiliaries in Greek from Homer up to the Present Day (Amsterdam: Adolf M. Hakkert, 1965), 2–3. 15 At the same time Aerts extends (as the title of his work already suggests) the category of auxiliaries to the verb G¢ZY. At this point we do not agree with Aerts. He deals with this type of periphrasis under the influence of modern languages in which the verb “to have” clearly functions as auxiliary in the perfect tenses (Periphrastica, 128: “the obvious question from the point of view of the modern languages is how G¢ZY is used as auxiliary verb in Greek”). The construction G¢ZY used in connection with a participle with the same sense as the perfect, appears, according to the examples quoted by Aerts, mostly in tragic poets, with a few examples in prose (Periphrastica, 128–77). Kühner / Gerth also deal withG¢ZY plus participle, but they refuse to call it a periphrasis; it is a false periphrasis (“eine scheinbare Umschreibung”) that is not equivalent to a simple form, but it does justice to a continuous situation as the result of an action (R. Kühner / B. Gerth, Ausführliche Grammatik der griechischen Sprache. II: Satzlehre 2 [Hannover: Hahnsche Buchhandlung, 31992], 61–2). Porter too does not agree to considerG¢ZY in connection with a participle a periphrastic construction (Verbal Aspect, 490: “Thus it is linguistically implausible to posit the use of G¢ZY in periphrastic constructions in classical Greek“). He criticizes the label periphrasis attributed to the examples quoted by Aerts; in most of the cases there is no question of a weakened sense of the (alleged) auxiliary G¢ZY (Verbal Aspect, 489–90). The alleged periphrasis G¢ZY with participle being equivalent to a perfect tense is all but a characteristic in classical Greek. This construction with G¢ZY is absent from the Septuagint of Ezekiel; therefore we do not investigate this construction in more detail here. 16 Porter, Verbal Aspect, 452–3. This paragraph resumes pp. 452–4. 13

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

130

Katrin Hauspie

between which only complements modifying the participle can appear. Other complements related to the auxiliary inserted between the participle and auxiliary make the latter more independent, undermining the perihrastic construction. The definition of Fanning comes close to ours. We do not strictly follow the vision of Evans, but we also countIKIPQOCK with participle among the periphrases.17 Some of them act as periphrasis equivalent to a simple verb form in the Septuagint of Ezekiel. We adopt from Porter the claim that the verb being an auxiliary has a weakened sense; its function is reduced to a purely grammatical tool (serving to form a composed tense). As a result the present study does not cover periphrases like OCKPQOGPQKGK™UKP=OCKPQP VCK. The distinction, made by Aerts, between substitute periphrasis, suppletive periphrasis and expressive periphrasis,18 may offer an answer to this aporia. Periphrases like OCKPQOGPQK GK™UKP = OCKPQPVCK are not fairly compatible with an equivalent simple verb form, as if they would communicate exactly the same sense (substitute periphrasis), and neither do they supplement a monolectic form which is no longer (if ever) extant in the verbal paradigm (suppletive periphrasis). They disconnect the verb from its complements, and thus create a more independent verb with a stronger emphasis given than a simple verb form could do (expressive periphrasis). This is definitely true when the simple verb form is the common form. Fanning notices in such periphrases a sense which is absent in the corresponding simple form.19 This semantic difference between periphrasis and simple form is enough for Evans to disregard this category in his study.20 Equivalence with a simple verb form is a fundamental point in his definition of periphrasis. However, Porter, insisted that this criterion should be abandoned. Due to the formal correspondance between periphrasis (participle with GK™OK and IKIPQOCK) and likely constructions with (the copula) GK™OK andIKIPQOCK, it is not always easy to categorically classify a periphrasis as substitute or suppletive (which we shall henceforward call substitute periphrasis)21 or as

17

The objections put forward by Porter against this type of periphrasis are treated below. The terminology ‘substitute’, ‘suppletive’ and ‘expressive’ are taken from Aerts, Periphrastica, 3. 19 Verbal Aspect, 310: “Finally, there is ‘expressive’ periphrasis, when the periphrastic phrase provides a sense which a parallel monolectic form does not possess.” 20 Evans, Verbal Syntax, 221–2. 21 We already demonstrated that suppletive periphrasis in fact draws back on a non-, or no longer extant simple verb form, and so constitutes a substitute periphrasis. Our choice for ‘substitute periphrasis’ is also motivated by the fact that in Ezekiel only such type of periphrasis occurs. 18

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Periphrastic Tense Forms withGK™OK andIKžIPQOCK in Ezekiel

131

expressive. The label “such or such periphrasis” cannot escape being somewhat subjective.22 In summary, we regard the combination of GK™OK orIKIPQOCK with participle, equivalent to a simple verb form, as being a “periphrasis”. When the verbs GK™OK andIKIPQOCK act as auxiliaries (grammatical tool), we call their combination with a participle equivalent to a simple verb form “substitute periphrasis”. In most cases the periphrasis concerns some forms in the verbal paradigm. There are uses of GK™OK andIKIPQOCK with participle where periphrasis is not limited to some forms of the verbal paradigm, but where it affects the verb itself in its entirety. We call them “expressive periphrases”; they are used with a special purpose, particularly to emphasize the verb.23 In these expressive periphrases GK™OK andIKžIPQOCK mostly act as copula. Substitute periphrasis (with the auxiliariesGK™OK andIKžIPQOCK) replaces a form in the verbal paradigm; expressive periphrasis formed with the copula concerns the verb itself, e.g. C UGDYPGK™OK is equivalent to C UGDGY, but RNG MY tolerates periphrasis only for a few forms in the verbal paradigm (C UGDYP GK™OK is a periphrasis of the verb C UGDGY, the form RGRNGIOGPQK GK UK is a periphrasis for the perfect indicative third person plural of RNGMY).

22 Evans more than once points to the element of subjectivity involved in the identification of some periphrastic constructions (Verbal Syntax, 233 and 247). Fanning too puts forward the difficulty to decide whether a construction is periphrastic or not (Verbal Aspect, 311). 23 We have said that the periphrastic construction disconnects the verb from its complements, thus creating a looser connection with them, in opposition to a simple verb form, so that the focus is on the verb. This point of view counters the thesis of Rosén who compares the periphrasisGK™OK plus present participle to the Coptic second tense. The second tenses (distinct from the primitive tenses by a prefix) in Coptic, in opposition to the primitive tenses, disconnect the verb from the context, and thus bring the adverbial complements forward. These second tenses are best translated in modern languages by a cleft sentence (French “phrase coupée”), e.g. “It is the man who hears” (H. J. Polotsky, Études de syntaxe copte [Le Caïre: Institut français d’archéologie orientale, 1944], 25–7; 57). Rosén notices in the periphrastic construction GK™OK plus present participle traces of and similarities with the coptic second tenses; he investigated such periphrastic construction in Herodotus, e.g. C NN +UKQFOGPG™UVKRGTK? 8RGTDQTGYPGK™TJOGPC (4.32) (H.B. Rosén, “Die ‘zweiten’ Tempora des Griechischen. Zum Prädikatsausdruck beim griechischen Verbum”, MH 14 [1957], 133–54). Gonda agrees with the thesis of Rosén to a certain degree, but restates it: the periphrasis, likeGK OK plus present participle, is best translated with a cleft sentence (J. Gonda, “A Remark on ‘Periphrastic’ Constructions in Greek”, Mnemosyne 12 (1959) 97–112, on p. 97); the periphrasis moves the attention from the verb to the complements (on p. 99). The periphrasis however, in opposition to the simple verb form, is mainly used in descriptive parts rather than in narratives that continue the story (on pp. 101–3). Rydbeck criticized the article of Rosén in detail: many periphrases in Herodotus do not underline the adverbial complement, and the theory of the second tenses applied to the Greek periphrases can not be maintained as a general rule for the classical and postclassical period (L. Rydbeck, “Bemerkungen zu Periphrasen mit GK¤PCK + Präsens Partizip bei Herodot und in der Koine”, Glotta 47 [1969] 186–200, on pp. 188– 196). Rydbeck clearly says the theory put forward by Rosén is not true of the Septuagint (on p. 198). At this point we agree with the criticism of Rydbeck.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

132

Katrin Hauspie

2. Periphrasis with GK™OK Ezek 1:12 J¤PVQƒRPGWOCRQTGWQOGPQP Ezek 2:10 IGITCOOGPCJ¤P Ezek 16:7 UW?FG?J¤USCC UZJOQPQWUC Ezek 16:14 UWPVGVGNGUOGPQPJ¤P Ezek 16:22 J¤USCC UZJOQPQWUC Ezek 20:49 RCTCDQNJG™UVKNGIQOGPJCWŸVJ (MT 21:5)24 Ezek 22:18 C PCOGOGKIOGPQLG™UVKP Ezek 23:29 MCK?G¢UC UZJOQPQWUC Ezek 24:17 QW™MG¢UVCKUWORGRNGIOGPQP Ezek 33:28 GK¤PCKFKCRQTGWQOGPQP Ezek 34:29 MCK?QW™MGVKG¢UQPVCKC RQNNWOGPQK Ezek 36:13 MCVGUSQWUCGK» Ezek 44:1 MCK?CWŸVJJ»PMGMNGKUOGPJ Ezek 44:2 MGMNGKUOGPJG¢UVCK Ezek 44:2 MCK?G¢UVCKMGMNGKUOGPJ Ezek 46:1 G¢UVCKMGMNGKUOGPJ Ezek 48:12 MCK?G¢UVCKFGFQOGPJ

V™MGNHNCZ8THJJH0HJG[JK[ JHD8VM [KJY JH[T  G Y …7C Y 8JN[KNH. JH[T G Y…?V2[J% K 8J][KNHONCOO 8[HJ…][KIKU JH[T  G Y 2D$Z…JG $V C  QN TD2 [RU W $ FQY8[J[K  QNY [7H …VGNGMQ T8IHU 8JY JG[JK[T8IHU T8IHUJH[HJY T8IHUJG[JK[ JHV[HJY

This list covers all cases ofGK™OK with participle being a periphrasis or interpreted as such in the Septuagint of Ezekiel. The phrases QK»MQL RCTCRK MTCKPYPG™UVKP and G¢UVCKW˜GVQƒLMCVCMNW\YP are not incorporated in this list. OK»MQLRCTCRKMTCKPYP is considered as a unit (Ezek 2:5, 6, 7; 3:9, 26, 27), in which RCTCRKMTCKPYP is taken adjectivally, modifying QK»MQL, analogously to (Q) QK»MQLQ RCTCRKMTCKPYP (Ezek 2:8; 12:9, 25, 27) and VQƒP QK»MQP VQƒP RCTCRKMTCKPQPVC (Ezek 17:12; 24:3; 44:6). ¹(UVCK W˜GVQ?L MCVCMNW\YP in Ezek 13:11 and W˜GVQƒL MCVCMNW\YP  G¢UVCK in Ezek 13:13 leave the possibility of periphrasis open. The same phrase in the dative in Ezek 38:22, W˜GV MCVCMNW\QPVK, combines MCVCMNW\YP with W˜GVQL as attributive adjective. In analogy to this case WGVQ?LMCVCMNW\YP in Ezek 13:11, 13 is taken as a unit, excluding the periphrastic construction MCVCMNW\YPG¢UVCK. None of the indicative periphrases mentioned above complies with classical usage, in which periphrastic constructions are limited to perfect and pluperfect forms of mute and liquid verbs and verbs whose perfect stem ends in a consonant for the third person plural. These verbs are represented four times in Ezekiel: Ezek 2:10; 22:18 (mute verbs), Ezek 16:14; 44:1 24

Ezek 21:5 of the MT corresponds to Ezek 20:49 in the edition of Ziegler 1977.

© 2011, Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen ISBN Print: 9783525532614 — ISBN E-Book: 9783647532615

Periphrastic Tense Forms withGK™OK andIKžIPQOCK in Ezekiel

133

(verbs with perfect stem ending in a consonant, unlike the classical usage in the third person singular). The active future perfect indicative always appears as periphrasis, G¢UQOCK plus active perfect participle, but does not occur in the Septuagint of Ezekiel; the passive future perfect indicative appears as a simple verb form.25 However, the periphrastic construction for the passive future perfect appears as well for verbs that do not support the passive future perfect as a simple verb form.26 In Ezek 24:17; 44:2(bis) and 46:1 this periphrastic future perfect indicative is used. Since mute, liquid and verbs with perfect stem ending in a consonant allow – to a certain degree – periphrastic construction in their indicative paradigm, the abovementioned periphrases in Ezekiel do not raise problems. 2.1 Perfect and pluperfect substitute periphrasis As far as the perfect and the pluperfect are concerned, simple verb forms are not absent from the Septuagint of Ezekiel; they even appear alongside a periphrasis. So in Ezek 2:10 IGITCOOGPCJ¤P and G IGITCRVQ (pluperfect indicative third person singular) stand close to one another. Ezek MCK?C™PGKNJUGPCW™VJ?PG™PYRKQPG™OQW 2:10 MCK?G™PCW™VXIGITCOOGPCJ¤PVC? G¢ORTQUSGPMCK?VC?Q¢RKUSGPMCK? G™IGITCRVQSTJPQLMCK?OGNQLMCK?QW™CK

[KJY[•PHRN+HV2  QTRK