Balaam in Text and Tradition (Forschungen Zum Alten Testament) 9783161563553, 9783161563560, 3161563557

Citation preview

Forschungen zum Alten Testament Edited by

Konrad Schmid (Zürich) · Mark S. Smith (Princeton) Hermann Spieckermann (Göttingen) · Andrew Teeter (Harvard)

131

Jonathan Miles Robker

Balaam in Text and Tradition

Mohr Siebeck

Jonathan Miles Robker, born 1980; 1999-2003 studied History and Philosophy, with a concentration in Religious Studies; 2006 Master of Theological Studies from Duke Divinity School; 2011 PhD from the Faculty of Protestant Theology at the FAU Erlangen, Germany; since 2013 at the Faculty of Protestant Theology of the WWU Münster, Germany; 2018 Habilitation. orcid.org/0000-0002-9793-3530

ISBN 978-3-16-156355-3 / eISBN 978-3-16-156356-0 DOI 10.1628/978-3-16-156356-0 ISSN 0940-4155 / eISSN 2568-8359 (Forschungen zum Alten Testament) The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliographie; detailed bibliographic data are available at http://dnb.dnb.de. © 2019 Mohr Siebeck Tübingen, Germany. www.mohrsiebeck.com This book may not be reproduced, in whole or in part, in any form (beyond that permitted by copyright law) without the publisher’s written permission. This applies particularly to reproductions, translations and storage and processing in electronic systems. The book was printed on non-aging paper by Gulde Druck in Tübingen, and bound by Buchbinderei Spinner in Ottersweier. Printed in Germany.

Foreword This monograph presents a reformatted version of my Habilitationsschrift, which was accepted at the University of Münster in January 2018. Prof. Dr. Reinhard Achenbach and Prof. Dr. Reinhard Müller served as reviewers for the committee. Other than formatting and some minor typographical issues, the content of this volume is identical with the manuscript submitted to the university. The initial impetus for this research stemmed from the project “Traditionsund Redaktionsprozesse im Buch Numeri und ihr Zusammenhang mit der Entstehung des Pentateuch” under the auspices of Christian Frevel (Bochum), Thomas Pola (Dortmund), and Aaron Schart (Essen) during my time as a researcher in Essen from 2011–2013. Through the aforementioned research project, I gained substantial insight about the peculiarities of the book of Numbers and developed an interest in how the pericope about Balaam relates to these issues. Having perused a number of attempts to explain the passage in Numbers 22–24, I found myself somewhat dissatisfied with earlier theses about this story’s genesis and how it fits into its current literary context. Beyond that, the connection to other biblical and the singular relevant extrabiblical attestations of the figure Balaam, son of Beor, had not been, in my opinion, sufficiently explicated. At the root of all of this, I developed an interest in the character Balaam, both as a literary figure, but also as a potentially historical personage. Questions about this figure, whether he was historical or not, guided me through this research and motivated this study. Along this path, a number of people instructed and aided me. To them I owe much and, for their guidance, I offer my thanks. After the completion of my dissertation, Prof. Dr. Siegfried Kreuzer (Wuppertal) and Prof. Dr. Aaron Schart (Essen) found positions for me as a researcher and instructor at their institutions. Prof. Kreuzer shared my strong interest in text-historical questions and encouraged me to continue this line of research in the Pentateuch. Prof. Schart brought me into the research project on Numbers and helped me to narrow down and focus the study on Numbers 22–24 as a specific problem in the book of Numbers. Without their initial input and support, this study would not have been possible. After moving to Münster in 2013, I found continuing interest and vigorous discussion with Prof. Dr. Reinhard Achenbach, one of the current experts on

VI

Foreword

Numbers in particular and the Pentateuch more generally. With the addition of Prof. Reinhard Müller to the faculty in 2014, I was able to engage with another exegete of great repute. Even though I was often of an opinion distinct from theirs, these scholars served as the whetstone on which I was able to sharpen my theses. With their extremely detailed observations and poignant questions, they engaged my research critically, helping me to refine it. In this capacity, I must also thank the Alttestamentliche Sozietät in Münster, which provided a productive forum to proffer observations, debate their meanings and evaluations, and synthesize theses. In particular, I would like to note and thank, beyond the aforementioned professors, Lars Maskow, who took time both during and outside of the colloquium to discuss and engage with my ideas. I would like to thank the editors of the series Forschungen zum Alten Testament for their willingness to accept this monograph into their series. The team at Mohr Siebeck, as well, deserves my praise for their helpful technical support and editing advice. Specifically, my thanks go to Dominika Zgolik and Katharina Gutekunst. Outside of a professional capacity, I remain indebted to my friends and colleagues at the faculty in Münster and elsewhere in Germany, who supported me with friendly words and plenty of coffee and sweets. Noteworthy were the contributions of Patrick Bahl, Sabine Joy Ihben-Bahl, Eike Herzig, and Rudi de Lange. For time away from the office, I thank “The Holy Rollers” for affording me with the regular opportunity to clear my head by bowling down as many pins as we could. Finally, I wish to thank my family: my wife Anja for her continual support in virtually every imaginable capacity, even at the most stressful times during this project, and our daughter Miriam, who permitted Anja to stay home from work, granting her the time to read and correct my manuscript. Both Anja and Miriam taught and continue to teach me what joy truly means. To them and all of the aforementioned, I express my deepest gratitude. Jonathan Miles Robker In Münster February 2019

Table of Contents Foreword ..................................................................................................... V List of Abbreviations ............................................................................... XIII

Chapter 1: Introduction ........................................................................ 1 1.1 The Question ......................................................................................... 1 1.2 Is the Balaam Narrative in Numbers 22–24 Uniform? .......................... 10 1.2.1 Walter Gross ...................................................................................... 10 1.2.2 Alexander Rofé .................................................................................. 11 1.2.3 Meshullam Margoliot ........................................................................ 12 1.2.4 Andreas Schüle .................................................................................. 14 1.2.5 Uwe Weise ........................................................................................ 15 1.2.6 Summary of Unified Models .............................................................. 17 1.3 Is the Story in Numbers 22–24 the Product of Multiple Sources? ......... 17 1.3.1 Julius Wellhausen .............................................................................. 19 1.3.2 August Freiherr von Gall ................................................................... 19 1.3.3 Heinrich Holzinger ............................................................................ 20 1.3.4 George Buchanan Gray ...................................................................... 21 1.3.5 Hugo Greßmann ................................................................................ 22 1.3.6 Otto Eissfeldt ..................................................................................... 23 1.3.7 Martin Noth ....................................................................................... 25 1.3.8 Jules de Vaulx ................................................................................... 27 1.3.9 Baruch Levine ................................................................................... 27 1.3.10 Axel Graupner ................................................................................. 29 1.3.11 Ludwig Schmidt .............................................................................. 32 1.3.12 Horst Seebass .................................................................................. 34 1.3.13 Joel Baden ....................................................................................... 34 1.3.14 Summary of Source-Critical Models ................................................ 37

VIII

Table of Contents

1.4 Redaction-Historical Models for the Development of Numbers 22–24 ................................................................. 38 1.4.1 Hedwige Rouillard ............................................................................. 38 1.4.2 Erhard Blum ...................................................................................... 40 1.4.3 Christoph Levin ................................................................................. 41 1.4.4 John Van Seters ................................................................................. 46 1.4.5 Reinhard G. Kratz ...............................................................................50 1.4.6 Markus Witte ..................................................................................... 52 1.4.7 Reinhard Achenbach .......................................................................... 57 1.4.8 Rainer Albertz ................................................................................... 66 1.4.9 Summary of Redaction-Historical Models ......................................... 67 1.5 Conclusion ............................................................................................ 67

Chapter 2: The Text of Numbers 22–24 .......................................... 69 2.1 Introduction .......................................................................................... 69 2.2 Textual Criticism .................................................................................. 73 2.2.1 Num 22:5–6 ....................................................................................... 73 2.2.2 Num 22:10–11 ................................................................................... 76 2.2.3 Num 22:21–35 ................................................................................... 76 2.2.4 Num 22:2–5 and 15–17 ...................................................................... 78 2.2.5 Num 23:10 ......................................................................................... 80 2.2.6 Num 23:18–21 ................................................................................... 81 2.2.7 Num 24:6–7 ....................................................................................... 83 2.2.8 Num 24:17–24 ................................................................................... 88 2.3 The Restored Text of Numbers 22–24 and Its Translation ................... 101 2.4 The Emendations to M in Numbers 22–24 .......................................... 118 2.5 The Character of the Textual Traditions ............................................. 119 2.5.1 The Septuagint Tradition ................................................................. 119 2.5.2 The (Pre-)Samaritan Tradition ......................................................... 122 2.5.3 The (Proto-)Masoretic Tradition ...................................................... 123 2.6 Conclusions ........................................................................................ 125

Table of Contents

IX

Chapter 3: Literary Criticism ........................................................... 128 3.1 Structural Analysis ............................................................................. 128 3.2 Literary-Critical Remarks and the Unity of the Composition .............. 131 3.3 The Narrative’s Exposition(s): Numbers 22:1–4 ................................. 133 3.4 The Main Narrative Body: Numbers 22:5–23:6, 11–17, 25–30 ........... 139 3.5 The Oracles: Num 23:7–10, 18–24; 24:3–9, and 15–24 ..................... 156 3.6 The Narrative’s Conclusion: Numbers 24:1–2, 10–14, and 24 ........... 171 3.7 Summary of Literary-Critical Results ................................................. 174 3.8 Redaction History ............................................................................... 195 3.9 The Story of Balaam in the Redaction Version(s) ................................ 197

Chapter 4: The Bible’s Balaam beyond Numbers 22–24 .......... 207 4.1 Genesis 36:32–33 // 1 Chronicles 1:43–44 ......................................... 207 4.2 Numbers 31 ........................................................................................ 209 4.3 Deuteronomy 23:4–6 .......................................................................... 218 4.4 Joshua 13:21–22 ................................................................................ 229 4.5 Joshua 24:9–10 .................................................................................. 233 4.6 Judges 11:25 ...................................................................................... 241 4.7 Micah 6:5 ........................................................................................... 243 4.8 Nehemiah 13:1–3 ................................................................................ 246 4.9 Conclusions: Balaam in the Hebrew Bible .......................................... 250 4.10 Balaam at Qumran and in the New Testament .................................. 253 4.10.1 Balaam at Qumran ......................................................................... 254 4.10.2 Matthew 2 ...................................................................................... 256 4.10.3 2 Peter 2:15–26 .............................................................................. 259 4.10.4 Jude 11 .......................................................................................... 261 4.10.5 Revelation 2:14 .............................................................................. 262 4.11 Conclusions ...................................................................................... 263

X

Table of Contents

Chapter 5: An Inscription from Deir ʿAlla. Balaam in Transjordan ....................................................................... 271 5.1 Introduction ........................................................................................ 271 5.2 The Inscription’s Language ................................................................ 276 5.3 The Text, Its Reconstruction, and Its Translation ................................ 279 5.4 Commentary and Evaluation .............................................................. 288 5.5 The Inscription’s Form ....................................................................... 293 5.6 The Inscription’s Sitz im Leben .......................................................... 295 5.6.1 The Scribe ....................................................................................... 296 5.6.2 The Location ................................................................................... 297 5.6.3 The Function ................................................................................... 298 5.7 The Relationship to the Bible’s Balaam(s) .......................................... 300 5.8 Conclusions: What Does the DAPT Reveal about the Bible’s Balaam? ............................................................ 304

Chapter 6: Balaam as a Character. His Origin and Background .............................................................. 306 6.1 What’s in a Name? ............................................................................. 306 6.2 Balaam’s Heritage .............................................................................. 308 6.2.1 ‫ פתורה אשׁר על־הנהר‬.......................................................................... 309 6.2.2 (‫ ארץ בני־עמו)ן‬................................................................................... 311 6.2.3 ‫ ארם‬.................................................................................................. 312 6.2.4 ‫ מהררי־קדם‬........................................................................................ 314 6.2.5 ‫ מדין‬.................................................................................................. 316 6.2.6 ‫ מפתור ארם נהרים‬.............................................................................. 317 6.2.7 Conclusions about Balaam’s Ethnicity and Geopolitical Background .................................................... 318 6.3 Balaam’s “Profession” ...................................................................... 319 6.3.1 ‫( פותר‬Num 22:5) .............................................................................. 320 6.3.2 ‫( ארר‬Num 22:6, 12; 23:7; and 24:9) ................................................. 321 6.3.3 ‫( קסם‬Num 22:7; 23:23; and Josh 13:22) ........................................... 324 6.3.4 ‫( קבב‬Num 22:11, 17; 23:8, 11, 13, 25, 27; and 24:10) ...................... 326

Table of Contents

XI

6.3.5 ‫( עלה‬Num 23:2 and 4) ...................................................................... 328 6.3.6 ‫( זעם‬Num 23:7–8) ............................................................................ 330 6.3.7 ‫( נחשׁ‬Num 23:23 and 24:1) ............................................................... 331 6.3.8 ‫( ותהי עליו רוח אלהים‬Num 24:2) ....................................................... 334 6.3.9 ‫( הגבר שׁתם העין‬Num 24:3 and 15) ................................................... 336 6.3.10 ‫( שׁמע אמרי־אל‬Num 24:4 M and 16) ............................................... 337 6.3.11 ‫( מחזה שׁדי יחזה‬Num 24:4 and 16) .................................................. 337 6.3.12 ‫( נפל וגלוי עינים‬Num 24:4 and 16) ................................................... 338 6.3.13 ‫( וידע דעת עליון‬Num 24:16) ............................................................ 339 6.3.14 ‫ קלל‬and ‫( קללה‬Deut 23:6; Josh 24:9; and Neh 13:2) ....................... 340 6.3.15 Balaam’s Profession in the Inscription from Tell Deir ʿAlla .......... 342 6.3.16 Conclusions about Balaam’s “Profession” ..................................... 344 6.4 Balaam’s “Religious Affiliations” ...................................................... 347 6.4.1 ‫ יהוה‬.................................................................................................. 347 6.4.2 ‫ אלהים‬/ ‫ אלהן‬..................................................................................... 349 6.4.3 ‫ אל‬.................................................................................................... 350 6.4.4 ‫ שׁדי‬................................................................................................... 353 6.4.5 ‫ עליון‬.................................................................................................. 356 6.4.6 Conclusions about Balaam’s Religion .............................................. 357 6.5 Conclusions about the Character Balaam ........................................... 358

Chapter 7: Conclusions and Impetus for Further Research ...... 361 Bibliography ............................................................................................. 367 Source Index ............................................................................................ 387 Author Index ............................................................................................ 395 Subject Index ............................................................................................ 398

List of Abbreviations Generally, the abbreviations in this volume follow the SBL Handbook of Style, Second Edition. Abbreviations that do not follow or appear in SBL are present here. DAPT The Deir ʿAlla Plaster Texts, specifically Combination A G

The Septuagint

HexRed The Redactor of the Hexateuch La

The Vetus Latina

M

The Masoretic Text

ML

Codex Leningradensis

PentRed The Redactor of the Pentateuch Q

Qumran

R1

The first redactor / redaction of a biblical text

R2

The second redactor / redaction of a biblical text

R3+

The third redactor / redaction of a biblical text

S

The Peshitta

S1

The oldest source text behind Numbers 22–24 and cognate texts

S2

A second, fragmentary source text attested in Numbers 22–24

Smr

The Samaritan Pentateuch

T

Targum Version(s)

V

The Vulgate

Chapter 1

Introduction 1.1 The Question 1.1 The Question

Who was Balaam, and what did he do? This curious figure is most well known from Numbers 22–24, but he appears in many other texts as well. Interested readers encounter him in Numbers 31; Deuteronomy 23; Joshua 13 and 24; and Micah 6, as well as in the New Testament. The biblical material presents a broad spectrum about this enigmatic character, who has fascinated readers since Antiquity. Philo spends time commenting on him and interpreting his undertakings.1 The community at Qumran cited as messianic one of his supposed prophecies recounted in the book of Numbers,2 while at the same time including him in a list of false prophets.3 Josephus proffers a lengthy recounting and explication of the biblical material.4 No fewer than four New Testament authors obliquely allude to or expressly refer to either him or his prophecy.5 Others around the transition between the eras refer or allude to him or his prophecies, such as the community at Qumran, the author of 1 Enoch, PseudoPhilo, and Philo.6 The Targums demonstrate further analysis of this figure in 1 Cf. Praem. 91–97 and 163–72; Mos. 1.290–1. Regarding Philo’s interpretive engagement with Balaam, cf. Herbert Donner, “Balaam pseudopropheta,” in Beiträge zur Alttestamentlichen Theologie: Festschrift für Walther Zimmerli zum 70. Geburtstag, ed. Herbert Donner, Robert Hanhart and Rudolf Smend (Göttingen: Vandenhoeck & Ruprecht, 1977), 118–19 and Peder Borgen, “‘There Shall Come Forth a Man’: Reflections on Messianic Ideas in Philo,” in The Messiah: Developments in Earliest Judaism and Christianity, ed. James H. Charlesworth (Minneapolis: Fortress Press, 1992), 341–61. 2 Cf. Florentino García Martínez, “Two Messianic Figures in the Qumran Texts,” in Current Research and Technological Developments on the Dead Sea Scrolls, ed. Donald W. Parry and Stephen D. Ricks (Leiden: Brill, 1996), 14–40. 3 Cf. 4Q339 and Aharon Shemesh, “A Note on ‘4Q339’ ‘List of False Prophets’,” RevQ 20, no. 2 (December 2001): 319–20. 4 Cf. Ant. 4.102–58. 5 Namely, the authors of Matthew, 2 Peter, Jude, and the Revelation. 6 For Qumran, cf. the discussion in Chapter Four. For 1 Enoch, cf. Eibert Tigchelaar, “Balaam and Enoch,” in The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, ed. George H. van Kooten and Jacques van Ruiten, TBN, vol. 11 (Leiden: Brill, 2008), 87–99. For Pseudo-Philo, cf. Jacques T.A.G.M. van Ruiten, “The Rewriting of Numbers 22–24 in Pseudo-Philo, Liber Antiquitatum Biblicarum 18,” in The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, ed. George H. van Kooten

2

Introduction

the Aramaic-speaking Judaism of Antiquity, sometimes in common with Philo or other interpreters.7 The engagement with this figure continued also in Medieval Judaism and Christianity.8 Yet, even the most superficial reading of the biblical materials about Balaam demonstrates disparate images of this peculiar personality.9 This confused and confusing characterization has left an impressive mark even into the twenty-first century in the form of the extensive secondary literature devoted to Balaam. Much of the modern fascination with Balaam, particularly before the 1970s, focused on the identification of sources behind the biblical Balaam material. Exegetes sought to explain why Numbers characterizes Balaam in several different manners, why Balaam in Deuteronomy and Joshua remains distinct from Balaam in Numbers, and what Micah might have known about any literary or historical Balaam figure. For the material in the Hexateuch, such discussions made Balaam more or less a pawn in iterations of the Urkundenhypothese (the Documentary Hypothesis). Often this process began already with regard to the

and Jacques van Ruiten, TBN, vol. 11 (Leiden: Brill, 2008), 101–30. For Philo, cf. George H. van Kooten, “Balaam as the Sophist par Excellence in Philo of Alexandria: Philo’s Projection of an Urgent Contemporary Debate Onto Moses’ Pentateuchal Narratives,” in The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, ed. George H. van Kooten and Jacques van Ruiten, TBN, vol. 11 (Leiden: Brill, 2008), 131–61. 7 Cf. Robert Hayward, “Balaam’s Prophecies as Interpreted by Philo and the Aramaic Targums of the Pentateuch,” in New Heaven and New Earth. Prophecy and the Millennium. Essays in Honour of Anthony Gelston, ed. Peter J. Harland and Robert Hayward (Leiden; Boston; Cologne: Brill, 1999), 19–36 and Alberdina Houtman and Harry Sysling, “Balaam’s Fourth Oracle (Numbers 24:15–19) According to the Aramaic Targums,” in The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, ed. George H. van Kooten and Jacques van Ruiten, TBN, vol. 11 (Leiden: Brill, 2008), 189–211. 8 For Rabbinic Judaism, cf., the overview of material and the comments in, e.g., Geza Vermes, Scripture and Tradition in Judaism. Haggadic Studies. Second, Revised Edition., StPB, vol. 4 (Leiden: Brill, 1973), 127–76 and Ronit Nikolsky, “Interpret Him as Much as You Want: Balaam in the Babylonian Talmud,” in The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, ed. George H. van Kooten and Jacques van Ruiten, TBN, vol. 11 (Leiden: Brill, 2008), 213–30, as well as the literature cited there. For an introduction to Patristic comments on Balaam, cf. Johan Leemans, “‘To Bless with a Mouth Bent on Cursing’: Patristic Interpretations of Balaam (Num 24:17),” in The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, ed. George H. van Kooten and Jacques van Ruiten, TBN, vol. 11 (Leiden: Brill, 2008), 287–99. 9 Though, some have gone to remarkable lengths to conform the retelling of Balaam’s story in the Bible. Cf., e.g., Rufus Phineas Stebbins, “The Story of Balaam,” The Old Testament Student 4, no. 9 (May 1885): 385–95, who regarded the whole story of Numbers 22– 24 as Balaam’s self-serving and deceitful autobiographical report. However, Stebbins paraphrasing the tale does not conform to the strictures of critical study. Nor does the retort of Stebbins’ report; cf. B.F. Simpson, “The Story of Balaam Reconsidered,” The Old Testament Student 5, no. 3 (November 1885): 125–28.

1.1 The Question

3

textual history of Numbers 22–24 (and other passages about Balaam), as Wevers noted: “Most commentaries on Num concentrate on obvious inconsistencies in the text, and resort to source analysis. The Alexandrian translator of course knew nothing of Yahwists, Elohists and Priestly writers. He certainly did not distinguish between a source using ‫ יהוה‬and another using ‫ ;אלהים‬he was faced with a completed text, much like a consonantal BHS text. Oddly, he seems not to have been concerned about the inconsistencies which trouble modern scholars, though some of them are ob|vious. Thus that for the second visit of Moabite dignitaries, divine approval for Balaam’s journey to Moab was given, though at the first visit it was not. Nor does the translator show concern at the uneasy fit of the angel’s barring the way to Balaam’s ass in spite of permission to go to Moab having been granted. He made no attempt at reconciling such difficulties, but simply translated what was before him.”10

Genuflections on the literary background of the figure of Balaam continue today, albeit often (though by no means exclusively) quite divorced from the source-critical epistemology of the nineteenth and twentieth centuries. Some more recent studies have considered the redactional characterizations of the literature and its Balaam figure. Others have sought to affirm the general unity of Numbers’ portrayal of Balaam, at least as found in Numbers 22–24. The literary issues have by no means been entirely resolved, but they are also not the only table at which Balaam is discussed.11 After the deciphering of cuneiform and the subsequent availability to modern audiences of Mesopotamian literature and the customs attested therein, interest in Balaam renewed with a new nuance. No longer could he only be compared and contrasted with biblical prophets or those known from the Hellenistic and Roman world. The opportunity arose to compare him with equivalents found in the Akkadian sources. Exegetes and students of Oriental culture could reflect on Balaam’s mantic background and practices, in what ways the biblical image of Balaam suggests or affirms his supposed Mesopotamian background. 12 The mysterious city of his origin – simply called “Pethor on the river” in the Hebrew Bible – could be recognized and equated with a city found in Akkadian sources, namely Pitrû.13 This discussion in turn left its traces on 10

John William Wevers, “The Balaam Narrative According to the Septuagint,” in Lectures et Relectures de la Bible. Festschrift Pierre-Maurice Bogaert, ed. Jean-Marie Auwers, André Wénin (Leuven: Peeters, 1999), 136–37. 11 Chapter Three will address these literary-critical and redactional-historical issues. 12 Cf., e.g., already Samuel Daiches, “Balaam – A Babylonian Bārū,” in Hilprecht Anniversary Volume (Leipzig: J.C. Hinrichs, 1909), 60–70, who attempted to present ten common features between Mesopotamian bārū and Balaam. Against this position, cf. Leonhard Rost, “Fragen um Bileam,” in Beiträge zur Alttestamentlichen Theologie: Festschrift für Walther Zimmerli zum 70. Geburtstag, ed. Herbert Donner, Robert Hanhart and Rudolf Smend (Göttingen: Vandenhoeck & Ruprecht, 1977), 377–87. 13 This identification goes back to at least 1885; cf. the translation of Shalmaneser’s Monolith Inscription in Archibald Henry Sayce, Assyria. Its Princes, Priests and People (London: The Religious Tract Society, 1885), 147, though George Buchanan Gray, A Critical

4

Introduction

biblical exegetical discourse about Balaam, even about what this new data implied about the sources’ engagement with this traditional figure. And yet, this certainly was not the final aspect of the discourse about Balaam.14 In 1967, Balaam received renewed interest with the recovery of an ancient Transjordanian inscription – found at Tell Deir ʿAlla – that mentions him by name, even with the same patronymic as that of the Bible.15 Now attention could turn to Balaam as a Transjordanian personage or epigraphic literary figure. Exegetes could contrast the Bible with a new source of material about this fabled figure. That the inscription also featured a vision and foretold some forthcoming destruction hardly went unnoticed. But the poorly preserved inscription required more attention simply to decipher what it said. With more time to appreciate the inscription, more can be said about it, particularly regarding the text’s composition and the circumstances behind its creation. Any relationship it might have to the biblical tradition, a relationship which was expounded quite vociferously shortly after the inscription’s discovery, can also be appreciated more fully.16 Each of these matters – the biblical text, the traditions behind it, their relationships to the world of the ancient Orient, the specific nature of any common background between the biblical text and the inscription from Deir ʿAlla – still merits discussion. None of the problems have been resolved with anything approaching certainty or scholarly consensus. Particularly in the case of the biblical materials, continued interest and the development of fundamentally distinct literary-historical models in the past several decades mandate that a new approach to this old discussion be advanced. This work will attempt to cover the various features of the debate around Balaam, including a strong focus on the biblical materials. Methodologically, the traditional canon of historical criticism, with reference to other methods where appropriate, guides this study. The monograph will approach Balaam from several perspectives, but the primary focus remains the biblical text, particularly regarding questions of 1) its textual transmission; 2) its literary inception; 3) its literary transmission and redactional history; 4) its tradition-historical background; and 5) its theological and Exegetical Commentary on Numbers, Impression from 1986, ICC (Edinburgh: T&T Clark, 1903), 325 dated it to some publication of Sayce’s from 1878 that I have been unable to identify. 14 The discussion of Balaam’s background and any potential relationship to Mesopotamia follows in Chapter Six. 15 The initial publication followed only in 1976 (Jacob Hoftijzer and Gerrit van der Kooij, eds., Aramaic Texts from Deir ‘Alla, DMOA, vol. 19 [Leiden: Brill, 1976]), though a notice about the discovery occurred already in the same year; cf. Hendricus Jacobus Franken, “Texts from the Persian Period from Tell Deir ‘Allā,” VT 17 (1967): 480–81. 16 Chapter Five discusses this inscription and its implications for our understanding of Balaam. Some additional tradition-historical considerations of this inscription follow in Chapter Six.

1.1 The Question

5

or religious-historical impetus and development. This provides a framework for the discussion at hand. Ultimately, this work will identify the ancient background of this historical or fictional figure as an Aramean with some metaphysical capability. Due to Aramean influence on the Cisjordan and Transjordan, Israelite familiarity with this character developed such that they incorporated him into their literary engagement with the neighboring kingdom of Moab. The historical background for this earliest literature must have been during the ninth or eighth centuries BCE, when Israel and Moab stood as opposed militant combatants and AramDamascus occupied the Transjordan. Both the biblical and Transjordanian epigraphical accounts of Balaam reflect this historical and cultural background. Though the historical background of the oldest literary Balaam tradition belongs to the monarchic period in Israel, even this primary version was retrojected into a narrative about Israel’s origin from the time of the exodus. This created the impression that Moab and Israel had not been amenable since before Israel arrived in the land, according to one of their biblical origin stories. This first biblical account about Balaam existed from its literary inception as a written source, which currently stands in Numbers and can be reconstructed with some reliability. This recovered source cannot be identified as one of the Pentateuch sources traditionally postulated and reconstructed in the Documentary Hypothesis (J, E, D, or P). It may have initially consisted essentially only of the Balaam narrative and oracles in an abbreviated form as found in Numbers 22–24. The oldest biblical material about Balaam viewed him unequivocally positively. Perhaps scribes at the royal court of the late ninth or first half of the eighth century BCE (the Jehu dynasty) composed this piece. The loose integration of this material suggests that its place within Numbers stems from a later editorial integration into its current context, though it could have represented a portion of a longer contiguous source from its inception. After its initial composition, this Balaam source was edited, expanded, and combined with other materials now found in Numbers. Its incorporation into a larger Deuteronomic/Deteronomistic exodus-eisodus narrative or even some kind of early Enneateuch present the most likely scenarios. This later composition afforded the Israelite entrance into the land from the east to accommodate the incorporation of the Balaam material. This “edition” must have contained at least portions of Exodus, Deuteronomy, and Joshua.17 Yet, others also continued to emend and append other material to this Balaam story now found in Numbers. At least one layer of these later redactions present part of a priestly composition that expounded on the Deuteronomistic composition including the

17 A satisfactory engagement with that material in a literary-critical and redactional-historical perspective goes generally beyond the bounds of this study. For this reason, I will only superficially address them here.

6

Introduction

Balaam story and oracles. Other additions may have been part of larger redactional undertakings, but probably only represent specific, context-oriented Fortschreibungen. At the same time, other literature about Balaam now found in the Hebrew Bible (Deuteronomy 23; Joshua 13 and 24; Judges 11; Micah 6; and Nehemiah 13) reflected on the various versions of the story in Numbers 22–24, interpreting what they found and transforming Balaam’s image in Israelite literature. Gradually these interpretations attest the development of negative sentiment toward Balaam. Material about Balaam in the Hebrew Bible continued to adapt even into the stage of transmission attested by the manuscripts. These latest impressions of Balaam were quite influential on Jewish authors, who demonstrate some ambivalence towards Balaam, including some New Testament authors, who all view him negatively, though one – Matthew – at least appropriates one element of an oracle ascribed to Balaam in Numbers through a positive reception.18 With that, we can trace an ancient Oriental figure from ninth or eighth century BCE and his development into the Roman Period and witness the transitions in attitudes toward him.19 The primary interest of this work rests in the biblical materials, particularly that found in Numbers 22–24. These chapters are the longest about Balaam and the most important in any discussion about him. For that reason, Chapters Two and Three focus on Numbers 22–24 from text-historical, literary-critical, and redactional-historical perspectives. These chapters present my reconstruction based on many impetuses found in the secondary literature published to date. In order to familiarize the reader with the various literary-critical and redaction-historical positions about Balaam proffered, a brief cross section of the history of scholarship will open this work below. Particularly those unfamiliar with the development of German literary-historical and redaction-critical models in the past few decades will find this opening section helpful. At the same time, the growing split between some North American and Israeli models with those of continental Europe will be addressed. This introductory chapter focuses primarily on the discussion surrounding Numbers, but also naturally includes some reflections on the Balaam materials in Deuteronomy and Joshua, as these texts frequently appear along with Numbers in models that reconstruct the development of the Pentateuch or Hexateuch. 18 This study only obliquely addresses the works attested at Qumran, Philo, and Josephus. The New Testament texts, which have become biblical – albeit to a particular audience – receive somewhat more attention. 19 The further reception history of this character in art and literature, even Rabbinical literature, remains outside of this study’s scope. Cf., however, Stefan Beyerle, “‘A Star Shall Come Out of Jacob’: A Critical Evaluation of the Balaam Oracle in the Context of Jewish Revolts in Roman Times,” in The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, ed. George H. van Kooten and Jacques van Ruiten, TBN, vol. 11 (Leiden: Brill, 2008), 163–88 and Nikolsky, “Interpret Him”.

1.1 The Question

7

Following this introductory history of scholarship, Chapter Two will address a variety of text-critical issues in the primary text, Numbers 22–24, focusing on distinctions with the textual traditions and translations, such as the Masoretic textual tradition (M), the Samaritan Pentauch (Samaritanus; Smr), and the Septuagint (G). Since much of the debate about the place of Numbers 22–24 (as well as the other biblical texts discussed in this volume) in the developing literary corpus of the Pentateuch, Hexateuch, or Enneateuch has occurred somewhat divorced from text-critical discussions, this chapter seeks to bring these two approaches closer together. The necessity of this remains conspicuous, as the text of Numbers 22–24 attests a number of significant variants in the manuscripts. From this survey, it will become apparent that variants in the Numbers text demonstrate its development even into the Roman Period, into the time from which manuscripts still exist. At the same time, the text-critical analysis demonstrates the remarkable stability of the textual tradition of Numbers over centuries of transmission. Having plumbed the depths of the text-critical issues in the passage, Chapter Three addresses literary-critical issues in Numbers 22–24. This chapter proposes a new reconstruction of the literary development of Numbers 22–24. Several elements in the version reconstructed at the conclusion of Chapter Two demonstrate diachronic development behind even that oldest reconstructed version. That is, several hands expressed themselves in the composition now found in Numbers 22–24; we should reckon with at least four. That being said, the majority of material in Numbers 22–24 appears to have existed from its literary conception as a unity. Having identified the secondary, tertiary, and later additions to Numbers 22–24, Chapter Three then continues, briefly turning to the larger problem of the redactional development of Numbers 22–24 in the context of larger literary compositions. The focus here remains primarily on the developing literary context within the book of Numbers, but this cannot be viewed entirely divorced from the rest of the narrative literature in (Genesis or) Exodus through Kings. First and foremost, the redaction-historical study advances the thesis that the Balaam story of Numbers 22–24* in its oldest form either existed as an independent literary composition outside of some exodus– eisodus composition, though presuming some such historical or – more accurately, narratological – context or as part of a collection of exodus material. This oldest version presumably dates back to the monarchic period in the Northern Kingdom, i.e., Israel. Later editors and scribes incorporated new elements over the course of transmission. These included additions that appear to stem from Deuteronomistic (late preexilic or exilic) and Priestly (exilic or postexilic) backgrounds, as well as even later editorial developments that some have identified with redactions bearing monikers like “Hexateuch Redaction”, “Pentateuch Redaction”, or “Theocratic Editing”. These considerations affirm the developing negative attitude toward Balaam described in the literary-critical examination of these chapters.

8

Introduction

Having approached those issues and hopefully having proffered some plausible new solutions, Chapter Four turns attention away from Numbers 22–24 to the other biblical texts about Balaam. These are studied in the same way as the text of Numbers 22–24, first text-critically, then literarily, and redactionhistorically. Many of the same issues occur in these texts as in Numbers 22– 24. Many of the proposed theses from the preceding chapters will echo here. This survey will demonstrate that some other biblical traditions demonstrate affinity with distinct phases of the development of Numbers 22–24. Others demonstrate attitudes distinct from some versions of Numbers 22–24 that might have impacted its development. This chapter concludes with an overview of Balaam’s reception history at Qumran and in the New Testament. This reception again affirms Balaam’s development as a literary figure, with his negative reception coming to dominate later interpretations of his activities. Having covered the biblical material about Balaam, Chapter Five addresses the relevant epigraphic inscription from Tell Deir ʿAlla. Here, the focus is first on the inscription itself. What can we read from the surface? What does it mean? How old is it? Does it demonstrate diachronic development? What does its Balaam look like and how does it express information about him? Then this chapter compares and contrasts its Balaam from the one in the Bible. While the amount of common material between the biblical and epigraphical Balaam figure remains manageable, it will become clear that they share some common elements in their historical and tradition-historical backgrounds. That informs our reconstruction of any plausible earlier or common literary or historical Balaam figure. From here, Chapter Six addresses the tradition-historical backgrounds of the changing images of Balaam, generally appraising the terminology applied to him. The backgrounds reflected in all of the material about Balaam in the Hebrew Bible and the Tell Deir ʿAlla Inscription flow into this survey. This discussion will demonstrate and elucidate the distinct and often disparate backgrounds of literary material about this figure. It will conclude with an appreciation of whether we should reckon with Balaam as a historical or literary figure and what the cultural background for such a supposed figure might be, though any conclusion achieved here must remain necessarily speculative. Finally, Chapter Seven reviews the conclusions of each element of this study, summarizes them, and reflects on their interrelatedness. At the same time, it will present matters that remain open for future study, particularly the development of the biblical literature in the Pentateuch, Hexateuch, or Enneateuch. However, before diving into the examination of Numbers 22–24, I would like to reiterate my objectives and specify my theses, as well as offer an overview of developments in the history of studies about Balaam. Several theses will be proffered and defended in this study. First, an older version of the Ba-

1.1 The Question

9

laam story in Numbers 22–24 will be reconstructed based on manuscript evidence. This reconstructed version has been lost, but stood in some fashion behind the various biblical versions of Numbers 22–24 currently known to us (Smr, G, Q, and M). Text-historically, it will become apparent that this story about Balaam in Numbers 22–24 continued to develop and change in a limited manner well into the Roman era, as demonstrated by the manuscripts and the versions. Secondly, this final layer of adaptation will be shown to present the culmination of earlier editorial processes, here theoretically mapped and reconstructed. The tale in Numbers 22–24 began as a smaller core, consisting of both narrative and oracular material. This core, which should be dated tentatively to the ninth or – more likely – early eighth century BCE, was expanded and adapted on a number of occasions. One, the first redaction, demonstrates affinity with material and theology that can be described as Deuteronomistic. This first redaction added some narrative and oracular material, and recontextualized the whole by incorporating it into an exodus narrative, a Deuteronomistic composition, to borrow the vernacular of Blum.20 At a later date, scribes inserted this expanded story into other material, commonly identified as characteristically Priestly, following in the wake of some priestly tradition. Later material can also be identified, the final elements of which appear remarkably similar to those revisions apparent in the manuscript traditions and the variants attested by the ancient translations. The other biblical texts about Balaam affirm this redaction-historical reconstruction and evince many of the same phenomena. The inscription from Tell Deir ʿAlla provides an external datum supporting the date of the oldest reconstructed Balaam material and suggests that a wider Balaam corpus was known in the southern Levant before and during the eighth century BCE. This extrabiblical tradition permits the postulation of a historical figure behind the distinct Balaam traditions, but more importantly demonstrates that Balaam was not merely a creation of the biblical authors’ imaginations, even though they certainly filled out his figure with more data than we can find outside of the Bible. Finally, the tradition-historical data demonstrate divergent attitudes towards Balaam and affirm the development in the complex literary figure we find in the biblical materials at present. Perhaps he bases on some historical figure, but little could be said about such a personage. A concluding chapter will reflect on the possibility and need for further related study based on the features identified here, particularly those dealing with the text-history and redactional development of the Enneateuch, Hexateuch, or Pentateuch. With that, we can turn to our survey of scholarly research on the figure of Balaam, beginning with modern literary and source-critical approaches.

20

Cf. Erhard Blum, Studien zur Komposition des Pentateuch, BZAW, vol. 189 (Berlin: de Gruyter, 1990).

10

Introduction

1.2 Is the Balaam Narrative in Numbers 22–24 Uniform? 1.2 Is the Balaam Narrative in Numbers 22–24 Uniform?

For some time, some scholars have argued for the general unity of Numbers 22–24; that is, Numbers 22–24 is not the product of two or more sources. Some go so far as to suggest that it did not come about through the expansion of one (or more) primary narrative(s) with redactional material. Often, as a necessary caveat, scholars genuflect on the narrative’s unity as a sign of its independence from its context. Only a few exegetes have argued that the text of Numbers 22– 24 came to exist as a uniform narrative without any recourse to postulated sources in the sense of the Documentary Hypothesis or redactional embedding and/or expansion (to each of these, see below); the following discussion covers some important examples.21

1.2.1 Walter Gross In 1974, Walter Gross published a dissertation describing Numbers 22–24 as consisting of several units in contrast to being the product of two sources.22 In this study, he focused primarily on literary-historical and form-critical concerns, thus concentrating exclusively on the prose portions of the text. His working principle is that texts that do not mandate division, should be regarded as uniform.23 The primary unit, Num 22:4b–21* (without ‫וזקני מדין וקסמים בידם‬ in 22:7a); 22:36–23:25* (without 23:4b and 13agd); and 24:11 and 25 was the oldest version, a literary unit about Balaam. Later editors expanded this unit on a few occasions: the first expansion (= Unit 2, in Gross’s nomenclature) added material from Num 23:26–24:10 (without 24:1ag); and 24:11–15; Unit 3 added 22:2–3a, and 4a; finally, Unit 4 added the narrative about Balaam’s interaction

21 The position and reconstruction of Sutcliffe will not be addressed here, since it requires filling too many narrative gaps with mere speculation; cf. Edmund F. Sutcliffe, “De Unitate Litteraria Num XXII,” Bib 7, no. 1 (1926): 3–39 and Edmund F. Sutcliffe, “A Note on Numbers XXII,” Bib 18, no. 4 (1937): 439–42. Rather than recognize the tensions in the text as such, Sutcliffe goes to great lengths to explain why they are indeed not tensions, filling in substantial narrative and quasi-historical information to fulfill this need. Since Timothy R. Ashley, The Book of Numbers, NICOT (Grand Rapids, MI; Cambridge, UK: Eerdmans, 1993), 454–55 did not really argue the model, but mostly presumed it, particularly in the case of the donkey narrative, it will also not be detailed here. László Pákozdy, “Az istennevek használata a Bileámperikópában,” Theologiai Szemle 14 (1938): 160–65 argued for the consistency of a single source in Numbers 22–24 for theological reasons. The narrator used distinct divine names to demonstrate with certainty YHWH’s superiority over the mantic practices of other peoples, as well as over oracles, magic, and prophetic undertakings. 22 Cf. Walter Gross, Bileam: Literar- und formkritische Untersuchung der Prosa in Num 22–24, SANT (Munich: Kösel-Verlag, 1974). 23 Cf. Gross, Bileam, 16: “Textteile, die nicht zur Zertrennung zwingen, gelten als zusammengehörig”.

1.2 Is the Balaam Narrative in Numbers 22–24 Uniform?

11

with his donkey, 22:22–35.24 To arrive at this reconstruction, he considered several elements that traditionally led to the division of Numbers 22–24 into the sources J and E and demonstrated their insufficiency to explain the condition of the text.25 Gross’s whole reconstruction relies essentially on the identification of 23:25 + 24:11 and 25 as the conclusion of the story, with all of the material in 23:26– 24:10 as a later interpolation. However, his reasoning for this reconstruction is insufficient. He names two criteria for this decision to regard 23:25 as the real conclusion of the narrative: syntax and duplication.26 Regarding the first point, the curious phrase ‫גם־קב לא תקבנו גם־ברך לא תברכנו‬, with its curious parallel syntax, demands that something else follows. Otherwise, Balak commands Balaam to do something, or rather – more accurately – forbids him from doing something, without its being fulfilled. Such would be the case nowhere else in this story. N.b. particularly the case in 24:11 + 25. That alone suggests a weakness in Gross’s reconstruction. The duplication of information in 22:41–23:2, 13–14 that he recognizes in 23:27–30 is not, in fact, a duplication. Rather, it presents a progression of the narrative. It has been edited to some degree, as will become apparent in Chapter Three, but it does not represent the copying of an original now found in 22:41–23:2, and 13–14. Reading 24:11 and 25 immediately after 23:25 provides an unsatisfactory conclusion to the narrative. The reader reckons with Balaam telling Balak something else, something that Balak has even explicitly demanded of him. Without at least some of the intervening material in 23:26–24:10, the climax would be substantially reduced and the story without a satisfactory denouement. On the other hand, Gross’s observations regarding the donkey story and the opening of the pericope in Num 22:1–5 merit further consideration. He did note that editors incorporated this material into an older version, an observation worth remembering.

1.2.2 Alexander Rofé Rofé regards Numbers 22–24 generally as an independent narrative unit. For him, that means, that it has no constitutional connection to the surrounding Numbers material. 27 Nonetheless, Rofé recognizes one piece of the Balaam 24

Gross does not provide provenience for the “additions” in 22:3b, 7a*; 23:4b, 13aγδ; and 24:1aγ. For his observations about the secondary status of the donkey story, cf. Gross, Bileam, 121–23. 25 Among other elements, he considered duplication of material, the divine nomenclature, the verbs used for cursing, the terminology used for Balak’s messengers, and the various terms used for the group undertaking the exodus; cf. Gross, Bileam, 64–88. 26 Cf. Gross, Bileam, 136–37. 27 Cf. Alexander Rofé, “The Book of Balaam” (Numbers 22:2–24:25). A Study in Methods of Criticism and the History of Biblical Literature and Religion. With an Appendix: Balaam in the Deir ʿAlla Inscription, JBS (Jerusalem: Simor Ltd., 1979), 12–36.

12

Introduction

story that does not belong to this uniform scheme: the episode about Balaam’s ass represents a later interpolation into the story.28 His argumentation covers particularly form-critical and tradition-historical matters and regards the variant attitude toward Balaam in the donkey pericope as evidence for this episodes distinct background from the rest of Numbers 22–24. Additionally, he identifies the usage of divine epithets as divergent in the two types of material: the main body of Numbers 22–24 relies on ‫ אלהים‬in the narrative and ‫ יהוה‬in the oracles, with only the story of Balaam’s ass referring to ‫ יהוה‬in narrative material. The few other appearances of the Tetragrammaton outside of the donkey story resulted from cross-contamination from the donkey episode in its surrounding context over the course of textual transmission and are not relevant for source-critical undertakings.29 Otherwise, the rest of Numbers 22–24 and its general independence overlooks the presupposed narrative background of the story: why is Israel in Moab? Freed from the context of some exodus–eisodus narrative, the Balaam story of Numbers 22–24 lacks sufficient prerequisites to have existed independently, presuming that the beginning has not gone missing. Either that, or it must have belonged to a larger collection of stories reflecting the exodus period. At the same time, Numbers 22–24 mandates that the story occur during the period of the exodus (Num 22:5, 11; 23:22; and 24:8). The initial reference to a historical period in Rofé’s reconstructed version mandates a missing antecedent: to what does ‫ בעת ההוא‬refer in Num 22:4? In turn, therefore, it could not have existed as an independent unit, but would have had to have been composed for some extant context within the book of Numbers (i.e., a Fortschreibung). The addition of v. 22:2, the appending of which Rofé regards as the editorial embedding of the rest, does create some tension with the subsequent material.30 At the same time, this explanation remains insufficient for other tensions and duplications within Numbers 22–24, the most obvious example of which appears in the next verse: Num 22:3. Nonetheless, Rofé’s appreciation of the general unity of Numbers 22–24, excepting the donkey story, presents a welcome impetus for the evaluation of this pericope.

1.2.3 Meshullam Margaliot Another attempt to regard Numbers 22–24 as a literary unit, in this case entirely devoid of editorial insertions or expansions appeared in two presentations in 1973 and 1989. In the subsequently published essays,31 Margaliot argued based

28

Cf. Rofé, Book of Balaam, 42–57. Cf. Rofé, Book of Balaam, 37–40. 30 Cf. Rofé, Book of Balaam, 34–36. 31 Cf. Meshullam Margaliot, “The Connection of the Balaam Narrative with the Pentateuch,” in Proceedings of the Sixth World Congress of Jewish Studies, Vol. 1 (Jerusalem: 29

1.2 Is the Balaam Narrative in Numbers 22–24 Uniform?

13

on theme, that Numbers 22–24 must be regarded as a whole, conceived of by a single author. The theme that unites both pieces is that of the faithless prophet. Over the course of the story, Margaliot’s presumed author recounts the story of Balaam’s transformation from a true prophet (‫ )נביא‬to a mantic (‫)קוסם‬. Within this transition, he afforded the donkey story a prominent role: after the two occasions upon which God visited Balaam in the night, God allowed Balaam to go with the officers in order to graciously provide Balaam four more chances to change his mind.32 In order to allow this reading, Margaliot equated the donkey story’s ‫ מלאך יהוה‬with the ‫ אלהים‬of the rest. To this end, the story of Balaam’s ass remains a necessary and integral part of the whole. Finally, he appreciated the final oracle complex as an important piece, since the story would otherwise lack an appropriate response to Balak’s attempted cursing. A sense of retribution, in Margaliot’s opinion, mandates that the fourth oracle stand in this context.33 The complex structure of the whole demonstrates the unity in Margaliot’s opinion. The Pentateuch transmits this story, in Margaliot’s conception, with a threefold purpose: 1) to balance Numbers’ otherwise negative attitude toward Israel; 2) to tell the story of the first prophets of the nations, i.e., Balaam as the first non-Israelite prophet; and 3) to demonstrate that prophet’s failures, making him a kind of anti-Moses, and justifying for that reason God’s rejection of all other non-Israelite prophets.34 A number of issues in this reconstruction merit brief evaluation. First, neither of the terms Margaliot identified as referring to Balaam and describing his transformation even appears in Numbers 22–24, particularly not about him. That makes it unlikely that the story must be understood as Margaliot interpreted it. Second, the equation of the ‫ מלאך יהוה‬with ‫ אלהים‬is forced in order to demand the unity of the composition. Should it have been the author’s or editor’s intention to equate these two, he or she certainly could have done it in a more conspicuous fashion. Third, the recognized sense of retribution could have easily been an editorial addition to an earlier whole. Nothing from the story’s inception demands that the Israelites should destroy Balaam. For that matter, the story ends without them even doing it. Fourth, the complexity of a passage hardly demonstrates its unity. It must not necessarily affirm its disunity, but complexity seems an odd choice to mandate uniformity. Finally, his comments about nature of prophecy belonging to exclusively to Israel after Balaam’s failure presumes that what Balaam said was inaccurate or reads it a priori in light of Deuteronomy 23. Neither of these is necessary or particularly World Union of Jewish Studies, 1977), 279–90 and Meshullam Margaliot, “Literary, Historical and Religious Aspects of the Balaam Narrative, Numbers 22–24,” in Proceedings of the Tenth World Congress of Jewish Studies, Jerusalem, August 16–24, 1989. Division A: The Bible and Its World (Jerusalem: World Union of Jewish Studies, 1990), 75–82. 32 Cf. Margaliot, “Literary, Historical and Religious Aspects,” 77–80. 33 Cf. Margaliot, “Literary, Historical and Religious Aspects,” 81. 34 Cf. Margaliot, “Literary, Historical and Religious Aspects,” 82.

14

Introduction

likely. His comments about prophecy among non-Israelites read more like apologetic than exegesis.

1.2.4 Andreas Schüle At the beginning of the twenty-first century, Andreas Schüle proposed a more canonical approach to uncover the unity of the Balaam pericope. He began his diachronic analysis discussing the story of Balaam’s ass, describing it as a more or less necessary portion of the story. To his mind, the episode between Balaam and his donkey should clarify to whose will Balaam will submit, Balak’s or God’s.35 Regarding this as a necessary component of the plot, he then assumed that any multiplicity of versions of the Balaam story must have included it. However, that both wildly overestimates the integration of this passage into the whole and presumes that Balaam the character cannot be taken at his word. To the latter point, the audience has no reason to doubt whom Balaam will follow; he has repeatedly demonstrated to that point in the story that he does what God says (Num 22:11–13, 18, and 20–21). Thus there is no narrative basis before Num 22:22 for regarding Balaam as anything other than obedient to the divine will. To suggest otherwise, one must read a malicious characterization of Balaam into this story, since it is not immanent to the text. Add to this the conspicuous Wiederaufnahme in Num 22:21 and 35, and this point fails to convince. Schüle proceeded from this observation as his starting point to argue (cogently) against the differentiation of source documents from the Urkundenhypothese in Numbers 22–24.36 He recognized the problems of cogency in the sources (see the examples below) and affirmed the observation of Noth (see below) that, without the premise that the sources J and E must stand behind J and E, no one would come up with the idea of J and E from Numbers. With that in mind, he then turned to discuss the model of Gross (see above). He rejected Gross’s evaluation of Numbers 22–24 on the grounds that Numbers 22–24 must have existed as an independent narrative and that it could not have existed as such in the form that Gross reconstructed.37 However, Schüle has not demonstrated that Numbers 22–24 must have existed as an independent narrative. He just states that such was the case. He continues against Gross’s observations, by suggesting that Gross did not accurately recognize how the text of Numbers describes Israel in this pericope, generally using “the people”, but later in the story’s recounting also using “Israel”. While arguing against Gross,

35

Cf. Andreas Schüle, Israels Sohn – Jahwes Prophet: ein Versuch zum Verhältnis von kanonischer Theologie und Religionsgeschichte anhand der Bileam-Perikope (Num 22–24), Altes Testament und Moderne, vol. 17 (Münster: Lit, 2001), 501–51. 36 Cf. Schüle, Israels Sohn, 51–59. 37 Cf. Schüle, Israels Sohn, 60.

1.2 Is the Balaam Narrative in Numbers 22–24 Uniform?

15

however, Schüle himself overlooked one of the primary distinctions in the nomenclature about the people leaving Egypt: he stated unequivocally that Israel never appears before Num 23:7,38 which is simply false (cf. Num 22:2). This oversight becomes all the more ironic and problematic when he paraphrases this verse, leaving out “Israel”, on the very next page.39 Since Schüle then interpreted this as an important motif for the rest of the story and the incorporation of the oracles, namely that Balaam is the one who first really sees “Israel”, but it is patently incorrect, we can reject this portion of Schüle’s argumentation. Schüle then continued his approach by combining both of the aforementioned points to discern the development of a motif over the course of Numbers 22–24.40 Since neither of these points appears to be accurate, his reconstruction can be rejected. Additionally, one could raise the question as to whether a competent editor could not have created such a development in terms of sight and seeing, as Schüle has postulated. Must such a motif be the creation of a single author of a uniform text? Possibly, but that must be demonstrated. Rather, in general Schüle’s initial argumentation awakens the impression that he must argue for the uniformity of the pericope Numbers 22–24 in order to approach it theologically as an aspect of the whole canon.41 To that end, his argumentation would have probably functioned better with a purely synchronic evaluation of Numbers 22–24 without any recourse to diachronic processes behind the text.

1.2.5 Uwe Weise A more recent primary proponent who identifies Numbers 22–24 as a unity is Uwe Weise in his volume Vom Segnen Israels.42 Fundamentally, one must recognize that Weise’s objective is not some description of the figure Balaam, nor of a literary-historical analysis of Numbers 22–24. Rather, he approached Numbers 22–24 on the basis of text and communication theory. For the most part, thus, this study reads like a synchronic evaluation of the Balaam story that attempts theoretical explanations for the text as it stands that must stem from one hand. The unity is more presumed than argued; the argumentation focuses only on maintaining the unity. To that end, various marked breaks in the text are simply explained away as intentional. There are several problems with Weise’s model and his thesis. Some examples should aid in demonstrating this.

38

Cf. Schüle, Israels Sohn, 60. Cf. Schüle, Israels Sohn, 61. 40 Cf. Schüle, Israels Sohn, 61–65. 41 Cf. Schüle, Israels Sohn, 69–119. 42 Uwe Weise, Vom Segnen Israels: eine textpragmatische Untersuchung der BileamErzählung Num 22–24, Textpragmatische Studien zur Literatur- und Kulturgeschichte der Hebräischen Bibel (Gütersloh: Gütersloher Verlagshaus, 2006). 39

16

Introduction

From the outset, he smoothed elements of the text that one might consider diachronically relevant. For example, in one of the most curious matters in his evaluation of the text of Numbers 22–24, he simply correlates Moab and Balak in the first four verses of Numbers 22.43 In my opinion, this appears to be an unsatisfactory answer to the question of the duplication in verses 2 and 4 with the transition of subject to Moab (and its duplication, as well) in order to grade the story’s opening. He applied the same logic with the terminology for Israel, which changes a few times in Num 22:1–4, a matter that Weise ignored, regarding them as uniform. He overlooked conspicuous duplications. These oversights become all the more poignant, since he described the syntax of the text with precision and attention to detail. Why should that attention to detail stop at the lexical level? That presents a serious deficit to his thesis. More likely, there is secondary material to be found in Num 22:1–4. However, following Blum’s model (see below), Weise ultimately recognized that these four verses stemmed from an editorial hand that compiled the Balaam story with the preceding material.44 Does his constitution of the rest of Numbers 22–24 as a unit fare any better? The most poignant example of a literary-critical break in Numbers 22–24 and the one that has thus far received the most consensus is the interpolation of the story of Balaam’s ass. However, Weise did not regard this as an insertion. Nevertheless, Weise’s rejection of the secondary nature of 22:21aβb–35a also fails to convince. His points are: 1) other texts demonstrate similar tensions; 2) the transition in the prepositions from ‫ את‬in 22:20 to ‫ עם‬in 22:21; 3) the repetition of the root √‫ יסף‬in 22:15a, 19b, 25b, and 26a.45 To the first point: other texts demonstrating such tension could also evince editing. To make this claim stick, Weise would have to demonstrate that no tension in the Bible is the result of editing, particularly in the cases he cited. That is a monumental task, even in the cases he cited. Therefore, this argument does not appear particularly strong. To his second point: this transition in prepositions possibly transports meaning and might be significant, suggesting that God permitted Balaam to go with them, but he joined them instead. Should that have been the case, one wonders if the preposition carries the same meaning in, e.g., 22:22, and why Balaam’s servants, who apparently joined the movement causing anger did not receive any warning or punishment. Distinct prepositions, particularly those with extensive semantic overlap hardly present overwhelming evidence of narratological development. Such a transformation, of one preposition to the other could have happened at any stage in the text’s transmission and need not affirm the unity of the composition. This argument is pretty weak. 43

Cf. Weise, Vom Segnen Israels, 72. Cf. Weise, Vom Segnen Israels, 187–96. 45 Cf. Weise, Vom Segnen Israels, 105–6. 44

1.3 Is the Story in Numbers 22–24 the Product of Multiple Sources?

17

To his third point: The √‫ יסף‬constructed with an infinitive in sentences describing repeated activity is normal Hebrew syntax. Is there some other way to express this sentiment? Are all sentences in the HB that use such constructions products of a single source? That hardly seems likely. This argument is as flimsy as the other two. The text in Num 22:22–35 demonstrates obvious tensions with the surrounding material that cannot be readily explained in the conception of a single author. It remains implausible that the same author whose God commanded Balaam to go with Balak’s intercessors should be the author whose God then enflames in anger at Balaam doing as he was commanded. The appearance of an intermediary between the deity and Balaam in 22:21aβb–35a, where none was needed in 22:9–20 makes it unlikely that these episodes are from the same hand. The theology of 22:21aβb–35a is distinct from that of the surrounding material, which implies that it is not from the same composer as the surrounding material. That example, and the difficulties with the text’s exposition should suffice to demonstrate the difficulties with Weise’s reconstruction, even ignoring for the moment other tensions, like in 24:14. With that, I conclude the overview of some positions that commend regarding Numbers 22–24 as a unit.

1.2.6 Summary of Unified Models The fundamental matter that scholars emphasizing the unity of the text have demonstrated remains the generally unified character of Numbers 22–24. They recognize that, in general, the text recounts one narrative with some connected oracles. That is a welcome observation, as we will see in the following study. However, they fail to account in all cases for the various tensions and breaks within the text. Some of these are quite conspicuous, such as the donkey episode, for which – admittedly – some, like Rofé, are willing to make an exception. These models often tend to emphasize the independent character of the narrative and its freedom from the surrounding material. Because of the remaining tensions and the current factual interdependence between Numbers 22–24 and its surrounding, however, many exegetes have sought to reconstruct a plurality of sources, usually two, behind the text of Numbers 22–24. To this we now turn.

1.3 Is the Story in Numbers 22–24 the Product of Multiple Sources? 1.3 Is the Story in Numbers 22–24 the Product of Multiple Sources?

Since the uniform coherence of the text must be called into question, other compositional models for Numbers 22–24 must be sought. In the development and advancement of the Urkundenhypothese(n), many scholars postulated or subsequently affirmed the existence of source documents – usually four in

18

Introduction

number – behind the Pentateuch.46 These documents were combined, presumably keeping as much material from each source as possible, in order to form the whole of the Torah. These positions dominated the landscape of Pentateuch research for most of the 19th and 20th centuries and have enjoyed a renaissance among some contemporary scholars.47 The sources, generally postulated based on material in Genesis, Exodus, and Deuteronomy, were traced with greater or lesser finesse through the whole of the Pentateuch. To arrive at the content of a specific source, duplicate or triplicate material often played an important role, as did the divine nomenclature. That is, scholars often regarded the sources as reporting the same things in different versions with distinct terminology for God. When such tactics were applied to Numbers 22–24, scholars found duplicate material in Balaam’s visions and oracles as well as inconsistencies within the naming of God (as either ‫=[ אלהים‬E] or ‫=[ יהוה‬J]) and thus developed several similar models for dividing the story into two narrative strands, each a part of a respective Pentateuchal source: J or E.48 However, few have reflected on whether Numbers 22–24 inherently justifies this approach and whether either of these postulated, reconstructed strands can be understood in and of itself or connected to the preceding and subsequent material in their postulated source documents.49 These problems are particularly acute in older version of the Documentary Hypothesis, but remain in newer iterations as well.

46 N.b. this discussion of various iterations of the Documentary Hypothesis neither seeks nor desires to be exhaustive. Such an exercise would be, due to the massive amounts of literature and the long-running popularity of this model, tedious and superfluous to the discussion here. For anyone seeking a more exhaustive list, I would point them to the table in Gross, Bileam, 419–27, which covers the material well, at least until 1974. In this discussion, I have focused on representative models that have especially impacted scholarly reception. 47 Cf., e.g., Joel S. Baden, The Composition of the Pentateuch: Renewing the Documentary Hypothesis (New Haven; London: Yale University Press, 2012). 48 For Nöldecke, Num 22:1 was part of his Grundschrift (later identified as the Priesterschrift) and 22:1 alone in Numbers 22–24 belonged to this composition in his reconstruction, whereas the rest stemmed from other sources; cf. Theodor Nöldeke, Untersuchungen zur Kritik des Alten Testaments (Kiel: Schwers’sche Buchhandlung, 1869), 86–87. Adherents to the documentary hypothesis generally followed Nöldecke in identifying P in 22:1 and only in 22:1 in Numbers 22–24. 49 Cf., e.g., Max Löhr, “Bileam, Num 22,2–24,25,” AfO 4 (1927): 86: “Wie will man angesichts dieser Textüberlieferung ernstlich eine jahwistische und elohistische Quelle feststellen?” Even earlier, Heinrich Holzinger, Numeri, KHC, vol. 4 (Tübingen; Leipzig: Mohr Siebeck, 1903), 107 noted that “…der Gebrauch von ‫ ַי ְהוֶ ה‬und ‫ֹלהים‬ ִ ‫ ֱא‬ist, wie Sam. und LXX nahelegen, in der Bileamperikope zum Teil gründlich in Verwirrung geraten und leistet keinen Führerdienst.” Nonetheless, while regarding Holzinger as advancement over previous models, Rudolph still rejected his thesis that divine terminology cannot lead to literary-critical decisions in this pericope; cf. W. Rudolph, Der “Elohist” von Exodus bis Josua, BZAW, vol. 68 (Berlin: de Gruyter, 1938), 103–5. In this, he relied exclusively on Bruno Baentsch, Exodus, Levitikus, Numeri, HKAT, vol. 1.2 (Göttingen: Vandenhoeck & Ruprecht,

1.3 Is the Story in Numbers 22–24 the Product of Multiple Sources?

19

1.3.1 Julius Wellhausen Though not its first proponent, the starting point for much of this discussion about source criticism can be found with Julius Wellhausen, the most widely known version. In a very few pages of his Composition des Hexateuchs, he identifies the sources J and E in Numbers 22–24 and ascribes their combination to the Jehowist, a redactor whose primary function in Wellhausen’s model was the conflation of these two oldest sources.50 After more or less carefully dividing the material into the sources J and E, Wellhausen implies inability to achieve this goal: “Die Erzählung namentlich scheint ganz aus einem Guss und ist wol auch wirklich vom Jehovisten neu gegossen worden.”51 In this way, Wellhausen weakens his own thesis and demonstrates its improbability. He implies that the complete reformulation of the sources must have made their recovery impossible. The more responsible resolution would have been to reject the sources outright as an explanation, rather than create a productive Jehowist as a redactor. Nonetheless, this model continued to find followers for more than the next century. Still, many of them were able to operate in Numbers 22–24 without recourse to a Jehowist, as Wellhausen did. Others often were not as reluctant to reconstruct source documents as Wellhausen was. On the other hand, some needed even more editors to fit their supposed reconstructed sources into the grander scheme.

1.3.2 August Freiherr von Gall An early proponent of a source-critical model for the development of the Pentateuch with recourse to a number of redactions is August von Gall.52 Ultimately, he distributed the material in Numbers 22–24 into three unequal sources (J, E, and P, with P, of course, covering the least material) and several redactions.

1903), 595. More recently, and based on both text-critical and literary-historical considerations, Gross, Bileam, 69–80 demonstrated that dividing Numbers 22–24 into J and E sources based on the terminology used for the divinity remains inefficacious and insufficient. 50 Cf. Julius Wellhausen, Die Composition des Hexateuchs und der historischen Bücher des Alten Testaments, Dritte Auflage. (Berlin: Georg Reimer, 1899), 109–11 and 347–52. 51 Wellhausen, Composition des Hexateuchs, 111. 52 Cf. August Freiherr von Gall, “Zusammensetzung und Herkunft der Bileam-Perikope in Num. 22–24,” in Festgruss Bernhard Stade zur Feier seiner 25jährigen Wirksamkeit als Professor, ed. Wilhelm Diehl, et al. (Giessen: Ricker, 1900), 1–47.

20

Introduction

P: E: J: RJE: RII: RIII: RIV: RV: RVI: Gloss:

22:1 22:3b, 4* (without ‫)אל־זקני מדין‬, 5*, 6*, 7*, 9–10 ,12, 20, 36, 38, 40; 23:1, 3*, 5, 6a, 12 22:2, 3a, 5*, 6aγ, 7aβ, 8, 11, 13–17, 18*, 19* (from ‫)שׁבו‬, 21 (to ‫)וילך‬, 22–34, 37, 39, 41; 23:1, 3*, 2a, 4b, 2b, 4a; 24:10a; 23:11; 24:11, and 25 22:6* (‫)אוכל‬, 18* (from ‫)אל־עבדי‬, 21*, 35* (‫ ;)עם שׂרי מואב‬23:7–10 23:13–24 23:25–24:9 24:12–19 24:20–22 24:23–24 23:6b

Table 1: August von Gall (1900)

His leading criterion for the division into sources remains the name of the divinity; in cases that present problems, he simply looked for a manuscript or textual tradition with the reading he required for his model.53 At the same time, he recognized that different names for Israel could demonstrate distinct provenience for the literary material.54 In order to fit his program, he requires no fewer than three source documents.55 To these, he adds no fewer than six editors.56 Nonetheless, none of his sources is legible or understandable and his six redactions can possibly be more readily and easily explained. He thus created more literary-critical problems than he solved, a common problem with these models.57 His final redaction requires a dating in the Roman Period in order to conform to his interpretation.58 That requires him in turn to note that G and Smr were constantly updated to accord with proto-M,59 which, while a possible explanation for their consistency, could be more readily explained through an earlier dating of that editorial expansion. Yet, it remains noteworthy that by 1900 at least some authors had begun to reckon with more complex redactional processes behind the Pentateuch.

1.3.3 Heinrich Holzinger In 1903, Holzinger, while admitting that “with some goodwill” [my translation] one could read Numbers 22–24 (without 22:22–34) as a uniform text,60 divided

53

N.b. examples of this in Gall, “Zusammensetzung und Herkunft,” 6–7, 9, 11, and 13. Cf. Gall, “Zusammensetzung und Herkunft,” 4–5. 55 Cf. the division into sources in Gall, “Zusammensetzung und Herkunft,” 9. 56 Cf. particularly Gall, “Zusammensetzung und Herkunft,” 13–16. 57 Cf. the critique of precisely this point in Schüle, Israels Sohn, 57–59. 58 Cf. Gall, “Zusammensetzung und Herkunft,” 43–46. 59 Cf. Gall, “Zusammensetzung und Herkunft,” 47. 60 Cf. Holzinger, Numeri, 104. 54

1.3 Is the Story in Numbers 22–24 the Product of Multiple Sources?

21

the text into narrative strands of J and E. In this he essentially followed Wellhausen’s division.61 Distinct from many of the other exegetes reflecting on this passage, Holzinger suggested that the episode describing Balak’s interaction with Balaam may be incomplete in both sources: no description of what happens to Balak after this incident remains.62 In this he disregards Num 24:25 as a sufficient conclusion.63 Like Wellhausen, Holzinger identified a creative editor RJE, who was able to craft a narrative that precluded the reconstruction of the pure source documents. Nonetheless, he attempted to do so. E:

J: RJE: P: Gloss:

22:2b, 3b, 4, 5* (including Pethor and bγ), 6b, 7* (without Midian), 8–11a, 12–16, 19–21aαb, 36, 38; 23:13aαb*, 14a, 1–2a, 4bα, 2b, 5a; 24:10aβb*; 23:12; and 24:13a 22:3a, 5* (Ammon + bβ), 6a, 11b, 17–18, 21aβ, 22–34, 37, 39–41; 23:3, 4a, 5b, 6a; 24:10aα; 23:11; 24:11b–12, and 13b 22:2a, 35; 23:4bβ, [6b,] 13aγδb, 14b–17 22:1, 4*, and 7* (Midianites) 23:6b, 13a*; 23:7–10; and 23:25–24:2

Table 2: Heinrich Holzinger (1903)

The most obvious issue in his reconstruction is the apparent and superfluous transposition of material within chapters 23–24. These transpositions remain implausible, needlessly complex, and do not even further the cause of reconstructing plausible sources. The extensive amount of material that Holzinger is forced to ascribe to “glosses” hardly impresses literary-critical sensibilities. The incomplete nature of his reconstructed sources commend their rejection and do not serve to aid in affirming the plausibility of source-critical resolutions to the tensions and supposed duplications in Numbers 22–24.

1.3.4 George Buchanan Gray Nor is this matter exclusive to German literature. George Buchanan Gray reflected a similar vein in his commentary on Numbers. He identified Num 22:1 as P and disregarded its relevance as a connecting element between 21:33–35 or 21:21–32 and 22:2. Yet, he failed to explain how this transition might have otherwise occurred.64 Generally, he divided the narrative such that Numbers 23 stems from E, whereas Numbers 24 generally stems from J, while chapter 22 presents an amalgamation of both.65 Much of this division relies on the names

61

Cf. Holzinger, Numeri, 105. Cf. Holzinger, Numeri, 106. 63 Cf. Holzinger, Numeri, 126. 64 Cf. Gray, Numbers, 306–7. 65 Cf. Gray, Numbers, 310–13. 62

22

Introduction

of God and the invention of new tensions in the text. For example, Gray presumed that multiple narratives were compiled here, which is why he suggests that 24:1, “attaches to nothing that precedes it”, which of course is only true if one precluded its attachment to the immediately preceding verses. Yet his inability to adequately divide the sources led him to identify passages as, e.g., “mainly J”.66 In this, he follows Wellhausen. Rather than reject the model of dividing the narrative into two strands, he continued to favor reconstructing narratives that remain senseless when regarded independently of each other. Thus, this model fares no better than similarly constructed ones.

1.3.5 Hugo Greßmann Still before the first World War, Hugo Greßmann suggested the redactional development of a particular source.67 Having divided the text of Num 22:2–40 into J and E strands,68 Greßmann regarded the whole of 22:41–24:25 as E, a segment of E without a J pendant.69 However, within this E tradition, he identified two editorial expansions exclusive to E. The first, E2 (23:25–24:3), was added to incorporate the third poem (24:4–9). The editorial character of this material becomes apparent from the distinct manner of revelation in 24:2 and the independent character of the fourth oracle, attested by the introduction of its author.70 An even later E editor, E3, appended the fourth oracle (24:16–24) through the addition of 24:12–15. Greßmann’s evidence for this E-internal expansion is the boring repetition (“langweilige Wiederholung”) of 22:18 in 24:12–13 and the “stilwidrig[er]” character of Balaam’s lengthy final oracular complex (24:17–24). Balaam’s lengthy answer to Balak contradicts 24:10–11; a king would not tolerate such a verbose answer.71 That last point rests on theoretical historical plausibility and cannot be regarded as literary-critically relevant: the literary critic of the Bible should not be interested in the historical plausibility of what some actual Balak may or may not have permitted in his presence, but rather with the rigor of his consistent image in the primary literature. None of Greßmann’s versions presents a real advance in terms of legibility over that of his predecessors. To some degree, he does represent a turn from the Jehowist of Wellhausen.

66

Cf. Gray, Numbers, 324. Cf. Hugo Greßmann, Mose und seine Zeit. Ein Kommentar zu den Mose-Sagen, FRLANT, vol. 18 (Göttingen: Vandenhoeck & Ruprecht, 1913), 318–34. 68 Cf. Greßmann, Mose und seine Zeit, 318–19. 69 Cf. Greßmann, Mose und seine Zeit, 328–31. 70 Cf. Greßmann, Mose und seine Zeit, 329. 71 Cf. Greßmann, Mose und seine Zeit, 329. 67

1.3 Is the Story in Numbers 22–24 the Product of Multiple Sources? P: E: E2: E3: J: R:

23

22:1 22:3b, 4* (without ‫)אל־זקני מדין‬, 5*, 7*–10, 12–21, 36, 38, 40–41; 23:1–24, 24:10–11, and 25 23:25–24:9 24:12–24 22:2, 3a, 5* (‫)בני־עמון‬, 6, 7a*, 11*, 22–34, 37, 39; 23:28; 24:1aαb, 2aαb, 3– 10aαb, 11–12a, 14aβb–17 (18–19) 22:5* (‫)פתורה אשׁר על־הנהר‬, 35

Table 3: Hugo Greßmann (1913)

1.3.6 Otto Eissfeldt Another example of a source-critical division without a Jehowist can be found in Otto Eissfeldt’s Hexateuch-Synopse.72 Eissfeldt distributes the material as in the following table. P: E: J: R:

22:1 22:2, 3b, 8*, .9–10a, 12*, 13*, 19a, 20, 21*, 36*, 38b, 40–41; 23:1–26 22:3a, 4–8*, 10b–11, 12–13*, 14–18, 19b, 21*, 22–35, 36*, 37–38a, 39; 23:28; 24:2–19, 25 22:4*, 7*; 23:27, 29–30; 24:1, 20–24 (Possibly also 24:18–19, since one anticipates finding only an oracle regarding Moab)

Table 4: Otto Eissfeldt (1922)

The redactor in this model combined all of the sources at the same time. While P could match its postulated surroundings in Eissfeldt’s model (preceded by Num 20:22–30 and followed by Num 25:19–26:65), J and E present difficulties. Neither of Eissfeldt’s reconstructed J and E narratives can be understood by itself; they mutually rely on each other, a matter that contradicts their originally coming from separate narratives. In other words, their mutual dependency precludes their independent backgrounds. Some examples should help to clarify this. 72 For this discussion, cf. Otto Eissfeldt, Hexateuch-Synopse. Die Erzählung der fünf Bücher Mose und des Buches Josua mit dem Anfange des Richterbuches in ihre vier Quellen zerlegt und in deutscher Übersetzung dargeboten samt einer in Einleitung und Anmerkungen gegebenen Begründung, reprint, 1922 (Darmstadt: Wissenschaftliche Buchgesellschaft, 1973), 183*–189*; Otto Eissfeldt, “Die Komposition der Bileam-Erzählung. Eine Nachprüfung von Rudolphs Beitrag zur Hexateuchkritik,” ZAW 57 (1939): 212–41 (= Otto Eissfeldt, “Die Komposition der Bileam-Erzählung. Eine Nachprüfung von Rudolphs Beitrag zur Hexateuchkritik,” in Kleine Schriften. Zweiter Band, ed. Rudolf Sellheim and Fritz Maass [Tübingen: Mohr Siebeck, 1963], 199–226); and Otto Eissfeldt, “Sinai-Erzählung und BileamSprüche,” HUCA 32 (1961): 179–90 (= Otto Eissfeldt, “Sinai-Erzählung und Bileam-Sprüche,” in Kleine Schriften. Vierter Band, ed. Rudolf Sellheim and Fritz Maass [Tübingen: Mohr Siebeck, 1968], 21–31).

24

Introduction

The fractions of verses from Num 22:8 and 19 that Eissfeldt attributes to E cannot be sensibly read as part of that narrative without presuming that substantial material is missing. Rather, they would better fit his other postulated source, J. The same is true of the fractions of verses in 22:12–13 that he attributes to J: without the material attributed to E, the J material in his reconstruction is nonsense (“and he said to them: stay here, as YHWH speaks to me. Balak ben Zippor, king of Moab spoke to me, saying…” and “Come now, curse it for me. Perhaps I could engage it and defeat it. You should not go with them…”).73 Beyond this, the reconstructed J and E sources in Eissfeldt’s model do not readily match their reconstructed narrative context. The J source reads “and Israel dwelt in the land of the Amorites. And Moab greatly feared the people because it was large…”.74 The transition to Moab presents a non sequitur and implies that these two elements do not belong together. Eissfeldt’s E in Numbers 22 begins with “and Balak ben Zippor saw everything that Israel did to the Amorites” in 22:2, which presents a problem, as Israel has done nothing to the Amorites in Eissfeldt’s E before 22:2.75 He consequently created an improbable textual link. The end of the narrative fares no better in Eissfeldt’s model. After Balaam returns home in Eissfeldt’s J, the reconstructed narrative continues “and the people began to whore with the daughters of Moab” (25:1b). This revelation surprises Eissfeldt’s audience, as Israel was not in Moab in Eissfeldt’s J, but was rather in the land of the Amorites (21:31).76 While such a transition is theoretically possible – granted, the rest of Eissfeldt’s J material in Numbers 22– 24 deals with the Moabites – it can hardly be regarded as obvious. E fails even to fulfill such a low standard of narrative quality by jumping from Num 23:26 (23:27, 29–30; 24:1 are regarded as coming from a redactor in Eissfeldt’s model) to Num 32:1: “And Balaam answered and said: did I not say thus to you: all that God will say to me, that I shall do? And the Reubenites and the Gadites had many animals…”. Thus, both reconstructed narrative strands in Numbers 22–24 fail to conjoin with the reconstructed sources preceding and following them. Such a reconstruction fails to convince or satisfactorily interpret the evidence. Rather, the guiding principle behind it – and the only reason that someone could legitimately attempt to reconstruct such sources – is the variant nomenclature for the deity: Eissfeldt sought to keep texts about ‫יהוה‬ distinct from texts about ‫אלהים‬. These examples demonstrate two things: 1) Num 22–24 cannot be separated solely based on the nomenclature for God into two original strands that could be regarded as even remotely complete in

73

Cf. Eissfeldt, Hexateuch-Synopse, 184*. All translations from the German are mine. Cf. Eissfeldt, Hexateuch-Synopse, 183*. 75 Cf. Eissfeldt, Hexateuch-Synopse, 182*–183*. 76 Cf. Eissfeldt, Hexateuch-Synopse, 183* and 190*. 74

1.3 Is the Story in Numbers 22–24 the Product of Multiple Sources?

25

these cases: they would be nonsensical; and 2) neither version could be a Fortschreibung of the other, since each remains incomplete without the other. Thus, the narrative is better understood as a unity in these cases. These examples from Eissfeldt’s reconstruction could be readily multiplied and repeated with the preceding cases.

1.3.7 Martin Noth Martin Noth, although recognizing the limits of the Documentary Hypothesis in the book of Numbers and expressing some reservation, still favored this model above all others in his commentary of the book. He famously noted: “Nimmt man das 4. Mosebuch für sich, so käme man nicht leicht auf den Gedanken an ‘durchlaufenden Quellen’, sondern eher auf den Gedanken an eine unsystematische Zusammenstellung von zahllosen Überlieferungsstücken sehr verschiedenen Inhalts, Alters und Charakters (‘Fragmentenhypothese’). Aber es wäre eben […] unsachgemäß, das 4. Mosebuch zu isolieren. Es hat im alttestamentlichen Kanon von Anfang an zu dem größeren Ganzen des Pentateuch gehört; und auch die wissenschaftliche Arbeit an diesem Buch hat immer wieder nur bestätigen können, daß es in diesem größeren Zusammenhang gesehen werden muß. Es ist daher gerechtfertigt, mit den anderwärts gewonnenen Ergebnissen der Pentateuchanalyse […] an das 4. Mosebuch heranzutreten und die durchlaufenden Pentateuch‘Quellen’ auch in diesem Buche zu erwarten, selbst wenn, wie gesagt, der Sachverhalt im 4. Mosebuch von sich aus nicht gerade auf diese Ergebnisse hinführt. Doch muß angesichts der besonderen Art des Buches diese Anwendung gewonnener Ergebnisse mit Zurückhaltung und Vorsicht vorgenommen werden; und es geht keinesfalls an, einfach nur eine Aufteilung des Textbestandes auf die Pentateuchquellen J, E und P (und allenfalls noch sekundäre Wucherungen dieser Quellen) durchführen zu wollen.”77

Nonetheless, his identification of the sources in the Balaam narrative presents a series of problems similar to those identified in the preceding iterations. Comparing all of the notices that Noth proffered, his reconstructed sources in Numbers 22–24 distribute the material as in Table 5: Martin Noth (1966).78 P: E: J: R:

22:1b 22:2, 9–12a, 17b?, 20, 38aβb, 41; 23:1–2, 3–6?, 7–26 22:4–8, 12b?, 13–19, 21–35, 36–38aα, 39–40; 23:28*; 24:1–19 22:1a, 4*, 7*, 23:27–30*; 24:1aβ, 10*, 20–24

Table 5: Martin Noth (1966)

It should be immediately apparent that Noth mentioned no specific source(s) for anything in 22:3, though he did identify it as a doublet.79 The distinction 77 Martin Noth, Das vierte Buch Mose: Numeri, übersetzt und erklärt, ATD, vol. 7 (Göttingen: Vandenhoeck & Ruprecht, 1966), 8. 78 To the following table, cf. Noth, Numeri, 150–64. 79 Cf. Noth, Numeri, 155.

26

Introduction

from Eissfeldt’s reconstruction are limited, but the problems identical. As an example, we can consider the first sentences of Noth’s E: “And Balak ben Zippor saw everything that Israel did to the Amorites. And God came to Balaam and said: who are these men with you?” J fares no better in this regard; the connection of 22:8b to 12b–13 should serve exemplarily: “And the princes of Moab stayed with Balaam. You shall not curse the people, for it is blessed. And Balaam arose in the morning and said to the Princes of Balak: return to your land…” Attributing these verses to these sources (and redaction) fails to eliminate the problems identified in Eissfeldt’s source-critical undertaking. The reader is left with two portions of incoherent text, each illegible without the other, and a handful of minor additions included to smooth the current narrative. Such a solution is hardly desirable when better alternatives can be found. Elsewhere, Noth identified the common feature of every tradition about Balaam in his affiliation with Peor, going so far as to describe that as virtually the only feature readily recognizable about this character. “Allen Zweigen der Bileamerzählung gemeinsam und darum dem ältesten Bestand der Überlieferung zuzurechnen ist außer dem Namen die Verbindung mit dem Heiligtum des Baal Peor und das Bild von einem mit ‘Macht’ des Wortes begabten ‘Wahrsager’. Darüber hinaus ist nichts einigermaßen Sicheres mehr zu ermitteln.”80

This image must be refined, particularly as Peor plays no role at all in Deuteronomy 23, Joshua 24, Judges 11, and Micah 6. In the same context, Noth recognized two variations on themes about Balaam. In the first, he identified three stages of the integration of the initially foreign Balaam story into the Pentateuch. 1) Israel’s God’s greater power over the curses of a foreigner and God’s transformation of the curse into blessing (Deut 23:5–6); 2) God prevented Balaam from even invoking the curse; 3) Instead of the curse, God forced him to utter a blessing. The other line of development turned into the material known in Num 31 and Josh 13, in which Balaam remained a dangerous foreigner.81 “Damit war der ehedem heidnische Zauberer tatsächlich in einen israelitischen Gottesmann verwandelt und die Voraussetzung dafür geschaffen, ihm weissagende Segenssprüche für Israel in den Mund zu dichten, wie es die beiden Bileamlieder von Num. 24 sind, die auch ihrerseits die in der erzählenden Weitergabe seit alters überlieferte Verbindung des Bileam mit dem Heiligtum des Baal Peor wenigstens andeuten.”82

However, as will become apparent in the discussion below, the transformation of this character must have followed different lines than those Noth preliminarily sketched.

80 Martin Noth, Überlieferungsgeschichte des Pentateuch, 3. reprint, 1948 (Darmstadt: Wissenschaftliche Buchgesellschaft, 1966), 83. 81 Noth, ÜGP, 83–85. 82 Noth, ÜGP, 85.

1.3 Is the Story in Numbers 22–24 the Product of Multiple Sources?

27

1.3.8 Jules de Vaulx Jules de Vaulx generally followed these source-critical designations in the Balaam pericope.83 However, he added stages in the development of the texts before their incorporation into the sources or into the larger Balaam composition. He recognized two sources with two parenthetical processes (i.e., one at each end, the beginning and the conclusion) behind the whole. The process began with a pre-Israelites, probably Gadite, legend about the Mesopotamian Balaam. From this legendary tradition, the Yahwistic and Elohistic oracles developed. These oracles then became part of their respective sources. Finally, an editor combined the traditions. This presents a traditional view about the Balaam narrative evincing the same problems as the other source-critical explanations with the added insecurity about the material preceding the sources on some literary or even pre-literary stage.

1.3.9 Baruch Levine At the outset of the twenty-first century, Baruch Levine commended another source-critical appreciation of Numbers 22–24. He regarded Numbers 22–24 as generally E’s product.84 This E material was embedded in its current context through the addition of verses 22:1 and 2, the former being “a priestly postscript to Numbers 21” and the latter presenting the primary connection of Numbers 22–24 to its current context.85 There is obviously some truth in this, as reiterated here in Chapter Three. Levine also identified the oracles as independent and coming from another source and the “Tale of the Jenny” as an interpolation.86 While this more or less uniform image may be correct for much of the matter in Numbers 22–24, the identification of the oracles as separate from the narrative remains problematic, as do some of his other observations. For example, Levine – having identified 22:2 as redactional – regarded 22:4 “as a gloss, linking Numbers 22:3–21 to the introductory verse, Numbers 22:2.”87 This interpretation cannot be accurate, as removing one verse of a parallel set from the narrative as part of a redaction and identifying the other as a gloss, results in the unfortunate circumstance of having no introduction for the character Balak in the original narrative. He would just appear from nowhere 83

Cf. Jules de Vaulx, Les Nombres, Source bibliques (Paris: Gabalda, 1972), 256–65. Cf. Baruch A. Levine, Numbers 21–36, Anchor Bible, vol. 4A (New Haven: Yale, 2000), 138. For Levine’s general adherence and particular take on the Documentary Hypothesis (including the addition of his poetic source “T” [as in Transjordan] in Numbers 21; 23– 24), cf. Baruch A. Levine, Numbers 1–20, Anchor Bible, vol. 4 (New York: Doubleday, 1993), 48–50. 85 Cf. Levine, Numbers 21–36, 139 to the former, and Levine, Numbers 21–36, 137 to the latter. 86 Cf. Levine, Numbers 21–36, 138. 87 Levine, Numbers 21–36, 145. 84

28

Introduction

at the end of 22:7, a verse that mentions his words. Removing 22:4 as a gloss would also leave v. 5 without an obvious subject: who is the “he” implied in the verb ‫ ?וישׁלח‬Is it Moab? Such circumstances remain most implausible. Levine regarded Num 22:21 as the original conclusion of the material preceding the donkey story, with 22:22 presenting the opening of that insertion. In this model, the story of the donkey then concludes with 22:35. The problem appears in the reference to the donkey in 22:21, but the lack of one in 22:35. His model regards the story of Balaam’s donkey as having existed in some form independently from its current context. Yet he failed to offer compelling evidence to commend this postulation. At no point does the text of Numbers 22:21–35 identify Balaam by any other means than his name; it provides no data about him. That is, the narrator presumes the audience’s familiarity with this character. Following such a detailed, multifaceted introduction in Num 22:5, such a paucity of information would not be a problem, but without any such data, the lack of Balaam’s background seems implausible within an independent source. One particularly acute problem with Levine’s theory remains conspicuous: the lack of an antecedent for ‫ הוא‬in 22:22 should one remove 22:21 from before it. In order to accommodate this transition, Levine suggested that a redactor removed the name Balaam from 22:22 in order to afford a better reading.88 However, that would be completely unnecessary; a reading with Balaam in 22:22 would not in any way disturb the narration, and such minor editing seems difficult to substantiate. His sole purpose was to maintain 22:22– 35 having previously existed as an independent narrative. Should one regard the donkey’s tale as a Fortschreibung or a redactional addition, this problem disappears entirely. The expansion of the text – and not the insertion of an independent tradition – presents the best model for understanding the compositional embedding of the story of Balaam’s donkey. I will further develop this matter in Chapter Three. Levine’s suggestion that the oracles come from an independent source remains problematic.89 First and foremost, the story of Numbers 22–24* without the oracles neither climaxes nor makes sense. Balak’s interactions with Balaam remain without explanation or merit, and the story would thus be unable to advance any further. At the same time, the oracles reference their narrative surroundings. Both the oracles (Num 23:22 and 24:8) and the narrative framework (Num 22:5) refer to the exodus from Egypt. The oracles presume Balak’s (23:7 and 18) summoning Balaam (24:3 and 15) to come do his bidding. Without the narrative, the circumstances of these oracles remain just as unclear. Perhaps Friedman phrased it best: “The poetry is embroidered more intimately in its prose context than most biblical poetry that is housed in the narrative books. Neither the narrative nor the poetry is complete without 88 89

Cf. Levine, Numbers 21–36, 153. Cf., e.g., Levine, Numbers 21–36, 41.

1.3 Is the Story in Numbers 22–24 the Product of Multiple Sources?

29

the other. The narrative context sets the scene and creates the ironic character of the poetry: the Moabite king has hired Balaam to curse Israel, but Balaam blesses them instead. The poems themselves refer back to the narrative context and are referred to in turn in the course of the narrative: Balak hears the poems and complains about them. The Balaam episode is a splendid example of the merging of poetry and prose in the Hebrew Bible.”90

For these reasons, the story necessitates the oracles and the oracles demand a connection to their surrounding material. Neither makes sense without the other and both harmonize together, presenting a generally consistent whole. However, with these addenda, Levine’s position can be followed to a greater degree than other systems that generally operate on Numbers with recourse to the Documentary Hypothesis. The Balaam narrative tends to present a largely consistent story with connections to the preceding material, as can be seen in, e.g., Num 22:2. At a later stage, someone expanded the narrative with the tale of the donkey and later oracles.91 While one should disagree with some of the specifics of his model, Levine’s tendency at least seems the most correct, especially when compared to the other proponents of the Documentary Hypothesis. His focus on the general unity and independence of the narrative, while in need of modification and a divorce from the Urkundenhypothese, presents the most acceptable iteration of source-critical research.

1.3.10 Axel Graupner A few years later, Axel Graupner attempted a reconstruction, his being one of the most complicated presented here. Graupner introduced the Balaam pericope thus: “Eine Überleitung […] muß das Werk enthalten haben; denn die Bileamperikope Num 22– 24 enthält wieder sämtliche Anstöße, die bisher zu einer Quellenscheidung nötigten: den Wechsel von Elohim und Jahwe, die Doppelung des Stoffes und die Koinzidenz beider Phänomene, außerdem Widersprüche, die so massiv sind, daß sie sich mit einem traditionsgeschichtlichen oder einem Fortschreibungsmodell kaum erklären lassen.”92

Divine nomenclature as a problem has been sufficiently addressed in the preceding models and will return as an issue in Chapter Three. The duplications he referred to are 22:2 with 4b, 22:3a with b, and 22:41–23:26 with 23:28– 24:19. However, his last – and most significant – parallel texts are not really parallels. Rather, the one (23:28–24:19) builds upon and advances the logic of 90 Richard Elliott Friedman, Commentary on the Torah with a New English Translation and the Hebrew Text, reprint, 2001 (San Francisco: Harper Collins, 2003), 507. The contrast between the embedding of poetry in prose material in the Balaam story is probably most stark when compared to the poetic materials in Numbers 21. 91 To the addition of the final oracles, cf. Levine, Numbers 21–36, 237–38. 92 Axel Graupner, Der Elohist: Gegenwart und Wirksamkeit des transzendenten Gottes in der Geschichte, WMANT, vol. 97 (Neukirchen-Vluyn: Neukirchener Verlag, 2002), 159– 60.

30

Introduction

the other (22:41–23:26).93 Since refuting these two problems, divine monikers and duplications, as relevant to source criticism precludes their coincidence, Graupner remains without much evidence. Rather than immediately reference the contradictions to which he referred (as in the other proffered evidence), he suggested that tradition-critical and Fortschreibung models are impossible because their progenitors each determined differing amounts of Fortschreibung and distinct dates for their reconstructed texts.94 Yet, does the mere existence of various opinions qualify as evidence? Have proponents of source-critical undertakings in the Balaam passage overcome these differences? According to the preceding (and following) survey, they have not. Graupner even admitted this (“Allerdings wird die Aufteilung auf J und E verschieden vorgenommen.”95) without noting the irony. Graupner followed Wellhausen in identifying tension in Num 22:36–38.96 However, like Wellhausen, Graupner ignored the fact that without verse 36, Balak would not be in Balaam’s presence. Such a hole in the text precludes the division into sources in this case. Again, this time following Seebass,97 Graupner disregarded the third and fourth oracles as the natural continuation of the first two. He based this interpretation on Seebass’ assessment that Balaam’s introducing himself becomes superfluous in a text that had already introduced him.98 This introduction, particularly in the third oracle, focuses the reader’s attention on Balaam and does not mandate an origin distinct from the preceding oracles. Rather, a narratological transformation has taken place, setting the third oracle apart from those that preceded it. 99 While the first two oracles begin with Balak, the third prophetic injunction changes the perspective to Balaam in its exposition. In his reconstruction of J and E, Graupner permitted J one group of emissaries,100 but noted that there were two in the E account.101 It remains entirely unclear how he could reconstruct three groups going to Balaam from Num 22–24

93

Cf. the discussion to this point in Levine, Numbers 21–36, 210–12. Cf. Graupner, Elohist, 160 Fn. 17. 95 Graupner, Elohist, 160. 96 Cf. Wellhausen, Composition des Hexateuchs, 348–49. 97 Cf., e.g., Horst Seebass, Numeri 22,2–36,13, BKAT, vol. 4.3 (Neukirchen-Vluyn: Neukirchener, 2007), 46. 98 Cf. Graupner, Elohist, 161. 99 Cf. Jacob Milgrom, Numbers = [Ba-Midbar]: The Traditional Hebrew Text with the New JPS Translation, commentary by Jacob Milgrom, The JPS Torah Commentary, vol. 4 (Philadelphia: Jewish Publication Society, 1990), 467: “The poetry was composed for the sake of the prose. Without the narrative, the poetic oracles would make no sense, and all their allusions to personalities, nations, and events would be incomprehensible.” 100 Cf. Graupner, Elohist, 161. 101 Cf. Graupner, Elohist, 162. 94

1.3 Is the Story in Numbers 22–24 the Product of Multiple Sources?

31

in any extant version. Graupner’s evidence for E consists primarily of the name Elohim and God’s appearing to Balaam.102 Further, Graupner regarded the elements reflecting on God’s prohibiting Balaam’s travel as additions (“Zusätze”) with a twofold purpose: characterizing Balaam as an obedient prophet of YHWH (!) from the outset and focusing on God’s identity presented in his name.103 He failed to include 22:21 in any of his sources or redactions and divided 23:1–6 pretty weakly, again based largely on the appearance of the Tetragrammaton (while at the same time ignoring the text-critical issues which could have supported him in this case). He identified E in 23:1–2, 3aα, 4a and Z in 3aβbc, 4b, and 5–6.104 He regarded 24:8a as an addition,105 but did not list it anywhere as such in his results,106 where it must be regarded as part of J. According to Graupner’s reconstruction, RJE demonstrates an inconsistent tactic in the Balaam pericope: “In anderer Hinsicht weicht der Befund jedoch ab. Während RJE | üblicherweise die jahwistische Darstellung der elohistischen vorzieht und E in J einfügt, hat er in der Bileamperikope die jahwistische Darstellung nur fragmentarisch, die elohistische dagegen geschlossen erhalten.”107 How probable is it, that a compiler or redactor would change his or her approach to the text for a single piece of a longer whole? Nevertheless, Graupner’s postulated editor maintains another method in this passage that is otherwise atypically RJE: “Außerdem ist der Anteil der jehowistischen Redaktion […] erheblich höher als gewöhnlich.”108 The primary editor changed his favored source and increased his editorial undertaking only for this passage in Graupner’s model. That hardly seems likely. In sum, one can reconstruct Graupner’s as in Table 6.109

102

Cf. Graupner, Elohist, 163. Cf. Graupner, Elohist, 16. 104 Cf. Graupner, Elohist, 165–66. 105 Cf. Graupner, Elohist, 170. 106 Cf. Graupner, Elohist, 171–72. 107 Graupner, Elohist, 171–72. 108 Graupner, Elohist, 172. 109 Graupner does not clarify if the reference to Midianites are from J or not. He leaves this open. His Z should be generally regarded as JE in the nomenclature of Ludwig Schmidt. 103

32 P: E: J: R: Z: RJE: RJEP: N:

Introduction 22:1b 22:3a, 4b–6, 7b, 8aαb, 9–13a, 14–17, 20–21, 36, 38, 40–41; 23:1–2*, 3aα1, 4a, 7–10a, 13–15, 18–23a, 24–26; 24:25 22:2, 3b, [4a*], 7a*, 22–34, 37, 39; 23:28; 24:1aαb, 2aαb, 3–10aαb, 11–12a, 14aβb–17 (18–19) 22:1a 22:4a*, 7a*; 24:1aβb, 8a 22:8aβ, 13b, 18–19, 35; 23:3aβb, 4b–6, 11–12, 16–17, 27, 29–30; 24:10bβ, 12b–14α [?] 22:1a; 24:2aβ 23:10b, 23b; 24:20–24

Table 6: Axel Graupner (2002).

Beyond the two illegible sources, he proffered no fewer than five redactions before the culmination of this story into its final edition. The problems with his reconstruction should be evident from the preceding discussion. Presumably a simpler reconstruction than Graupner’s can be found.

1.3.11 Ludwig Schmidt Ludwig Schmidt presented a somewhat more complicated reconstruction than the oldest iterations.110 Schmidt maintained the two sources J and E. To this end, Schmidt identified cross-references suggesting that the Balaam story contains three structures that can only be explained with two sources, combined and expanded upon by a first redaction. 1) Between the third oracle (24:3b–6a, 7–9a) and Balak’s message (22:5b–6). “Da in den Sprüchen 22,5b.6 nur in 24,8*f. vollständig berücksichtigt wird, stammen die Botschaft Balaks und der dritte Spruch von demselben Verfasser.” 111 The relationship between these verses makes it improbable that the currently third oracle was originally any oracle other than the first. The preparation for this first oracle can be found in 23:28 and 24:2–3a. The connection with the oracle is also visible in the root ‫ שׁכן‬as found in 24:2 and 5. Numbers 24:9 is related to Gen 27:29 and 12:1–3, both of which are J texts in the Documentary Hypothesis.112 2) The references to God placing a word in Balaam’s mouth and Balaam doing only what God says are related to each other and place the first two oracles in a two-fold framework between 22:20b and 23:26, as well as 22:38 and 23:5*, 12, and 16. Nothing could have followed Balak’s statement in 23:25 and Balaam’s answer to it 110 Cf. already Ludwig Schmidt, “Die alttestamentliche Bileamüberlieferung,” BZ 23 (1979): 234–57. 111 Ludwig Schmidt, “Bileam: Vom Seher zum Propheten Jahwes. Die literarischen Schichten der Bileam-Perikope (Num 22–24),” in Gott und Mensch im Dialog. Festschrift für Otto Kaiser zum 80. Geburtstag, ed. Markus Witte (Berlin: de Gruyter, 2004), 335. 112 Schmidt, “Bileam,” 335–36.

1.3 Is the Story in Numbers 22–24 the Product of Multiple Sources?

33

in 23:26. Since these passages all belong in a larger context, they must be attributed to the Elohist.113 3) The verses 23:27, 29–30; and 24:1* serve as a redactional bridge between the two versions. Numbers 23:29–30 copies from 23:1–2* and demonstrates that Balaam still thought it possible to curse Israel, which does not make sense following his second oracle (23:19–20). With 24:1* the redactor explains why Balaam did not leave Balak on this occasion and that he will bless Israel with this third oracle. The addition of the “edge of Israel” in 23:13 also comes from the redactor, as it interrupts the context of 13aα and 13b. Its addition only makes sense when considered with 24:1, in which Balaam sees the whole of Israel. Since these verses are redactional parentheses, there must have been two Balaam stories.114 The material in 22:8–19 must be the work of a redactor because it refers to both versions. The message from J (22:5–6) plays a role, but 8–19 also anticipate v. 20, which comes from Schmidt’s E. 115 The passage in 22:8–19 identifies Balaam as the obedient prophet of YHWH, which also connects this passage to 24:11b–13.116 Instead of only one redactor, Schmidt suggested that there were two: the Jehowist (JE) and the redactor of the Pentateuch (PentR, represented in the following table with R). By increasing the number of redactors responsible for additions to the text, he – like others before him – both increased the complexity of the development of the text vis-à-vis models like Wellhausen’s and decreased the intelligibility of the sources. Schmidt divided the text of Numbers 22–24 as in Table 7: Ludwig Schmidt (2004).117 P: E: J: JE: R:

22:1 22:2–3a, 20–21, 36*, 38, 41; 23:1–4a, 5*, 6a, 7–13aα, 13b–17aα, 18–26 22:3b–7*, 22–34, 37, 39–40a; 23:28; 24:2–9. 10a, 11a*, 12a, 14*, 15–17, 25 22:8–19, 35, 40b; 23:27, 29–30; 24:1aαb, 11b, 21b–13 22:4*, 5*, 7*, 36*; 23:4b, 6*, 10*, 17*, 22–23; 24:1*, 6*, 18–24

Table 7: Ludwig Schmidt (2004)

From even the first glance, it becomes clear that this process fractures the text significantly. This fracturing decreases the plausibility of reconstructing anything that could be called a “source” behind the material. Even in this splintering of the text into more redactional levels, some of the tensions in the text remain. For example, Schmidt does not include 24:14b–17 in one of his redactional levels. Rather, he ascribes it to J. The oracle’s interruption of Balaam’s departure does not seem to bother Schmidt. Combining such issues with the 113

Schmidt, “Bileam,” 336–37. Schmidt, “Bileam,” 337–38. 115 Schmidt, “Bileam,” 338–39. 116 Schmidt, “Bileam,” 338. 117 To the following, cf. his discussion in Ludwig Schmidt, Das 4. Buch Mose. Numeri, ATD, vol. 7/2 (Göttingen: Vandenhoeck & Ruprecht, 2004), 122–44. 114

34

Introduction

illegibility of the remaining sources suggests that a more plausible solution should be sought.

1.3.12 Horst Seebass One other attempt at the beginning of the 21st century in German literature, that of Seebass, maintained this model, with the same problems as those noted above.118 Still, it merits mention. Seebass’s “sources” remain just as illegible as Schmidt’s, although Seebass has the simplifying advantage of only one redactor. Nonetheless, this single redactor was responsible for significant portions of the text, just as was the case in Schmidt’s model. P: E: J: R:

22:1 22:2–3a, 5–6, 7b–21, 36–41; 23:1–4, 5b–13aα, 13b–15, 17b–23a, 24, 28; 24:1– 7, 8b–10bα, 11–19, 25 22:3b–4*, 7a*, 22–35 22:4*, 7*, 35; 23:5a, 13aβ, 16–17a, 23b, 25–27, 29–30; 24:8a, 10bβ, 20–24

Table 8: Horst Seebass (2007)

1.3.13 Joel Baden Joel Baden proffered the most recent attempt to divide the Pentateuch into sources as postulated in the Documentary Hypothesis. Unlike many of his precursors however, Baden left Numbers 22–24 largely intact, the product of a single hand, essentially taken from E without interpolation or redaction. Yet, his reconstruction ultimately fares no better than previous attempts under even modest scrutiny. That being said and to be fair, Baden admittedly barely dealt with Numbers 22–24 in his monographs. And when he did, it was generally from the perspective of Deuteronomy. In his first monograph on the Documentary Hypothesis, he offered the following about Balaam: “Though the Balaam pericope is notoriously difficult to analyze source-critically, most scholars [here, he references only Gray (see above) and Friedman, JMR]119 see the majority, if not all, of the pericope as deriving from E. As we have seen to this point, the author of D tends to use the E narrative as much as possible, including stories from J only when they are

118 Cf. Seebass, Numeri III, 1–107 and his earlier articles Horst Seebass, “Einige vertrauenswürdige Nachrichten zu Israel Anfängen: Zu den Söhnen Hobabs, Sichon und Bileam im Buch Numeri,” JBL 113 (1994): 577–85 and Horst Seebass, “Zur literarischen Gestalt der Bileam-Perikope,” ZAW 107 (1995): 409–19. 119 I.e., Gray, Numbers and Richard Elliott Friedman, The Bible with Sources Revealed: A New View Into the Five Books of Moses (San Francisco: HarperSanFrancisco, 2003).

1.3 Is the Story in Numbers 22–24 the Product of Multiple Sources?

35

found in that source alone and are necessary for his historiographical or theological argument. In this case, because the references to Ammon and Moab in Deut 23:4–5 are based on E, it is even more likely that the references to the Balaam pericope are based on E as well.”120

Leaving aside the qualification of two scholars as a majority, Baden failed to discuss the theological and historical distinction between the two presentations of Balaam in Numbers 22–24 and Deuteronomy 23. His argumentation consists essentially of “D knows E; D knows Balaam; therefore, Balaam is E”. However, he completely disregarded the distinct portrayals of this character in Numbers 22–24 and Deuteronomy. Never at any point does Numbers suggest that Balaam was willing to curse Israel, as the text of Deuteronomy indicates. Quite the opposite: the Numbers text repeatedly qualifies Balaam as presenting only that which he was instructed (Num 22:18, 38; 24:13). Beyond this, the connection to the Ammonites is missing in Numbers 22–24, unless one follows Smr. Such a reference appears to be a novelty in Deuteronomy. Baden’s argumentation here fails even to appreciate the overt distinctions between the witnesses and thus fails to convince. The perspective from Deuteronomy does not present the only problem with Baden’s reconstruction, however. For example, most material on Numbers 22– 24 presents part of Baden’s discussion of Christoph Levin’s Der Jahwist.121 Otherwise, he only really mentioned 22:1, a verse he ascribed to P, in passing.122 Baden made one oblique reference to the Balaam narrative, ascribing it to E: “The exclusion of Ammonites and Moabites from the community in Deuteronomy 23:4–5 refers explicitly to the lack of hospitality those nations showed Israel during the wilderness wandering, that is, in E (Num 20:14–21; 22–24).”123 Baden also identified the material about Sihon and Og immediately preceding Numbers 22–24 as E.124 That is, he attributed both Num 21* and 22– 24* to E, but Num 22:1 to P. The itinerary notice from P is necessary between these texts to cover the distance from Bashan to Moab, implying that some such notice must have been present in E as well. It is narratologically necessary in its contemporary context. Yet, is it really likely that a compiler – particularly as in Baden’s understanding – would delete any such notice from its extant E context and replace it with an element from a context in P when he or she could have just kept the E notice? The insertion of a single P element into a broadly E narrative structure seems most improbable, making the acceptance of this model at this point problematic. It even generally contradicts Baden’s own vision of his compiler: “Along the same lines, the compiler did not rearrange his 120

Joel S. Baden, J, E, and the Redaction of the Pentateuch, FAT (Tübingen: Mohr Siebeck, 2009), 184. 121 Cf. Baden, Composition, 61–62 presents the relevant discussion of Christoph Levin, Der Jahwist, FRLANT, vol. 157 (Göttingen: Vandenhoeck & Ruprecht, 1993). 122 Cf. Baden, Composition, 147. 123 Baden, Composition, 137. 124 Cf. Baden, Redaction, 130–41 and Baden, Composition, 119.

36

Introduction

source materials, but rather kept the contents of the documents in their original order.”125 Then why should some scribe insert a P verse into an otherwise uninterrupted E context? That would make no sense. Yet, this is not the only problem with the divisions in Baden’s model here. Several of the elements that he identified as typical of or even unique to J can be found in Numbers 22–24, which he has identified as E, e.g., Cain (Num 24:21)126 and Israel’s arrival at Pisgah in Moab (cf. Num 23:14).127 It is improbable that E and J would deliver such consistent data, should they have been entirely independent works that had not been edited together. This is precisely what Baden thought of them, however. A more likely explanation in this case would be that one text knew the other and edited it to conform, a position that Baden rejected: “First and foremost, the fact that when the two sources [J and E] are intertwined, as in Genesis 37, the Horeb pericope, and Numbers 11, E is not a series of interpolations, but a complete and coherent narrative in its own right; E is comprehensible only when isolated from J, and vice versa. In addition, some elements of E cannot be understood as theological updatings of J, such as the identification of the mountain as Horeb rather than Sinai or that of Moses’s father-in-law as Jethro rather than Reuel. Rather, these differences must emerge from a variant tradition, as they have no theological value on their own, much less as a reworking of J.”128

Baden suggested that E was unfamiliar with Esau and the identification of Esau and Edom.129 However, he overlooked the parallel usage of Edom and Esau in Num 24:18a.130 If the aforementioned elements are typical of or unique to J, how did they find their way into E? Baden provided no answer. Finally, Baden overlooked the internal tensions within the text and refused to discuss the dates of the composition of his reconstructed individual sources. He failed to address the multiple introductions of Balak and the various expositions of Numbers 22–24, the variant style and genre of the donkey story in Num 22:21–35, the conspicuous interpolation of Num 24:14b–24, and other matters. Some hand distinct from the source he identified must have been responsible for these tensions. Yet, Baden’s compiler did not undertake such additions, leaving the question of these additions unaddressed.131 Since Baden

125

Baden, Composition, 223. Cf. Baden, Composition, 68. 127 Cf. Baden, Composition, 80–81. 128 Baden, Composition, 126. 129 Cf. Baden, Composition, 143: “…E never mentions Esau, and seems not to know of Esau’s existence, or the identification of Esau and Edom.” 130 This oversight probably also stems from an over reliance on M, which does not offer this parallel at this point; cf. the text-critical discussion of this verse in the following chapter. 131 Cf. Baden, Composition, 224: “It should also be noted that the compiler’s insertions are very brief and tend to used the language of only a single source, rather than a combination 126

1.3 Is the Story in Numbers 22–24 the Product of Multiple Sources?

37

regarded only Num 22:1 as foreign to Numbers 22–24 and did not remove, e.g., the final oracles in Num 24:14b–24, one is left to wonder what the date of E’s composition must have been. Since it apparently references Greek aggression (Num 24:23–24), E most likely would have been composed sometime during or after the late Persian period at the earliest. Since D was familiar with E as an independent document and relied on it as such in Baden’s model,132 D would most likely be a Hellenistic composition at the earliest. The compilation of the Pentateuch must have thus also occurred sometime immediately before or during the translation of the Septuagint, if one accords P a date later than D and allows time for the compilation of the various sources. This timing seems improbable. A better explanation would be the later addition of the pertinent verses into an already extant context.

1.3.14 Summary of Source-Critical Models Since none of these models provides a convincing source-critical reconstruction for Numbers 22–24, one must wonder if there is a common methodological problem in their background. The difficulties stem from the attribution of fractions of the text to different narrators on the basis of criteria such as divine sobriquets – sobriquets not even text-historically secure in many appearances.133 The aforementioned reconstructions all suffer from the application of a model developed from other texts, particularly in Genesis and Exodus, to a text for which it is entirely inappropriate. The described “duplicate” material often does not even really consist of real duplicates: the emissaries, Balaam’s encounters with God, and the first three oracles all build on the material preceding them and cannot be identified as duplication in any fashion. Since there are no a priori reasons to presume duplicate narrative traditions behind Numbers 22–24, we can disregard models based on the application of the Urkundenhypothese for the reconstruction of the development of Numbers 22– 24. Nonetheless, these scholars have emphasized one paramount quality of Numbers 22–24: it seems to be part of a larger whole or at least related to other texts in its current and earlier diachronic contexts. The recourse to sources has always served to incorporate the composite narrative now found in Numbers 22–24 and each of its constituent elements within longer strains of narrative. Yet, the diversity of the models and their general implausibility, in addition to

of elements from more than one. Thus some lengthier passages that seem to draw on language from multiple sources are, most likely, not to be attributed to the compiler.” 132 Cf., e.g., Baden, Composition, 128: “In Moses’s introductory speed in Deuteronomy 1–11, the history of Israel from Horeb to the plains of Moab is recounted, and in virtually every instance D follows the E narrative.” 133 See the discussion about text history in the following chapter.

38

Introduction

their often disregarding the generally unified character of Numbers 22–24 preclude such models as satisfactory explanations for the literary history behind that text. Having ruled out models regarding the text as uniform and as the product of the combination of distinct source documents, the other possibility to consider here are redaction-critical models. To these we now turn.

1.4 Redaction-Historical Models for the Development of Numbers 22–24 1.4 Redaction-Historical Models for the Development of Numbers 22–24

In recent decades, more criticism of source-critical models have developed, leading to a plethora of alternative descriptions of the editorial processes behind the development of the Pentateuch as a whole and for the Balaam pericope in particular. Proponents of redaction-historical models have instructed us about the iterative (and perhaps even durative) processes of editing that stand behind the composite text. They recognize a core of text that expanded on several occasions under the influence of multiple redactors. Some of this editing must be recognized as part of systematic endeavors, covering larger portions of what became scripture (i.e., the book of Numbers, the Pentateuch, etc.). Other editorial undertakings more likely demonstrate contextually limited Fortschreibungen that add information only at specific points of relevance for that redactor or those redactors. Two matters remain worth noting in this capacity for redaction-historical models: 1) they often recognize a more or less consistent core; and 2) they emphasize the continuing process of editing as more than the simple act of compiling sources.134

1.4.1 Hedwige Rouillard Hedwige Rouillard published one such undertaking, specific to the Balaam pericope, in 1985. 135 Rouillard reconstructed a basic layer consisting of Num 134

For a criticism of this model more generally, but argued on the evidence of Leviticus 26, cf. Jeffrey Stackert, “Distinguishing Innerbiblical Exegesis from Pentateuchal Redaction: Leviticus 26 as a Test Case,” in The Pentateuch. International Perspectives on Current Research, ed. Thomas B. Dozeman, Konrad Schmid, and Baruch J. Schwartz, FAT, vol. 78 (Tübingen: Mohr Siebeck, 2011), 369–86. A very basic version of such a thesis that still had not entirely escaped from source-critical analyses is Philip J. Budd, Numbers, WBC, vol. 5 (Waco: Word Books, 1984), 256–65, who recognized an E base story with oracles, that a J editor expanded. 135 Hedwige Rouillard, La péricope de Balaam: la prose et les “oracles”, EBib (Paris: Gabalda, 1985). Cf. already the earlier observations in Hedwige Rouillard, “L’anesse de Balaam: Analyse littéraire de Nomb., XXII, 21–35,” RB 87, no. 1 (January 1980): 5–37 and Hedwige Rouillard, “L’anesse de Balaam: Analyse littéraire de Nomb., XXII, 21–35 (fin),” RB 87, no. 2 (April 1980): 211–41.

1.4 Redaction-Historical Models for the Development of Numbers 22–24

39

22:2–21, 36–41; and 23:1–26. This was expanded under Deuteronomistic influence with 22:22–35 (exilic). Finally, editors added 23:27–24:19 and 24:20– 22 + 23–24 in the postexilic period.136 Some issues here merit further discussion. For Rouillard, Num 22:2–21 presented an essentially homogenous text.137 But there were exceptions to this. For example, she regarded 22:4b not as the original introduction to the story, but instead as presenting a parenthetical remark explaining the narrative context. In this way, it remains similar to Judg 14:4b, which has the same function in Judg 14:1–4.138 At first glance, this appears a plausible explanation, but Num 22:2 does not fulfill the role that Num 22:4b does. The latter of these verses attributes Balak with qualities not present in the former, qualities that are necessary for the advancement of the story and the commentary on kingship found in the oracles. Without Num 22:4, Balak never receives a proper introduction and some of the story’s poignancy would fade. Such lacking poignancy would become particularly acute when the Moabite circumstances are contrasted with the preceding peoples with which Israel interacts: the Canaanites, Edomites, and Amorites all have kings in Numbers 21. Without Num 22:4b, Moab would have had none. Regarding the narrative about Balaam’s ass, Rouillard noted that the donkey story does not conform to any sources, whether E or J.139 Rather, the story represents an interpolation into the older Balaam material. While this observation is welcome here, as above, this particular reconstruction presents some problems, as did those above. Without the material that follows in Num 22:22–35, the donkey’s appearance in Num 22:21 appears unmotivated. To this end, one can compare Num 22:21 and 35: Num 22:35b looks like a perfect Wiederaufnahme of Num 22:21aαb with the singular addition of the subject ‫בלעם‬. This commends regarding either 22:21aαb or 35b as original to the text. Perhaps 21aβ represents an interpolation to make 22–35a better fit their context. I only briefly mention this matter here, but will cover it in more depth in Chapter Three. A first edition ending with 23:26 remains improbable. Should this have been the conclusion, it would present more of a fading out than a real resolution. The story would open with Balak summoning Balaam and continue through their rocky interactions. Then it would just end with Balaam effectively stating, “told you so”, while the two of them stood there looking at each other. That hardly seems likely, and something must have followed. The reconstruction hardly does justice to the tension in Num 24:14, putting 14a and 14b on the same diachronic level. That seems most unlikely, given the 136

Cf. Rouillard, Balaam, 484–87. Cf. Rouillard, Balaam, 111–13. 138 Cf. Rouillard, Balaam, 40–43. 139 Cf. Rouillard, Balaam, 115–16. 137

40

Introduction

obvious break between these sentences and the conspicuous continuation of 14a in 24:25. The discussion in Chapter Three will also further develop this observation. However, the attribution of the final oracles to different hands does seem like a plausible scenario. Based on the content and the introductions to the final statements in Numbers 24:20–24, one could reckon with separate editorial expansions consisting of Num 24:20–22 and 23–23. Nonetheless, the possibility should be investigated as to whether 20 was added distinctly from 21–22 or even perhaps with some element of 19. Chapter Three will cover this further.

1.4.2 Erhard Blum A few years later, in 1990, Erhard Blum proposed a redaction-historical model for the development of the Pentateuch as a whole, tracing precursor editions of the Torah that he identified with a Deuteronomistic Composition (KD) and a Priestly Composition (KP).140 In this survey, he covered the Balaam story in his redaction-historical reconstruction of the Pentateuch in some twelve lines of text.141 Blum accepted the generally independent characterization of the pericope presented by Gross and Rofé, noting that only Num 22:2 conjoins the Balaam narrative with its context (namely to Numbers 21). This commends Num 22:3–4 as the remaining (though truncated) exposition of the story. At the same time, he noted the addition of the Midianites in verses 4 and 7 of Numbers 22 as specific and limited additions reminiscent of the priestly opinion of Balaam, referencing Num 25:16–18 and Numbers 31. Bearing these factors in mind, Blum cautiously ascribed the Balaam story in Numbers 22–24 to his KD, noting this solution may not be certain, but is certainly the simplest.142 There is something to be said for this solution. Blum recognized the somewhat loose connection of the Balaam pericope to the surrounding narratives, while at the same time appreciating two distinct characteristics in the material that embeds it in its context. On the one hand, Num 22:2 places it in the context of Numbers 21, with which it shares some similarities.143 On the other hand, the addition of the Midianites does suggest affinity with more priestly perspectives later in the book of Numbers. To this extent, one can follow Blum, with Numbers 22–24 evincing a connection to something like a Deuteronomistic composition, which was later edited to a priestly composition. Nonetheless, Blum’s overview certainly oversimplifies the disparate character of many elements still found in Numbers 22–24. Beyond the removal of verses 22:1–2 as redactional, he leaves the rest entirely intact. This fails to 140

Blum, Studien zur Komposition. For Blum’s contribution, cf. Blum, Studien zur Komposition, 116–17. 142 Cf. Blum, Studien zur Komposition, 130. 143 Blum, Studien zur Komposition, 130 focused on the geographical similarities in Num 21:20 + 23:13, and 28 (125), as well as matters of content and language. 141

1.4 Redaction-Historical Models for the Development of Numbers 22–24

41

cover substantial, even fundamental tensions found in these chapters. To name just two, he does not genuflect on the tensions found in the placement of the anecdote about the donkey (Num 22:21 vs. 22–35), nor even on the duplication of information in Num 22:3. At the very least, to this end, his model fails to clarify the specifics of the Balaam pericope. Some of his attributions to Deuteronomistic or Priestly compositions appear spurious, and his general reticence regarding any pre-Deuteronomistic Balaam material remains insufficient. This will become apparent in the following literary-critical remarks.

1.4.3 Christoph Levin A few years after Blum’s survey, Christoph Levin published a study that regarded J as neither early, nor a source, but as an editor compiling and supplementing the non-P material in the Pentateuch.144 In Numbers 22–24 he reconstructed an edition of the Balaam story that pre-dated J’s editing. Levin’s preYahwistic Balaam narrative consists of 22:4b, 5a (without ‫)על הנהר‬, 5bα (without ‫)הנה כסה את עין הארץ‬, 5bβ, 6a (without ‫)כי־עצום הוא ממני‬, 7a* (only (‫)וילכו‬, 7b, 21, 36 (without ‫)אשׁר על־גבול ארנן‬, 41; 23:2b (without ‫ ;)בלק ובלעם‬24:3, 4b (without ‫)אשׁר‬, 6a, 10a, and 25. This older version came from the eighth or seventh century, just like the inscription from Deir ʿAlla. With this final observation, Levin moved from the purely literary to the historical, blurring the edges to a small degree. For Levin, Balak was a historic figure, and did not belong to some “mythical prehistory of sagas”.145 The preredactional version of this story stemmed from Israel (i.e., the Northern Kingdom) and demonstrated historically accurate political circumstances: Moab was not able to ultimately destroy Israel (though Mesha cast them out of his territory), just as Balak recognizes his inability to drive them from the land.146 However, Levin’s historical reconstruction remains unlikely, since Mesha apparently was in fact able to drive Israel from Moab, if we are willing to believe the Mehsa Inscription (and I see no serious reason to doubt it in this point).147 That would in turn suggest that Balak’s endeavors should have been efficacious in some older version, but that remains impossible to reconstruct. No biblical tradition recognizes “historical” military engagement between Balak and Israel.148 At most, Josh 24:9–10 could imply that Balak sought battle, but there is no indication that blood was shed and certainly no indication in the Bible 144 Cf. Levin, Jahwist and, more recently reiterated, Christoph Levin, “The Yahwist: The Earliest Editor in the Pentateuch,” JBL 126, no. 2 (Summer 2007): 209–30. 145 “Nicht [aus] der sagenhaften Vorzeit”; cf. Levin, Jahwist, 384. 146 Cf. Levin, Jahwist, 385. 147 Cf. my evaluation of the Mesha Inscription and the literature cited there in Jonathan Miles Robker, The Jehu Revolution: A Royal Tradition of the Northern Kingdom and Its Ramifications, BZAW, vol. 435 (Berlin: de Gruyter, 2012), 232–40. 148 Cf. particularly Judges 11 and the discussion in Chapter Four.

42

Introduction

that Moab under Balak’s auspices was successful. Levin’s reconstructed historical background fails to appropriately match the circumstances behind the text as he described them. If Mesha succeeded, Balak should have too. Turning back to literary considerations, Levin regarded only one oracle as original to the Balaam narrative, the oracle currently found in the third position. Yet, at the same time, Levin avowed the possibility that nothing remains of the original oracle, but noted: “Sicher ist aber, daß Bileam schon in der vorgegebenen Fassung Israel nicht verflucht hat, sondern gesegnet.”149 The single oracle in this version was not originally an oracle of YHWH, but the editor transformed it into one.150 Even before his Yahwistic editor had transformed the oracle into an oracle about YHWH, a pre-Yahwistic editor converted the term ‫שׁדין‬, known from the Deir ʿAlla inscription to the singular in the pre-Yahwistic version in order to afford its affiliation with YHWH. To affirm this, Levin attempted to reinforce connections between his oracle and the inscription from Tell Deir ʿAlla, the Deir ʿAlla Plaster Text (DAPT). For example, Levin saw a remarkable similarity in the opening of his first oracle (Num 24:4b) and the reconstructed first line of the inscription from Tell Deir ʿAlla, with only the first term (‫ נאם‬vs. ‫ )ספר‬essentially distinct between the two.151 In his reconstruction, he overstated the similarities between these two texts generally, as will become clear in Chapter Five. His reconstruction does not remain free of internal inconsistencies, for that matter, either. For example, for some reason, Levin included Num 24:5 in his evaluation of the source, though he had previously removed it as redactional.152 The core of the only oracle recounted in Levin’s J must have been hollowed out, but he did leave one element: “Like vales that spread out, like gardens on the river” (Num 24:6a). The other conserved elements of the oracle (24:5, 6b, and 9b) present only redactional additions. That means that the oracle that J took from his source was not even an oracle, just a doubled idyllic metaphor. That’s implausible. To what end would someone incorporate only that? Levin’s only recourse involves postulating that something else must have stood there, hardly a satisfying – though admittedly – possible reconstruction.153 Considering the end of Levin’s reconstructed pre-Yahwistic Balaam material, the unedited source material ended with the phrase in 24:10a, which immediately proceeded to v. 25. The redactor added 24:10bαβ to demonstrate a final time that YHWH stood by Israel regardless of what its enemies attempted. To this he also added 11a to afford a better connection between 10abαβ and 25.154 This reconstruction also seems quite improbable, particularly regarding 149

Levin, Jahwist, 386. Cf. Levin, Jahwist, 385–86. 151 Cf. Levin, Jahwist, 386. 152 Cf. Levin, Jahwist, 386. 153 Cf. Levin, Jahwist, 383–84. 154 Cf. Levin, Jahwist, 384. 150

1.4 Redaction-Historical Models for the Development of Numbers 22–24

43

the reconstructed source material. The transition from 24:10a directly to 25 makes no sense and hardly presents a plausible conclusion to this episode. Balak gets angry and claps his hands, Balaam just wanders off. What an enigmatic conclusion that would be. An earlier, non-Yahwistic editor edited the whole of Levin’s reconstructed pre-Yahwistic version in Levin’s reconstruction as well. This person supposedly added 22:8aαβ, 8b, 12a, and 22 (up to ‫)הולך‬. This edition should have demonstrated that Balaam overstepped God’s commandment; God could not have wanted what Balak sought, even if it ultimately failed. This addition remains recognizable because Balaam’s disobedience has no consequences.155 However, if one removes these elements from the story, even in Levin’s limited pre-Yahwistic rendition, the narrative has no real point, particularly theologically. The first editor thus created a problem in Levin’s reconstruction that was not there in the original nor resolved in any edition within the Yahwistic redaction. This hardly speaks for the plausibility of the reconstruction. It additionally makes Balaam an irredeemable, negative figure in the Yahwistic version. Later editors obviously would have had to edit this version to be more affirming of Balaam, before still other, later editors in the Pentateuch turned him into a wretched figure. Certainly it is simpler to suggest a positive Balaam figure becoming more negatively charged over increasingly later stages of redaction than to permit evaluations of his undertaking to swing back and forth so substantially and radically. To this point, see Chapters Four and Five. Turning the material Levin described as redactional, he identified elements of Yahwistic redaction in Num 22:1, 3, 6aab (from ‫)כי‬, 8aγ, 12b; 24:1aα, 5, 6b, 9b, 10bαβ, 11a.156 Levin’s Yahwistic editor added some elements consistently in Levin’s reconstruction. For example, Levin regarded Num 22:3b as redactional, an addition borrowed from Exod 1:12. From the same pericope, Levin’s Yahwist borrowed from Exod 1:9 in Num 22:6a, when Balak comments that the people “is stronger than me”. God’s interjection to Balaam the first time Balak’s emissaries come to him, “Do not curse the people, for it is blessed”, demonstrates Levin’s Yahwist’s redactional reference to Genesis 12. The concept of blessing appears as a redactional addition favoring the divine promise, just as in Gen 24:31 and 26:29. The addition of the blessing and cursing in Num 22:6b (as in Gen 12:3a; 27:29b; and Num 24:9b) should both underscore the lasting quality of Balaam’s undertaking and increase the drama of the episode. Additionally, the redactor added Balaam’s reference to accurately reporting the divine speech (Num 22:8aγ, as he was wont to do (Gen 12:4; 24:51). God’s forbidding Balaam’s accompanying the men does not require the following sentiment that the people should not be cursed because it is already blessed (Num 22:12). He based this observation on the ‫ ו‬missing in ML, but 155 156

Cf. Levin, Jahwist, 387. Cf. Levin, Jahwist, 381.

44

Introduction

present in essentially every other relevant witness ever, but also noted that it matches the redactional sentiment in Gen 26:29 (not about “the people”) and 24:31 (also not about the people).157 Text-critically this remains a problem with Levin’s thesis. Beyond that, the evidence that something is not required hardly suffices to justify it as a secondary interpolation into its context. Finally in terms of the thematic of blessing, Levin’s Yahwist appended Num 24:9b to the third oracle in a citation of Gen 12:3a.158 The Balaam story in Numbers 22–24 represents the conclusion of Levin’s Yahwist, at least in terms of what was included in the Hebrew Bible. To afford the prerequisite geographical context within Moab, the editor added a note moving the Israelites from Kadesh to Moab. The phrase “across the Jordan by Jericho” demonstrates that the editor is in Cisjordan . The situation between the Israelites and Moabites matches that between the Israelites and the Egyptians in Exodus 1, thus the common language. He disregarded (in a footnote) the doublet in 22:3, noting that the thought (“Gedankengut”) and language are consistent with the Yahwistic redactor. This hardly seems a plausible resolution to that difficulty.159 Rather, he left this duplication unresolved. Regarding the opening of the oracle, still a singular event in Levin’s reconstruction, Levin’s Yahwistic redactor added a framework before the only oracle known to the pre-Yahwistic source. He added the phrase that Balaam saw that it pleased Yahweh to bless.160 The context of Levin’s reconstructed Yahwist in no manner permits an understanding whence Balaam gained this understanding. Suddenly, after awakening God’s wrath and then sacrificing, Levin permited him the apparent clairvoyance to know what pleases the deity, information he appreciates without recourse to revelation or observation. This hardly seems a plausible reconstruction. After the Yahwistic additions, later editors added substantially more material (apparently in this order in Levin’s conception): 1) The donkey story in 22:22–35*: Wiederaufnahme of 21 in 35. It even has layers: the donkey’s speaking comes from a later hand. The older version consisted of 22, 27, 31–32aαb, 34aαb–35. The same editor added 22:38 to demonstrate that Balaam had learned his lesson.161 2) Balaam’s obedience: this editor added 22:13–21aα, demonstrating that Balaam had not been disobedient. He or she added the words from v. 35aβ in the donkey episode to 20b. Thus, this editor created a contradiction (intentionally?). Levin provided no intention for the permission of this marked contradiction.162 The same editor added 22:40 to confirm Balaam’s obedience (quite 157

Cf. Levin, Jahwist, 382. Cf. Levin, Jahwist, 46. 159 Cf. Levin, Jahwist, 381. 160 Cf. Levin, Jahwist, 382. 161 Cf. Levin, Jahwist, 387. 162 Cf. Levin, Jahwist, 387–88. 158

1.4 Redaction-Historical Models for the Development of Numbers 22–24

45

ironically, one might add, given the circumstances) and provide the meal as an affirmation of Balak’s willingness to treat Balaam well. Balaam turns aside to receive God’s word during the sacrifice (23:3, 5–6). Obedience returns as a theme for the final time in 24:11b–14a, the conclusion of this edition.163 3) The fourth oracle: 24:15–19. For economical reasons, the first story only had one oracle. The fourth oracle is most clearly recognizable as an addition, since it is missing a narrative introduction. At least in this point, we can follow Levin. Following this is a threefold coda that concludes with words against Assyria / Syria, i.e., Seleucids, according to Levin.164 4) The first and second oracles: 23:7–10 and 18–24 are later interpolations. This becomes apparent from the introduction to the third oracle, which is selfevidently oldest. This postulation bases on Balaam’s self-introduction, which is curious in its current context. The core of the first oracle is a poetic recounting of the narrative (cf. Exodus 15 and Judges 5). The core of the second oracle consists of Num 23:18, 21–22, and 24, a more militant parallel of 24:7–9.165 Two questions remain regarding these supposed additions: 1) who would add such an empty oracle as we find in the current first oracle at a later date? 2) Who would place the more militant interpretation, i.e., a kind of climax in the oracles in Levin’s structure, before the oracle it supposedly revised? Neither of these options appear particularly likely. 5) The preparatory procedure and Balak’s reaction: the oldest preparations appear in Num 22:41–23:6 with Balak’s reaction in 24:10–11a. Those of the second oracle, 23:13–18 and 25–26, must be youngest. Between these two, someone composed the second reaction in 23:11–12 and the third preparation in 23:27–30. This was added with the first oracle. Numbers 24:1aβ–2 mention multiple oracles, meaning they must have been added after the introductions of the first two oracles.166 A number of matters decide the issue against Levin’s reconstruction. First, his oldest reconstructed source lacks either a point or a clear structure. The preYahwistic reinterpretation of the text seems poorly motivated. The Yahwistic editing, while limited in scope, still unlikely could be responsible for the main point of the whole story: Balaam blesses Israel. This is particularly true as the Yahwistic version, while mentioning the motif of blessing on a few occasions in Levin’s postulated editing, fails to include a real blessing within its edited oracle. Particularly the conclusion of his older versions fails to convince on a narrative level. The sheer volume of editorial material required, while admittedly not impossible, commends a more simplified approach as likelier. The late addition of the currently first and third oracles appear unmotivated and 163

Cf. Levin, Jahwist, 388. Cf. Levin, Jahwist, 388. 165 Cf. Levin, Jahwist, 388. 166 Cf. Levin, Jahwist, 388. 164

46

Introduction

inexplicable in his reconstruction. He overlooked tensions and duplications, even in his Yahwistic version, the most important of which is that found in 22:3. On the other hand, his recognition of the redactional character of the donkey episode and of the final oracle complex (Num 24:14b–24) represent important observations for the undertaking at hand.

1.4.4 John Van Seters Unlike Levin, but publishing his study of the Yahwist about the same time, John Van Seters was quite critical of the postulated relationship between the DAPT and Numbers 22–24. Whereas Levin saw the inscription as an opportunity to date his pre-Yahwistic version of Balaam, Van Seters regarded the biblical transformation of Balaam into YHWH’s prophet as “a much later modification”.167 That being said, he provided no date, by which time people had forgotten to which era Balaam should be ascribed. Simultaneously, he affirmed Rost’s recognition of the common Mediterranean elements in the sacrificial undertakings in the Balaam story and in Xenophon’s Anabasis.168 In contrast to Documentary hypotheses, Van Seters followed Gross and Rouillard in identifying the donkey story as a later addition with limited connection to the older story (v. 20 taken up in v. 35). Otherwise it lacks connections to the preceding and subsequent material. However, dividing up the rest of Numbers 22–24 is more complicated.169 Van Seters identified 22:1 as potentially part of his Yahwist, since it is part of “the broader framework and not part of the story itself”.170 This means it cannot be P, as generally in the Documentary Hypothesis. Numbers 22:2–4 connect to 21:21–35 and match Exod 1:7ff, which Van Seters attributed to his Yahwist. This means that Num 22:2– 4 looks like the Yahwist’s work. However, Van Seters failed to qualify in what way Num 22:2–4 precisely appends to Num 21:21–35. Nowhere does Numbers affiliate Og with the Amorites. This fact makes the unmitigated connection of Num 21:23–35 to 22:2–4 problematic at the least, and impossible at worst. P modified this Yahwistic account through its introduction of the negatively connoted Midianites (they are otherwise positively viewed in Van Seters’ Yahwist). The reintroduction of Balak in v. 4 seems to be a problem, but Van Seters rejected Gross’s contention that 4b really begins the older narrative.171 Instead, he followed Rouillard’s identification of 22:4b as a parenthetical remark, much like Judges 14:4b. This circumstance implies that 4b could not have been the

167 John Van Seters, The Life of Moses. The Yahwist as Historian in Exodus-Numbers (Louisville, KY: Westminster/John Knox Press, 1994), 412. 168 Cf. Rost, “Fragen um Bileam”. 169 Cf. Van Seters, Life of Moses, 413. 170 Van Seters, Life of Moses, 414. 171 Cf. Van Seters, Life of Moses, 414–15.

1.4 Redaction-Historical Models for the Development of Numbers 22–24

47

original introduction.172 As with Rouillard, however, the identification of 22:4b as an ex post facto parenthetical following 22:2 seems unlikely. Rather, they presumably stem from different hands, and, of Balak’s two introductions only the one in 4b matches the circumstances of the core of the story of Numbers 22–24. The strewn references to the exodus “must reflect complete familiarity with the Dtr exodus formula”.173 The metaphor about the ox licking up the green of the field must reflect the oracles in 23:22 and 24:8, each of which explicitly connects to the exodus.174 While that may be true, Van Seters ignored the distinct vocabulary in each case. That hardly makes his suggestion probable. On the other hand, one can support his observations regarding the terminology for cursing and Balak’s emissaries. Van Seters largely dismissed the possibility of distinct terminology for cursing demonstrating any literary-critical relevance.175 The distinct terminology for the delegates also does not distinguish different sources behind the text.176 Still, he could have added to this that there is no narratological way to divide up the usage of these terms into distinct strata. Van Seters did not recognize the distinction between the themes of Israel’s size and might.177 Nor did he appreciate the distinct structure of the narrative material reflecting on Balaam’s demanding that altars be built, suggesting that the building of altars was necessary for his divine practice.178 Yet, supporting evidence for this claim is lacking. For him, verse 21 and its resumption in verse 35 demonstrate that the donkey story presents a “post-J addition”.179 His – probably accurate – assertion fails to explain whether 22:21 or 35 belongs to the older layer. In his first heading, he even includes Num 22:22,180 suggesting that he perhaps thinks 21 in its entirety presents part of the older story and that 35 in its entirety was redactional. But he never clearly specifies it. Importantly, however, he at least recognized that the donkey episode presents a later interpolation. He reiterated this, noting that “[t]he section in Num. 22:36–23:6 is… a continuation of the previous 22:2–21, with the same principals represented in the same way. There is a steady progression toward the oracles, which are the focal point in the whole narrative.”181 172

Cf. Van Seters, Life of Moses, 415. Van Seters, Life of Moses, 415. 174 Cf. Van Seters, Life of Moses, 415–16. 175 Cf. Van Seters, Life of Moses, 417–18. 176 Cf. Van Seters, Life of Moses, 418. 177 Cf. Van Seters, Life of Moses, 416. 178 Cf. Van Seters, Life of Moses, 416–17. 179 Van Seters, Life of Moses, 419. 180 Cf. Van Seters, Life of Moses, 414. 181 Van Seters, Life of Moses, 421. 173

48

Introduction

In this way, Van Seters noted that the oracles and the narrative material must be related: “If the prose narrative, without the story of the ass in 22:22–35, is a literary unity, […] then the three oracles of blessing in 23:7b–10, 18b–24, and 24:3b–9 must be taken together as part of that literary work.”182 The oracles have much in common and must be related.183 They together build up to a certain climax, with the third oracle emphasizing this with its different form. “The third oracle makes some distinction from the previous two in its reception of the revelation and visionary experience (24:3b–4), and this relates directly to the prose introduction in which the spirit of God comes upon him.”184 However, Van Seters unnecessarily saw Ezekiel as the example from which this manner of revelation was borrowed. No need exists to postulate that distinction, particularly when he regarded the Deuteronomist and Jeremiah as the inspiration for the reception of the first two oracles. That is, he argued for their unity on the one hand, while attributing the spirit of the content and manner of revelation to two distinct backgrounds. That seems unlikely. Van Seters recognized no disjunction between 24:5–6 and 7, a commendable observation. He saw Ezekiel as a common source for these metaphors. He undertook a text-critical change (reading ‫ דליותיו‬for ‫ דליו‬in 24:7) and then recognized the proximity to a metaphor in Ezekiel. Here, he cited specifically Ezek 31:3–10. The presumed similarity makes his reconstructed, conjectural reading less plausible: if the similarity is as marked as he suggests, it seems likely that M would have retained it. As noted in Chapter Two (see below), the chapter about text history, M sometimes has consistent readings where other witnesses are lacking them. Therefore, if this were the text later editors had in mind, it seems likely that they would have corrected it again, presuming it had become corrupted. Or the connection is not nearly as strong as he thought.185 Let us consider this further. Van Seters noted: “It is in v. 7b of the oracle that there seems to be an awkward shift from reference to the people to the king. Comparison with Ezekiel, however, gives a possible explanation. In Ezek. 19:10 the prophet describes Israel (‘your mother’) as a vine planted by the water, ‘fruitful and full of branches by reason of abundant waters’ (mmym rbym). It then immediately shifts to a statement about the king within the same imagery, v. 11… The similarity in the thought sequence is so startling that it strongly suggests a dependence of the third oracle on this text. The fact that this oracle in Ezekiel is preceded in 19:1–9 by a long allegory on the theme of lion and lioness similar to Num. 24:9a seems to make the connection between the texts all the more likely.”186

This argumentation relies entirely on M, which does not attest the oldest reading in Numbers here. That means that any similarities he identified probably 182

Van Seters, Life of Moses, 424. Cf. Van Seters, Life of Moses, 424–25. 184 Van Seters, Life of Moses, 425. 185 Cf. Van Seters, Life of Moses, 426–27. 186 Van Seters, Life of Moses, 427. 183

1.4 Redaction-Historical Models for the Development of Numbers 22–24

49

attest later recensional undertaking within the M tradition rather than compositional intention of some author or editor. Let me reiterate, to arrive at this reconstruction, Van Seters first made a conjectural text-critical emendation to conform Numbers 24 to Ezekiel 19 before stating that the two texts must be related. This reconstruction must be rejected. Still, Van Seters recognized the consistent structure of the narrative framework of the oracles. They have a common pattern of going to a particular place to look at the people and of building seven altars for the offering of sacrifices. “This pattern does not precede the final oracle because it is no longer construed as a blessing; it is a prediction. Balak’s attempts to obtain a curse have ended with the third oracle.”187 Van Seters thus recognized that the first three oracles generally belong together and to the original narrative without expansions, while the final oracles at the conclusion of chapter 24 also present later additions to the basic text. This redaction-historical circumstance also explains the more dramatic expression of Balak’s anger after the third oracle (as opposed to after the second), while simultaneously demonstrating why he demonstrates no more anger after the fourth: it was not part of the same composition, but only added as an appendix.188 Van Seters concluded his determination of the Yahwist: “Thus, the whole narrative, excluding the ass story of 22:22–35, up to 24:13 belongs to the one author, J.”189 He identified 24:14–19 as not stemming from J and 20–24 as later additions to that. He did not, however, suggest whether these passage share any link to, e.g., the donkey story. Ultimately, he identified J in the Balaam story as comprising of Num 22:2–21, 36–41; 24:1–13, 25. In this list, he seems to have forgotten chapter 23, which otherwise was part of his J source.190 He included the whole of 24:14 in his redactional layer, but it seems improbable that one author stands behind 24:14a and 14b. Nonetheless, 24:13 could have segued into 24:25 with no problem; Num 24:14a would have perhaps made a more appropriate bridge between the two verses, however. In terms of dating his Yahwist, Van Seters proffered the following observations: “I have drawn attention in the treatment of the oracles to a number of parallels between the Balaam story of Numbers 22–24 and Second Isaiah. These deserve some emphasis because they are largely ignored… Since I have argued that J is a contemporary of Second Isaiah in the late exilic period, it may be helpful to review these similarities here. First, within the oracles the term ʾēl, ‘God,’ serves as an alternate designation for Yahweh, and this is also true for Second Isaiah [Isa 40:18; 42:5; 45:14, 21, 22; 46:9; contra usage for ‘a god’ in Isa 43:10, 12; 44:10, 15, 17; 45:15, 20; 46:6]. The use of ʾēl is otherwise rather rare in the rest of preexilic and exilic prophecy [Isa 5:16; 9:5; 10:21; 12:2; 14:13 (all late); 31:3, for which 187

Van Seters, Life of Moses, 422. Cf. Van Seters, Life of Moses, 422–23. 189 Van Seters, Life of Moses, 427. 190 Cf. Van Seters, Life of Moses, 428. 188

50

Introduction

compare Ezek 28:2, 9; Jer 32:18 (late); Hos 2:1; 11:9 (both late); 12:1 (?)]. It is noteworthy that in the royal inscriptions of the Neo-Babylonian period the kings sometimes refer to the principal deity simply as ilum, ‘god’, as an alternate for the name of the deity.”191

To these, he added Israel’s divine accompaniment in the wilderness.192 The desert blooming in 24:5–6 is reminiscent of Isa 43:19; 49:9; 51:3; 55:13). “In Isa. 44:4 Israel is compared to a green tamarisk and willows by flowing streams, similar to the comparison in Num. 24:6, and this is viewed as God’s ‘blessing’ (v. 3). Such gardens of exotic trees by streams of water are particularly characteristic of Babylonia, which could have influenced the choice of this imagery in Ezekiel, J, and Second Isaiah.”193

Yet, none of these observations are particularly strong. First, his observation about ‫ אל‬presumes that ‫ אל‬must be understood as a parallel for ‫יהוה‬. The original understanding of the text might not mandate that. One should not argue from the monotheistic tendency of Second Isaiah that Numbers had the same monotheistic tendency. Rather, one would have to argue from internal evidence within Numbers, evidence which, as far as I see it, Numbers 22–24 lacks. One should not presume that it was monotheistic from the outset. Turning to the metaphors: one need not regard divine accompaniment as part of a common tradition-historical background. Rather, the Numbers text (and Exodus) could have inspired the understanding of God remaining with Israel in wilderness found in Second Isaiah. Theoretically, the alternative would be possible: that Second Isaiah inspired the composition of a wilderness narrative. Centuries could have stood between these texts in either of these scenarios. Finally, the metaphors of fruitfulness expressed through idyllic metaphors do not require a characteristically Babylonian background. Any culture that lives from the land regards fruitfulness as a kind of blessing. And that is ignoring the text-critical issues in Num 24:6, which must first be secured in its transmission and understanding before being placed in its current form in a tradition-historical matrix it should supposedly share with Second Isaiah and Ezekiel. Ultimately, while Van Seters offered some poignant observations, his model as a whole fails to convince for Numbers 22–24.

1.4.5 Reinhard G. Kratz At the outset of the 21st century, Reinhard G. Kratz published a survey about the narrative books of the Old Testament.194 In his discussion of the Torah and 191

Van Seters, Life of Moses, 434–35. Cf. Van Seters, Life of Moses, 435. 193 Van Seters, Life of Moses, 435. 194 Cf. Reinhard G. Kratz, Die Komposition der erzählenden Bücher des Alten Testaments (Göttingen: Vandenhoeck & Ruprecht, 2000). The English translation, generally cited here, appeared a few years later; cf. Reinhard G. Kratz, The Composition of the Narrative Books of the Old Testament, trans. John Bowden (London; New York: T & T Clark, 2005). 192

1.4 Redaction-Historical Models for the Development of Numbers 22–24

51

Former Prophets, Kratz neither particularly dealt with the Balaam passage nor did he provide any concrete redactional context for the Balaam story. Nonetheless, he regarded it more or less as an independent narrative that was incorporated into some exodus composition consisting of portions of Exodus– Joshua.195 He remained generally reticent about what material in Numbers 22– 24 is redactional and what is original. One notes that in his table about the sources and editorial material in his reconstructed exodus narrative,196 he included the whole of Numbers 22–24 (with the singular exception of 22:1) both in the “source” and the “additions” column. He freed Numbers 22–24 from its surrounding context in order to maintain his postulated link between 22:1 and 25:1, which for him attests the oldest geographical (and narratological) bridge between the exodus and the eisodus. However, he did not so much demonstrate as stipulate the connection between 22:1 and 25:1. These verses say barely different things: the Israelites camped in Moab and Israel lived in Shittim. Joining these two verses together hardly reads like complementary information, as Kratz would have it. Rather, this reconstruction creates a duplication in the text, one that would be highly indicative of editorial activity. One notes that each verse has its own subject referencing the same group (Israelites and Israel), each has its own terminology for dwelling (√‫ חנה‬and √‫)ישׁב‬, and each represents a distinct location, or at the very least the same general vicinity with different terminology (the fields of Moab vis-à-vis Jericho vs. Shittim). 197 Taken together, one can more readily postulate against Kratz that 22:1 and 25:1 are not the product of a single hand. Any literary critic would take one look at that and divide it into two distinct pieces. With that, his entire particular reconstruction for Numbers as the bridge between the exodus and eisodus falters. The independent Balaam narrative belonged to a roughly ninth or eighth century context for Kratz.198 The evidence he cited for this, exclusively in a footnote (in German) or endnote (in English),199 consists of the conflict with Moab attested in the Mesha inscription and the Deir ʿAlla Plaster Text (DAPT). In order to afford this dating, Kratz must divorce the Balaam story from its narrative context, since he did not regard this oldest exodus story as coming from that early of an era. In order to accommodate this splitting of the traditions, he removed the phrase ‫ יצא ממצרים הנה‬from Num 22:5, identifying this 195

Cf. Kratz, Komposition, 295–96 (= Kratz, Composition, 287). This oldest narrative embedding was expanded into a Deuteronomistic composition, covering Exodus–2 Kings before becoming part of the Yahwistic Enneateuch with the addition of Genesis. Cf. the tables in Kratz, Composition, 326 = Kratz, Komposition, 331. 196 Cf. Kratz, Composition, 294 = Kratz, Komposition, 303. 197 Cf. the statements, “I’m in East Saint Louis” as opposed to, “I’m in the fields of southwest Illinois across the river from Saint Louis”. Both express overlapping geographical information. 198 Cf. Kratz, Composition, 310. 199 Cf. endnote 63 in Kratz, Composition, 298–99.

52

Introduction

second ‫ הנה‬as evidence of a Wiederaufnahme. However, this is literary-critically impossible: 1) Kratz failed to account for the other witnesses against ML that attest a ‫ והנה‬in the second case in this verse, making it improbable as a Wiederaufnahme (cf. the comments to this verse in the following chapter); 2) the phrase ‫ יצא ממצרים הנה‬hardly presents a likely interpolation. Why would a redactor add this second superfluous ‫הנה‬, when ‫ ויכס‬would more appropriately fulfill the syntactical requirements? Kratz had to undertake this stopgap measure (in this case I would prefer to refer to it with the German Verlegenheitslösung), since either the phrase ‫ הנה עם יצא ממצרים‬or the phrase ‫הנה כסה‬ ‫ את־עין הארץ‬could be diachronically secondary. But without the first, ‫הנה כסה‬ ‫ את־עין הארץ‬would have no subject and Balak’s message would become nonsensical, devoid of content. This reconstruction is impossible. But Kratz had to remove the reference to the exodus to make his reconstruction work. This carves the text into an improbable form and forces the literary-critical evidence into an inappropriate redaction-historical model. His evidence is not immanent to the text and he forcefully conformed his literary-critical decision to his redaction-historical prerequisite in this case. The obvious improbability of this reconstruction precludes its viability.

1.4.6 Markus Witte Just two years after Kratz proposed his general model for the diachronic processes behind the Torah and Former Prophets, Markus Witte presented a detailed redaction-historical analysis of the Balaam pericope.200 Rather than focus on the narrative portions of the text, Witte sought to tackle the redactionhistorical problems through recourse to the poetic material: “Zielpunkt der Bileamerzählung, zumindest in ihrer kompositionellen Endgestalt, sind die poetisch gefaßten Spruchreihen in 23,7–10; 23,18–24; 24,3–9 und 24,15–24. Insofern sollten diese, und nicht eine literarkritische Untersuchung der Prosa von Num 22–24, der Ansatzpunkt einer redationsgeschichtlichen Interpretation des literarischen Gesamtbefundes der Bileamperikope sein.”201

Witte began at the end, looking at the final oracles in Numbers 24. To this end, Witte described “eschatologische Fortschreibungen” in Num 24:20–24 based on four characteristics: 1) the distinct introductory phrase; 2) the expansion of the geopolitical horizon; 3) their form; and 4) the explicit eschatological character of the oracles.202 Witte regards the fourth oracle as eschatological because 200 Cf. Markus Witte, “Der Segen Bileams – eine redaktionsgeschichtliche Problemanzeige zum ‘Jahwisten’ in Num 22–24,” in Abschied vom Jahwisten. Die Komposition des Hexateuch in der jüngsten Diskussion, ed. Jan Christian Gertz, Konrad Schmid, and Markus Witte, BZAW, vol. 315 (Berlin; New York: de Gruyter, 2002), 191–213. 201 Witte, “Segen Bileams,” 197. 202 Cf. Witte, “Segen Bileams,” 198.

1.4 Redaction-Historical Models for the Development of Numbers 22–24

53

Edom and Seir will become “endzeitliches Erbland (‫ )יְ ֵר ָשׁה‬Israels”.203 However, he did not really address in what way the end times play a role in this identification. Nor did he adequately address the text-critical issue in the oracle against Edom. He viewed the phrases ‫ ולא עתה‬and ‫ ולא קרוב‬as eschatological, as well as the introductory form ‫באחרית הימים‬, 204 but nonetheless failed to demonstrate that the phrases must be understood eschatologically.205 Regarding the fourth oracle complex, Witte concluded: “Hier genügt die Erkenntnis, daß es sich bei 24,14b–24 insgesamt um eine ‘spätprophetische’ Fortschreibung handelt.”206 In regarding the fourth oracle as a late conception, Witte recognized that it was younger than Balaam’s third speech. That is, these pieces came from distinct diachronic levels, contrary to what – particularly – proponents of the Urkundenhypothese had suggested. Nonetheless, the third oracle was not a unified text and also included secondary interpolations: “…der dritte Bileamspruch bestand ursprünglich aus 24,3b–6.9b. Er ist literargeschichtlich älter als der vierte Bileamspruch.”207 Witte’s reasoning: the second person versus the third person and the chiastic form of 24:5–6, 9b. The content of 7–9a supposedly reminds the reader of the eschatological material that follows and thus it is distinct from the rest of the oracle. This suggestion hardly convinces, however, as the supposed “eschatological” character of 7–9a refers to particular historical or least “historiographical” circumstances (the exodus, the victory over Agag) and not some anticipated future eschaton. Should one remove elements from this oracle, it should not be for reasons of some supposed eschatology. The similar introductions in 24:4 and 16 resulted from the editor responsible for v. 16 copying from v. 4. In this case, Witte overlooked the weighty witnesses Smr and G and even went so far as to expand v. 4 to be more consistent with v. 16 in spite of all text-critical witnesses opposing this decision. For that matter he completely ignored the variants of G and Q in this passage. The division into sections in the second and third person are not as precise as he would suggest. Nor does his reconstructed original have a particularly chiastic structure. For that matter, there is no real middle to the proposed chi, and his suggested structure reads more like ABB’C. Nonetheless, his observation about the layering of material within the third oracle is certainly noteworthy. Turning to the first oracle, Witte regarded Balaam’s wish to die the death of the righteous (Num 23:10b) as a secondary addition to an original composition consisting of two stanzas of three bicola each. He proffered no arguments fa203

Witte, “Segen Bileams,” 198. Cf. Witte, “Segen Bileams,” 199. 205 Cf. Edward Lipiński, “‫ באחים הימים‬Dans Les Textes Préexiliques,” VT 20, no. 4 (1970): 445–50. 206 Witte, “Segen Bileams,” 200. 207 Witte, “Segen Bileams,” 201. 204

54

Introduction

voring his regarding that half verse as an addition. He suggested that both stanzas are constructed as parallels: 7bα // 9a; 7bβ // 9b; 8 // 10a.208 With the exception of the rhetorical questions in 8 and 10a, these parallels are hardly obvious in terms of form or content. At least regarding the oracle’s object, one can follow Witte’s position: “Ziel des ersten Spruchs ist die Beschreibung der Sonderstellung Israels unter den Völkern.”209 In the second Balaam oracle, Witte regarded 23:18b–19 as a perfect parallel to the first stanza of the first oracle (23:7b–8) and 23:21–22 as corresponding to 23:9b–10a (the second stanza).210 Only two aspects of the parallel can be affirmed: 1) 18b and 17bα both reflect the narrative circumstances of the oracles; 2) and 19b and 8 are both rhetorical questions. Verses 19a and 7bβ are only similar in that they are in the middle, hardly overwhelming evidence of parallel structure or – more significantly – content. Numbers 23:20 corresponds with 24:9a and thus must be secondary in his model. To this level, one must add 23:23a, since the idea of ineffective mantic undertakings against Israel hardly matches the context following 23:22. Verse 23b is a later addition, as demonstrated by the terminal note “at that time” (‫)כעת‬. These considerations led him to postulate multiple layers within the second oracle: 1) an eschatological addition in 23b–24; 2) two additions that emphasize the impossibility of diminishing Israel’s blessing; 3) an original layer consisting of two stanzas, each with three bicola (18b–19 and 21–22). However, since Witte did not sufficiently demonstrate the secondary character of 24:9a, there is no common reason to remove 23:20 and 23:23a. This observation becomes particularly acute, as Witte did not place 23:20 and 23:23a in his eschatological layer, though he previously had commented on the eschatological nature of 24:9a. His division of the redactional layers remains quite problematic in the oracles and should be reconsidered. In his estimation, the similar form of the reconstructed first and second oracles matches to such a degree that they presumably come from the same author. The form of the third oracle does not match the first two and thus must come from a later hand, possibly related to the additions in 23:20 and 23a.211 That is, the oracles come from several different layers, with only the first and the second oracles stemming from the same hand, according to Witte. Three layers become apparent in Witte’s reconstruction: 1) the oldest oracles in 23:7–10a + 23:18–19, 21–22; 2) the additions about blessings in 23:20 and 23a that match the oldest material in the third oracle (24:5–6, 9b); 3) the youngest, the eschatological additions in 23:10b, 23b–24; 24:7–9a, 14b–17, 18–19, and 20–24. 208

Cf. Witte, “Segen Bileams,” 201. Witte, “Segen Bileams,” 201. 210 For Witte’s argumentation regarding the second oracle, cf. Witte, “Segen Bileams,” 201–2. 211 Cf. Witte, “Segen Bileams,” 201–2. 209

1.4 Redaction-Historical Models for the Development of Numbers 22–24

55

The fourth oracle and its related material was not part of an older source, but was independently added. However, since we have identified several problems with his reconstructed diachronic layering of the oracles, this issue must be examined anew in Chapter Three. The question remains whether the first two oracles, which are clearly related to the narrative material, and the third oracle in its oldest version can be ascribed to two independent narrative strands in Numbers 22–24. According to Witte, Numbers 23:27–24:2 should be regarded as a literary unit, the lead up to the third oracle, since there is no literary-critical basis for severing the text into constituent parts. Num 23:25–26 should be included as part of this composition, but it does not present the original continuation of the second oracle, since Balaam neither blessed nor cursed in the original version (!).212 This addition (23:25–24:9) belongs to the same layer as 23:20 and 23a. “Die literargeschichtliche Zusammengehörigkeit der prosaischen Hinführung zum dritten Bileamspruch und der Segensnotizen zeigt sich deutlich an dem Gegenüber von 23,23a und 24,1f.: Weil gegen Israel keine Mantik (‫ )נחשׁ‬hilft, wendet sie Bileam im Folgenden nicht 213 mehr an, sondern wartet auf die Inspiration durch die ‫ֹלהים‬ ִ ‫רוּח ֱא‬.” ַ

In order to make this work, Witte had to remove 23:11bβ.214 This in turn required the removal of 24:10bβγ from its context, allowing him to regard 24:10bα + 11a as the original reaction to the second oracle. However, Witte overlooked the ‫ ו‬at the beginning of 11a, which suggests that this phrase continued the speech and did not begin it. This then went directly into 11b–13a and then concluded with 24:25. The repetition of the names in 24:10bα – according to Witte – indicates that 10a belonged to a different redactional layer.215 I would suggest alternatively, that the names certainly ease the syntax, making better sense in this case. The notices of blessing are also secondary in chapter 22 in Witte’s conception. Numbers 22:6b is secondary and presumes the blessing in 24:9b. Perhaps these two verses are related, but it can hardly be regarded as certain that either of them is an interpolation based on Witte’s evidence presented thus far. On the other hand, by comparing 22:6 and 22:11 Witte presumed that only the demand to curse is original and that even the comment about Israel’s strength is secondary.216 That presents a possibility worth further consideration. Numbers 22:12b, that Israel is blessed and not cursed, should be regarded as secondary when compared with the notices in 22:13b, 14, 16, and 20.217 Thus he 212

Cf. Witte, “Segen Bileams,” 203–4. Witte, “Segen Bileams,” 204. 214 Cf. Witte, “Segen Bileams,” 205. 215 Cf. Witte, “Segen Bileams,” 205. 216 Cf. Witte, “Segen Bileams,” 205. 217 Cf. Witte, “Segen Bileams,” 205–6. 213

56

Introduction

regarded all of the material about blessing as secondary. This layer consisted of 22:6aα2b, 12b; 23:11bβ, 20, 25; 24:1 and the inclusion of the third oracle with its literary embedding, 23:25–24:3a + 24:3b–6, and 9b.218 However, the recognition of 22:12b as secondary when contrasted with 22:13b et al., as Witte suggested, remains spurious. One has nothing to do with the other, and all of these phrases can stand next to each other without evincing any particular tension or contradiction. In Num 22:13b, Balaam states merely that God has not allowed him to go, a message the officials repeat to Balak in v. 14. Balaam never states why God did not permit him to go with the others, only that he did. No obvious tension exists between 22:12b and 13b et al. Ultimately, Witte settled on three primary layers, noting that the latest may have had multiple strata within it. The youngest layer is the eschatological layer.219 The second layer presents the material about Balaam as someone who blesses Israel and the donkey pericope (based on the motifs of anger and opened eyes, the phrase ‫בעיני‬, the hypostases of God, the motif of knowing/not knowing, Balaam’s recognition of his failing, and the root ‫)נטה‬. This second layer presents things in threes and is Witte’s J.220 The oldest material is from the original story, which consisted of one source and not two, as in the older Documentary Hypothesis, but does contain some glosses (‫וזקני ;אל־זקני מדין‬ ‫ ;מדין‬23:4b and 24:13b).221 The older layer dealt in twos. The narrator uses ‫ אלהים‬exclusively, whereas the characters use ‫ יהוה‬and ‫אל‬.222 The layer dealing with blessing can be affiliated with J, for whom this supposedly represented an important motif.223 However, these texts cannot be compiled into anything like a source and were certainly expansions of an original layer.224 At the same time, Witte recognized a number of connections between the oldest layer and the exodus and conquest materials: “Diese Querverbindungen [dieser Grundschicht zur Exodus-, Wüsten- und Landnahmeerzählung] deuten darauf hin, daß der Verfasser der ‘Grundschicht’ eine Exodus-Eisodus-Erzählung redigiert hat.”225 Since J apparently edited some version of this, J cannot be the creator of a Tetrateuch composition, as has often been suggested. For this reason, Witte decided that the older version must be post-Dtr and post-P.226 “Kompositions- und redaktionsgeschichtlich bemerkenswert ist, daß die ‘Grundschicht’ von Num 22–24 vor allem in einem redaktionellen Verweissystem mit den Büchern Exodus, Nu-

218

Cf. Witte, “Segen Bileams,” 206. Cf. Witte, “Segen Bileams,” 206. 220 Cf. Witte, “Segen Bileams,” 206–7. 221 Cf. Witte, “Segen Bileams,” 207. 222 Cf. Witte, “Segen Bileams,” 208. 223 Cf. Witte, “Segen Bileams,” 208. 224 Cf. Witte, “Segen Bileams,” 209. 225 Witte, “Segen Bileams,” 209. 226 Cf. Witte, “Segen Bileams,” 209–11. 219

1.4 Redaction-Historical Models for the Development of Numbers 22–24

57

meri und Josua steht, während die ‘Segensschicht’ explizite terminologische Bezüge zwischen der Bileamerzählung, der Genesis und dem Deuteronomium intensiviert bzw. selbst herstellt.”227

That is, different diachronic layers of the Balaam pericope evince connections to different literary works or corpora. From Witte’s observations we can glean several impulses. First, the layering of the poetic material in Numbers 22–24 may play a more significant role than has often been noted. Second, the oldest material in the Balaam pericope demonstrates affinity with material about the exodus and Joshua. Third, that some editorial material could indicate an editorial connection to Deuteronomy not present in the original. That being said, many problems appear regarding the specifics of Witte’s divisions, as Chapter Three will demonstrate.

1.4.7 Reinhard Achenbach A common feature to the redaction-historical models thus far is the general consensus that some original layer and its redactional additions can be reconstructed from the text of Numbers 22–24 as we have it. In his magisterial work on Numbers from 2003,228 Reinhard Achenbach presented himself as less sure about the details, while not denying the possibility entirely. For him, the source of the narrative can only be fragmentarily reconstructed in the best case.229 However, recovering the original was not Achenbach’s primary purpose in his study. Rather, he wanted to clarify the redaction history of Numbers, of which Numbers 22–24 presents only a part and for which Numbers 22–24 only plays a limited, though important role. In order to reconstruct the redaction history of Numbers and the place of Numbers 22–24 in it, Achenbach began with some reflections on its relationship to Deuteronomy. Since DtrL does not reference the Balaam tradition in Deut 3, it becomes at least permissible to regard its final form as “eine nachdtr. Auseinandersetzung mit der Überlieferung”.230 Further, regarding Deuteronomy and Balaam in the Bible more generally, he noted: “Die ‘Deuteronomisten’ hatten jedoch an der Bileamsgestalt kein Interesse, denn in Dtn 1– 3 wird sie nicht erwähnt, wohl aber in dem nach-dtr. qāhāl-Gesetz Dtn 23,4f. (HexRed*) und in der Rekapitulation der Heilsgeschichte Jos 24,9f. (HexRed). Davon abhängig sind Neh 13,2 und Micha 6,5 (vgl. auch zu Miriam!). Num 31,8.16 und in Folge Jos 13,22 sind

227

Witte, “Segen Bileams,” 212. Cf. Reinhard Achenbach, Die Vollendung der Tora. Studien zur Redaktionsgeschichte des Numeribuches im Kontext von Hexateuch und Pentateuch, BZABR, vol. 3 (Wiesbaden: Harrassowitz, 2003). 229 Cf. Achenbach, Vollendung, 391. 230 Achenbach, Vollendung, 390. 228

58

Introduction

ThB zuzurechnen. So muß zumindest erklärt werden, wie es zu dieser massiven Neubelebung der älteren Überlieferung gekommen sein kann.”231

In this way, he sought to demonstrate that the character of Balaam, regardless of where he appeared in the Hebrew Bible, was a product of late editorial activity. Significant for the study at hand is primarily Achenbach’s recognition that the discussion about Balaam in Numbers cannot be divorced from his appearances in Deuteronomy and elsewhere in the Bible. I will further develop these matters in Chapter Four. Regarding the background of the Balaam tradition, Achenbach recognized the combination of a Balaam legend with the conflict with Moab. He seems to imply that this was not an original element of the literary tradition. “Traditionsgeschichtlich liegt die Geschichte der Konflikte mit Moab einerseits, die BileamGestalt andererseits im Vorfeld der Erzählung, die auf der Ebene der Legendenbildung miteinander verbunden worden sind. Die theologische und literarische Ausrichtung steht aber schon ganz im Dienste der Landnahmeerzählung.”232

At the same time, as noted here, he regarded the whole of the story as part of the conquest narrative. But one must ask: how plausible is this? The Moab conflicts were over quite a long time before the Persian-Hellenistic period that he envisioned for the post-Dtr. historical background of the narrative. By that time, the Moabites had ceased to exist.233 So this would create the same problem he identified with the reconstruction of an old source. We have no appropriate historical context during the period he envisioned. Beyond that: to what end does the story commend a connection to the conquest narrative? Connections to the conquest only play a role in the final oracles, if even there. And these final oracles are probably substantially later than the rest. Therefore, it would be inappropriate to presume the later dating of the whole, while focusing on and reinterpreting in a special manner, what may be the youngest portion of the text. The issue of the dating of the whole must be further and more particularly addressed than that proposed by Achenbach. While Achenbach did not seek to reconstruct the literary-historical background of the text,234 focusing instead on its initial location in the Hexateuch 231

Achenbach, Vollendung, 390 n. 8. Achenbach, Vollendung, 392. 233 For the history of Moab and the Moabites and their interactions with Israel and Judah in the Iron Age, but their evaporating relevance for Yehud during the Persian period, cf., e.g., Stefan Timm, Moab zwischen den Mächten. Studien zu historischen Denkmälern und Texten, ÄAT, vol. 17 (Wiesbaden: Harrassowitz, 1989); Udo Worschech, Die Beziehungen Moabs zu Israel und Ägypten in der Eisenzeit, ÄAT, vol. 18 (Wiesbaden: Harrassowitz, 1990); Erasmus Gaß, Die Moabiter – Geschichte und Kultur eines ostjordanischen Volkes im 1. Jahrtausend v.Chr., ADPV, vol. 38 (Wiesbaden: Harrassowitz Verlag, 2009); and the literature cited there. 234 Cf. Achenbach, Vollendung, 392 n. 17. 232

1.4 Redaction-Historical Models for the Development of Numbers 22–24

59

composition, he did recognize that neither the story nor the oracles make sense without the other.235 This finding certainly makes sense, but the question remains to what degree the narrative and the poetic material were cast at the same time, of the same mold. After its initially embedding in a Hexateuch complex, later editors added further material and intertextual references over the course of the Pentateuch redaction. 236 Because of this new version of the story, Achenbach presumed that it had become impossible to reconstruct older versions of the text, though it still remained possible to postulate them.237 Each of the redactional layers, suggested Achenbach, relied on its literary context and cannot be divorced from it. Verses 2 and 4b came from the Hexateuch redactor, verses 1 and 3a from the Pentateuch redactor. This second editor also added 3b and 4a from a Vorlage of another version.238 This fails to convince since in v. 2, which Achenbach regarded as older than v. 4b, Balak appears, though nothing otherwise introduced him in this story to this point. Balak would have apparently viewed Israel from Moab although Achenbach’s Hexateuch redaction placed them in Bashan, should 22:2 have immediately followed 21:35 as he presumed. This also maintains the duplicate introductions of Balak within one redaction, hardly a feasible consideration syntactically (“Balak ben Zippor saw all that Israel… and Balak ben Zippor was king of Moab at this time”). Why would the author or editor repeat the subject in this manner? Without these verses, Achenbach’s original version, i.e., that which preceded the Hexateuch composition, would begin without an explicit subject: “he sent messengers…”, hardly a convincing exposition. Because of this, he had to include v. 5 in his Hexateuch redaction and saw the sending of messengers (as in Num 20:14; Deut 2:26; cf. Num 21:21) as a satisfactory condition for this ascription. 239 His ascription of vv. 3b–4a to another version seems problematic, at least should one regard this as another version of some Balaam or Balak narrative. Neither of these characters appears in these sentences, making a parallel version unlikely or at least inconspicuous. A simpler explanation seems likely, as I will propose in Chapter Three. Achenbach identified a number of topics that played a role in the Balaam narrative and oracles, each of which belongs to the Hexateuch composition in his reconstruction. “In ihr [i.e., in der Bileam-Legende] bündeln sich die großen Themen der Überlieferung von Segen und Fluch, Verheißung und Erfüllung, Gottesoffenbarung an Israel und an die Völker.”240 Only the first of these really applies. There is no revelation to Israel in this passage, nor is any promise fulfilled. Certainly nothing promised in Numbers 22–24 finds its fulfillment 235

Cf. Achenbach, Vollendung, 392–93. Cf., e.g., Achenbach, Vollendung, 394. 237 Cf. Achenbach, Vollendung, 398. 238 Cf. Achenbach, Vollendung, 395–96. 239 Cf. Achenbach, Vollendung, 396. 240 Achenbach, Vollendung, 393. 236

60

Introduction

within the Hexateuch, the context in which Achenbach sought to incorporate it. Significantly for his whole approach, Achenbach read Numbers 22–24 through the lens of Deuteronomy 23, noting that Balaam was forced (“bezwungen”) by YHWH to bless Israel.241 That reading is not immanent to Numbers 22–24, however, which in no way indicates that a deity forced Balaam to do anything. The connection to Deuteronomy 23, if any, and the direction of influence, if any, must be demonstrated and not presumed. I will reconsider this matter in Chapter Four. Achenbach saw Num 22:2–3a as the continuation of the Og pericope and 4b as a later addition explaining that Balak was the king of Moab, just as Sihon and Og had been. This is problematic in light of Numbers’s testimony about the figure Og. Nowhere does Numbers regard him as an Amorite, which Achenbach’s reading presumes. Rather, this comes from his reading Deuteronomy (e.g., Deut 3:8) as the interpretive lens for Numbers from the outset. As stated above, this must be demonstrated in this case, and not merely accepted. In his reconstruction, the Hexateuch redactor ignored the information about Moab in DtrL (Deut 2:17–18, which in turn had already replaced an older version) and reported this long story to replace it.242 On the other hand, perhaps these various data about Moab stem from distinct hands that were ignorant of each other and did not merely ignore each other. Further, he suggested: “In Num 22,2f. wird konstatiert, daß schon im Ostjordanland begann, was die dtr. Rede Dtn 7,17–24* für die dtr Landnahmeerzählung angesagt hatte.”243 This seems unlikely, since the Balaam narrative and oracles, particularly in their final form, make no recourse to God destroying the peoples. Rather, some future king will undertake this action against them. That seems to preclude a particularly strong relationship between these texts. Theologically, Achenbach suggested that Num 22:9–12 can only be told conforming to P. “Die Texte suggerieren, daß Jahwe sich in seiner Gottheit als Elohim den Nicht-Israeliten Abimelech, Laban und Bileam offenbart. Als solcher kommt er und redet er auch zu den Israeliten, die ihn noch nicht kennen (Ex 20,1.20).”244 This makes all of these Gentiles to faithful people of YHWH in the sense of Josh 2:11b, according to Achenbach.245 The comment that Israel is blessed remains unusual in the Pentateuch, but hopes for the Dtr. exhortation in Deut 7:14.246 The problem with this is that Deut 7:14 looks to the future, whereas Num 22:12 looks to the past. 241

Cf. Achenbach, Vollendung, 393. Cf. Achenbach, Vollendung, 394. 243 Achenbach, Vollendung, 395. 244 Achenbach, Vollendung, 398. 245 Cf. Achenbach, Vollendung, 398. 246 Cf. Achenbach, Vollendung, 398–99. 242

1.4 Redaction-Historical Models for the Development of Numbers 22–24

61

“Daß Israel von Jahwe gesegnet sei, trifft nur unter der Voraussetzung zu, daß i.S. der dtr. Paränese die Verheißung des Segens bei der Eroberung des Landes nun beginnt eingelöst zu werden (Dtn 7,13f.), andererseits nur unter der Voraussetzung, daß mit dem Auftreten Israels an dieser Stelle die Jakobsverheißung Gen 35,9–15 schon eingetroffen ist.”247

Following Levin, Achenbach regarded this passage in Genesis as P since it matches Gen 17:1–8. That Num 22:12 applies to Israel (here called simply “the people”) seems evident, but how this is related explicitly to the blessing of Jacob does not appear conspicuously in Numbers. The increase of Jacob’s dust (Num 23:10), the veracity of God’s promises (Num 23:19), his victory over the peoples (Num 24:8–9), and the deeds of Jacob’s royal progeny (Gen 35:11b; Num 24:7, 17–19) are transmitted to and through Balaam in his oracles. “Der Gottesbefehl an Bileam Num 22,12 ist also seinem Stile wie seiner Theologie und Aussage nach formuliert vor dem Hintergrund einer Verarbeitung vor-dtr., dtr. und priesterlichen Gedankengutes durch eine dieses vereinigende Redaktion. Die redaktionelle Durchdringung der traditionellen Stoffe im Zug ihrer Réécriture scheint also weitreichender zu sein, als man bisher zu vermuten bereit war.”248

That is, the original meaning of the text cannot be identified due to the various editorial reworkings the text underwent. Achenbach seemed to be thinking of the relationship of Num 22:12 with Gen 12:3.249 To this end, he suggested that the Hexateuch Redaction used the Balaam story to connect the promise to the ancestors (particularly Genesis 12) with the conquest narrative. Starting with the ancestors, we can easily dismantle this proposal. Other than the motifs of blessing and cursing, Gen 12:3 and Num 22:12 do not have as much in common as Achenbach might like. For one, Gen 12:3 describes the destiny of those who bless or curse Abram and provides no insight about the status of Israel’s blessedness. At the same time, Gen 12:3 reflects √‫ – קלל‬admittedly in addition to √‫ – ארר‬a root that never appears in Numbers 22–24. The connection there, thus in contradiction to what Achenbach suggested, appears at best tenuous, and the connections to any supposed Deuteronomistic and Priestly editing at best ephemeral. Turning to the conclusion of any supposed Hexateuch, the lack of material about the conquest in Num 22:15–21 permits at least the suggestion that some of its material may have come from a Vorlage that predated the Hexateuch redaction. This Achenbach admits. 250 Nonetheless, nothing in all of Numbers 22–24 connects unequivocally to the conquest, making this text at best irrelevant to that narrative arc. Rather than a Hexateuch composition, the anticipation of the monarchy (cf. Num 24:7b and 17) better commends a context in an Enneateuch composition. Any connection to Genesis 12 would only strengthen this suggestion. 247

Achenbach, Vollendung, 399. Achenbach, Vollendung, 399. 249 Cf. Achenbach, Vollendung, 400. 250 Cf. Achenbach, Vollendung, 400. 248

62

Introduction

In Achenbach’s reconstruction, as in many others, the story of the donkey is a fragment that does not demonstrate any linguistic or literary-critically relevant connection to other sources. Editors inserted it at a late point in the composition history of the narrative.251 This seems like a probable scenario, as will become clear in Chapters Three and Four. However, his description of the ‫ מלאך יהוה‬seems distinct from the others he cited or remains incorrect: “Die Rede ist vom ‫ מלאך יהוה‬im Pentateuch stets dann, wenn die geschilderten Personen noch aufgrund ihrer Herkunft (wie Hagar, Gen 16,7.9.10f.) oder ihres Erkenntnisstandes (wie Abraham, Gen 22,11.15, oder Mose Ex 3,2) einer unmittelbaren Begegnung mit der Gottheit (‫אל‬, Gen 16,13; ‫אלהים‬, Gen 22,1ff.; Ex 3,1b.4b) noch nicht (Hagar), oder gerade eben (Mose) ausgesetzt werden.”252

Balaam has already engaged with God in this story and thus presumably none of these conditions apply to him. The addition of the donkey story must have come from circles around the Pentateuch redaction in Achenbach’s opinion, but clear evidence for this is lacking, as nothing in the passage evinces a clear vision of the Pentateuch Narrative. Regarding the distinctive formulation of Num 22:38b, Achenbach suggested: “So ist anzunehmen, daß in v. 38b PentRed den vorgegebenen Text des HexRed entweder erweitert oder – was nach 23,26 wahrscheinlicher ist – umformuliert hat. Dementsprechend ist mit Eingriffen auch in 23,5.12.16 zu rechnen.”253 This is true because the phraseology does not match 22:20b.254 This distinction raises the question as to why the text of 22:20b (and 35) was not emended to match the other usage of PentRed in Achenbach’s reconstruction. However, this observation merits further discussion, to which we will return in Chapter Three. Regarding Num 23:10b, Achenbach suggested: “Er wünscht sich zu sterben wie Jakob und wie dieser im Angesicht seiner 12 gesegneten Söhne zu den Ahnen versammelt zu werden (Gen 49,28f.). Dies ist der Tod der ‘Rechtschaffenen’.”255 Further: “Der Text setzt also die Einheit von Väter-, Exodus- und Landnahmeerzählung voraus. Es handelt sich also in Num 23,7–10 also insgesamt nicht um ein ehemals selbständiges Orakel, sondern um ein literarisch auf den hexateuchischen Kontext der Erzählung hin konzipierten Maschal, der den theologischen und literarischen Vorgaben und Interessen des HexRed entspricht.”256

This connection can only be implied if one presumes that Balaam refers to Jacob’s death in this oracle and connects it to the root ‫ ישׁר‬as found in Josh 10:13 251

Cf. Achenbach, Vollendung, 403–4. Achenbach, Vollendung, 404. 253 Achenbach, Vollendung, 407. 254 Cf. Achenbach, Vollendung, 406. 255 Achenbach, Vollendung, 411–12. 256 Achenbach, Vollendung, 412. 252

1.4 Redaction-Historical Models for the Development of Numbers 22–24

63

and 2 Sam 1:18, neither of which are used in those instances as something narrative. They refer to a book of songs. Only by presuming these connections can one find them. No text in the Pentateuch or Hexateuch refers to Jacob as “upright”. The plural used in this verse makes it particularly improbable that the oracle should refer to him. Achenbach viewed the Balaam story in its current context as the illustration of God’s refusal to disavow the promise to Abraham [sic] in Gen 12:1–3. However, Gen 12:1–3 has nothing to do with Israel.257 Any connection to Gen 12:1– 3 in Numbers 22–24 genuflects on the status of the Moabites and Arameans: Balak is the one who seeks to curse Israel and Balaam is the one who blesses them. Israel should be blessed, according to Achenbach, before the renewal of the covenant in Moab (Deut 5–30*), which places the maintenance of this blessing under different conditions. Therefore Num 23:11 fulfills a literary function within the interests of the Hexateuch redaction.258 This argument does not really reflect on the direction of the relationships between the texts, but rather presumes that all of them must be on the same diachronic level. Further, he suggested that, “[d]ie offene Perspektive, unter der die Erzählung steht, kommt nicht aus ihrem Erzählkern, sondern aus ihrer redaktionellen Positionierung am Beginn der Landnahmeerzählung.”259 While that possibly may be true, Achenbach presumed that this connection must be redactional rather than compositional.260 Yet, by removing Balaam’s response to Balak in Num 23:12 as an interpolation from PentRed, Achenbach created a narrative in which Balak begins speaking twice, in vv. 11 and 13, the Hexateuch version at the latest.261 That seems both superfluous and unlikely. When considering the uniformity of the oracles, Achenbach described the transition to the third person in Num 23:21 as signalling the beginning of a commenting Fortschreibung.262 Achenbach saw the second oracle as a reference to the monarchy of YHWH, but failed to consider the problems with such an interpretation. Nowhere does the text (particularly in the versions that predate M) make reference to YHWH as king.263 When reflecting on the somewhat mysterious verse Num 23:23 Achenbach commented: “V. 23 [von Numeri 23] fügt hinzu: Israel hat nach Dtn 18,14 nicht die Wahrsagepraktiken anderer Völker nötig, weil Jahwe zu ihm geredet hat und redet – durch Mose. Zauberei und Wahrsagerei gilt als heidnisch. Daß damit Bileam seine eigene Position und die Position der

257

Cf. Achenbach, Vollendung, 413. Cf. Achenbach, Vollendung, 413. 259 Achenbach, Vollendung, 414. 260 Cf. Achenbach, Vollendung, 414. 261 Cf. Achenbach, Vollendung, 414. 262 Cf. Achenbach, Vollendung, 416. 263 Contra Achenbach, Vollendung, 416–17. 258

64

Introduction

Bileamorakel in Israel relativiert, ist eine subtile Nebenwirkung, die der Interpolator hier beabsichtigt. Er liegt also auf der Linie des PentRed.”264

Again, this presumes the attested background connections. He presumed the negative characterization of Balaam without demonstrating that it is immanent in the text. Nowhere has Numbers 22–24 suggested that Balaam attempted ‫נחשׁ‬ or ‫ קסם‬against Israel. He argued that the scene following the second oracle and connecting to the third presents a logical continuation of the narrative. “Die dritte Orakelszene ist zunächst wieder ganz analog den ersten beiden gestaltet. Es gibt von daher keinen Grund, die Passage einem anderen, neueren Autor als dem HexRed zuzuordnen.”265 This seems likely. However, Achenbach went on to suggest that, by referencing the opening of the third oracle with ‫נאם בלעם‬, the author seeks to diminish the oracle’s quality and dignity; it is not a ‫נאם יהוה‬, but only from Balaam.266 He did not evaluate David’s final words (2 Sam 23:1) in the same way, though he noted that they are on the same level (“auf eine Ebene”), whatever that means in this case.267 That seems less likely, and certainly unclear. Are David’s words in 2 Samuel 23 of diminished quality and dignity? To the introduction to the third oracle, in which Achenbach saw connections to Joshua (HexRed! Josh 24:27), Abraham (Gen 17:1, 17), and Moses (Deut 32,1), he stated: “So scheint es sich bei dem Vers keinesfalls um eine alte Orakeleinleitung zu handeln, sondern vielmehr um eine antikisierende Stilisierung des legendären Sehers, in der dieser als der größte Prophet zum Werkzeug Gottes geworden ist und unter den Völkern nachgeordnet den Vätern Israels und Mose die Verkündigung der Gerichte Jahwes an diese Völker auszurichten hat. Die Annahme ist nicht unplausibel, daß nun PentRed in dem dritten Orakel, welches aus dem bisherigen Verlauf heraus den Höhepunkt der Erzählung markiert, massiv interpretierend schon in der Disposition eingegriffen hat.”268

He saw a contrast between the desert and the land that implies the conquest as a theme in the third oracle. For this reason, he attributed Num 24:5–7a to HexRed, especially as it attests no familiarity with the blessing of Moses.269 Numbers 24:7b–9 are additions from PentRed or later. 7b references Agag, which Achenbach connected to 1 Sam 15:8 and Deut 25:17–19.270 The second case makes no mention of Agag and Num 24:7b does not note that Agag was an Amalekite. This contextualization is thus tenuous. Numbers 24:8a “picks up” 23:22, whereby Achenbach ignored the distinct syntax. He regarded Num 264

Achenbach, Vollendung, 417. Achenbach, Vollendung, 418. 266 Cf. Achenbach, Vollendung, 419. 267 Cf. Achenbach, Vollendung, 419–20. 268 Achenbach, Vollendung, 420. 269 Cf. Achenbach, Vollendung, 420–21. 270 Cf. Achenbach, Vollendung, 421. 265

1.4 Redaction-Historical Models for the Development of Numbers 22–24

65

24:8b as “independent” (“eigenständig”) in its formulation, but suggested that it anticipates David’s dynasty’s victory over their enemies.271 Such an interpretation is hardly consistent with either a Pentateuch or Hexateuch narrative context, but once again could reflect an Enneatuch composition. He regarded Num 24:9a as interacting with Gen 49:9 (just like Num 23:23 did) and Deut 33:20, implying its belonging to the same editorial layer. This was undertaken to further embed the prophetic principle in the Torah, an undertaking of the PentRed in Achenbach’s reconstruction. 272 The return to the second person in Num 24:9b closed the circle opened in 5a and programmatically references back to Gen 12:3.273 The same sentence appears in Isaac’s blessing Jacob in Gen 27:29, which Achenbach identified as presuming Judah’s domination of Edom. 274 “Der Text kehrt also i.S. des HexRed zu seinem Grundgedanken zurück: der Segen auf Abraham und Jakob-Israel ist durch Bileam nur zu bestätigen gewesen.”275 That is, Achenbach recognized that a third oracle was part of the original story, but he attributed no element of the transmitted text to it. Every part of Num 23:4–9 belonged to the Hexateuch redaction, the Pentateuch redaction, or was a part of the Hexateuch composition edited by the Pentateuch redaction. That seems problematic to say the least. We will return to the structure, content, and diachronic analysis of the third oracle in Chapter Three. For Achenbach, the basic version of the story concluded with 24:14a followed by 25.276 The material between these verses, 24:14b–24 represent quite late expansions on the text. “Sie [Num 24,14–24] sind wohl jünger als ThB [theokratische Bearbeitung], denn sie vertiefen das eindrückliche Bild vom Seher Bileam. ThB hingegen wertet die Figur des Bileam vorwiegend negativ (vgl. Num 31,8.16). So bereiten sie die radikalisierte Auslegung der qāhāl-ordnung in Dtn 23,4–6 vor, wie sie die chronistische Schilderung Neh 13,3 formuliert.”277

This seems an unusual contradiction to every other biblical text about Balaam. Are there any other indications that late attempts were made to improve Balaam’s standing in the biblical literature? I know of none, but we will have to return to this issue in Chapter Four. It seems more likely that the appending of these oracles must have at the least begun before the negative additions of Achenbach’s Pentateuch Redaction and Theokratische Bearbeitung. Perhaps one could accept his reconstruction if some earlier addition added the positive elements in 24:14b–23 with the perhaps negative 24 (who or what is ‫)?עבר‬ 271

Cf. Achenbach, Vollendung, 421. Cf. Achenbach, Vollendung, 421. 273 Cf. Achenbach, Vollendung, 421. 274 Cf. Achenbach, Vollendung, 421. 275 Achenbach, Vollendung, 421. 276 Cf. Achenbach, Vollendung, 421. 277 Achenbach, Vollendung, 422. 272

66

Introduction

coming from a negatively reacting editor. That would however imply that there must have been an older version than that of HexRed, which Achenbach appears to reject in Numbers 22–24. At least he did not maintain that these verses must have been part of a later redaction; he identified them as Fortschreibungen. That at least leaves a backdoor open that a later editor could have added such sentiments about Balaam here, and only here. The plausibility of such an assertion remains highly speculative in my opinion. Thus, in total, Achenbach’s study of the Balaam pericope yielded the following results: The editor HexRed formed the story by redacting an earlier, irretrievable narrative from legend about war with Moab that had been connected to the Balaam-tradition. This HexRed version consisted of Num 22:1– 3a, 4b–6, 7b, 8–20, 36–38a; 23:1–4, 5b, 12, 16aαb, 17–20, 25–30; 24:1a, 2–3a, 5–7a, 10–14a, and 25. The PentRed added the parodying and polemically pointed donkey narrative and the material in Num 22:3b–4a, 7a, 21–35, 38b; 23:5a*, 12, 16aβ, 21–24; 24:1b, 3b–4, and 7b–9. The youngest editors added Num 24:14b–24. 278 As described above, this reconstruction contains both a number of helpful observations and a number of problems. In contrast to, e.g., Witte, Achenbach remained much more skeptical of our ability to reconstruct a pre-Hexateuch version of Numbers 22–24. The discussions in Chapters Three and Four will reflect on whether that is the case.

1.4.8 Rainer Albertz In an article in two pieces, Rainer Albertz argued for the post-P introduction of the Balaam pericope into its ultimate context.279 Here, I will touch only briefly on the relevant argumentation. To argue this point, he suggested that a non-P editor used the P notice from Num 22:1 and then composed 22:2 in order to incorporate the Balaam story into this context. Since Num 22:2 references a text that Albertz identified as post-P (Num 21:21–35), Num 22:2 must stem from the same post-P author. Further, this same editor included the material in Num 24:14b–19 with its references to the Edomites. He previously identified the Edomite material in Num 20:14–21 as post-P, which means that Num 24:14b–19 should be post-P as well. Ignoring for the moment whether the identification of a people suffices for the ascription of elements to the same author, one must question whether 20:14–21 in itself is even of one piece.280 Should more than one party have 278

Cf. Achenbach, Vollendung, 424. Cf. Rainer Albertz, “Das Buch Numeri jenseits der Quellentheorie. Eine Redaktionsgeschichte von Num 20–24 (Teil I),” ZAW 123, no. 2 (2011): 171–83 and Rainer Albertz, “Das Buch Numeri jenseits der Quellentheorie. Eine Redaktionsgeschichte von Num 20–24 (Teil II),” ZAW 123 (2011): 336–47. 280 N.b. the duplications in 14–17 // 19 and 18 // 20 in addition to the distinct terminology of “Israel” in 14 and 21, but “Israelites” in 19. 279

1.5 Conclusion

67

been responsible for these verses, any one party could have been responsible for the material in 24:14b–19. That party could have been pre-P, P, or post-P, at least theoretically. Albertz’s failure to recognize the duplications in 20:14– 21 presume that one author must have been responsible for all of this material and that single author must have worked after P. However, that is hardly guaranteed. Turning to the opening of the Balaam preicope in 22:1–2, the diachronic relationship of these verses is probably exactly the opposite of what Albertz proposed. Should the Balaam narrative have existed in an early version that began with 22:3, which Albertz apparently presupposed when he ascribed 22:1 to P and 22:2 to a post-P editor who inserted the Balaam tale, no need for Num 22:2 exists. It does not clarify the situation; nor does it include any relevant or necessary information after 21:21–22:1. It would be a purposeless addition. That means that either the original Balaam narrative must have begun with 22:2, which Albertz cannot allow, since he previously identified the whole of 21:21–35 as a distinct compositional unity and dated it after P. Or Num 22:2 followed on some form of Num 21:21–35* in an earlier addition, meaning that 22:1 is in fact the interpolation. Should either of those scenarios be the case – and both of them indeed appear more likely than Albertz’s reconstruction – than the insertion of the Balaam story into its context in Numbers 22–24 cannot represent a post-P interpolation in the manner than Albertz reconstructed.

1.4.9 Summary of Redaction-Historical Models When taken as a group, however, what advantages do redaction-historical models for the development of the Balaam pericope offer? The recognition that texts grow in different manners under the influence of various texts over the course of time seems the best response. These models present the refreshing of texts, the actualization of their contents to address new audiences at different times. They permit the addition of new materials of a variety of forms over the course of longer eras. Most importantly, they demonstrate the high estimation of the texts that were redacted (otherwise older versions would have just been rejected and not simply incorporated and expounded), while simultaneously explaining the disparate elements that appear within a textual corpus without artificially restricting the number of “sources” that might have stood behind the final product.

1.5 Conclusion 1.5 Conclusion

Having covered several models for the development of Pentateuchal texts in general and Numbers 22–24 in specific, we can summarize several elements that merit further discussion below. That is, what are the desiderata of literary-

68

Introduction

critical and redaction-historical undertakings in Numbers 22–24? The preceding survey has hopefully at least made clear, that there are only limited elements upon which scholars agree. One example would be the growing consensus on the generally uniform character of the whole, with the exception of particularly the donkey episode and the verses Num 24:14b–24. Yet even that simple statement would not find universal support. However, most of the discussion revolves ultimately around two poles. First, to what degree is Numbers 22–24 a uniform composition? The question in scholarship is generally not whether Numbers 22–24 contains elements from more than one source, whatever exegetes might mean with that term. Rather, the question reflects on to what extent one finds distinct material within Numbers 22–24. Whether dividing the text among redactional layers or even in the proposed sources of the newest iteration of the Documentary Hypothesis, it seems hardly anyone regards Numbers 22–24 consisting entirely of one piece, even if only Num 22:1 alone did not belong to the rest. The natural extension of this first question regards the number of occasions that such editing might have occurred. Second, when did Numbers 22–24 become part of a larger narrative structure or was it always extant as such? Was the author of some source responsible for its location in Numbers? Or did an editor decide to add it at that point? Beyond these questions, this study will address the other biblical texts about Balaam. On the one hand, this should serve as a kind of control for the results of the text-critical, literary-critical, and redaction-historical results presented here. That is, to what degree, if any, do the other texts about Balaam affirm the observations made here about Numbers 22–24? What information might the epigraphic source about Balaam add to the discussion and how does it fit into this scheme? In what way does the understanding of the character Balaam reflected in the language used about him affirm or contradict the observation made in Numbers 22–24, the other biblical Balaam traditions, and the inscription from Tell Deir ʿAlla? Finally, might the textual history of Numbers 22–24 and the variants attested in the manuscript traditions inform this literary-critical undertaking? To this matter we currently turn.

Chapter 2

The Text of Numbers 22–24 2.1 Introduction 2.1 Introduction

This chapter focuses on the text of Numbers 22–24 and its transmission history as reconstructed with the aid of various manuscripts and early translations. The most relevant witnesses for this study – as with just about any study focused on the Torah / Pentateuch – are the Masoretic text (M) attested in the outstanding witness of Leningradensis (ML; Codex B19A of the state library in St. Petersburg), the various Greek manuscripts commonly identified under the rubric of Septuagint (G), the Torah of the Samaritans (Smr), and the biblical manuscripts from the Dead Sea (Q), particularly 4QLev–Numa and 4QNumb. Interaction with non-biblical texts from Qumran, as well as the translations of the Targumin (T), the Vetus Latina (La), the Vulgate (V), and the Peshitta (S) play a lesser, though significant role for the appreciation of the history of the text. Focusing on the available witnesses, this study seeks to reconstruct the consonantal text as far back as one might chronologically risk. I would venture to suggest that we can follow this textual chronology to a greater or lesser degree of certainty even behind what might be called “the canonical Hebrew text” that became authoritative in late Antiquity or over the course of the early Middle Ages. Substantial differences between the Masoretic (and presumably therefore proto-Masoretic) textual tradition and especially the Septuagint and Samaritanus indicate that the text of Numbers remained in a somewhat fluid state well into the Hellenistic or Roman era, presumably beyond the separation both of the Samaritan community and early Christianity from their Judaic origins. Parallel versions apparently existed concurrently. At the same time, the variance of the textual witnesses should not be overstated. The most important witnesses all attest a remarkably consistent text. It is precisely due to this similarity that one must address the differences: how did they come about and can we determine their relationship to one another? It is improbable that only one text – or even one textual tradition – always attests the oldest reading; each case must be handled on its own merits and only secondarily addressed in the context of other variants. What this means is that, e.g., neither M nor G will be assumed to always present the older reading in cases where variants appear. Addressing the criterion for deciding for or against a particular reading, I am convinced by the evaluation of Emanuel Tov:

70

The Text of Numbers 22–24

“Common Sense, rather than textual theories, is the main guide, although abstract rules are sometimes also helpful.”1 Tov’s comment demonstrates the subjective nature of the undertaking and reminds us that such reconstructions are based on the limited evidence available and therefore steeped in insecurity, remaining entirely a matter of plausibility. Accepting this insecurity, an attempt to reconstruct the oldest possible text, and its further development, still can be fruitfully undertaken. To a degree, this study returns some focus onto the Septuagint version of Numbers.2 In general, the Septuagint of Numbers must have translated a text quite similar to the consonantal text relied on by the Masoretes. Mostly it remains close to its Hebrew Vorlage, often to the detriment of Greek syntax and grammar. For this reason, some scholars considered the sometimes peculiar language of G, particularly in cases like Numbers, to evince a particular Jewish Greek dialect.3 In spite of this, G and M evince definite distinctions, raising the question as to how these distinctions developed. “The differences [between G and M] may be due to a series of facts. Some are by no means intentional. The Greek text may have been corrupted in the process of copying or the translators may have worked with a Vorlage that differed from the MT. Or they may have misunderstood the Hebrew, or understood it in a way differing from that of the Masoretes. One should not forget that they worked with unvocalised texts. The identification of the root of some Hebrew word forms may have caused problems. It should be added that the Masoretes had to deal with similar difficulties. Their solution is not necessarily the best. Moreover, we are not always sure that we understand the MT.”4

The following discussion of the variants will illuminate, and – where possible – explicate these differences. 1 Emanuel Tov, Textual Criticism of the Hebrew Bible, Third ed. (Minneapolis, MN: Fortress, 2012), 281. Italics in original. 2 For general information on the philosophical and ideological backgrounds of translation in antiquity, cf. the helpful introduction in Theo van der Louw, Transformations in the Septuagint. Towards an Interaction of Septuagint Studies and Translation Studies, CBET, vol. 47 (Leuven; Paris; Dudley, MA: Peeters, 2007), 25–55. Regarding more generally the use of G in textual criticism, cf. Emanuel Tov, The Text-Critical Use of the Septuagint in Biblical Research, Third ed. (Winona Lake, IN: Eisenbrauns, 2015), 1–40. 3 John A.L. Lee, A Lexical Study of the Septuagint Version of the Pentateuch, SCS (Chico, CA: Scholars Press, 1983) has satisfactorily refuted this position; cf. his summarizing comment in Lee, Lexical Study, 30: “…in order to account for the peculiarities of LXX Greek it is sufficient to refer to the fact that the work is a translation, and unnecessary to posit the existence of a living ‘Jewish-Greek’ dialect.” 4 Johan Lust, “Messianism and the Septuagint, with Special Emphasis on the Pentateuch,” in Messianism and the Septuagint: Collected Essays by J. Lust, ed. Katrin Hauspie, BETL, vol. 178 (Leuven: Peeters, 2004), 136 = Johan Lust, “Messianism and the Septuagint, with Special Emphasis on the Pentateuch,” in Theologische Probleme der Septuaginta und der hellenistischen Hermeneutik, ed. H. Graf Reventlow, Veröffentlichungen der Wissenschaftlichen Gesellschaft für Theologie, vol. 11 (Gütersloh: Chr. Kaiser – Gütersloher, 1997), 32.

2.1 Introduction

71

Nonetheless, many scholars have viewed the Septuagint version of Numbers as being of particularly poor quality. For example, Frankel regarded Numbers as the weakest volume of the Pentateuch.5 When contrasting it with the other books of the Pentateuch, he commented that, “Numer. hat sehr viele (und oft sinnlose) Abweichungen vom masor. T., fast mehr als die anderen vier Bücher des Pentat. zusammengenommen.”6 More recently, Wevers developed such an estimation of the text of Greek Numbers during the process of this editing the volume for the Göttinger Septuagint. In his Notes on the Greek Text of Numbers, Wevers opened his evaluation of G with the following comment: “The Greek translation of Numbers […] is without a doubt by far the weakest volume in the Greek Pentateuch. What makes work on the book so frustrating is that side by side one can find gross failures to follow ordinary rules of grammar, i.e. of apparent incompetence, as well as acute and even subtle distinctions betraying an active mind engaged in the interpretation of sacred scripture, ready not only to clarify obscure passages, but even to correct what might appear to be factual errors or contradictions within the text.”7

In both of these cases, the scholars consider the translator’s inability to be the background of differences between M and G. That is, differences in the versions implied incompetence in the translation (or its transmission). More recently, Aejmelaeus has commented generally on the translators of G that, “[t]heir general intention was not directed towards the formal representation of items in the original. They did not consciously aim at word-for-word translation. Retention of the original word-order or consistency in lexical choices was not striven after.”8 Her conclusion suggests that perhaps scholars have been measuring the quality of the translation with the wrong ruler. Some of the distinctions within the Greek text of Numbers represent intentional stylistic variations created by the translator to better fulfill the needs of the target language.9 Specifically in regard to Numbers, Ausloos has suggested that, “[…] the very well | thought-out manner in which the translator renders topological etiologies implies that the characterisation of the LXX-translation of Numbers as the weakest of the Pentateuch needs to be nuanced somewhat.” 10 Ziegert has 5

Cf. Zacharias Frankel, Ueber den Einfluss der palästinischen Exegese auf die alexandrinische Hermeneutik (Leipzig: Barth, 1851), 168: “Diese Version [die griechische Version des Numeri-Buches] gehört zu den verfehltesten der heil. Schrift.” 6 Zacharias Frankel, Einfluss, 170. 7 John William Wevers, Notes on the Greek Text of Numbers, SBLSCS (Atlanta, Ga.: Scholars Press, 1988), ix. 8 Anneli Aejmelaeus, “Translation Technique and the Intention of the Translator,” in VII Congress of the International Organization for Septuagint and Cognate Studies, Leuven 1989, ed. Claude E. Cox (Atlanta, GA: Scholars Press, 1991), 26. 9 Cf. Nechama Leiter, “Assimilation and Dissimilation: Techniques in the LXX of the Book of Balaam,” Text 12 (1985): 79–95. 10 Hans Ausloos, “The Septuagint’s Rendering of Hebrew Toponyms as an Indication of the Translation Technique of the Book of Numbers,” in Textual Criticism and Dead Sea

72

The Text of Numbers 22–24

demonstrated – relying on, among other things, one portion of the Balaam pericope – that the translator of Numbers was not as poor as previously suggested. For example, he noted that, “[d]er Übersetzer stand vor der Aufgabe, einen schwierigen, stellenweise sogar unklaren Text ins Griechische zu übersetzen. Dabei hat er sich bemüht, den Text verständlich wiederzugeben und gleichzeitig den Inhalt zu verdeutlichen.” 11 Further, he commented, “Offensichtlich hatte der Übersetzer das Ziel, den Text einerseits grundsätzlich wörtlich, andererseits klar und verständlich wiederzugeben.”12 That is, the translator’s inability was not the problem. Other matters often must have been responsible for these distinctions. Many of these will become apparent in the following discussion. Especially since the discovery of the texts from the Judean Desert, positions hypercritical of the Greek translation of Numbers must be reevaluated; the variegated nature of the texts there demonstrates that variants previously attributed to “the translator” can at times be reconstructed with more certainty as having been part of G’s Vorlage. We now know, for example, that there were Hebrew texts bearing readings similar to those present in G. The commonalities between Smr, Q, and G are particularly telling in these cases and have frequently been overlooked or regarded as insignificant in previous studies. These matters will be addressed extensively below in a reversal of the trend to always presume the priority of M against G and/or Smr, particularly with the support of Q.13 The discussion begins with a discussion of the most difficult text-critical issues in Numbers 22–24. Thereupon follows a translation of Numbers 22–24 that primarily reflects M, but does reflect the emendations undertaken both in the preceding text-critical analysis and in the footnotes. The footnotes in the translation consider the remaining text-critical matters. These text-critical matters dealt with in the footnotes are of limited importance (such as a missing

Scrolls Studies in Honour of Julia Trebolle Barrera. Florilegium Complutense, ed. Andrés Piquer Otero, Pablo A. Torijano Morales, JSJS (Leiden; Boston: Brill, 2012), 49–50. 11 Carsten Ziegert, Diaspora als Wüstenzeit. Übersetzungswissenschaftliche und theologische Aspekte des griechischen Numeribuches, BZAW, vol. 480 (Berlin: de Gruyter, 2015), 262. 12 Ziegert, Diaspora, 262. 13 The subsequent chapter will use these results, then consider the inappropriateness of literary-critical models that follow M exclusively for the basis of their reconstructions. David Lee Phillips, The Samaritan Version of the Book of Numbers with Hebrew Variants: A Close Textual Study (Lewiston, NY: Edwin Mellen Press, 2014) regarded the text of Smr as “the autograph”, i.e., the original version of the composition of Numbers, a position which must be rejected a priori for methodological reasons. To the particular insufficiencies of this model regarding Numbers 22–24, cf. Jonathan Miles Robker, review of The Samaritan Version of the Book of Numbers with Hebrew Variants: A Close Textual Study, Antiguo Oriente 12 (2014): 234–40.

2.2 Textual Criticism

73

conjunction at a phrase’s opening) and often could not be settled with any certainty anyway.Finally, the chapter will close with a reconstruction of an older Hebrew text and a brief summary of the results of this study and considerations about the character of the various primary textual traditions. These considerations about documented changes in the textual traditions should aid in postulating developmental stages behind the reconstructed text. Significantly more reflection on the issue of documented changes of texts and their impact on theoretical models has developed in recent years.14 Hopefully this reconstruction will add fruitfully to that debate.

2.2 Textual Criticism 2.2 Textual Criticism

The textual tradition of Numbers 22–24 remained generally stable. While several differences between M (more specifically ML) and other traditions (and manuscripts) make themselves conspicuous, most of them are of limited relevance and of the variety that could be identified by David Carr’s “memory variants”.15 Even the Septuagint translation of Numbers, “the weakest volume in the Greek Pentateuch”, 16 demonstrates strong affinity with M, especially when contrasted with, e.g., the versions of Exodus. Yet, this very textual stability demands further exploration of the instances attesting substantial differences. A number of verses merit special attention due to conspicuous, extensive text-critical issues in the various textual traditions. These will be dealt with in the following discussion.

2.2.1 Num 22:5–6 The text of Num 22:5–6 demonstrates manifold problems, mostly revolving around Balaam’s location and origin in v. 5 and whether the first person plural or singular should be used in v. 6. Relatively speaking, the problems in v. 6 are easily dealt with: the plural in some of the G witnesses would seem to be a later depersonalization of the conflict between Balak and Israel; it better matches a conflict between the people of Moab and the people of Israel. The attestation

14 Cf., e.g., David M. Carr, The Formation of the Hebrew Bible: A New Reconstruction (New York: Oxford University Press, 2011), 3–149 and Reinhard Müller, Juha Pakkala, and Robert Barend ter Haar Romney, Evidence of Editing (Atlanta: SBL Press, 2014), as well as the edited volumes Raymond F. Person, Jr., ed., Robert Rezetko, Empirical Models Challenging Biblical Criticism, AIL, vol. 25 (Atlanta: SBL Press, 2016) and Reinhard Müller and Juha Pakkala, eds., Insights Into Editing in the Hebrew Bible and the Ancient Near East, CBET, vol. 84 (Leuven; Paris; Bristol, CT: Peeters, 2017). 15 Cf. David M. Carr, The Formation of the Hebrew Bible: A New Reconstruction, 57–65. 16 Wevers, Notes on the Greek Text of Numbers, ix.

74

The Text of Numbers 22–24

of the singular in the pre-Samaritan text of 4QNumb as well as 4QLev–Numa strongly favors the singular as the older reading.17 Numbers 22:5a presents problems of a significantly more complicated nature in that they involve the textual histories of both Numbers and Deuteronomy. The Hebrew texts of M and Smr are not clear in Num 22:5a; literally they read: “he sent messengers to Balaam ben Beor to Pethor that is upon the river land of the sons of his kin (Smr: ‘land of the Ammonites’) to call him”. What does that even mean? According to Deut 23:5 M, Balaam came “from Pethor, Mesopotamia” (‫)מפתור ארם נהרים‬. Again there are variants in the textual tradition: Smr reads “Pethor” in the same form as Num 22:5 (‫ ;)פתורה‬G does not record “Pethor” at all in Deut 23:5, nor does it seem that V includes “Pethor”.18 How did this come about? While the notes in BHQ to Deut identify this as a process of influence from Num (Smr) or simplification (G and V),19 it seems more likely that the opposite was true. If one were to regard G and V as attesting the older text of Deut 23:5, the picture becomes more readily explainable: scribes rectified the absence of Pethor by adding it in pre-Smr or Smr, using the form copied from Num 22:5 into Deut 23:5. On the other hand, M more adequately incorporated it into the text by changing the form to match the syntactical demands of the narrative context. Such an explanation is clearer than postulating why G and V chose not to incorporate into, or even deleted Pethor from their texts; it is not recorded in any of their known manuscripts. Thus, at an earlier stage, Deut 23:5 only reported that Balaam came from Mesopotamia (‫)ארם נהרים‬. This distinguishes the Deuteronomy tradition from Numbers.20 So, the form ‫ פתור‬was apparently copied into Deut 23:5. But how did it come to be in Num 22:5 and what was its original form? Both V and S understood the term differently from M, Smr, and G, namely as “the interpreter”, i.e., ‫( הפותר‬ariolum or ‫ܪ‬ respectively).21 While this Hebrew root is somewhat

17

For the generally, though not exclusively, pre-Samaritan character of 4QNumb, cf. Tov, Textual Criticism, 91; Martin Rösel, “Die Textüberlieferung des Buches Numeri am Beispiel der Bileamerzählung,” in Sôfer Mahîr. Essays in Honour of Adrian Schenker Offered by Editors of Biblia Hebraica Quinta, ed. Yohanan A. P. Goldman, Arie van der Kooij, and Richard D. Weis (Leider; Boston: Brill, 2006), 218; and Armin Lange, Die Handschriften biblischer Bücher von Qumran und den anderen Fundorten, vol. 1 of Handbuch der Textfunde vom Toten Meer (Tübingen: Mohr Siebeck, 2009), 80–82. 18 The absence of “Pethor” in Deut 23:5 was not even noted in BHS, a matter that BHQ thankfully rectified. 19 Cf. Carmel McCarthy, ed., Deuteronomy, BHQ, vol. 5 (Stuttgart: Deutsche Bibelgesellschaft, 2007), 115*. 20 Cf. the discussion about the passage in Chapter Five for a more detailed discussion. 21 Michael L. Barré, “The Portrait of Balaam in Numbers 22–24,” Int 51 (1997): 256 understood M’s reading as an Aramaic form meaning “the seer” or “the diviner”. The postpositive definite article seems unlikely, though.

2.2 Textual Criticism

75

rare, 22 S and V probably made a wordplay explicit. The town of Balaam’s origin sounds like “interpreter”, a vocation that S and V state that Balaam fulfilled.23 It is clear from G that the translator either did not understand the wordplay or did not attempt to reflect it in any way in the translation: the form found in M and Smr must have been the one he or she read (note the α on the end of Φαθουρα).24 He or she may not have understood it as a city,25 though that seems unlikely. Numbers, in contrast to Deuteronomy, never attested Balaam coming from Mesopotamia. It is also unclear how the next phrases should be understood. G looks like a direct and slavish translation of the Hebrew: “upon the river land sons of his people”. All traditions are probably missing the preposition (‫ ;)ב‬it vanished as a result of haplography caused by the similarity between the preceding ‫ ר‬and ‫ב‬, perhaps even in paleo-Hebrew in which these letters are particularly similar (b and r).26 Originally the text must have read “Pethor by the river in the land (‫ )בארץ‬of his people”. The Torah of the Samaritans regarded Balaam as an Ammonite.27 Since S and V and several Hebrew manuscripts also attest his Ammonite heritage, it is entirely possible that it is older than that in ML and G.28 Alternatively and more likely, Smr, S, and V may have included an Ammonite heritage for Balaam in order to make the text better accord with Deuteronomy 23, which forbids both the Ammonites and the Moabites from becoming part of the assembly ( ‫קהל‬ ‫)יהוה‬. Since the Ammonites do not otherwise figure in Numbers 22–24, scribes could have consciously or unconsciously read ‫ עמון‬instead of ‫עמו‬. If M and G 22

The limited number of other occurrences of √‫ פתר‬in the Hebrew Bible appear exclusively in the Joseph Novella: Gen 40:8, 16, 22; 41:8, 12–13, 15. 23 So also Vermes, Scripture, 129 and Schüle, Israels Sohn, 25. 24 A misunderstanding seems likely as the translator transparently transcribed etiological toponyms in his or her target language on several other occasions; cf. Ausloos, “Septuagint’s Rendering”. 25 This is contra Wevers, Notes on the Greek Text of Numbers, xii, who presumed that the translator recognized this as a city and who thus saw the translator as incompetent, having set a neuter relative pronoun when grammatical strictures would have demanded a feminine. However, the masculine (and feminine) are also attested in the Greek tradition, even though not by the greatest witnesses. Perhaps the translator and later tradents did not know what to make of it and thus set a neuter relative pronoun. 26 The series of consonants ‫( ר באר‬rab r) make this confusion even more likely. 27 Seebass, Numeri III, 15 saw the contradiction between the city Pitrû and the region Ammon as a reason to accept ‫ בני עמון‬as the older reading. However, he failed to consider problems with identifying Pethor with Pitrû. His position can consequently be rejected. Cf. already William F. Albright, “The Home of Balaam,” JAOS 35 (1915): 389, who recognized that ‫ עמון‬clearly presents a corrective emendation. 28 Omar al-Ghul, “The Question of the Homeland of Balaam Again: The Contribution of the Arabic Sources,” WO 36 (2006): 94–103 argued that the Arabic evidence supported such a reconstruction, but I find this doubtful.

76

The Text of Numbers 22–24

do not represent the older reading, than presumably the ‫ ן‬was overlooked due to its similarity to ‫ו‬, a mistake that also could have been made in the paleoHebrew script (n vs. w).29 Ultimately it must remain unclear which texts attest the older reading, but M and G seem the most plausible.

2.2.2 Num 22:10–11 In Num 22:10, the word ‫ לאמר‬is missing in M and Smr. The longer reading in G, V, and S is probably older than that of M and Smr. Scribes presumably overlooked ‫ לאמר‬due to homoioarchton, between ‫ אלי‬and ‫לאמר‬. The attestation at Qumran in the generally pre-Samaritan text of 4QNumb offers significant support to this postulation. Therefore, it will be regarded as the older reading and not merely as a harmonization with v. 5.30 Basically every text-critically relevant witness other than M reads “a people has come out” and not “the people coming out” in 22:11. The Masoretic reading presumably developed as a result of dittography from ‫ הנה‬to ‫העם‬, with the following participle then being harmonized with the subject.31 Further, after ‫ עין הארץ‬there is a sizable plus in G (καὶ οὗτος ἐγκάθηται ἐχόμενός μου = “and is dwelling across from me”), a matter not even noted in BHS. While it might be tempting to ignore this as a late harmonization during the course of reception history, it is attested in 4QNumb, adding credence to its representing the older reading. It will be regarded as such here.32 It was presumably removed in the Hebrew traditions of M and Smr as redundant.33

2.2.3 Num 22:21–35 Historically, probably the most important text-critical matter in this pericope is the usage of ‫ אלהים‬or ‫ יהוה‬in the story of Balaam’s donkey in Num 22:21– 35. While this may seem like a somewhat trivial matter, it is quite important due to the literary-critical operations that have been undertaken on this passage

29 Possibly another reading stood behind both variants: ‫“( עמנו‬our peoples”). This conjecture has the merits of both explaining each variant and matching the context, since the Moabites (one people) are speaking to the Midianites (another people). That attractive, this conjecture must of needs remain speculative. 30 Contra Martin Rösel and Christine Schlund, “Arithmoi / Numeri / Das vierte Buch Mose,” in Septuaginta Deutsch: Erläuterungen und Kommentare Band I. Genesis bis Makkabäer, ed. Martin Karrer and Wolfgang Kraus (Stuttgart: Deutsche Bibelgesellschaft, 2011), 482. 31 Contra Seebass, Numeri III, 16. 32 Contra Seebass, Numeri III, 16. 33 Contra Rösel and Schlund, “Arithmoi,” 482.

2.2 Textual Criticism

77

because of it.34 Several supporters of the Urkundenhypothese used this transition from the general usage of ‫ אלהים‬in Num 22–24 to the usage of ‫ יהוה‬in Num 22:21–35 as an opportunity to cut this narrative into two strands and incorporate those strands into larger narrative documents.35 That is, text-critical operations were led by literary-critical models and the text itself was neither the basis of the model nor used to question the model’s veracity in this pericope. For example, one can consider Martin Noth, who replaced M’s ‫ אלהים‬in 22:22 with ‫( יהוה‬reading with Smr, one M manuscript, and three G manuscripts)36 in order to identify the provenience of the passage beginning in 22:22 as the source document J.37 However, both these literary-critical models and other explanations of this transition fail to convince. It is impossible to determine which of these readings was older in each case since the semantic fields were essentially identical by the period of transmission and reception under consideration; scribes could exchange one for the other without creating a theological problem.38 Thus, it would be inappropriate to determine which reading should be given priority and would be even more inappropriate to use the postulated older reading as the primary basis for literary-critical dissections, especially when these constructions cause serious narratological and redactional-historical problems. The literary-historical implications will be handled more expansively in the subsequent chapter. Numbers 22:32b presents a number of text-critical problems. First, the object suffix (‫ )ך‬should be restored to ‫ שׂטן‬with Smr, G, and the other witnesses; it was lost in the Masoretic tradition, presumably due to haplography influenced by the following ‫כי‬.39 Further, it is entirely unclear what ‫ ירט הדרך‬should

34 Cf. Jonathan Miles Robker, “The Balaam Narrative in the Pentateuch / Hexateuch / Enneateuch,” in Torah and the Book of Numbers, ed. Christian Frevel, Thomas Pola, and Aaron Schart (Tübingen: Mohr Siebeck, 2013), 334–66 and the literary-critical discussion in the next chapter. 35 See the discussion in the preceding chapter. 36 Cf. Noth, Numeri, 146. 37 Cf. Noth, Numeri, 156–57. 38 Cf. Greßmann, Mose und seine Zeit, 324. Contra Rösel, “Textüberlieferung,” 221, who saw Smr as a harmonization and thus favored M in every instance. Martin Rösel, “Die Septuaginta und der Kult. Interpretationen und Aktualisierungen im Buch Numeri,” in La double transmission du texte biblique. Etudes d’histoire du texte offertes en hommage à Adrian Schennker, ed. Yohanan Goldman and Christoph Uehlinger, OBO (Göttingen: Vandenhoeck & Ruprecht, 2001), 39 and Rösel, “Textüberlieferung,” 224 saw G as a devaluation of Balaam, who now uses ‫ יהוה‬less often as a name for God. Yet, this postulated recension failed to change every occurrence; Numbers 24:1 is particularly relevant in that the narrator refers to Balaam’s satisfying ‫יהוה‬, even in G (κύριος), a matter inconsistent with Rösel’s reconstructed recension. Cf. Seebass, Numeri III, 17, who wrote of G’s “allgemeinen Vorliebe für θέος”. 39 Cf. Seebass, Numeri III, 17.

78

The Text of Numbers 22–24

mean.40 It is pretty clear that the object suffix appended to ‫ דרך‬should be restored, as in the various witnesses (it irrefutably vanished due to haplography) and that the article ‫ ה‬should be removed from the beginning of ‫דרך‬. The NRSV translates ‫ ירט‬as “perverse”, which matches (quite literally) the Vulgate (perversa); however, the Latin means something like “turned around” and does not carry the same connotations as the English. Samaritanus reads “your path is the bad one (‫”)הרע‬.41 G reads οὐκ ἀστεία, “not well-formed” or “not good”; cf. Exod 2:2, where G translates ‫ טוב‬with ἀστεῖον, the only other attestation of this Greek word used in a translation of a known Hebrew biblical text in G being Judg 3:17.42 It is thus possible that the G-Vorlage and Smr matched here: οὐκ ἀστεία is an unusual, but viable translation of ‫הרע‬. The conjecture of BHS (‫)ירע‬ reads “shaky” and may be accurate.43 The confusion came about due to different errors in the traditions: Smr read ‫ ה‬as ‫י‬, whereas M read ‫ ט‬instead of ‫ ;ע‬both of these errors are more likely in paleo-Hebrew. Thus, 22:32baα reads “for your path is shaky before me” in the oldest possible reconstruction.

2.2.4 Num 23:3–5 and 15–17 Chapter 23 attests a number of related problems in vv. 3–5 and 15–17. Many of these verses demonstrate phenomena identical to the variants in 22:21–35, namely variation in the nomenclature used for the deity or communicant with Balaam. In 23:3 Smr and G, Balaam suggests that “God” (‫ אלהים‬and ὁ θεός) might come to him. Numbers 23:4 in Smr identifies the subject as “the messenger of God” (‫)מלאך אלהים‬. In 23:5, “the messenger of YHWH” (‫)מלאך יהוה‬ is the subject in Smr, “God” (ὁ θεός = ‫ )אלהים‬is the subject in G, and YHWH (‫ )יהוה‬is the subject in M. The limited remains of 4QNumb seem to attest the same reading as Smr in every instance. What implication does this have for reconstructing the oldest possible version? Similar observations to those above prevail. Once the story of Balaam’s donkey had been added into the narrative, not only ‫ יהוה‬and ‫ אלהים‬could be used as subjects, as encountered in the narrative prior to Num 22:22; now the text could reference ‫ מלאך יהוה‬and ‫מלאך‬ ‫אלהים‬, just as Smr does.44 Samaritanus’ usage even makes more sense in this 40 August Dillmann, Die Bücher Numeri, Deuteronomium und Josua, Kurzgefasstes exegetisches Handbuch zum Alten Testament (Leipzig: Verlag von S. Hirzel, 1886), 147 proposed ‫ יָ ָר ְט ָתּ‬as found in the BHS. Cf. Holzinger, Numeri, 114. 41 Seebass, Numeri III, 17 considered this a hiphil form, which would not make sense in this context. 42 There for ‫ ;בריא‬the other attestations in G are Jdt 11:23; 2 Macc 6:23; and Sus 1:7 43 Contra Noth, Numeri, 147, who identified the reading ‫ ירע‬as a “Verlegenheitsauskunft”. 44 Rofé, Book of Balaam, 37–40 regarded the usage of ‫ יהוה‬throughout Numbers 22–24 as a contamination of the text after the insertion of the donkey episode. This observation works just as well for the additional appearances of ‫ מלאך יהוה‬in Smr.

2.2 Textual Criticism

79

context, since it was ultimately the messenger of YHWH who told Balaam only to report what he – the messenger – said in the final literary form (22:35a). This readily explains why Smr reads “messenger of YHWH” in 23:5: it is a narratological harmonization with the preceding material. We encounter identical phenomena in 23:15–17. As the aforementioned cases (also in 22:21–35) demonstrate, it is impossible to state with any certainty whether ‫ אלהים‬or ‫יהוה‬ presents the older reading; it does seem clear that readings including ‫מלאך‬, as in 23:5 Smr and in 4QNumb, represent later harmonization with the content in 22:21–35. Smr evinces such harmonization already in 22:20, where “the [divine] messenger” appears for the first time. The editors of the apparatus of BHS encourage the exegete to increase the harmonization of 23:3 with 23:15 through the addition of particles or the usage of the plural for the number of offerings. Because these matters do not match their context, it remains advisable not to change them and to read with M (which also matches G against Smr and/or S in these instances). This harmonization should be rejected as unnecessary. The conclusion of Num 23:3 in G is longer than in M et al. and reads καὶ παρέστη Βαλακ ἐπὶ τῆς θυσίας αὐτοῦ καὶ Βαλααμ ἐπορεύθη ἐπερωτῆσαι τὸν θεὸν καὶ ἐπορεύθη εὐθεῖαν (“and Balak stood by his offering and Balaam went to enquire of God”). 4QNumb affirms it as an old reading, for which reason it should be restored to the text. It cannot be regarded as a harmonization in that v. 15 does not match it, even though G attests a similar, abbreviated comment in v. 15. At some point the Hebrew traditions may have eliminated it intentionally as a redundancy or unintentionally through parablepsis. These words follow the obelos in the Syro-Hexapla, implying that they had disappeared from the Hebrew textual tradition by the seventh century CE,45 if not already by the time of Origen in the third century CE46 The same should be said for the longer reading in 23:15, which is essentially ubiquitously attested in the Greek tradition. Thus, in 23:3 one should add ‫וילך ויתיצב בלק על עלתו ובלעם נקרה אל אלהים‬ with 4QNumb and G. The final major point to be dealt with in M Num 23:3–5 and 15–17 is the verb ‫ויקר‬, presumably from ‫קרה‬. Samaritanus reads ‫מצא‬, which has also been reconstructed in 4QNumb. It is unclear what G read, but a good candidate would be ‫ ;זרח‬cf. Isa 60:2 and the discussion about 24:17 below. A misreading seems unlikely, since there is no clear way that a ‫ ק‬could be confused for a ‫ ז‬or vice versa. This suggests that one should either postulate a Vorlage that varied from M and Smr or that the translator is responsible for the distinction. Since all three great witnesses attest distinct readings, it is possible that a corrupt tradition stands somewhere in the background of all three, which precludes a 45

Cf. Tov, Textual Criticism, 151 for the date of the composition of the Syro-Hexapla. V, attesting the shorter reading, would support Origen’s probable familiarity with the shorter reading. 46

80

The Text of Numbers 22–24

secure reconstruction of any older reading. The most probable solution is that G translated this root inconsistently. At the same time, Smr probably replaced this somewhat rare form (‫ קרה‬occurs otherwise only in Niphal in Exod 3:18 and 2 Sam 1:6) with the more common ‫מצא‬. Thus, M should be maintained. Having dealt with the most difficult issues of the narrative material, we can turn our attention to Balaam’s prophecies. In this area one finds the most and greatest variation among the textual traditions. It remains unclear who was responsible for the substantial differences between Smr and G and M. Further, the relative chronology of the variants remains unclear, but I will attempt at least to relate them to one another and discuss their distinctions. Several verses will be used exemplarily, these being the verses that vary most.

2.2.5 Num 23:10 At the conclusion of Balaam’s first oracle in Num 23:10, M and Smr inquire: “who can count the dust of Jacob?” (‫ )מי מנה עפר יעקב‬whereas G inquires: “who can count the seed of Jacob?” (τίς ἐξηκριβάσατο τὸ σπέρμα Ιακωβ). Since it is highly improbable that someone confused ‫ זרע‬for ‫ עפר‬or vice versa, either the translator of G had a different word in his or her Vorlage, or the translator created the variant reading. Here it seems likely that the translator of G may be responsible: rather than leave the metaphor as it stands in the Hebrew, he or she made the interpretation explicit in the sense of Gen 13:16 and 28:14. Jacob’s dust is his seed (and that of Abram).47 In the second stich of 23:10, Smr and G read both differently than M and from each other. Smr and G agree that another question is being asked and that the stich is not merely the completion of the foregoing: ‫ מי ספר‬and τίς ἐξαριθμήσεται. These present the preferable reading.48 Further, both Smr and G attest different readings than M’s ‫“( את־רבע‬quarter”); Smr reads ‫מרבעת‬ (“quadrilateral/square”), which is close enough to M that one cannot preclude a misreading or intentional change within either tradition; M seems the more likely candidate for the postulation of an intentional change in that its text is sensible. However, the translator of G either again changed the meaning or the 47 Cf., e.g., Vermes, Scripture, 147; Martin Rösel, “Jakob, Bileam und der Messias. Messianische Erwartungen in Gen 49 und Num 22–24,” in The Septuagint and Messianism, ed. Michael A. Knibb (Leuven: University Press, 2006), 164; and Wevers, Notes on the Greek Text of Numbers, 44. 48 Rösel, “Jakob,” 164 seemed to doubt this being the G-Vorlage, but he ignored Smr. The relationship to Gen 13:16 in the Septuagint remains difficult, mostly because of the unusual translation of √‫ מנה‬with ἐξαριθμέω there. Such a translation can only be found in Gen 13:16 and Ps 89:12 G; otherwise, ἐξαριθμέω always serves as a translation √‫ספר‬. Cf. Num 31:5, in which √‫ ספר‬also should be reconstructed. Gilles Dorival, Les Nombres, La Bible d’Alexandrie, vol. 4 (Paris: Cerf, 1994), 159 recognized a G-Vorlage distinct from M in Num 23:10, a position I support.

2.2 Textual Criticism

81

Vorlage read ‫ ;משׁפחת‬δήμος is G’s standard equivalent for this root (147 times in Num). Though the situation remains somewhat cloudy, the text most likely was corrupted away from something like the postulated G-Vorlage to something like Smr. Proto-M or M would have changed the Smr reading, which is somewhat nonsensical (to what extent does Israel have squares?), into something more sensible. Thus, the following development can be postulated: ‫( משׁפחת‬G-Vorlage) → ‫( מרבעת‬Smr) → ‫( את־רבע‬M).49 While it remains difficult to determine how Smr came to confuse the consonants it did or why it changed them, this seems the most likely development of the text. One can note here as well, that M’s ‫ את‬in this verse presents the only usage of this particle in all of Balaam’s oracles and that it disrupts the parallelism in this verse: the preceding stich contained no object marker.50 In 23:10b, G delivers a much nicer parallelism than M and Smr. That better structure could indicate later reworking. “May my spirit die among the spirits of the upright; and may my seed be like their seed” (ἀποθάνοι ἡ ψυχή μου ἐν ψυχαῖς δικαίων καὶ γένοιτο τὸ σπέρμα μου ὡς τὸ σπέρμα τούτων); thus reads G, implying a Vorlage that read ‫ תמת נפשׁי בנפשׁות ישׁרים ותהי אחריתי כאחריתם‬or something similar. It is more likely that the scribe of the Vorlage or the translator of G created this parallelism than that so much of it was lost or edited out in the Hebrew tradition. Yet the possibility of textual corruption does remain;51 cf. 23:10a.

2.2.6 Num 23:18–21 Turning to Balaam’s second prophecy, it is clear that G understood the text both differently than M and probably correctly in 23:18bβ. The verb ‫ אזן‬is never otherwise used with ‫ עד‬as the marker of the object to what one listens.52 The Septuagint (μάρτυς) understood ‫ עד‬to mean “testimony”, which seems the best reading, even though G records it in nominative and not the appropriate accusative and apparently overlooked the ‫ י‬attested in Smr and M. In principle however, G presupposes the same consonantal text as Smr and M in this verse. Yet, this was later misunderstood and incorrectly pointed by the Masoretes. 49

Cf. similarly Seebass, Numeri III, 19. N.b., however, the narrative introduction to the oracles in Num 24:20–21 (+ 23 [reconstructed]). 51 Samuel E. Loewenstamm, “The Death of the Upright and the World to Come,” JJS 16 (1965): 183–86 suggested that the text originally must have read ‫תתם נפשׁי תם ישׁרם ותהי‬ ‫אחריתי כמהו‬, a reading lost due to a number of scribal errors and – more significantly – interpretations. While this theoretically remains possible, he speculated this without any supporting evidence. 52 The only other attestation of √‫ אזן‬+ ‫ עד‬is Job 32:11, whereby in Job the preposition should be understood temporally (“I listened until you understood, while you sought words”) in contrast to the usage in Num 23:18, where it would represent the object. The reading “until I/me” does not make sense. 50

82

The Text of Numbers 22–24

The Samaritan text matches M, and approximates G. Thus, Smr – supported by the understanding of G – provides the oldest understanding. Numbers 23:19 “…must have sounded very shocking to the commentators, and the various attempts to paraphrase it are intended to soften its apparent disrespect”. 53 Theological and ideological considerations apparently guided G’s translation of Num 23:19a, in which G reinterpreted the text, removing any suggestion that God could lie and strengthening the understanding of God’s omnipotence (God cannot be held back; God cannot be forced).54 Presumably in order to bring Num 23:19aβ better in line with other biblical understandings of God, in which God does regret or relent (cf., e.g., Gen 6:6 and Jonah 3:10), G translated the text inappropriately, even though its Vorlage probably consisted of the same consonants as M.55 The Greek translation prevents an obvious contradiction to other texts and thus attests a distinct understanding, not a variant Vorlage. In Num 23:20, “[t]he great variety of small divergences in the exegesis of this passage is due to textual difficulties. The affinity between the Fragmentary Targum, the Peshitta, and the Vulgate is worth noting.”56 A ‫ ל‬is missing in 23:20a M before ‫ברך‬, but recorded in Smr and S (and possibly in G). The Masoretic text should be emended to match these other witnesses. Further, 20b should be read in the first person with Smr and G. The (proto-)Masoretic tradition changed the text to preclude Balaam’s blessing of Israel. This fits with the negative evaluation of him that developed in Judaism and Christianity.57 The reading in the third person represents a theological reevaluation of Balaam, whom M records saying, “He has blessed and I will not take it back”, instead of the more intentional (from Balaam’s perspective), “I have blessed and will not take it back”. The Masoretic Text absolutely affirms that God – and not Balaam – blesses. Again, in Num 23:21 G translates ‫“ און‬sin” in an atypical fashion. The Septuagint regularly uses μόχθος as a translation for ‫( עמל‬Deut 26:7 and Qohelet [21x!]). Otherwise, μόχθος is used as a translation for ‫( תלאה‬Exod 18:8; Num 20:14; Neh 9:32; Lam 3:65), ‫( יגיע‬Isa 55:2; Jer 3:24; Ezek 23:29), or ‫( פרך‬Lev 25:43, 46, 53; Ezek 34:4). Other unique equivalents, such as the present case, include ‫( פעלה‬Isa 61:8) and ‫( חמס‬Jer 28:35). Still, the majority of usages equate μόχθος with ‫עמל‬. Note however, the usage of ‫ עמל‬later in the verse. This second occurrence explains the curious translation in G-Num 23:21a, which does not imply a variant Vorlage from M or Smr. The translator simply reversed the order. 53

Vermes, Scripture, 151. Cf., e.g., Wilhelm Rudolph, “Zum Text des Buches Numeri,” ZAW 52 (1934): 119. 55 Cf. Rösel, “Jakob,” 165. 56 Vermes, Scripture, 152. 57 Cf., e.g., Num 31:16; Josephus, Ant. 4.104–131; 2 Pet 2:15–16; and Rev 2:14. Cf. also Vermes, Scripture, 173–77 and Chapter Five, below. 54

2.2 Textual Criticism

83

Finally, in 23:21b metathesis stood behind the misreading in M and Smr. The text should read ‫ ותורעת‬instead of ‫ותרועת‬: the glorious deeds of kings are with Israel, not the noise of kings.58 The possibility of an intentional change should not be completely ignored: the ‫ ותרועת‬of M reminds the reader of Amos 2:2: Moab shall die in an uproar (‫)שׁאון‬,59 with shouting (‫)ותרועה‬, and the voice of the trumpet. One can hardly overlook the commonalities in the expectations of these texts regarding Moab. It is worth noting that only M maintains these similarities.

2.2.7 Num 24:6–7 In Balaam’s third oracle, it is impossible to reconstruct a variant Vorlage for G in Num 24:6, even though its understanding of the verse clearly differs from M. The Greek translator(s) of Numbers otherwise used σκίαζω in the sense of the cloud of God overshadowing the tabernacle (cf. G-Num 9:18, 22; 10:36). The sense of expanding shade connected to the root is not unfamiliar to the Hebrew Bible, though exceptionally rare: in Jer 6:4 the Niphal form of ‫ נטה‬is attested in the sense of the evening shadows lengthening.60 It is possible that G understood its Vorlage, which in this case would have matched M, in this way. Consequently, there is no need to emend the text.61 The reading of ‫ נטה‬for ‫ נטע‬in G, Smr, and V should be accepted in 24:6b and the text of M changed. 4QNumb attests this reading. Changing the verb also means that the object should be changed from “aloe-wood” to “tents”, a matter of merely repointing the text and thus not strictly a text-critical undertaking.62 The scribes responsible for the M reading presumably changed the text to conform to the idyllic similes of the rest of the verse (vales, gardens, and cedars). Here, M represents a later interpretative reading than G, Smr, Q and V.63 In Num 24:7a, G does not appear at first glance to be a translation of a text similar to M. Many commentators attribute these differences to translational 58 Though the translation of “kings” with ἄρχων is unusual, it is not unheard of. According to Takamitsu Muraoka, A Greek-English Lexicon of the Septuagint (Louvain; Walpole, MA: Peeters, 2009), 70 it occurs seven times in G. The translator probably felt this lemma was contextually more appropriate, as the kingdom of Israel had not come into existence yet in the narrative; cf. Wevers, Notes on the Greek Text of Numbers, 405. Alternatively, one should not entirely dismiss this as evidence of an anti-monarchic tendency in G. Rösel, “Jakob,” 166 believed the plural suggests that God is king. 59 Cf. the parallel to Num 24:17 in JerM 48:45. 60 Cf. Vermes, Scripture, 158. 61 Cf. Dorival, Nombres, 139 and Ziegert, Diaspora, 249–50. 62 Contra Seebass, Numeri III, 22–23. 63 Cf. Wevers, Notes on the Greek Text of Numbers, 404–5, who seemed somewhat unsure about G being the older reading, even though Q supports it unequivocally. Contra Milgrom, Numbers, 204, who retained “aloes”, but identified it as a wordplay with “tents”.

84

The Text of Numbers 22–24

interpretation, a loose translational style from those behind the Greek tradition.64 The Septuagint offers a variant, yet plausible text in 24:7a, but generally matches M in 24:7b. I would argue that proto-M et al. was most likely corrupted and subsequently emended after the translation of G. A number of typical scribal errors could account for the M reading developing from the G-Vorlage, but not the alternative. The G-Vorlage must have read something like: ‫יאזל אישׁ מזרעו ומשׁל בעמים רבים‬.65 It is especially conspicuous that the first and last words are virtually identical in M and my reconstructed G-Vorlage. But can this text be related to M? We will proceed word for word. Either the translator of G misinterpreted ‫ יזל‬as coming from ‫אזל‬, or the textual tradition of M (and Smr) lost an ‫ א‬over the course of transmission and transformed ‫ יאזל‬into ‫יזל‬.66 It is possible that the G-Vorlage matched M in this instance and read ‫ יזל‬or ‫יאזל‬. It is impossible to regard the following four words in M as a possible Vorlage for G; there is no way that ‫ מים מדליו וזרעו במים‬can be translated as “man”, “his seed”, “and he will reign”, and “over people” respectively. Either one of

64 Cf., e.g., Arie van der Kooij, “Perspectives on the Study of the Septuagint. Who Are the Translators?” in Perspectives in the Study of the Old Testament and Early Judaism. A Symposium in Honour of Adam S. Van der Woude on the Occassion of His 70th Birthday, ed. Florentino García Martínez and Ed Noort, VTS (Leiden; Boston; Cologne: Brill, 1998), 227: “It may well be that specific ‘readings’ in cases where the same Hebrew text was shared, varied in different circles or milieus within Judaism because of a different interpretation.” Against such interpretative postulations, cf. Ziegert, Diaspora, 251: “Der Übersetzer hat seine Vorlage [in Num 24,7] wohl im Sinne eschatalogischer und messianischer Aussagen verstanden, hat aber wahrscheinlich keine eigene und neue Interpretation in den Text hineingetragen.e [sic]” 65 Cf. Julius A. Bewer, “The Literary Problems of the Balaam Story in Numb., Chaps. 22–24,” AmJT 9, no. 2 (April 1905): 260, who proposed precisely this translation. He even suggested that the Aramaic ‫ אזל‬was chosen specifically to make Balaam sound foreign. This is contra Eugene Ulrich, The Biblical Qumran Scrolls: Transcriptions and Textual Variants. Volume 1: Genesis–Kings (Leiden/Boston: Brill, 2013), 155, who reconstructed the text of the G-Vorlage as ‫יצא איש מזרעו וזרועו בעמים רביב‬. This reconstruction is curious in that κυριεύω is under no circumstances an appropriate translation of √‫זרע‬. Either the editor presumed the last three words of the verse must match M or he himself fell victim to dittography, writing √‫ זרע‬twice when it only should have been recorded once. It should be noted that ‫אדם‬ presents a plausible alternative to ‫ אישׁ‬in this reconstruction. 66 Cf. Johan Lust, “The Greek Version of Balaam’s Third and Fourth Oracles. The ἄνθροπος in Num 24,7 and 17: Messianism and Lexicography,” in VIII Congress of the International Organization for Septuagint and Cognate Studies. Paris 1992, ed. L. Greenspoon and O. Munnich, SBLSCS, vol. 41 (Atlanta: Scholars Press, 1995), 236. To the former, cf. Rösel and Schlund, “Arithmoi,” 490–91. To the latter, one should note that quiescent ‫ א‬often elided in Aramaic and later texts. One possible example of such a phenomenon with precisely this verb is Jer 2:36. For another example of a missing ‫ א‬in this pericope, cf. Num 24:24; cf. Ziegert, Diaspora, 252–53.

2.2 Textual Criticism

85

the texts – proto-M or the G-Vorlage – must have been corrupted (the G-Vorlage before its translation into Greek or proto-M et al. thereafter), one of them was intentionally changed, or the G-translator took an extensive level of liberty with the text. The second word in the Vorlage of 24:7a, ‫ אישׁ‬or perhaps ‫( אדם‬cf. Num 23:19), could have been the word behind a variant reading in M, namely ‫מים‬.67 A scribe at some point confused ‫ ש‬and ‫מ‬, a mistake that is not unlikely in paleoHebrew (v vs. m), particularly when one notes the subsequent ‫ מ‬beginning the next lexeme, whether in M or my reconstructed Vorlage. N.b. also the ‫ מים‬at the conclusion of the preceding verse. Then ‫ א‬was intentionally changed from an original ‫ מ‬in order to make sense of the word. These errors and emendations best explain the subsequent transition from ‫ יאזל‬to ‫יזל‬: it was changed in order to make the verb better match the new corrupted and edited subject. In this context, one must note that the verb ‫ נזל‬does not normally refer to pouring or flowing from artifacts, but rather natural or supernatural sources, often with theological and soteriological undertones.68 The only non-metaphorical usage beside this one that could be understood as presenting flowing from a source made by humans is Prov 5:15, which refers to a well. However, this usage is by no means clear: the well could also be natural. At any rate, it seems an unlikely choice to describe an action undertaken with “buckets”. Considering the buckets, one notes further that the lexeme ‫ דלי‬is only attested twice in the Hebrew Bible, the other occasion being Isa 40:15.69 Leaving the ending aside for a moment, one notes that this third word, ‫מדליו‬, shares the first and third consonant with the reconstructed fourth word of G, namely ‫משׁל‬.70 It is possible that this was emended or transformed due to a scribal error, 67 William Horbury, “Monarchy and Messianism in the Greek Pentateuch,” in The Septuagint and Messianism, ed. Michael A. Knibb (Leuven: University Press, 2006), 122 regarded the later common reading between the Jewish recensions (Aquila, Symmachus, and Theodotion) and M as evidence of M’s priority. However, the Jewish recensions could just as easily attest a later proto-M recension away from a version like the G Vorlage. 68 Cf. Exod 15:8; Deut 32:2; (Judg 5:5;) Job 36:28; Ps 78:16, 44; 147:18; Song 4:15–16; Isa 44:3; 45:8; 48:21; Jer 9:17; 18:14. 69 Others have tried to trace this back to a corrupted form of “from his branches” (‫)מדליותיו‬. This root was then interpreted in light of other texts in order to achieve a satisfactory meaning for this oracle. Cf. Milgrom, Numbers, 204 and Levine, Numbers 21–36, 197, who connected Num 24:7 with Ezek 17:7 and Jer 19:13. Vermes took a slightly different route. “This disconcerting exegesis results from the following midrashic associations. ‫יזל‬ ‫ מים‬recalls ‫( יזלו צדק‬Is. xlv. 8), where righteousness is symbolically expressed as water. Cf. Am. v. 24. Also, ‫ צדק‬is associated with the Messiah from Jeremiah xxxiii. 15 and xxiii 5.” (Vermes, Scripture, 159) “In short, water = righteousness = Messiah.” (Vermes, Scripture, 160) All of these solutions could be indicative of later recensional work within the Masoretic tradition. The Septuagint does not connect these texts in a similar fashion. 70 ‫ ב‬+ ‫ משׁל‬is often translated with κυριεύω; cf., e.g., Gen 3:16; 37:8; Judg 9:2; 14:4; 15:11B; 2 Chron 20:6; PsG 105:41; Isa 3:4, 12; 19:4; and Lam 5:8.

86

The Text of Numbers 22–24

though no particular errors readily come to mind. The most likely reconstruction would have to seek recourse to the paleo-Hebrew alphabet: the original reading was lvmw, which became lmw (haplography), then lwm (metathesis), before becoming ldm (confusion), and ultimately being edited to wyldm, the current reading in M. The number of errors required for this reconstruction is high, but not impossible. Add to these reflections the fact that the third word in the reconstructed G-Vorlage is identical to the fourth word in M, and the situation begins to become more clear. It is entirely possible that the third and fourth words were reversed in the Hebrew textual tradition, which led to some restructuring of the verse. At this point it is possible that “and he will rule” (‫ )ומשׁל‬became “from his buckets” (‫)מדליו‬. The final difference is the missing ‫ ע‬in M’s fifth word, which would easily have occurred at any stage in the transmission, presumably after the transition to “water” as the subject of the first clause. Thus, the following corruption of this verse can be postulated: ‫יזל מים מדליו וזרעו במים רבים → יאזל אישׁ מזרעו ומשׁל בעמים רבים‬ As demonstrated here, this reconstruction requires the change or deletion of six consonants and the reversal of two words. The process of transition probably began as a result of a scribal error that ultimately manifested into an intentional changing of the text to achieve as much sense from the newly corrupted text as possible. Possibly this variance took place under the influence of Isa 40:15 M, the only other occurrence of the lexeme ‫ דלי‬in the Hebrew Bible. But what relates Num 24:7 to Isa 40:15? In Isaiah the bucket symbolizes YHWH’s greatness visà-vis the nations; one can hardly miss the theological undercurrent. The postulation of a Vorlage for G that was distinct from M seems preferable and methodologically sounder than making the translator responsible for all of these variations, particularly when many other cases demonstrate the translator’s careful and generally precise technique.71 From another perspective: what would the translator of G gain by making these changes with no regard for his or her Vorlage? The standard answer in scholarship is, to paraphrase, “he is making the text messianic or eschatological”.72 Yet, one must question to what extent this verse is eschatological or

71 Contra Wevers, Notes on the Greek Text of Numbers, 406, who denied the possibility of distinct parent texts. 72 Cf., e.g., Marilyn Frances Collins, Messianic Interpretation of the Balaam Oracles, University Microfilms International Dissertation Service (Ann Arbor, Mich., 1990), 166; Gilles Dorival, Marguerite Harl, and Olivier Munnich, La Bible grecque des Septante. Du judaïsme hellénistique au christianisme ancien, Initiations au christianisme ancien (Paris: Cerf, 1988), 288; Vermes, Scripture, 59–60 and 159–66; Dieter Vetter, Seherspruch und Segensschilderung. Ausdrucksabsichten und sprachliche Verwirklichungen in den Bileam-

2.2 Textual Criticism

87

messianic in G. Honestly, I don’t see it. G merely states that a person will come from the seed of Jacob and rule many nations. Rather, “…die großen Unterschiede, die der griechische Text von V.7 [in Numeri 24] im Vergleich zu 𝔐 bietet, [lassen sich] eher durch textliche Phänomene auf der Ebene der Vorlage erklären als durch eine bewusst vorgenommene messianische Neuinterpretation des Übersetzers.”73

One need not look to the eschaton to fulfill this prophecy; one need look no further than David, at least no further than David as envisioned in Chronicles and the Deuteronomistic History.74 “The translator does not use such words as ‘king’, or ‘anointed’ [in Num 24,7.17]. But he does take the passage to refer to a future savior figure, who has a kingdom.”75 While this may be true, the question remains unresolved whether the translator regarded this person as still belonging in his future, or whether he translated consistently with the internal logic of his text. Rather than presume that this text presupposes some eschatological messiah or the end of time, which it does not explicitly do in G or any other version, we could consider looking for the fulfillment of this text within the confines of the literature we have available to us.76 This matter will be dealt with in the next chapter, under the rubric of literary criticism. Sprüchen von Numeri 23 und 24, Calwer Theologische Monographien, vol. 4 (Stuttgart: Calwer, 1974), 39; Rouillard, Balaam, 363–74, 415–66; Seebass, Numeri III, 23; and Rösel and Schlund, “Arithmoi,” 490–91. 73 Ziegert, Diaspora, 248. 74 Cf., e.g., 1 Chr 18:1–13 // 2 Sam 8:1–14. N.b. also the application of √‫ משׁל‬in regard to David and Solomon in 2 Sam 23:3 and 1 Kgs 5:1. They present the only monarchs described in this manner. 75 John J. Collins, “Messianism and Exegetical Tradition. The Evidence of the LXX Pentateuch,” in The Septuagint and Messianism, ed. Michael A. Knibb (Leuven: University Press, 2006), 146. 76 “Is the Greek translation more messianic than the MT? Using the star and the sceptre as symbols the Hebrew text clearly foretells the coming of a new king in Israel. The victories over the enemy in verse 17, and especially over Edom in verse 18, call to mind the reign of David, or an eschatological messianic king to be compared with David as in Amos 9,11–12. The Septuagint replaces the sceptre symbol by the vague term ἄνθρωπος. This appears to do away with the royal character of the expected figure. The contrary can be held only when one can demonstrate that the translator used the term ἄνθρωπος as a messianic title.” (Johan Lust, “The Greek Version of Balaam’s Third and Fourth Oracles. The ἄνθροπος in Num 24,7 and 17: Messianism and Lexicography,” in Messianism and the Septuagint: Collected Essays by Johan Lust, ed. Katrin Hauspie, BETL, vol. 178 [Leuven: Peeters, 2004], 76 = Lust, “Greek Version [1995],” 241.) Cf. Lust, “Messianism and the Septuagint, Pentateuch,” 149 = Lust, “Messianism and the Septuagint, Pentateuch,” 44. Also, “In the LXX in general, ἄνθρωπος is a more neutral term without messianic connotations. It often simply means ‘someone’. [...] Philo’s use of the term in Num 24,7 confirms this. Indeed this author avoids clearcut messianic notions. […] ‘Man’ for him was ‘mankind’. At the end of the days there would be peace. ‘Man’ was going to subdue the world and fulfil the task for which he was created. This eschatological ‘man’ corresponds to the primeval ‘man’. […] The Christian

88

The Text of Numbers 22–24

The only other difference between M and G in 24:7 that mandates a textcritical evaluation is the distinction between ‫ אגג‬in M and Γωγ in G (and Smr).77 The relative insignificance of the character Agag (1 Samuel 15) suggests this as the older reading vis-à-vis the eschatological (in this case legitimately so) character Gog (cf., e.g. Ezekiel 38–39).78 The reference to Gog also moves in the direction of divine kingship away from human kingship, as the one who shall defeat Gog in Ezekiel 38–39 is none other than God.79 At a later stage of textual transmission, someone probably changed the text of G and Smr away from Agag to Gog.80 Further, G presumably apocopated ‫ מלכותו‬to ‫מלכו‬, probably under the influence of the preceding form. The comparison only makes sense if Agag (or Gog for that matter) is compared to a king and not a kingdom.81 Thus it appears that G presents the older text for v. 24:7a, but that M provides it for 24:7b.

2.2.8 Num 24:17–24 The final text-critical matter that must be handled consists of essentially the entire fourth oracle, Num 24:15–24, but here I will concentrate on the major variants in 17–24. The various traditions differ – sometimes dramatically – here, such that it may remain impossible to achieve any kind of text that could authors knew Philo. Origen in Alexandria certainly did. The Christian Fathers may have accepted Philo’s reading and inserted it into the Septuagint. On the other hand it is not excluded that the original version of the Septuagint, based on a vision similar to Philo’s, already had it.” (Lust, “Greek Version [2004],” 85 = Lust, “Greek Version [1995],” 250) 77 Cf. Vermes, Scripture, 161. Seebass, Numeri III, 23 incorrectly failed to consider Smr, and his position must be rejected. Cf. Ziegert, Diaspora, 256. 78 No reason exists to regard Agag as the general title for Amalekite kings, as Carl Friedrich Keil, Leviticus, Numeri und Deuteronomium, Biblischer Commentar über die Bücher Mose’s (Leipzig: Dörffling & Franke, 1870), 335 did. 79 This distinction could even imply a Masoretic reinterpretation of Num 24:7 in light of texts like Isa 41:21 and 44:6, which stress that God is Israel’s king. The same cannot be said of the G-reading of this passage. 80 Cf. Wevers, Notes on the Greek Text of Numbers, 405; Rösel, “Textüberlieferung,” 222; and Milgrom, Numbers, 204. Contra, e.g., Gall, “Zusammensetzung und Herkunft,” 35–36. There is no need to emend the text based on 1 Chr 14:2, as Lust proposed: “Both the Hebrew text and the translation of 1 Chron 14,2 probably imply an allusion to the oracle of Balaam. The author’s choice of the term ‫‘ למעלה‬highly’ probably suggests that he understood or read ‫ מאגג‬in Num 24,7 as ‫מהגג‬.” (Lust, “Greek Version [2004],” 73–74 = Lust, “Greek Version [1995],” 237–38.) Horbury, “Monarchy and Messianism,” 108 followed Lust’s interpretation. 81 Contra Seebass, Numeri III, 23, who saw this comparison as a logical progression. The other alternative would be to postulate that the Vorlage of G originally read something like ‫“( וירם ממגג מלכותו‬his kingdom will supersede Magog”), which left no traces in the Greek or Hebrew traditions, however. This lack of witnesses makes it unlikely as the oldest attainable reading.

2.2 Textual Criticism

89

be responsible for all of them. This fact more than any other demonstrates that the biblical text continued to be updated beyond what could be called an Endtext, even into the period of textual transmission. It may be possible, however, to elucidate the historical contexts presupposed by some of these variants. These will be dealt with in the following. In Num 24:17a, the translator of G apparently understood the form ‫ אראנו‬as the Hiphil imperfect of ‫ ראה‬and ‫ אשׁורנו‬as a first-person Piʿel imperfect form from ‫אשׁר‬, presumably ‫ אשׁרנו‬or ‫ אשׁרהו‬or something similar.82 Since the first variant (‫ )ראה‬would only require a different vocalization, it is not strictly a text-critical matter. Similarly in the second case, only plene vs. defective forms define the distinction. It would be understandable that M was changed toward the sense of 23:9 in both of these cases, an argument that cannot be made for G.83 Further evidence supports M’s understanding being secondary: the form ‫ אשׁורנו‬appears only in Num 23:9 and 24:17, increasing the likelihood that one text influenced the other. However, G translates the two forms differently. M more likely adjusted 24:17 to be more consistent with 23:9 in which M and G still match.84 To make sure that no one interpreted 24:17 with G or its Vorlage again, someone in the proto-Masoretic tradition presumably added the ‫ ו‬to the middle of the lexeme intentionally, though it could have been an accident. At the latest, the Masoretes precluded anyone going back to the version found in the Septuagint by noting in the Masorah parva that the form ‫שׁוּרנּוּ‬ ֶ ‫ ֲא‬occurs only in two instances, here at 24:17 and in the preceding 23:9. This could suggest that the Masoretes included this note to secure the text precisely against the version found in the Septuagint. Regardless, it remains indisputably more likely that someone emended the proto-M tradition for consistency in this case in 24:17 than that the generally competent and consistent Septuagint translator of Numbers provided two translations for an identical form within a few verses, a form that only appears in these two instances. Thus, proto-M or M most likely is responsible for the variants in these cases, whereas G preserves an older understanding in the first case and an older reading in the second. M also presents 82

This should be reconstructed with exceptional caution, since √‫ אשׁר‬is not attested in the first person; cf. Gen 30:13; Mal 3:12, 15; Ps 41:3; 72:17; Prov 3:18; 31:28; Job 29:11; Song 6:9 and Sir 11:28; 34:9; and 37:24 for every biblical usage of this verb. A cognate noun, “joy”, is known from Punic; cf. KAI 145:11. 83 Note Wevers, Notes on the Greek Text of Numbers, 412: “…I would conclude that he [the translator of G] read the lines here [Num 24:17] to contrast with 23:9.” Wevers’ interpretation insisted that the translator must be the responsible party without even considering the possibility that M might have changed the text for the sake of consistency. The Masorah parva, that this form only occurs twice in the Hebrew Bible, implies that the Masoretes recognized this form as unusual and perhaps sought to preclude precisely the reading affirmed in G. 84 Contra Leiter, “Assimilation and Dissimilation,” 84, who believed that the G Vorlage presented identical forms in these cases.

90

The Text of Numbers 22–24

a better parallelism and the explicit connection to 23:9 could be seen as strengthening the de-personalized understanding of this verse. Numbers 23:9 mentions a people (‫ )עם‬and 24:17 a tribe (‫)שׁבט‬. There are manifold problems in 24:17b.85 I will begin with 17bβ. Here, G, S, and V read “is rising” or “will rise”. The BHS offers ‫ זרח‬as a possible translation back from the Greek,86 since ἀνατέλλειν is a common translation for ‫זרח‬ in G; cf., e.g., Gen 32:32 and Judg 9:33. Such a transition could have been the result of a simple error. The confusion (in either direction) could be best explained aurally; cf. Aram. ‫ דהב‬and Heb. ‫*( זהב‬z < *d). G also apparently understood the verb in the future tense, not an obvious translation of the Hebrew perfect, suggesting that a variant should be identified here: M = ‫ ;דרך‬G-Vorlage = ‫יזרח‬. The New Testament apparently implies G’s reading; cf. Matt 2:2: ἐν τῇ ἀνατολῇ, which suggests that G’s reading was known in the first century CE and could not be a later recensional element.87 One must also note that M often uses ‫ דרך‬as a verb to emphasize God’s power vis-à-vis the nations and that it can often take on a soteriological aspect.88 The same cannot be said of G’s ἀνατέλλειν. The “king” (‫ )מלכא‬who should come out of Jacob according to the Targums is probably a later interpretation, as is the Targums’ reading of “messiah” (‫ )משׁיחא‬for ‫( שׁבט‬or whatever word appeared there are the time; see below) in the following clause. This interpretation is clearly messianic, a quality that commentators often ascribe to G (particularly in this verse).89 The explicitly messianic readings of the Targums should serve to differentiate their reading from the non-messianic, or at best implicitly messianic reading of G. Here, one might also mention that the text of M was understood messianically at Qumran (e.g., in 4Q175 = 4QTestimonia and CD VII, 18–21).90 Even moderns understand it this way.91 85 Vetter, Seherspruch, 44 proposed reading the present tense here, which neither M nor G really merits. 86 As suggested by Wellhausen, Composition des Hexateuchs, 351. 87 NT manuscripts universally attest the form in Matt 2:2. 88 Cf. the comments and attestations in Jonathan Miles Robker, “Bileam messianisch gelesen?” in Textgeschichte und Theologie, ed. Frank Ueberschaer, Jonathan Miles Robker, and Thomas Wagner (Tübingen: Mohr Siebeck, 2018). 89 Cf., e.g., Zacharias Frankel, Einfluss, 183; M.F. Collins, Messianic Interpretation, 166; Dorival, Harl, and Munnich, La Bible grecque des Septante, 288; Vermes, Scripture, 59–60 and 159–66; Vetter, Seherspruch, 39; and Rouillard, Balaam, 363–74, 415–66. 90 Cf. Gall, “Zusammensetzung und Herkunft,” 36–38, who argued the messianic character of M, logically without any recourse to its use at Qumran. Cf. also Robker, “Bileam messianisch gelesen?”. 91 Cf., e.g., Gilmore H. Guyot, “The Prophecy of Balaam,” CBQ 2, no. 4 (October 1940): 335, “In this prophecy to Israel is promised a king clothed with splendor (star) and endowed with authority (sceptre); this king will destroy the enemies of Israel, Moab and Edom; he will rule over his enemies.”

2.2 Textual Criticism

91

That being said, there is no way that G’s ἄνθροπος can be a translation of the Hebrew ‫שׁבט‬.92 Either G had a Vorlage distinct from proto-M, or the translator drastically and dramatically changed the sense of the text. In my opinion, three matters suggest that G represents a legitimately alternative reading and not merely a translational interpretation. First of all, Philo attests the G reading on two occasions (Mos. 1.290 and Praem. 95).93 Second, several ancient textual witnesses attest personalities to whom this text is referring: ‫ משׁיחא‬in the Targums, ‫( ܪ‬ryš’) in S, and dux in Irenaeus (= La); these could all be further interpretations of who this “person” in G or its Vorlage was, but are hardly probable reflections of Hebrew ‫שׁבט‬.94 The Vulgate, on the other hand, does reflect the Hebrew with its reading virga, a reading supported by Symmachus (σκῆπτρον). These circumstances at least make it possible that prior to Symmachus and the translation of the Vulgate, there were Hebrew manuscripts that reflected a reading of a person in Num 24:17. Translations outside of the Greek tradition as well as the NT made interpretative judgments about who this person was: the messiah (T), a prince (S and La), or the “king of the Jews” (βασιλεὺς τῶν Ἰουδαίων; Matt 2:1–2). 95 Another Hebrew tradition (probably later than that responsible for the G-Vorlage) depersonalized this prophecy and understood it as referring to a whole tribe and not one figure undertaking a military campaign against the Moabites. This in turn reminds the reader of Num 23:9 (“…a people alone…”), suggesting M represents a later revision in 24:17 to maintain a better internal consistency with 23:9. None of the other witnesses reflect this.96 These observations about the potential depersonalization of the prophecy in favor of a tribe brings us to the third point favoring G representing an old Hebrew Vorlage and not being merely a translational interpolation: M offers a brilliant pun in the form of the scepter that will strike Moab, being at the same 92 Contra Wevers, Notes on the Greek Text of Numbers, 413, who found this substitution “not really far-fetched” and remained readily convinced of the interpretive agility of the translator of G. This is contra his opinion elsewhere, in which he noted that, “Num […] preferred stereotypes and seldom created fresh ways of rendering materials into Greek.” (John William Wevers, “The Göttingen Pentateuch: Some Post-Partem Reflections,” in VII Congress of the International Organization for Septuagint and Cognate Studies, Leuven 1989, ed. Claude E. Cox [Atlanta, GA: Scholars Press, 1991], 59.) He also identified this translation as anti-monarchical: “man” instead of “scepter”. 93 Origen also followed it in In Numeros Homilia 13,7. 94 One possible explanation for an error could be found in the paleo-Hebrew script: the paleo-Hebrew writing of ‫( אישׁ‬vya) is not that different from ‫( ראשׁ‬sar). The similarity is even closer between ‫( אדם‬mda) and ‫( ראשׁ‬var). It is quite possible that such a confusion stood behind these variants. Nonetheless, the priority would remain with G. 95 The S and La reading might also support G: it would be an understandable scribal error to mistakenly read ‫ ראשׁ‬for ‫אישׁ‬: ‫ראשׁ → יאשׁ → אישׁ‬. A similar argument could be made for ‫ אדם‬as the oldest reading. 96 Contra Vermes, Scripture, 165.

92

The Text of Numbers 22–24

time a tribe that will strike Moab. It is highly improbable that the translator found this pun and willfully destroyed it, even if he or she could not reproduce it in the target language.97 What kind of translator would do that? One could argue, alternatively, that the translator did not understand what ‫ שׁבט‬meant, but the translator of Numbers translated it accurately in every other occurrence in Numbers with a form of φυλή (Num 4:18; 18:2; 24:2 [!]; 32:33; and 36:3). In this case, M’s use of “tribe” would imply a direct reflection about Balaam’s viewing Israel “according to their tribes” in Num 24:2. The transition is apparently from one expected person – whoever that might be – to a whole tribe. The reading in G is by no means eschatologically “messianic”; rather it seems to anticipate a specific situation in which Moab and Edom were conquered or destroyed. Some have suggested that the translational interpretation of ἄνθροπος for ‫ שׁבט‬reflects a connection to Gen 49:10, a verse in which we encounter a similar understanding. However, this solution honestly presents more problems than any resolution. M G V S

Gen 49:10 ‫שׁבט‬ ἄρχων sceptrum (sbṭ’)

Num 24:17 ‫שׁבט‬ ἄνθροπος virga ‫( ܪ‬ryš’)

Table 9: Staffs in Gen 49:10 + Num 24:17

Among all of the alternatives, only one textual tradition maintains consistency between these verses: M. All of the other witnesses either knew other readings or failed to harmonize these two texts.98 Though it is not impossible, this evidence – combined with the later witness of, e.g., Symmachus – suggests that the most likely tradition to have undergone later editing, if any, was M. The Masoretic Text undoubtedly presents the most consistent and thus presumably latest witness in these instances. It is at least as likely as the alternative that later readers and copyists in the Hebrew tradition might want to depersonalize this prophecy so that it could not be understood as referencing a specific messiah, which the Targums obviously had. Judaism in the Roman period had enough difficulties with messianic figures that it would be understandable that they might consistently seek to avoid understanding the text in a personalized way. One need not consider this a 97

See the etiological examples in Ausloos, “Septuagint’s Rendering”, in which the translator showed particular affinity in this regard. Cf., e.g., Num 11:1–3 in G. 98 This situation would be particularly acute if we presume that G-Numbers knew G-Genesis. In that case, the translator of Numbers must have explicitly chosen not to reflect the text of G-Gen 49:9!

2.2 Textual Criticism

93

change to avoid Christian interpretation – the readings at Qumran preclude this evaluation; one need consider the possibility of anti-monarchic tendencies during the Hasmonean era, as suggested by the Qumran readings. By the end of the later Bar Kochba revolt in the second century CE the proto-Masoretic text probably prevailed as the exclusive Hebrew version. Rather than focus on any single person being responsible for the destruction of Moab and Edom, as anticipated in G, T, S, and La, later recensors adapted the understanding to a “tribe” – Judah would be the most likely candidate (cf. precisely Gen 49:10) – and created a delicious wordplay at the same time: that a tribe is the blunt instrument that will bash in the skulls of the Moabites. Such an interpretation is at least as likely as the alternative and will be favored here. Again, it is possible that G and many other witnesses were familiar with a different Vorlage than proto-M in the last two phrases of 24:17, though this Vorlage is difficult to reconstruct. The first verb, ‫מחץ‬, could have been the same or another related verb; this cannot be assessed with any certainty. It again seems likely that G (here with the Targums, S, and V) reads a noun related to a person instead of ‫פאה‬: G reads ἀρχηγός, which is most often used as a translation for the Hebrew ‫ראשׁ‬.99 This probably presents the older reading. M’s second verb, ‫וקרקר‬, is obviously a mistaken reading for ‫וקדקד‬, as recorded in, e.g., Smr. 100 The Septuagint records προνομεύσει, from the root προνομεύω, for this verb. This Greek root most often translates ‫ בזז‬or ‫שׁבה‬, neither of which could be reconstructed here by postulating a readily identifiable, commonly attested scribal error. Most likely, G translated ‫ קדקד‬in this manner here to create consistency with Num 31:9, 32, and 53. Further, since Num 24:17bβ generally seems to match Jer 48:45bβ, it seems probable that the two texts are related. The most likely scenario is that G had a different Vorlage and that M’s version was later adapted to be more consistent with Jer 48:45. This adaptation must have taken place very late in the transmission of the text, since Jer 48:45 does not even exist in G.101 The possibility also should not be excluded that Num 24:17 M was redacted to better match Ps 68:22, a text in which anticipates God’s militantly striking God’s own enemies down. 99

Cf. Muraoka, Lexicon, 69 for the attestations. Such a meaning for ‫ קרקר‬is otherwise insecure or even unattested; cf. Wilhelm Gesenius, et al., Hebräisches und Aramäisches Handwörterbuch Über das Alte Testament: Gesamtausgabe, 18 (Berlin; Heidelberg: Springer, 2013), 1196. The Damascus Document VII, 20–21 also attests the mistaken reading. 101 Nonetheless, Achenbach, Vollendung, 422–23 regarded the relationship between Jer 45:48 M and Num 24:17 as both strong and with the texts coming originally from Jeremiah 48. This hardly seems likely, since that text in Jeremiah is missing in G, apparently evincing its late addition to the text. Cf. Pierre-Maurice Bogaert, “Heshbon entre Moab et Ammon: La finale ajoutée à l’oracle sur Moab en Jr 48,45–47 TM,” in Interpreting Translation: Studies on the LXX and Ezekiel in Honour of Johan Lust, ed. Florentino García Martínez and Marc Vervenne, BETL, vol. 192 (Leuven: University Press, 2005), 54. 100

94

The Text of Numbers 22–24

This lexical overlap presumably explains the presence of “his enemies” (‫ )איביו‬found in both Ps 68:22 and Num 24:18a M. Psalm 68:22 precludes entirely the need for a human interlocutor, which could plausibly parallel M’s tendency to depersonalize Balaam’s fourth oracle (see above). Therefore, with caution, we can construct the Vorlage of G in Num 24:17bβ as: ‫ומחץ ראשׁי‬ ‫מואב וקדקד כל־בני־שׁת‬, though it must be stressed that there are several other possible reconstructable permutations. This text (or whatever the G-Vorlage precisely looked like) was reworked in the proto-Masoretic or Masoretic tradition to resemble Jer 48:45, a very late text in Jeremiah M, and Ps 68:22. The Septuagint of Num 24:17bβ is closer to Ps 68:22, whereas M is closer to Jer 48:45. Without covering the matter in too much detail here, it appears probable that scribes redacted a network of texts at or about the same time in order to refocus some theological and messianic tendencies. They will be noted briefly here. The following texts demonstrate some relation to Num 24:17M: Gen 49:10M; 2 Sam 7:7, 14; Isa 11:1–4; JerM 10:16; 48:45; 51:19; Hab 3:13; PsM 68:22; 74:2; 78:67–68; 89:33; 110:5–6. What is most striking in many of these texts is their parallel nomenclature (most especially regarding the term ‫ )שׁבט‬and similar theology, found most often only in the Masoretic version; cf. especially Gen 49:10 and the various verses in Jeremiah.

In Num 24:18 ‫ יְ ֵר ָשׁה‬should be repointed as ‫ יְ ֻר ָשּׁה‬in both occurrences, as suggested by the Masoretic notation that this form of the word occurs only twice. Both appear in this verse. The standard form should be reconstructed, better matching Smr and S. However, this is not technically a text-critical matter, since the consonantal text remains the same; only the errant Masoretic pointing must be adapted. The G translation of ‫ – יְ ֵר ָשׁה‬or rather ‫ – יְ ֻר ָשּׁה‬with κληρονομία in 24:18 is unusual; in G-Num κληρονομία is the standard equivalent for ‫( נחלה‬cf., e.g., Num 18:20; 26:54; 36:4, 7, 9, and 12). On the one hand, this variance could be the product of the translator, as no plausible mistaken reading can be easily reconstructed, regardless of whether one would postulate proto-M or a variant G-Vorlage as the older reading. Further evidence for this can be found in G’s occasional translation of the verb ‫ ירשׁ‬with κληρονομέω (Cf. Num [13:30;] 14:24; 21:35; and 27:11). On the other hand, this potential distinction between the G-Vorlage and M could evince late recensional undertakings in this pericope in M, bringing this prophetic proclamation into greater consistency with Deuteronomic / Deuteronomistic terminology: cf. Deut 2:5, 9, 12, 19; 3:20; Josh 1:15; 12:6–7; and Judg 21:17.102 In this case, either G translated the noun found in M with a term used regularly elsewhere in the translation of Numbers as a semantic equivalent, or the text of proto-M was changed after the translation of G in order to incorporate a more dtr / dtn lexeme. Ultimately, the matter cannot be decided with certainty, as the Hebrew terms ‫ נחלה‬and ‫ ירשׁה‬overlap 102

The only remaining attestations of this noun are in 2 Chr 20:11; Ps 61:6; and Jer 32:8.

2.2 Textual Criticism

95

to a great degree semantically. Therefore, M will be maintained as the preferable reading. The Greek ἐξεγερθήσεται is an inappropriate translation for the Hebrew ‫רדה‬ as found in M in Num 24:19. This distinction suggests that G may have misunderstood its Vorlage or its Vorlage contained another reading, a reading essentially impossible to reconstruct. The most likely candidate for an alternative reading would be ‫עור‬3; G occasionally reads with forms of ἐξεγείρω for this Hebrew root.103 Where the passive occurs, as in Num 24:19, it most often reflects the Hebrew Nifal (as in, e.g., Jer 6:22), though the Hiphil is also attested on occasion (as in, e.g., Dan 11:25).104 Thus the G-Vorlage could have read something like ‫ ויעור‬or ‫ ויער‬or simply the imperfect ‫ יעור‬or ‫יער‬. The usage of waw-perfect forms in the preceding verses and the subsequent stich would commend ‫וער‬, but the simple perfect in 17b suggests it as a legitimate possibility as well. This reconstructed reading would imply that there was a pun in the G Vorlage that could not be recreated in Greek: ‫ עיר‬and ‫וער‬. That is, this variant could not be the work of the translator. Perhaps either confusion or intention lay behind the reading in M, Smr, T, and S. Both the Targum and the Peshitta affirm the consonantal reading in M (see below), but could affirm the transition in the understanding of this phrase in Num 24:19. The theologically loaded consonantal term in M commends an intentional change, whereas the rather normal reading in T and S could suggest an error. If the distinction in G resulted from an error, it most likely occurred in the paleo-Hebrew script. It remains probable that the G Vorlage reading was misconstrued into the S and T reading, which was then repointed for theological reasons into the text now found in M. The Targums and S commend reading a Hiphil from ‫ ירד‬instead of ‫דרה‬ as in M, which requires no consonantal emendation, only the rejection of the Masoretic pointing. On no other occasion in M does ‫ רדה‬precede the preposition ‫ מן‬in the manner here.105 Later interpretation in the (proto-)Masoretic tradition should most likely be regarded as responsible for the incorrect pointing – and thus reinterpretation of ‫ ירד‬as ‫ – דרה‬in M.106 A number of cases in the Hebrew Bible elucidate the usage of ‫ דרה‬and aid in identifying its background. Several occurrences of this verb are found in P (Gen 1:26 and 28) and H (Lev 25:43, 46, 53; 26:17) contexts. Two Psalms feature it as well: Ps 72 (ostensibly about Solomon in its current form) and Ps 110, both of which focus 103 Cf., e.g., Jer 6:22; 50:42 (= 27:41G); Ezek 38:14 (conjecture); Joel 4:12, Zech 2:17; 4:1; Dan 11:25 and Wevers, Notes on the Greek Text of Numbers, 414. 104 Seebass, Numeri III, 25 incorrectly saw √‫ עור‬in Nifal as the only possibility. 105 The usage in Ps 72:8 reflects the territory which will be ruled (“from sea to sea”) and not where the ruler shall sit in dominion as in Num 24:19. One should also note that Ps 72:8 represents the only other instance in M where ‫ דרה‬is pointed like a jussive, just as in Num 24:8. 106 Rudolph, “Zum Text Des Buches Numeri,” 115 sought to disregard it as a jussive because of its context in prophecy. Instead, he commended reading ‫וְ יִ ְר ֶדה‬.

96

The Text of Numbers 22–24

on the might of God’s chosen ruler. Its usage in Kings and Chronicles remains conspicuous: it applies only to Solomon’s reign and occurs only in presumably late redactional / recensional contexts (1 Kgs 5:4, 30, 9:23; and 2 Chr 8:10).107 These texts even go so far as to identify Solomon as the ruler of all territory west of the Euphrates (1 Kgs 5:4). Bearing these factors in mind, the reading in M is likely a later recensional adaptation of the text in order to focus on the might of this monarch from Jacob: no longer will he merely be raised up from Jacob as in G, rather he will rule from his capital in Jacob. Could this even be an oblique reference to Jerusalem or Zion?108 The reading in M could be the product of emendation over the course of transmission (possibly to be consistent with Ps 72 and Ps 110 and their expectations of ruling saviors). Considering Num 24:20, the phrase ‫ עדי אבד‬has not been adequately explained and its meaning remains unclear. The circumstances are particularly acute since this phrase only occurs in Num 24:20 and 24. The lack of clarity probably caused the variants in the witnesses. While M reads ‫ אבד‬as a participle, Smr reads it as an imperfect: ‫יאבד‬, which also explains the future tense in G.109 The (proto-)Masoretic Text apparently divided the words after the wrong consonant over the course of its transmission.110 G fails to offer an explicit parallel for the Hebrew ‫עד‬, implying that it might be a later addition. Why this would have been added and by whom would remain a mystery. It is possible that G had a Vorlage similar to Smr and tried to achieve the best possible translation. The ‫ עד‬in Smr should probably be read as a defective spelling of ‫עוד‬, something that M misunderstood. A final problem can be found in G’s translation of ‫ אחריתו‬with σπέρμα. While this is unusual, it is by no means impossible; cf. e.g., Num 23:10. It remains impossible to reconstruct with certainty precisely how the text read and what it was supposed to mean, but Smr – supported by the reconstructed Vorlage of G – appears to offer the most likely possibility (‫)אחריתו עד יאבד‬, which should be translated as “his progeny will perish yet”.111 Though G differs significantly from M (and Smr) in 24:22, the Vorlage 107 To the late recensional character of the verses: one need only note that two of the three usages in 1 Kgs are found in different locations in G; 1 Kgs 5:4 = 3 Kgdms 2:46f and 1 Kgs 9:23 = 3 Kgdms 2:35h. 108 Cf. Gray, Numbers, 322. 109 Wevers, Notes on the Greek Text of Numbers, 415 ignored the common reading between Smr and G. One should probably still favor the Qal stem as opposed to the Piʿel in this case. Cf. the discussion to this verb (particularly the Piʿel and the Hiphil) in Ernst Jenni, “Faktitiv und Kausativ von ‫‘ אבד‬zugrunde gehen’,” in Hebräische Wortforschung. Festschrift zum 80. Geburtstag von Walter Baumgartner, ed. Benedikt Hartmann, et al. (Leiden: Brill, 1967), 143–57. 110 Contra Seebass, Numeri III, 25. Cf. already William F. Albright, “The Oracles of Balaam,” JBL 63 (1944): 221. 111 Cf. Gall, “Zusammensetzung und Herkunft,” 41: “und sein Ende (wird kommen) bis zum Untergehenden”. In this way, he understood ‫ עדי‬as an artistic archaism.

2.2 Textual Criticism

97

of G was presumably almost identical to M. Three differences can be identified: M’s ‫ קין‬and ‫ עד־מה‬apparently read ‫ קן‬and ‫ ערמה‬respectively in the G-Vorlage and G read ‫ ואם‬instead of ‫כי אם‬.112 All of the other distinctions represent interpretive differences between the two texts. It appears that G may have preserved the older reading and that M tried to emend the text to make better sense of it, dividing the verse differently. One notes in this respect that Smr does not match M in the question beginning with ‫עד־מה‬. It is also possible that verses 21b–22 read as a chiasm in an older version, whereby a and a’ would reflect the pun ‫ שׁבה‬and ‫ ישׁב‬and b and b’ would both reflect on ‫קן‬:113 a b b’ a’

‫איתן מושׁבך‬ ‫ושׂים בסלע קנך‬ ‫ ערמה‬114‫ואם־יהיה לבער קן‬ ‫אשׁור תשׁבך‬

Thus, at an older stage, the text of the oracle in 21b–22 presumably read “permanent is your dwelling and set in the boulder is your nest. When a nest of treachery will be consumed, Ashur will take you prisoner”. This chiasm contrasts the permanence of the dwelling with its being taken prisoner and the metaphor of the stability of the rock with the coming destruction of the nest.115 Og’s cameo in Num 24:23 G should be regarded as older than M’s reading lacking him.116 Since this appearance apparently confounded later scribes, they dealt with it in two manners: 1) scribes behind the Hebrew traditions of M and Smr simply deleted it, probably because Og had been killed at the end of Num 21; and 2) later tradents of G changed it from Og to Gog, making it in a sense eschatological; cf. the witnesses that read “Gog” in 24:7. Other tradents in the 112 Cf. Wevers, Notes on the Greek Text of Numbers, 416: the ‫ כ‬was the product of dittography from the previous word, which inspired the confusion of ‫ י‬and ‫ו‬. Rudolph, “Zum Text des Buches Numeri,” 115 followed G for the consonants, but read the form ‫ ֲע ֶר ָמה‬because of Neh 3:34. 113 It remains a distinct possibility that ‫ ישׁב‬was the root in both 21b and 22b, but that this was misinterpreted, even by G, although some Smr manuscripts do reflect it; the Hebrew may have originally read ‫תושׁבך‬, in plene, which was then interpreted as coming from √‫שׁבה‬. Should that have been the case, the wordplay would be even more conspicuous. The conjectures of Rudolph, “Zum Text des Buches Numeri,” 115 remain without merit. 114 No need to transpose ‫ קן‬to the first line presents itself, as Albright, “Oracles of Balaam,” 222 proposed. The parallelism is possible without this transposition. 115 Is it possible that an older form of the oracle opened with Balaam’s seeing ‫ קנת‬instead of ‫ ?הקיני‬Some other texts suggest the identification of this group with the Midianites in some way; cf. Num 32:42 and Judg 8:11. Additional evidence for understanding this as a threat against the Midianites can be found in the identification of Moses’ father-in-law as both a Midianite (cf., e.g., Exod 18:1) and as a Kenite (cf., e.g., Judg 1:16), regardless of what his name may have been (cf. Judg 1:16 GBA vs. M). 116 Contra Zacharias Frankel, Einfluss, 172 and Seebass, Numeri III, 25, who favored the lectio facilior as the older reading.

98

The Text of Numbers 22–24

Greek tradition have made the verse match M: “A popular F V text (possibly hex in origin) has omitted the opening reference to an address; the omission equals MT, and may well be recensional in origin. On the other hand, it could easily be a case of homoioteleuton leading to parablepsis.”117 These variants in M, Smr, and several Greek witnesses presumably represent later developments in the text. The absence of Og or Gog in the Vulgate supports this hypothesis. Numbers 24:24 is clearly corrupt in M. The poor condition of the text becomes apparent both when one tries to translate M, which does not make sense and which fails to match subjects and verbs, and when one compares it to the other ancient witnesses. Even bearing this in mind, it is possible that the text was corrupted before the development of the variant traditions; BHS’s conjectural reading, ‫“( מירכתי ים‬from the furthest edge of the sea”) for M’s ‫מיד‬ ‫כתים‬,118 is entirely plausible, though not attested in any form in any of the relevant witnesses.119 Still, we must make the best of the evidence we have at hand. Many of the witnesses – especially M in my opinion – seem to have reworked this text such that it might better reflect the events of Dan 11:30. Here one finds the same motif of the ships coming from Kittim. In Num 24:24 one finds Italians (in the Vulgate) and Romans (in the Targum, which matches the non-θ’ text of Dan 11:30 in G) coming. Even M’s ships in Num 24:24 parallel Dan 11:30, making “ships” the lectio facilior. M appears to have sought consistency after the translation of G.120 But why am I so certain of this? One must recall that the textual tradition of Daniel developed over time as attested by the distinct versions of the text attested in Greek. So, while already the Old Greek version of Daniel 11:30 remains devoid of ships (just as the G and Smr of Num 24:24 are), the Theodotion text of Daniel, perhaps a century or more later, still lacks them.121 As the 117

Wevers, Notes on the Greek Text of Numbers, 416. Cf. Albright, “Oracles of Balaam,” 222–23. The paucity of evidence for this and his more drastic changes commends disregarding them as plausible older readings when other alternatives remain available. The same must be said for the proposals of Bewer, “Literary Problems,” 256 and Rudolph, Der “Elohist”, 126. 119 Contra Vetter, Seherspruch, 50: “Die Interpretation von MT als mijjărkăt-jam überzeugt aus drei Gründen: Erstens läßt sie sich gegenüber MT gewaltlos durchführen; zweitens paßt sie in die aus ägyptischen Nachrichten bekannte Situation [der Zeit vor dem Auszug aus Ägypten]…; drittens hat sie in späteren at.lichen Texten Entsprechungen.” To this end, he also cited Jer 6:22 = 50:41. None of these arguments can be maintained. Not damaging the text presents insufficient basis for emending the text; the text has not been transmitted from any “pre-exodus” period; and the fact that similar prophecies can be found in Jeremiah cannot be used a priori as the basis for emending Numbers, especially since the prophecies do not otherwise match in this instance. 120 There are some difficulties with reading these two verses too closely; cf. Brian Schultz, “The Kittim of Assyria,” RevQ 23, no. 1 (June 2007): 63–77. 121 Regarding the Theodotion text of Daniel, cf. Natalio Fernández Marcos, The Septuagint in Context. Introduction to the Greek Version of the Bible, trans. Wilfred G.E. Watson 118

2.2 Textual Criticism

99

Theodotion text represents a later translation towards proto-M and, as a general rule, quite slavishly follows its Vorlage, we can see that the Hebrew/Aramaic text of Daniel must have developed after the Old Greek translation but before the creation of the so-called Theodotion text, but still not to the level of consistency found in M. Thus, one should reckon with the reinterpretation of Num 24:24 and Dan 11:30 in M. This harmonization appears only in that version, such that it seems impossible to regard any version other than M as an interpretation. M remains consistent against the other witnesses in this case and is therefore unambiguously the lectio facilior. Further, one notes that Smr reads a form of ‫יצא‬, which presumably also stood behind the G reading of ἐξελεύσεται. The most likely candidates to explain both readings would be the plural masculine participle: ‫ ;ויצאים‬or the defective Hifil with an object suffix: ‫ויצאם‬, from which the reading of Smr (‫ )יוציאם‬and G (‫ )ויצא‬both can readily be reconstructed.122 Most likely, Smr added the suffix unintentionally (dittography) under the influence of the following ‫מ‬. This makes the proposed G-Vorlage the most likely oldest, uncorrupted reading. Numbers 24:24aα M does not have a verb, yet is followed by two clauses that begin with conjunctions; that makes it hardly a smooth read.123 Jerome solved this problem by melding the readings of M and G in V, presumably being familiar with both other readings: he simply translated the problematic lexeme – ‫ וצים‬in M – twice (venient [= ‫ ]ויצאו‬in trieribus [= ‫)]צים‬.124 The verbs in 24aβ both demand a plural subject, which is missing in G and syntactically implausible in M, suggesting they represent corrupted readings. Smr’s simple imperfect (‫ )יענו‬may represent the older reading which was corrupted in different manners by both G and M. Samaritanus presents the only consistent text in 24:24a: “He (El/God) will bring them from the hand of Kittim and they will oppress Assyria and Eber.” However, understanding Assyria and Eber as objects of the verbs ‫ ענה‬is not without problems: ‫ענה‬2 as a verb almost always marks the object with ‫את‬, defines the object through the use of an object suffix, or is reflexive. None of these cases are present here, making them at least quite unusual.125 The Masoretic Text suggests the oppressors are the ships from Kittim, G leaves them unidentified, and Smr implies that they are the unnamed object “them” from the previous suffix. A corrupted text presents the best solution to all of these problems. Most likely the verbs from ‫ ענה‬were (Atlanta: SBL, 2000), 142–54; Tov, Textual Criticism, 142–43; and the literature cited in each of these. 122 Cf. already, e.g., Archibald Henry Sayce, “Balaam’s Prophecy (Numbers 24:17–24) and the God Sheth,” Hebraica 4 (1887): 3. 123 Contra Wevers, Notes on the Greek Text of Numbers, 417, who viewed M as a “pendant construction”. 124 Gall, “Zusammensetzung und Herkunft,” 44–45 translated with the Vulgate without even noting it, suggesting, however, that ‫ ויצאו‬disappeared due to a scribal oversight. 125 Cf. also Exod 22:21 and Num 30:14.

100

The Text of Numbers 22–24

previously singular and accidentally transformed into plurals over the course of transmission, assuming that ‫ ה‬was mistakenly read as ‫ ו‬somewhere along the way. This error must be very ancient, as it is reflected in essentially all witnesses. This possibly occurred under the influence of the preceding ‫נ‬: ‫← יענו‬ ‫יענה‬. It remains entirely unclear to whom or what the lexeme ‫ עבר‬as in M and Smr should refer.126 The Seputagint and V both understood it as the “Hebrews”, parallel to “Assyria”, which is conspicuous: “Hebrews” (‫ )עברים‬should parallel “Assyrians” (‫)אשׁורים‬. However, “Hebrews” seems improbable based on the form attested in M and Smr. Perhaps Balaam’s suggesting that the Hebrews would perish in these witnesses demonstrates a tendency to evaluate Balaam more negatively. This negativity could evince a later idiosyncratic interpretation in G. One possibility for the original sense of the term would be simply the meaning “beyond” or “the other side”. The coming oppressors will dominate “Assyria and beyond” (perhaps Persia?). This reading is however insecure and at best a guess. In my opinion it cannot be excluded that the text originally read ‫ עבר הנהר‬or something similar (cf. T’s ‫)לעבר פרת‬, though this thesis can also not be advanced beyond a hypothesis.127 Should this have been the original reading, it must have been lost very early in transmission, as it is no longer reflected in any known witness, though T is closest. The plural subject in 24b in the various translations (e.g., G and V, which imply ‫ )המה‬might also presented the older understanding. G’s use of ὁμοθυμαδὸν would be indicative of a Vorlage that read ‫( יחדו‬cf., e.g., Exod 19:8; Job 2:11; Jer 5:5; Lam 2:8) instead of ‫ עדי‬as found in M. That G did not translate M’s ‫ עדי אבד‬seems obvious, as the translator would have then translated the same phrase two different ways for no apparent reason within five verses (cf. 24:20);128 nor do any witnesses in the Greek tradition evince attempts at conformity between these two verses. The last verb in Num 24:24b can be reconstructed in the plural with G and the versions (the ‫ ו‬presumably having been overlooked due to its location at the end of the verse, because of the ‫ ו‬at the beginning of the next verse, because of the similarity between ‫ ו‬and ‫ד‬, or some combination of these) and as the imperfect of Smr (the ‫ י‬most likely having been overlooked due to the preceding ‫ ו‬as found in the supposed G-Vorlage). G’s translator was fairly consistent in the translation of the Hebrew ‫ ;אבד‬cf. Num 16:33; 17:27; 21:29–30; 24:19: all from ἀπόλλυμι as in Num 24:20 and 126 Cf. Wevers, Notes on the Greek Text of Numbers, 417–18. Vetter, Seherspruch, 51 even understood it as a verb. 127 Cf. already Gall, “Zusammensetzung und Herkunft,” 43; Wellhausen, Composition des Hexateuchs, 351; and Holzinger, Numeri, 125, as well as Levine, Numbers 21–36, 206, who also supported this reading. 128 Such a translational technique would contradict Wevers’ description “[…] the Num translator preferred set expressions” in Wevers, “The Göttingen Pentateuch: Some Post-Partem Reflections,” 56.

2.3 Translation of the Restored Text of Numbers 22–24

101

24. These attestations suggest that the translator was trying to remain true to his or her Vorlage. The G (and other versions) of this phrase presume a Vorlage ‫ וגם המה יחדו יאבד‬or similar, though it is impossible to tell whether the GVorlage included ‫וגם‬.129 The text of Num 24:24 thus perhaps originally read something like: ‫ויצא מיד‬ ‫“( כתים יענה אשׁור וענה עבר ]הנהר[ וגם־המה יחדו יאבד‬And [someone] will come out from the hand of Kittim. He will oppress Assyria and will oppress Eber [Hanahar], but even they will be destroyed together”). This reconstruction would seem to be the best that can be made of the corrupted evidence available, though it is still by no means certain. With this in mind, the possibility of reconstructing the oldest achievable text of Numbers 22–24 becomes reasonable.

2.3 The Restored Text of Numbers 22–24 and Its Translation130 2.3 Translation of the Restored Text of Numbers 22–24

The text reconstructed here reflects the consonantal text of M, corrected in the manners described above and in the footnotes to the translation. A catalogue of the changes to M follows the translation and the reconstructed text. In general, few changes have been made, with substantial revision only occurring in the verses 23:3; 24:7, 17 and 24. The text presented here is roughly one percent longer (5278 consonants total) than the consonantal text of M (5268 consonants total). Of course a more defective orthography may have stood in the background of the traditions, but this would be more difficult – even impossible – to ascertain with any degree of plausibility.131 1

‫ויסעו בני ישׂראל ויחנו בערבות‬ ‫מואב מעבר לירדן ירחו‬ ‫וירא בלק בן־צפור את כל־אשׁר־‬ ‫עשׂה ישׂראל לאמרי‬ ‫ויגר מואב מפני העם מאד כי‬ ‫רב־הוא ויקץ מואב מפני בני‬ ‫ישׂראל‬

2 3

129

Numbers 22 The Israelites set out and camped in the steppes of Moab across the Jordan from Jericho. And Balak ben Zippor had seen all that Israel had done to the Amorites. And Moab was very afraid of the people because it was larger than they. And Moab was disgusted by the Israelites.

N.b. G-Num never translates ‫ וגם‬with anything other than καί. Cf. Num 13:27–28; 18:2; and 24:25. 130 Italics denote apparent redactional elements. Strikethrough denotes late glosses. Consult the discussion in the following chapter for the basis of these designations. The singular “it” has been used to reference to the “people” throughout in order to reflect Hebrew grammar. 131 Cf. the noteworthy attempt in Albright, “Oracles of Balaam”.

102 4

The Text of Numbers 22–24 ‫ויאמר מואב אל־זקני מדין עתה‬ ‫ילחכו הקהל הזה את־כל־‬ ‫סביבתינו כלחך השׁור את ירק‬ ‫השׂדה ובלק בן־צפור מלך‬ ‫למואב בעת ההוא‬ ‫וישׁלח מלאכים אל־בלעם בן־‬ ‫בעור פתורה אשׁר על־הנהר‬ ‫בארץ בני־עמו לקרא־לו לאמר‬ ‫הנה עם יצא ממצרים והנה כסה‬ ‫את־עין הארץ והוא ישׁב ממלי‬

5

132

So Moab spoke to elders of Midian: Now, this132 assembly will devour133 all that is around us, as the bull devours what is green in the field. And Balak ben Zippor was king of Moab at this time. He sent messengers 134 to Balaam ben Beor to Pethor,135 which is on the136 river, in the land of the sons of his people137 to summon him, saying: look, a people has come out of Egypt, and look,138 it covers the visible landscape139 and is dwelling across from me.

Smr, G, S, and V all read “this” instead of merely “the”. The translation reflects this alternative reading based on the gravitas of the witnesses; M omitted ‫הזה‬, probably an oversight. Contra Seebass, Numeri III, 14. 133 Smr’s singular is the lectio facilior; this makes no difference in the English translation. 134 G mentions “elders” (πρέσβεις) here, an unusual translation for ‫ מלאך‬otherwise found only in Num 21:21 and Deut 2:26. Wevers, Notes on the Greek Text of Numbers, 361 considers this the over-arching term for the elders (γερουσία) in Num 22:4, and 7. This may be correct; cf. however, 24:12: ἀγγέλοις. ‫ מלאכים‬seems most probable in this verse; cf. the probable reconstructed ‫ מ‬in 4QNumb (Ulrich, Biblical Qumran Scrolls, 151). Thus, the Septuagint translator presumably was responsible for the variant in this case. 135 S and V read “the interpreter” instead of “to Pethor”. Cf. the Aramaic √‫ פשׁר‬and the Akkadian pašaru, which both suggest an interpretation that Balaam was known for countering the (negative) intents of others. Cf. Seebass, Numeri III, 15 and the preceding text-critical discussion of this verse. 136 It is unclear whether the Vorlage of G contained the definite article. 137 Instead of “his people”, a number of Hebrew manuscripts, Smr, S, and V read “Ammon”; cf. Johan Lust, “Balaam, an Ammonite,” ETL 54 (1978): 60. “Ammon” may be the original reading, but is more likely a corrective to make the text better match Gen 19:30–38 (which identifies Moab and Ammon as half-brothers and cousins) and Deut 23:4–6 (which precludes both the Ammonites and the Moabites from inclusion into the assembly). In this scenario, the Moabites would be guilty in that they hired Balaam; the Ammonites would be guilty in that Balaam was an Ammonite who tried to curse Israel, at least when corroborated with Deut 23. Contra Lust, “Balaam,” 60–61. Albright, “Home of Balaam,” 389 commended reconstructing ‫בית עדן‬, which remains fanciful speculation devoid of evidence. 138 Whether the conjunction should be added before ‫ הנה‬cannot be decided with any certainty. The number and quality of witnesses commend emending ML, as reflected in the translation above. N.b. neither the BHS (which references the Sebirin in the critical apparatus) nor ML contain a notice in the Masorah parva that the conjunction should be added here; cf. David Noel Freedman, ed., The Leningrad Codex: A Facsimile Edition, general editor David Noel Freedman, managing editor Astrid B. Beck, associate editors Bruce E. Zuckerman, et al. (Grand Rapids, MI: Wm. B. Eerdmans, 1998), ad loc. 139 Literally, “the eye of the land”. Also in v. 11. Achenbach, Vollendung, 396: regarded ‫ עין הארץ‬as the water-rich region of the Jordan valley. Cf. Seebass, Numeri III, 15 for a discussion of this phrase’s meaning.

2.3 Translation of the Restored Text of Numbers 22–24 6

7

8

9 10

11

12

‫ועתה לכה־נא ארה־לי את־העם‬ ‫הזה כי־עצום הוא ממני אולי‬ ‫אוכל נכה־בו ואגרשׁנו מן־הארץ‬ ‫כי ידעתי את אשׁר־תברך מברך‬ ‫ואשׁר תאר יואר‬ ‫וילכו זקני מואב וזקני מדין‬ ‫וקסמיהם בידם ויבאו אל־בלעם‬ ‫וידברו אליו דברי בלק‬ ‫ויאמר אליהם לינו פה הלילה‬ ‫והשׁבתי אתכם דבר כאשׁר‬ ‫ידבר יהוה אלי וישׁבו שׂרי־מואב‬ ‫עם־בלעם‬ ‫ויבא אלהים אל־בלעם ויאמר מי‬ ‫האנשׁים האלה עמך‬ ‫ויאמר בלעם אל־האלהים בלק‬ ‫בן־צפר מלך מואב שׁלח אלי‬ ‫לאמר‬ ‫הנה עם יצא ממצרים ויכס את־‬ ‫עין הארץ וישׁב ממלי עתה לכה‬ ‫קבה־לי אתו אולי אוכל להלחם‬ ‫בו וגרשׁתיו‬ ‫ויאמר אלהים אל־בלעם לא תלך‬ ‫עמהם לא תאר את־העם כי‬ ‫ברוך הוא‬

103

And now, please come; curse this people for me since it is stronger than I am.140 Maybe I141 will be able to strike it and drive it from the land. Indeed, I know that the one you bless is blessed and the one you curse is cursed.142 So the elders of Moab and the elders of Midian set out (and the143 prophecies144 were in their hand). And they came to Balaam and told him Balak’s words. And he spoke to them: spend the night here and I will bring back a word to you just as YHWH tells me. The officers of Moab stayed with Balaam. God came to Balaam and said: who 145 are these men with you? Balaam spoke to God: Balak ben Zippor, King of Moab, sent to me, saying:146 Look, a people has come out of Egypt and it covers the visible landscape and is dwelling across from me. Now come curse it for me. Perhaps I will be able to engage it in battle and drive it away.147 God spoke to Balaam: you shall not go with them. You shall not curse the people, because it is blessed.

140 The plural of G is to be rejected. Cf. Seebass, Numeri III, 15 and the text-critical discussion above. 141 M’s singular is preferable. See Seebass, Numeri III, 15 and above. 142 Smr appears to be lectio facilior in the last phrase of v. 6; see above. 143 Smr and S explicitly refer to the prophecies as coming from the messengers (i.e., “their prophecies”), probably a later explication. 144 The usage of ‫ קסמים‬here remains curious, but perhaps decipherable. It appears to reference, at least in its current context, some model or evidence of the situation, perhaps even extispicies undertaken by Balak’s diviners (though that reads information into this story not provided by the text explicitly); cf. Victor (Avigdor) Hurowitz, “The Expression ûqsāmîm beyādām (Numbers 22:7) in Light of Divinatory Practices from Mari,” HS 33 (1992): 5–15. Otherwise it could have, in its original context, suggested that the Moabites and Midianites sought to curse Israel with some kind of magic. Cf. the discussion of the term ‫ קסם‬in Chapter Six. 145 G does not differ from M here; τι is a legitimate translation of ‫ ;מי‬contra Seebass, Numeri III, 16. Smr’s ‫ מה‬probably came about as a result of misreading ‫ י‬for ‫ ה‬in either paleo-Hebrew or the Samaritan script. 146 The longer reading of G, V, and S is preferable; see above. 147 The variants in 22:11 have been addressed above.

104 13

14 15 16

17

18

19

148

The Text of Numbers 22–24 ‫ויקם בלעם בבקר ויאמר אל־‬ ‫שׂרי בלק לכו אל־אדנכם כי מאן‬ ‫יהוה לתתי להלך עמכם‬ ‫ויקומו שׂרי מואב ויבאו אל־בלק‬ ‫ויאמרו מאן בלעם הלך עמנו‬ ‫ויסף עוד בלק שׁלח שׂרים רבים‬ ‫ונכבדים מאלה‬ ‫ויבאו אל־בלעם ויאמרו לו כה‬ ‫אמר בלק בן־צפור אל־נא תמנע‬ ‫מהלך אלי‬ ‫כי־כבד אכבדך מאד וכל אשׁר־‬ ‫תאמר אלי אעשׂה ולכה־נא‬ ‫קבה־לי את העם הזה‬ ‫ויען בלעם ויאמר אל־עבדי בלק‬ ‫אם־יתן־לי בלק מלא ביתו כסף‬ ‫וזהב לא אוכל לעבר את־פי‬ ‫יהוה אלהי לעשׂות קטנה או‬ ‫גדולה בלבי‬ ‫ועתה שׁבו נא בזה גם־אתם‬ ‫הלילה ואדעה מה־יסף יהוה‬ ‫דבר עמי‬

In the morning Balaam got up and said to Balak’s officers: go to your lord.148 YHWH149 has refused to release me to go with you. So the officers of Moab arose and went to Balak and said: Balaam refused to come with us. But Balak sent back to Balaam officers who were greater and more important than those. And they came to Balaam and said to him: thus speaks Balak ben Zippor: do not be held back from coming to me. For I can make you very rich and all that you say I will do. Please come and curse this people for me. Balaam answered and said to Balak’s servants:150 if Balak should give me the contents of his house, silver, and151 gold, I would not be able transgress the word of YHWH, my God,152 doing something small or something large in my heart.153 And now, please, you too should stay here tonight. I will reveal to you what else YHWH speaks with me.

The text has been emended to reflect G. Contra Zacharias Frankel, Einfluss, 172, who regarded G’s reading as an error due to the translator’s hastiness. Contra Seebass, Numeri III, 16, who saw G as an attempt to conform this verse to v. 14. 149 It cannot be determined with certainty if the Vorlage of G read ‫ יהוה‬or ‫אלהים‬. Wevers seems to imply that it read ‫יהוה‬, which the translator then changed to ὁ θεὸς: “Ms 426 changed ὁ θεὸς to κύριος, obviously on the basis of the Hebrew; so did the b smg+ text; the reading probably derives from one of the Revisers.” (Wevers, Notes on the Greek Text of Numbers, 367). 150 Though impossible to securely ascertain, the “servants” recorded in M seem to be the most likely original reading, since the other alternatives (“princes” in G and “messengers” in S) appear to harmonize with the rest of the narrative which never reports that Balak has “servants”, but which notes that he had “messengers” (22:5, 22–27, 31–32, 34–35; and 24:12) and “officers” (22:8, 13–15, 21, 35, 40; 23:6 and 17). Cf. Seebass, Numeri III, 16. 151 The reading “or” in some manuscripts of Smr and S is probably secondary, since the overwhelming majority of weighty witnesses counter it. The use of “or” makes Balaam even cheaper (i.e., even easier to persuade). Rather than requiring all three items, any one would suffice in Smr and S. 152 The reading “my God” in M is probably the oldest since it is the most theologically problematic, coming as it does in this context from the mouth of a non-Israelite; cf. Rösel, “Textüberlieferung,” 224 n. 63. 153 G adds “for my benefit” (loosely translated) to the end of the verse, which might be an attempt to harmonize 22:18 with 24:13 as the editors of BHS and Wevers, “Balaam Narrative,” 138 suggested. However, the reading ‫ בלבי‬has to be reconstructed in 4QNumb, suggesting that G and Q in fact preserve the older reading, contra Seebass, Numeri III, 16–17. The Greek translation of Numbers sometimes reflects ‫ לב‬/ ‫ לבב‬with διάνοια; cf. Num 15:39 and 32:7.

2.3 Translation of the Restored Text of Numbers 22–24 20

21 22

23

24 25

26

27

28

29

30

‫ויבא אלהים אל־בלעם לילה‬ ‫ויאמר לו אם־לקרא לך באו‬ ‫האנשׁים קום לך אתם ואך את־‬ ‫הדבר אשׁר־אדבר אליך אתו‬ ‫תעשׂה‬ ‫ויקם בלעם בבקר ויחבשׁ את־‬ ‫אתנו וילך עם־שׂרי מואב‬ ‫ויחר־אף אלהים כי־הולך הוא‬ ‫ויתיצב מלאך יהוה בדרך לשׂטן‬ ‫לו והוא רכב על־אתנו ושׁני‬ ‫נעריו עמו‬ ‫ותרא האתון את־מלאך יהוה‬ ‫נצב בדרך וחרבו שׁלופה בידו‬ ‫ותט האתון מן־הדרך ותלך‬ ‫בשׂדה ויך בלעם את־האתון‬ ‫להטתה הדרך‬ ‫ויעמד מלאך יהוה במשׁעול‬ ‫הכרמים גדר מזה וגדר מזה‬ ‫ותרא האתון את־מלאך יהוה‬ ‫ותלחץ אל־הקיר ותלחץ את־רגל‬ ‫בלעם אל־הקיר ויסף להכתה‬ ‫ויוסף מלאך־יהוה עבור ויעמד‬ ‫במקום צר אשׁר אין־דרך לנטות‬ ‫ימין ושׂמאול‬ ‫ותרא האתון את־מלאך יהוה‬ ‫ותרבץ תחת בלעם ויחר־אף‬ ‫בלעם ויך את־האתון במקל‬ ‫ויפתח יהוה את־פי האתון‬ ‫ותאמר לבלעם מה־עשׂיתי לך כי‬ ‫הכיתני זה שׁלשׁ רגלים‬ ‫ויאמר בלעם לאתון כי התעללת‬ ‫בי לו ישׁ־חרב בידי כי עתה‬ ‫הרגתיך‬ ‫ותאמר האתון אל־בלעם הלוא‬ ‫אנכי אתנך אשׁר־רכבת עלי‬

105

And God154 came to Balaam in the night and said to him: if the men have come to you to call you, arise! Go with them! But only the word that I will say to you shall you tell them. Balaam arose in the morning and saddled his donkey and went with the officers of Moab. But God’s155 anger was enflamed because he was going156 and the messenger of YHWH157 stood in the way to harass him. He was riding on his donkey and his two servants were with him. And the donkey saw the messenger of YHWH standing in the path and his sword was drawn in his hand. So the donkey left the path and walked in the field. And Balaam struck the donkey for leaving the path. But the messenger of YHWH stood in a tight pass of vineyards with a stone wall on each side. And the donkey saw the messenger of YHWH and pressed herself against the wall and pressed Balaam’s foot against the wall. So Balaam struck her again. Again the messenger of YHWH went out and stood in a tight space where there was no way to turn to the left or to the right. And the donkey saw the messenger of YHWH and laid down under Balaam. And Balaam’s anger enflamed and he struck the donkey with the stick. So YHWH opened the mouth of the donkey and she spoke to158 Balaam: what have I done to you that you struck me these three times? And Balaam spoke to the donkey: because you are teasing159 me! If there were a sword in my hand I would kill you now! And the donkey spoke to Balaam: am I not your donkey, upon whom you have ridden from your

154 The reading “messenger of God” as preserved in Smr was secondarily inserted to remove some of the direct contact between God and Balaam. This change also better integrates 22:21–35a into the narrative; cf. also 23:4 and 16 in Smr. 155 It cannot be determined with certainty whether the reading ‫ יהוה‬or ‫ אלהים‬is older. 156 The distinction between the participle (M) and the perfect (Smr and G) of the verb is text-critically irrelevant; both forms could have developed from the same consonantal text: ‫הלך‬. 157 The question as to whether ‫( יהוה‬M) or ‫ אלהים‬/ ὁ θεός (G [and its Vorlage?]), here and in the following, is the older reading has been dealt with above. 158 It is neither possible nor relevant to determine which preposition is original: ‫ ל‬or ‫אל‬. 159 Smr’s first person singular is probably a scribal error, influenced by the following ‫בי‬.

106

31

32

33

34

35

The Text of Numbers 22–24 ‫מעודך עד־היום הזה ההסכן‬ ‫הסכנתי לעשׂות לך כה ויאמר‬ ‫לא‬ ‫ויגל יהוה את־עיני בלעם וירא‬ ‫את־מלאך יהוה נצב בדרך‬ ‫וחרבו שׁלפה בידו ויקד וישׁתחו‬ ‫לאפיו‬ ‫ויאמר אליו מלאך יהוה על־מה‬ ‫הכית את־אתנך זה שׁלושׁ רגלים‬ ‫הנה אנכי יצאתי לשׂטנך כי־ירע‬ ‫דרכך לנגדי‬ ‫ותראני האתון ותט מלפני זה‬ ‫שׁלשׁ רגלים ולולי נטתה מפני כי‬ ‫עתה גם־אתכה הכיתי ואתה‬ ‫החייתי‬ ‫ויאמר בלעם אל־מלאך יהוה‬ ‫חטאתי כי לא ידעתי כי אתה‬ ‫נצב לקראתי בדרך ועתה אם־‬ ‫רע בעיניך אשׁובה לי‬ ‫ויאמר מלאך יהוה אל־בלעם לך‬ ‫עם־האנשׁים ואפס את־הדבר‬ ‫אשׁר־אדבר אליך אתו תדבר‬ ‫וילך בלעם עם־שׂרי בלק‬

youth until today? Have I been known to do thus to you?160 And he said: no. So YHWH opened Balaam’s eyes and he saw the messenger of YHWH161 standing in the path and his sword was drawn in his hand. And he bowed and prostrated himself162 on his face. And the messenger of YHWH spoke to him: why did you strike your donkey these three times? Look, I came out to harass you because your path before me is shaky.163 And the donkey saw me and turned aside from164 in front of me these three times. Perhaps165 – had she not turned aside from before me – I now also would have struck166 you167 but let her live. And Balaam spoke to the messenger of YHWH: I have sinned, for I did not know that you stood across from me in the path. And now, if it is bad in your eyes, I will return myself. But the messenger of YHWH spoke to Balaam: go with the men. But nothing [except] the word that I will speak to you shall you say.168 And Balaam went with Balak’s officers.

160 G provides a different reading: “I have not utterly disregarded you thus”. Smr may read it as a question, but if so, it is missing the interrogative-‫( ה‬haplography). 161 In the Greek tradition, both Origen and Vaticanus (and a number of other manuscripts) match M, reading “messenger of YHWH”, whereas the reconstructed G-originalis of Göttingen reads “messenger of God”. The reading “messenger of YHWH” may be older, but this is hardly certain. 162 The verb form found in Smr and its origin remain unclear. 163 Verse 32 was handled in the preceding discussion. 164 Smr records the preposition “from”, which may be older, though this cannot be clarified with certainty. G does not expressly reflect this preposition, but neither is it necessary in G. 165 G et al. read “If she had not”, which may be older. The reading in M would have come about from mistaking ‫ ול‬as ‫א‬. Cf. Seebass, Numeri III, 17. 166 In M the verb here matches that used in Balaam’s threatening his donkey in 22:29 (√‫)הרג‬, whereas Smr reads “to strike” (√‫)נכה‬. This implies that M could be later, artistically adapting the narrative to literally match Balaam’s threat. G also distinguishes these terms in 29 and 33 and can thus be understood as supporting Smr here. 167 Since the orthography of Smr is usually fuller than M (cf. Tov, Textual Criticism, 89– 90), this reversed situation suggests that Smr may be the older reading. 168 The “you shall guard to say” of Smr and G could be a later harmonizing gloss; cf. 23:12 and Rösel, “Textüberlieferung,” 220–21. The same could be said of Peshitta’s “do”; cf. 22:20 and Seebass, Numeri III, 17–18.

2.3 Translation of the Restored Text of Numbers 22–24 36

‫וישׁמע בלק כי בא בלעם ויצא‬ ‫לקראתו אל־עיר מואב אשׁר על־‬ ‫גבול ארנן אשׁר בקצה הגבול‬

37

‫ויאמר בלק אל־בלעם הלא שׁלח‬ ‫שׁלחתי אליך לקרא־לך למה‬ ‫לא־הלכת אלי האמנם לא אוכל‬ ‫כבדך‬ ‫ויאמר בלעם אל־בלק הנה־באתי‬ ‫אליך עתה היכול אוכל דבר‬ ‫מאומה הדבר אשׁר ישׂים אלהים‬ ‫בפי אתו אדבר‬ ‫וילך בלעם עם־בלק ויבאו קרית‬ ‫חצות‬ ‫ויזבח בלק בקר וצאן וישׁלח‬ ‫לבלעם ולשׂרים אשׁר אתו‬ ‫ויהי בבקר ויקח בלק את־בלעם‬ ‫ויעלהו במת בעל וירא משׁם‬ ‫קצה העם‬

38

39 40 41

1

‫ויאמר בלעם אל־בלק בנה־לי‬ ‫בזה שׁבעה מזבחת והכן לי בזה‬ ‫שׁבעה פרים ושׁבעה אילים‬

107

When Balak heard that Balaam had arrived he went out to meet him at the city169 of Moab that is at the boundary170 of the Arnon, which is at the edge of the boundary. And Balak spoke to Balaam: did I not emphatically send to you to summon you171? Why did you not come to me? Am I not certainly able to make you rich? Balaam spoke to Balak: look, I came to you. Now, am I really able to say anything other than the word that God172 sets in my mouth? That [alone] I will say.173 And Balaam went with Balak and they arrived174 at Qiryat Chuzot.175 And Balak sacrificed cattle and sheep and sent to Balaam and to the officers who were with him. In the morning, Balak took Balaam and led him up to Bamath176 [or: “the pillar of”] Baal and he saw from there the edge of the people. Numbers 23 And Balaam spoke to Balak: build177 seven altars for me here and have seven calves and seven rams ready here.

169 While the conjecture of BHS to read ‫ ער‬instead of ‫ עיר‬is possible, there is no evidence to support it text-critically. Already the ancient translations understood it as a simple, not a proper noun (G, T, V). Contra Achenbach, Vollendung, 407. 170 No evidence supports deleting the first occurrence of “boundary” in this verse, as Rudolph, “Zum Text des Buches Numeri,” 115 and BHS suggested. 171 The difference in the Tiberian versus Oriental Masoretic pointing of ‫ לך‬is not textcritically relevant. 172 Rudolph’s conjecture “my God” based on v. 18 may be accurate, but it remains without supporting text-critical evidence; cf. Rudolph, “Zum Text des Buches Numeri,” 115. 173 Smr and a number of G manuscripts (including GA) read “I will guard to say”, probably again as a harmonization with 23:12; cf. 22:35 and the comments there. 174 The hi. sg. with object (as in Smr, S, T) better matches the preceding verb and is later. 175 The phrase ‫ קרית חצות‬was misunderstood by G and T. Cf. Wevers, Notes on the Greek Text of Numbers, 382 and Seebass, Numeri III, 18. 176 The translator of G apparently did not understand this to be a location, but a pillar of Baal (τὴν στήλην τοῦ Βααλ). Thus G translates the Heb. ‫ במת‬in 22:41, but transliterates ‫במות‬ in 21:19–20. This favors the pronunciation “Bamath” or even the meaning “pillar”, M being a harmonization with Num 21. Cf. already Holzinger, Numeri, 115. Contra Seebass, Numeri III, 18. 177 Smr reads “make”, which has also been reconstructed in 4QNumb, though both readings might be possible there. G supports M, always otherwise translating √‫ בנה‬with οἰκοδομέω in Numbers.

108

The Text of Numbers 22–24

2

‫ויעשׂ בלק כאשׁר דבר בלעם‬ ‫ויעל פר ואיל במזבח‬

3

‫ויאמר בלעם לבלק התיצב על־‬ ‫עלתך ואלכה אולי יקרה אלהים‬ ‫לקראתי ודבר מה־יראני והגדתי‬ ‫לך וילך ויתיצב בלק על עלתו‬ ‫ובלעם נקרה אל אלהים וילך‬ ‫שׁפי‬

So Balak did just as Balaam had said and Balak and Balaam178 he offered a calf and a ram as a burnt offering on the altar.179 Balaam spoke to 180 Balak: stand 181 by your burnt offering182 while I go.183 Perhaps God184 will come out185 to call me. And anything that he shows me I will report to you. So Balak stood by his altar and Balaam went to encounter God.186 And he went out to a clearing.187

178 The absence of an explicit subject in G and some Heb. manuscripts suggests that “Balak and Balaam” is a later gloss. This change makes it explicit that they both took part in the offering. Consequently, Balaam also became culpable for taking part in an illegitimate offering, though this understanding is at best implicit; cf. Seebass, Numeri III, 18. 179 The conjecture presented by BHS, to add v. 4b here, is both without evidence and not a text-critical matter; cf. Seebass, Numeri III, 18. 180 It cannot be clarified whether any older version read ‫ אל‬or ‫ל‬. 181 The addition of “thus” as suggested by BHS creates a harmonization, where none may have been present. It seems more likely that it should be deleted in 23:15; cf. G. 182 The plural of Smr and S is clearly a later harmonization based on context. 183 The addition of “thus” (as in v. 15) is unnecessary; cf. G. 184 4QNumb, G, and Smr attest the older reading “God”; contra Seebass, Numeri III, 18. 185 BHS notes that Smr reads √‫ קרא‬here. I would add that it is unclear what G read in its Vorlage, in that it never otherwise uses φαίνω as a translation for ‫ קרה‬in the Pentateuch. The best possibility would most likely be from the hiphil of √‫ראה‬, but these are also not the standard equivalents in the G-Vorlagen of the Pentateuch. Φαίνω and related terms are used as equivalents for √‫ ראה‬in other books of G, however; cf. the curious case of Amos 5:22. Is it possible that G read ‫יראה‬, perhaps as paleo-Hebrew hary, in its Vorlage? Such an original paleo-Hebrew reading could explain the variants: from hary (G-Vorlage) to arqy (Smr) to hrqy (M). 186 The longer reading of G and 4QNumb has been restored as older, cf. already Gall, “Zusammensetzung und Herkunft,” 11, who suggested this even before the discovery of 4QNumb, and Nathan Jastram, “The Text of 4QNumb,” in The Madrid Qumran Congress. Proceedings of the International Congress on the Dead Sea Scrolls. Madrid 18–21 March, 1991, ed. Julio Trebolle Barrera and Luis Vegas Montaner, STDJ, vol. XI (Leiden; New York; Köln: Brill, 1992), 184–86. Contra Seebass, Numeri III, 18–19. Parablepsis abbreviated M and Smr. 187 In G, Balaam did not go to a clearing. Rather, he went “straightaway”. This presumably presents an interpretive attempt to understand the otherwise unattested singular ‫שׁפי‬. Cf. Dorival, Nombres, 433: “une crête dénudée”. Paul Volz, review of Zusammensetzung und Herkunft der Bileam-Perikope in Num. 22–24, TLZ 26, no. 14 (1901): 385 and Bewer, “Literary Problems,” 251 suggested that ‫ שׁפי‬could be an abbreviation for ‫שׁאל פי יהוה‬. While an interesting proposition, it is both speculative and fails to deal with the other text-critical evidence. This interpretation would be commendable, if any of the other mentions of Balaam’s going included something like his postulation. Cf. already Holzinger, Numeri, 115 contra Volz.

2.3 Translation of the Restored Text of Numbers 22–24 4

5 6

7

8

9

‫ויקר אלהים אל־בלעם ויאמר‬ ‫אליו את־שׁבעת המזבחת ערכתי‬ ‫ואעל פר ואיל במזבח‬ ‫וישׂם יהוה דבר בפי בלעם‬ ‫ויאמר שׁוב אל־בלק וכה תדבר‬ ‫וישׁב אליו והנה נצב על־עלתו‬ ‫הוא וכל־שׂרי מואב‬ ‫וישׂא משׁלו ויאמר‬ ‫מן־ארם ינחני בלק‬ ‫מלך־מואב מהררי־קדם‬ ‫לכה ארה־לי יעקב‬ ‫ולכה זעמה ישׂראל‬ ‫מה אקב‬ ‫לא קבה אל‬ ‫ומה אזעם‬ ‫לא זעם יהוה‬ ‫כי־מראשׁ צרים אראנו‬ ‫ומגבעות אשׁורנו‬ ‫הן־עם לבדד ישׁכן‬ ‫ובגוים לא יתחשׁב‬

109

And God 188 met 189 with Balaam and he 190 said to him: the seven altars I have erected and I have offered a calf and a ram on the altar. And YHWH191 set a word in the mouth of Balaam and said: return to Balak and thus shall you speak. And he returned to him. Look, he was standing by his burnt offering,192 he and all of the officers of Moab.193 And he raised his saying194 and spoke: From Aram195 Balak led me; the king of Moab from the mountains of the east: come! Curse for me Jacob! And come! Execrate Israel! What shall I curse? El has not cursed! And what shall I execrate? YHWH has not execrated!196 From the tops of boulders I will see him. And from hills I will gaze. Look! A people dwells alone, and among the nations it does not count itself.197

188 In Smr and 4QNumb (cf. Ulrich, Biblical Qumran Scrolls, 154) the text reads “messenger of God”, which probably represents a harmonization with 22:22–35, a reworking for the context. 189 This verb has the same problems here as in v. 3, with the exception that God “finds” Balaam in Smr. But recourse to paleo-Hebrew fails to aid in the reconstruction of this variant: acmyw. Could Smr attest the oldest reading, with the others representing an attempt to harmonize vv. 3 and 4? 190 There is no need to move v. 4b to follow v. 2. Even if there were, this is not a textcritical consideration. The addition of the subject βαλααμ in G reads like a clarifying gloss; cf. Seebass, Numeri III, 19. 191 Smr attests the “messenger of YHWH”; G reads “God”. BHS’s suggestion to delete the subject is without evidence or merit. 192 The plural “offerings” as attested in Smr, G, and S is a harmonization with the number seven mentioned in 23:1. Alternatively, one could argue the case (which is less likely in context) that M is incorrectly pointed. 193 At the end of v. 6 G reads “and the spirit of God was upon him”, harmonizing with 24:2; cf. Seebass, Numeri III, 19. 194 Cf. Dorival, Nombres, 434 for a discussion of this term, which he translated with “parabole”. 195 The conjecture that “Edom” should be read is baseless; cf. Levine, Numbers 21– 36, 168. 196 Some Smr mss reads the last phrase as “YHWH is not his execration” or similar. This resulted from the dittography of ‫י‬, which was then subsequently read as ‫ו‬. 197 It is quite possible to understand this phrase as a statement of Israel’s strength and independence from other nations. Some support, though perhaps speculative, for such an

110 10

11

12 13

14

15

The Text of Numbers 22–24 ‫מי מנה עפר יעקב‬ ‫ומי ספר משׁפחת ישׂראל‬ ‫תמת נפשׁי מות ישׁרים‬ ‫ותהי אחריתי כמהו‬ ‫ויאמר בלק אל־בלעם מה עשׂית‬ ‫לי לקב איבי לקחתיך והנה‬ ‫ברכת ברך‬ ‫ויען ויאמר הלא את אשׁר ישׂים‬ ‫יהוה בפי אתו אשׁמר לדבר‬ ‫ויאמר אליו בלק לכה־נא אתי‬ ‫אל־מקום אחר אשׁר תראנו‬ ‫משׁם אפס קצהו תראה וכלו לא‬ ‫תראה וקבנו־לי משׁם‬ ‫ויקחהו שׂדה צפים אל־ראשׁ‬ ‫הפסגה ויבן שׁבעה מזבחת ויעל‬ ‫פר ואיל במזבח‬ ‫ויאמר אל־בלק התיצב על־עלתך‬ ‫ואנכי אקרה כה אלהים‬

Who can count the dust198 of Jacob? And who can number the tribes199 of Israel? May my spirit die the death of the upright, and may my end be like it.200 Balak spoke to Balaam: what have you done to me? To curse my enemy I brought you, but look! You have certainly blessed201 [it]! He answered and spoke: have I not guarded to speak that which YHWH202 set in my mouth? Balak spoke to him: please go203 with me to another place where204 you will see it from there. The end of its edge you will see, but its entirety you will not see. So curse205 it for me from there. And he took him to the field of the Zophim, to the pinnacle of the Pisgah. And he built seven [an] altars and offered a calf and a ram at the altar. And he206 spoke to Balak: stand207 by your sacrifice208 and I, I will go out to encounter God.209

understanding could be found at Mari; cf. A. Malamat, “ʿAmm Leḇāḏāḏ Yiškōn: A Report from Mari and an Oracle of Balaam,” JQR 76, no. 1 (July 1985): 47–50. 198 It is possible that “from the dust” and “number” got confused in the course of textual transmission, as the note in BHS implies. The most likely explanation would be the confusion of a paleo-Hebrew ‫ ע‬with a ‫ס‬. This explanation does not deal with how the ‫ מ‬would have disappeared and it seems unlikely, though not impossible. This is contra Vetter, Seherspruch, 11, who saw “Staubgewühl” as coming from Akkadian turbu’tu. G reads “seed”, an unlikely translation of either of the aforementioned options and possibly a reference to Abram in Gen 13:16; cf., e.g., Vetter, Seherspruch, 11; Milgrom, Numbers, 197; and Dorival, Nombres, 70. Unlike Levine, Numbers 21–36, 174–75, I do not understand v. 9b militarily. 199 With G (‫ ;משׁפחת‬contra Gall, “Zusammensetzung und Herkunft,” 25, who reconstructed ‫ רבבות‬and referenced Deut 33:17). Smr may support this reading. M reads: “and the number of a quarter of Israel.” See above. 200 G offers a substantially different reading than M for 23:10. And it is quite likely that the verse is corrupt. See the discussion above. 201 The verbal distinction between Smr and M is generally irrelevant (Qal pass. part. vs. Piʿel inf. abs.); however, the nominal understanding of the root as attested in G could support Smr as the older reading. 202 G attests “God”. 203 Read with Qere. 204 Smr reads separate clauses: “You will see from there.” G reads in the negative. 205 There is no need to repoint the final verb as per the suggestion of BHS; cf. G. 206 A few Hebrew manuscripts, G, and S add “Balaam”, which appears to be a clarifying gloss. 207 “Thus” is missing in some Hebrew manuscripts, Smr, and G and should be removed. 208 The plural of many Hebrew manuscripts, Smr, and S is a harmonization with the seven altars in 23:1 and 14. 209 The end of the verse should be reconstructed with G, adding ‫אלהים‬.

2.3 Translation of the Restored Text of Numbers 22–24 16

17

18

19

20 21

210

‫ויקר יהוה אל־בלעם וישׂם דבר‬ ‫בפיו ויאמר שׁוב אל־בלק וכה‬ ‫תדבר‬ ‫ויבא אליו והנה נצב על־עלתו‬ ‫ושׂרי מואב אתו ויאמר לו בלק‬ ‫מה־דבר יהוה‬ ‫וישׂא משׁלו ויאמר‬ ‫קום בלק ושׁמע‬ ‫האזינה עדי בנו צפר‬ ‫לא אישׁ אל ויכזב‬ ‫ובן־אדם ויתנחם‬ ‫ההוא אמר ולא יעשׂה‬ ‫דבר ולא יקימנה‬ ‫הנה לברך לקחתי‬ ‫אברך ולא אשׁיבנה‬ ‫לא־אביט און ביעקב‬ ‫ולא־ראה עמל בישׂראל‬

111

And YHWH210 met211 Balaam and set a word in his mouth and said: return to Balak and thus shall you speak. And he came212 to him and look!213 He was standing by his offering214 and the officers of Moab were with him. And Balak spoke to 215 him: what did YHWH say? So he raised his saying and spoke: Arise, Balak, and listen!216 Heed my testimony, son of Zippor.217 Not a man is El that he might lie, nor a mortal that he might regret. Does he speak and not do it? Has218 he spoken and not undertaken it? Look! To bless I was brought. I will bless and not take it back.219 I220 have not beheld any wickedness221 in Jacob.

Smr reads “messenger of YHWH”. G reads “God”. Both represent later harmoniza-

tions. 211

Smr reads “called”; cf. Num 23:3. Perhaps G’s reading supports M, commending it as

older. 212 The reading “returned” (found in one Heb. manuscript, G, and V) harmonizes with v. 6; cf. Seebass, Numeri III, 20. 213 The ‫ הנו‬of ML etc. most likely came about as a scribal error from ‫הנה‬. At least there is no reason that Balak should be so emphatically reintroduced; he is doing precisely what the reader expects of him. 214 See note 182. 215 It would be impossible to determine which preposition represents the older reading, ‫ל‬ or ‫אל‬. 216 Smr seems to read with an object suffix, dittography from the following word; contra Vetter, Seherspruch, 17. 217 The last phrase of the verse follows the understanding attested in G; cf. Albright, “Oracles of Balaam,” 214; Vetter, Seherspruch, 19 and Dorival, Nombres, 438. Contra Zacharias Frankel, Einfluss, 173 and Baruch A. Levine, “The Plaster Inscriptions from Deir ʿAlla: General Interpretation,” in The Balaam Text from Deir ʿAlla Re-Evaluated: Proceedings of the International Symposium Held at Leiden, 21–24 August 1989, ed. Jacob Hoftijzer, Gerrit van der Kooij (Leiden; New York; Copenhagen; Cologne: Brill, 1991), 181. 218 Though largely irrelevant, the best non-M witnesses (Smr, G, V) do not attest the conjunction, which is thus not reflected here. 219 M is corrupt; the translation reflects G et al.; contra Levine, Numbers 21–36, 182–83 and Seebass, Numeri III, 20–21, who maintain M. 220 Contra Albright, “Oracles of Balaam,” 214 who emends the verb to a third-person passive. The verb in the first person is supported by Smr, G, S, and T. M increases the focus on God, away from Balaam. Cf. Vetter, Seherspruch, 17. 221 It would be impossible to decide with certainty between M and Smr since the semantic ranges of the different words overlap. See the discussion above.

112

The Text of Numbers 22–24 ‫יהוה אלהיו עמו‬ ‫ותורעת מלך בו‬

22 23

24 222

‫אל מוציאו ממצרים‬ ‫כתועפת ראם לו‬ ‫כי לא־נחשׁ ביעקב‬ ‫ולא־קסם בישׂראל‬ ‫כעת יאמר ליעקב‬ ‫ולישׂראל מה־פעל אל‬ ‫הן־עם כלביא יקום‬

Nor has he222 seen any evil in Israel. YHWH, his God, is with him and a king’s223 jubilation224 is with him. El, who is bringing him225 out of Egypt, is like the horns226 of a wild bull for him.227 Yes, there is no soothsaying in/against Jacob and no divination in/against228 Israel. At that time it will be said to Jacob and to Israel: what will229 El do?230 Look! A people like a lioness231 will arise

The first person in S is most likely a harmonization with the preceding phrase. It is unclear why G reads “kings’” in the plural, however it is unlikely a messianic interpretation: “The reference [to rulers in Num 23,21 LXX] is not necessarily eschatological, and it does not focus on an individual messiah.” John J. Collins, “Messianism and Exegetical Tradition,” 142. 224 Metathesis is responsible for the Hebrew reading (‫)ותרועת < ותורעת‬. My translation follows G et al.; cf. Vetter, Seherspruch, 18. 225 “Him” in the singular, as often attested, refers to the people being brought out and is older. Contra Albright, “Oracles of Balaam,” 215, no need to emend the verb presents itself. 226 Assuming that G accurately translated the Vorlage here, it remains unclear how that Vorlage read. The most obvious reconstruction would be ‫( כבוד‬e.g. Num 14:10, 21–21; 16:19; 17:7 etc.), but it would be difficult to explain what processes stood behind the misunderstanding of either G or M. Since the attestations of ‫ תועפות‬in this pericope make up fifty percent of the occurrences in the Hebrew Bible (the others are in Job 22:25 and Ps 95:4), it is not unlikely that the translator had some issues coming up with an adequate equivalent and thus chose δόξα as the best (perhaps metaphorical) possibility; cf. Joachim Schaper, Eschatology in the Greek Psalter, WUNT 2. Reihe, vol. 76 (Tübingen: Mohr Siebeck, 1995), 111 and Ziegert, Diaspora, 257–58. For the identification of ‫ ראם‬in G (and other witnesses), cf. Ziegert, Diaspora, 258: “Das seltsam anmutende μονόκερως ist ein Standardäquivalent für ‫ …ראם‬Denkbar ist durchaus, dass die Ursache in der hellenistischen Mythologie zu finden ist und dass das Einhorn dem Übersetzer als Ersatz für den in orientalischen mythologischen Texten ebenfalls prominenten Wildstier diente.”; and Vermes, Scripture, 153: “Apart from the Septuagint and the Vulgate, which translate ‫ ראם‬as ‘unicorn’ and ‘rhinoceros’, Jewish interpreters find it unfitting to compare God to an animal, and in place of ‫ראם‬, they read ‫רום‬, ‘exaltation’.” 227 Even if there were a reason to regard this verse as an addition from 24:8, that is not a text-critical matter; cf. the literary-critical discussion in the following chapter. 228 For a discussion about whether the preposition should be understood as “in” or “against”, cf. Seebass, Numeri III, 21. In my opinion, the tradents of this text left this intentionally ambiguous, permitting both meanings. 229 The repointing suggested in BHS better represents the sense of the verse; cf. G. Cf. also Vetter, Seherspruch, 22, who reads it as a participle form, but translates it with the present tense. 230 There is no reason to follow BHS in presuming that this verse is an addition. See the discussion in the next chapter. 231 The variants for “lion” in this verse (as in 24:9 below) are not text-critically relevant. The distinction between these terms remains unclear; cf. Brent A. Strawn, What is Stronger Than a Lion? Leonine Image and Metaphor in the Hebrew Bible and Ancient Near East, 223

2.3 Translation of the Restored Text of Numbers 22–24

25

26

27

‫וכארי יתנשׂא‬ ‫לא ישׁכב עד־יאכל טרף‬ ‫ודם־חללים ישׁתה‬ ‫ויאמר בלק אל־בלעם גם־קב לא‬ ‫תקבנו גם־ברך לא תברכנו‬ ‫ויען בלעם ויאמר אל־בלק הלא‬ ‫דברתי אליך לאמר כל אשׁר־‬ ‫ידבר יהוה אתו אעשׂה‬ ‫ויאמר בלק אל־בלעם לכה־נא‬ ‫אקחך אל־מקום אחר אולי יישׁר‬ ‫בעיני האלהים וקבתו לי משׁם‬

28

‫ויקח בלק את־בלעם ראשׁ‬ ‫הפעור הנשׁקף על־פני הישׁימן‬

29

‫ויאמר בלעם אל־בלק בנה־לי‬ ‫בזה שׁבעה מזבחת והכן לי בזה‬ ‫שׁבעה פרים ושׁבעה אילים‬ ‫ויעשׂ בלק כאשׁר אמר בלעם‬ ‫ויעל פר ואיל במזבח‬

30

1

‫וירא בלעם כי טוב בעיני יהוה‬ ‫לברך את־ישׂראל ולא־הלך‬ ‫כפעם־בפעם לקראת הנחשׁים‬ ‫וישׁת אל־המדבר פניו‬

113

and like a lion it will raise itself up.232 It will not lie down until it eats prey, and blood of the pierced it will drink. And Balak spoke to Balaam: you should definitely not curse it again. 233You should definitely234 not bless it again. Balaam answered and spoke to Balak: Did I not speak to you, saying every word235 that YHWH236 says, that I will do? And Balak spoke to Balaam: come on!237 I will take you to another place. Perhaps it will be upright in the eyes of God so that you can curse it for me from there. So Balak took Balaam to the pinnacle of Peor which looks down over the surface of the Yeshimon. And Balaam spoke to Balak: build for me here seven altars and prepare for me here seven calves and seven rams. So Balak did just as Balaam said238 and offered a calf and a ram at the altar. Numbers 24 When Balaam saw that it was good in the eyes of YHWH to bless Israel, he did not go like the previous times to proclaim the239 divinations.240 Rather, he set his face to the desert.

OBO, vol. 212 (Fribourg; Göttingen: Academic Press and Vandenhoeck & Ruprecht, 2005), 311–16. 232 No need to append the conjectural ‫גי‬, as Albright, “Oracles of Balaam,” 216 suggested, exists. 233 Smr’s addition of the conjunction is text-critically insignificant. 234 See note 201. 235 M reads “all/every”. G reads “the word”. Smr reads “every word”. Smr is perhaps oldest, with one element missing in each other version. Alternatively, one could consider whether Smr attests a conglomeration of the other two. 236 Smr, G, and V attest “God”. 237 No emendation required; cf. 23:13. 238 It would be impossible to distinguish the older reading between Smr (‫ )דבר‬and M (‫)אמר‬, since the semantic values essentially entirely overlap. 239 The article has been added to reflect Smr and G. M lost it due to haplography caused by the similarity to ‫ת‬. Contra Seebass, Numeri III, 22, who saw the article as an attempt to make Balaam more wicked. 240 4QNumb contains a plus at this point vis-à-vis all other traditions. Cf. the proposed reconstruction of the lacuna’s missing text in Jastram, “The Text of 4QNumb,” 188–90. It should be rejected as secondary.

114 2

The Text of Numbers 22–24 ‫וישׂא בלעם את־עיניו וירא את־‬ ‫ישׂראל שׁכן לשׁבטיו ותהי עליו‬ ‫רוח אלהים‬ ‫וישׂא משׁלו ויאמר‬ ‫נאם בלעם בנו בער‬ ‫ונאם הגבר שׁתם העין‬

3

4 ‫אשׁר מחזה שׁדי‬ ‫יחזה נפל וגלוי עיניו‬ ‫מה־טבו אהליך יעקב‬ ‫משׁכנתיך ישׂראל‬ ‫כנחלים נטיו‬ ‫כגנת עלי נהר‬ ‫כאהלים נטה יהוה‬ ‫כארזים עלי־מים‬ ‫ אדם מזרעו‬/ ‫יאזל־אישׁ‬ ‫ומשׁל בעמים רבים‬ ‫וירם מאגג מלכו‬ ‫ותנשׂא מלכתו‬ ‫אל נחהו ממצרים‬ ‫כתועפת ראם לו‬

5 6

7

8

241

Balaam lifted his eyes and saw Israel camping according to its tribes and the spirit of God was upon him. So he lifted his saying and spoke: oracle of Balaam, son of Beor. Oracle of the young man of the opened241 eye, Oracle of the hearer of El’s speaking,242 the who sees the vision of Shaddai; one who falls,243 yet his244 eyes are open. How good are your tents, Jacob; Your245 bivouacs, Israel? Like vales that are stretched out,246 like gardens upon a river, like tents YHWH stretches, like cedars upon the water.247 A man will come from his seed and rule many peoples. Greater than Agag will his king be and his kingdom will be exalted.248 El led249 him from Egypt, like the horns250 of a wild bull for him.

It is unclear how G arrived at its reading. There is no need to emend the Hebrew; cf. Levine, Numbers 21–36, 191–93. For the possible interpretations and meaning of ‫שתם‬, cf. Vermes, Scripture, 156; Milgrom, Numbers, 203; and Ziegert, Diaspora, 246–47. 242 This late gloss cannot be found in G or Smr, nor can it be reconstructed in Q. It was added from v. 16 at the latest stage of textual transmission. N.b. It is sometimes behind the asteriskos in G-manuscripts and the Syro-Hexapla. This position is contra, e.g., Albright, “Oracles of Balaam,” 217 and Seebass, Numeri III, 22. The assertion that 16aβ should be inserted is baseless. 243 G’s Vorlage, assuming it was accurately translated, remains impossible to reconstruct. The translator apparently understood this as a reference to sleep (cf. 22:8–13; Levine, Numbers 21–36, 189–90; and Ziegert, Diaspora, 247–48), as opposed to ecstatic prophecy (cf. Num 12:6 and Dorival, Nombres, 70, 139, and 445). 244 Suffix added with G. M and Smr lack the ‫ ו‬due to haplography, presumably an oversight due to the ‫נ‬. 245 The conjunction was presumably added in Smr, S, T, and V based on poetic considerations. Contra Vetter, Seherspruch, 28, who saw it as original based on poetic meter. 246 It is unclear how G arrived at “shady”. Possibilities: Heb. √‫ סכך‬or Aram. √‫טלל‬. Cf. Vermes, Scripture, 158. 247 This verse is corrupt in M and was dealt with above. There’s no apparent reason to transpose “tents” and “cedars” here as Gray, Numbers, 362–63 and Albright, “Oracles of Balaam,” 217 have proposed. 248 The difficulties of this verse, which have led some to even abandon reconstruction (cf., e.g., Albright, “Oracles of Balaam,” 218), was addressed in the preceding. 249 This translation reflects G and Smr; M was presumably changed to reflect 23:22. Cf. Albright, “Oracles of Balaam,” 218; Vetter, Seherspruch, 28; and Ziegert, Diaspora, 257. 250 Cf. note 226.

2.3 Translation of the Restored Text of Numbers 22–24

9

10

11

12

13

251

‫יאכל גוים צריו‬ ‫ועצמתיהם יגרם‬ ‫וחצו ימחצו‬ ‫כרע שׁכב כארי‬ ‫וכלביא מי יקימנו‬ ‫מברכיך ברוך‬ ‫וארריך ארור‬ ‫ויחר־אף בלק אל־בלעם ויספק‬ ‫את־כפיו ויאמר בלק אל־בלעם‬ ‫לקב איבי קראתיך והנה ברכת‬ ‫ברך זה שׁלשׁ פעמים‬ ‫ועתה ברח־לך אל־מקומך‬ ‫אמרתי כבד אכבדך והנה מנעך‬ ‫יהוה מכבוד‬ ‫ויאמר בלעם אל־בלק הלא גם‬ ‫אל־מלאכיך אשׁר־שׁלחת אלי‬ ‫דברתי לאמר‬ ‫אם־יתן־לי בלק מלא ביתו כסף‬ ‫וזהב לא אוכל לעבר את־פי‬ ‫יהוה לעשׂות טובה או רעה מלבי‬ ‫אשׁר־ידבר יהוה אלהי אתו‬ ‫אדבר‬

115

He will devour nations,251 his enemies. Their strengths he will gnaw252 and his arrow253 will strike through.254 He crouches, he lays like a lion, and – like a lioness – who will raise him? Whoever blesses you is blessed, and whoever curses you is cursed. Balak’s anger burned against Balaam, and he slapped his palms together. Balak spoke to Balaam: to curse my enemy I called you,255 but look! You have indisputably blessed 256 [him] these three times.257 And now, flee for yourself to your place! I said that I would certainly make you rich, but look! YHWH has refused to honor you! So Balaam spoke to Balak: did I not also speak to your messengers whom you sent to me, saying: Were Balak to give me the contents of his house, silver, and gold, 258 I would not be able to transgress259 the mouth of YHWH to do good or wicked from my heart. That which YHWH my God260 said, I have said it.

There is neither evidence nor need for the deletion of “nations” as BHS suggests. One could consider the removal of the ‫ם‬, and reconstruct the cs. as presupposed by G; cf. Ziegert, Diaspora, 258. 252 It is unclear how G arrived at its reading; cf. Dorival, Nombres, 447; Rösel and Schlund, “Arithmoi,” 491 and Ziegert, Diaspora, 258–59. 253 S does not offer a compelling emendation. M should be emended so that the subject matches the verb in sg. 254 G reads “and with his arrows he will shoot an enemy” (‫ )ובחציו ירה צר‬instead of ‫וחציו‬ ‫ימחץ‬. The difficulties in M led Seebass, Numeri III, 24 to refuse to translate it. However, cf. Ziegert, Diaspora, 259 for a resolution to the problem. 255 A text from the Cairo Genizah reads “I took you”; Smr reads “I called to you”. 256 Smr reads the participle as in 23:11 and 25. 257 It would be impossible to determine whether M (‫ )פעמים‬or Smr (‫ )רגלים‬offers the older reading. 258 Cf. note 151. 259 Smr does not attest an actual variant (it is merely written plene). The first pers. impf. attested in some Smr manuscripts is an error, as is clear from the syntax. 260 G reads “God”. Smr adds “to me”; several manuscripts and V include “my God”; cf. 22:18. The combined evidence suggests that the oldest reading was indeed the conjecture ‫יהוה אלהי‬. Most witnesses of M deleted (intentionally or otherwise) ‫אלהי‬, which is still reflected to some degree in G’s ὁ θεός (= ‫ אלהים‬or ‫ )אל‬and Smr’s ‫( אלי‬the ‫ ה‬could have disappeared as a result of haplography; ‫ ה‬and ‫ י‬are quite similar in both the paleo-Hebrew and Samaritan scripts).

116 14

15

16

17

The Text of Numbers 22–24 ‫ועתה הנני הולך למקומי לכה‬ ‫איעצך אשׁר יעשׂה העם הזה‬ ‫לעמך באחרית הימים‬ ‫וישׂא משׁלו ויאמר‬ ‫נאם בלעם בנו בער‬ ‫ונאם הגבר שׁתם העין‬ ‫נאם שׁמע אמרי־אל‬ ‫וידע דעת עליון‬ ‫מחזה שׁדי יחזה‬ ‫נפל וגלוי עיניו‬ ‫אראנו ולא עתה‬ ‫אשׁרנו ולא קרוב‬ ‫יזרח כוכב מיעקב‬

Now, look! I am going to my place.261 Come,262 I will report263 to you what this people will do to your people in later days.264 And he raised his saying and spoke: oracle of Balaam ben Beor. And an oracle of the young man of opened eyes.265 An oracle of266 the hearer of El’s speeches and the knower of Elyon’s knowledge. A vision of Shaddai he sees, one who falls, yet his eyes are opened.267 I will see him,268 but not now. I will consider him blessed,269 but not nearby. A star270 will ascend271 from Jacob

261 M: “to my people”; G: “to my place”; S: “to my land”. G might present a correlation of this verse with 24:11 and 25. S’s reading is less secure, possibly relating back to 22:5. However, the same could be said of M’s ‫עמי‬. The two uses of ‫ עם‬otherwise in this verse, could have also led to an error in M (and Smr), something that cannot be said of G. M (and Smr) can also be seen as an attempt to focus the verse on Balaam’s ethnic background in contrast to Balak’s and Israel’s. This matter will be further developed in the comments on the literary history of the Balaam story. Ultimately, the issue cannot be resolved with any clarity, but since G does not demonstrate any ideological change and S seems improbable as the older reading, G will be accepted as the oldest reconstructable version. From this, S focused on Balaam returning to his land, whereas the traditions of M and Smr focused on him returning to his people, contrasting him with the groups mentioned in 14b. 262 Smr reads without the paragogic ‫ה‬. 263 The errant reading in Smr is the result of metathesis. 264 Nothing indicates that ‫ באחרית הימים‬must be translated eschatologically as “in the last days” (as in G, KJV and NKJV, Luther’s [1984] “zur letzten Zeit”, and Zürcher’s “am Ende der Tage”); cf. the similar usage in Gen 49:10 and Deut 4:30 and 31:29, none of which must be understood as a reflection on the end of time. Cf. Noth, Numeri, 168; Levine, Numbers 21–36, 199. It need neither be understood messianically, as Guyot, “Prophecy of Balaam,” 336–37. 265 Cf. note 241. 266 G is missing the first word and thus explicitly equates the “hearer” in v. 16 with the “young man” in v. 15. This reading has been followed here. The other traditions harmonized. 267 Cf. note 243. 268 G understood this as a Hiphil. Seebass, Numeri III, 24 curiously noted that, “G punktiert das erste Verb messianisch als hi.” He did not consider that there is no evidence of G translators pointing texts nor did he explain how a text could be messianically pointed. 269 G understood this as coming from √‫“ אשׁר‬I will count him blessed”; cf. Gen 30:13. M resulted from an attempt to make 24:17 more consistent with 23:9. See the comments to this verse in the preceding discussion. 270 T interprets here, proffering “king”. 271 G (S and V) read “is raising”. BHS offers ‫ זרח‬as a possible translation back from the Greek; cf., e.g., Gen 32:32 and Judg 9:33 and Wellhausen, Composition des Hexateuchs, 351. Contra Seebass, Numeri III, 24. See the preceding discussion. Albright’s suggestion to regard the prepositional ‫ מ‬before ‫ יעקב‬and ‫ ישׂראל‬is not convincing; cf. Albright, “Oracles of Balaam,” 219.

2.3 Translation of the Restored Text of Numbers 22–24

18

19 20

21

22

‫ אדם מישׂראל‬/ ‫וקם אישׁ‬ ‫ומחץ ראשׁי מואב‬ ‫וקדקד כל־בני־שׁת‬ ‫והיה אדום ירשׁה‬ ‫והיה ירשׁה עשׂו‬ ‫וישׂראל עשׂה חיל‬ ‫וער מיעקב‬ ‫והאביד שׂריד מעיר‬ ‫וירא את־עמלק וישׂא משׁלו‬ ‫ויאמר‬ ‫ראשׁית גוים עמלק‬ ‫ואחריתו עד יאבד‬ ‫וירא את־הקיני וישׂא משׁלו‬ ‫ויאמר‬ ‫איתן מושׁבך‬ ‫ושׂים בסלע קנך‬ ‫ואם־יהיה לבער קן ערמה‬

117

and a man272 will rise from Israel and shatter the heads273 of Moab and the forehead274 of all of the sons of Seth.275 Edom will be a possession276 and Esau277 will be a possession of his enemies.278 But Israel will be a doer of might279 and he will rise up280 from Jacob and he will destroy a survivor from a city.281 And he saw Amalek and raised his saying and spoke: first of the nations was Amalek, but his progeny will perish yet.282 And he saw the Cainites [Kenites?] and raised his saying and spoke: permanent is your dwelling and set in the boulder is your nest.283 When a nest of treachery will be consumed,

272 G: “a man”; M: “a tribe” or “a scepter” (probably a play on words; cf. 24:2); S: “a prince”; Targum: “the Messiah”. See the discussion to these variants above. 273 M: “temples (anatomical)”; G et al.: “heads (political)”. 274 M: “tear down (?)”; Smr and JerM 48:45: “the crowns/foreheads”. 275 JerM 48:45: “sons of tumult”; T: “all the sons of man”. Albright, “Oracles of Balaam,” 220 left the reading intact, but understood it as a reference to the Šutu, group that can hardly be identified, but which he related to proto-Moabites. Sayce, “Balaam’s Prophecy,” 5 recognized it as a parallel for Moabites, which is probably accurate, though his religioushistorical ruminations stemming from this evaluation do not merit further consideration. 276 There is no need to emend the text with Smr or accept the conjecture of BHS. 277 Whereas M reads “Seir”, Smr and G read “Esau”, likely the older reading. “Seir” functions as a synecdoche for Esau in M and was probably intentionally emended. Bert Dicou, Edom, Israel’s Brother and Antagonist. The Role of Edom in Biblical Prophecy and Story, JSOTSup, vol. 169 (Sheffield: JSOT Press, 1994), 171 noted that the Chronicler used Edom and Seir synonymously, which could indicate recensional activity conforming to this ideology. Alternatively, it could have resulted from reading errors in which ‫ ו‬was misread as ‫ י‬and ‫ שׂ‬and ‫ ע‬were transposed. The ‫ ר‬would have been added to make the text clearer. However, the intentional change is probable. Cf. Ps 68:22. 278 “His enemies” was inappropriately added to this verse under the influence of Ps 68:22. 279 The conjectures offered by BHS in this verse are without merit. 280 Read ‫ וער‬with the presumed G Vorlage; cf. the discussion above. 281 The textual conjectures of BHS in this verse are without merit, contra Achenbach, Vollendung, 423. He sought to correct Num 24:19 in the sense of BHS’s conjecture to ‫( והאביד שׂריד מסעיר‬sic Achenbach; he meant ‫ משׂעיר‬for the last word) in accordance with Obad 18. While this is certainly theoretically possible, no text-critically relevant evidence commends this reconstruction and it thus appears like a modern midrashic attempt to make texts read more consistently. However, it seems most likely that this element should be removed for literary-critical reasons. Cf. the discussion there. 282 Cf. the discussion to Smr and G above. 283 The BHS’s conjecture in this verse is without merit.

118

The Text of Numbers 22–24

Ashur will take you prisoner.284 And he saw Og and raised his saying and spoke: oy, who will remain alive whenever El sets this?285 24 And he will come out from the hand of Kittim, He will oppress Assyria and oppress Eber [Hanahar?], but even they will be destroyed together. 25 ‫ ויקם בלעם וילך וישׁב למקמו‬And Balaam got up and walked and returned to his ‫ וגם־בלק הלך לדרכו‬place and also Balak walked to his way. Table 10: Reconstruction and Translation of Numbers 22–24 23

‫אשׁור תשׁבך‬ ‫וירא את־עג וישׂא משׁלו ויאמר‬ ‫אוי מי יחיה משׂמו אל‬ ‫ויצא מיד כתים‬ [‫וענה אשׁור וענה־עבר ]הנהר‬ ‫וגם־המה יחדו יאבד‬

2.4 The Emedations to M in Numbers 22–24 2.4 The Emedations to M in Numbers 22–24

Based on the observations presented above, ML requires the following emendations: 22:4: add ‫ הזה‬after ‫ הקהל‬with Smr, G, S, and V. 22:5: add ‫ ו‬to ‫ הנה‬with the witnesses against ML. 22:10: add ‫ לאמר‬at the end of the verse with Q, G, S, and V. 22:13: read ‫ אדנכם‬with G for M’s ‫ארצכם‬. 22:18: add ‫ בלבי‬with Q and G. 22:33: read ‫ ולולי‬with G et al for M’s ‫אולי‬. 22:33: read ‫ אתך הכיתי‬with Smr (support for distinct verb in G) for M’s ‫אתכה הרגתי‬. 23:2: delete ML’s ‫ בלק ובלעם‬with G and other Hebrew manuscripts. 23:3: change M’s ‫ יהוה‬to ‫ אלהים‬with Q, Smr, and G. 23:3: insert ‫ ובלעם נקרה אל אלהים‬before M’s last two words with Q and G. 23:10: read ‫ משפחת‬with G (supported by Smr?) for M’s ‫את רבע‬. 23:13: emend to ‫ לכה‬with the Qere. 23:15: remove ‫ כה‬from ML with other Hebrew manuscripts, Smr, and G. 23:15: add ‫ אלהים‬to the end of the verse with G. 23:17: emend ML’s ‫ והנו‬to ‫ והנה‬with the other Hebrew manuscripts, Smr, and TJ. 23:19: delete the ‫ ו‬in ‫ ודבר‬with Smr, G, and V. 23:20: replace M’s ‫ וברך‬with Smr’s and G’s ‫( אברך‬which TJ also supports). 23:21: replace M’s ‫ הביט‬with ‫ אביט‬as in Smr, G, S, and T. 23:21: read ‫ ותורעת‬as in G, S, and T instead of M’s ‫ותרועת‬. 23:22: read ‫ מוציאו‬instead of ‫ מוציאם‬with one Hebrew and many Greek Manuscripts, La, and V. 24:4: delete 24:4a with Smr and G.

284

See the discussion to vv. 22–24 below. Cf. G and V. The G-Vorlage may have read ‫ ב‬instead of ‫מ‬, as the translator of Num often uses ὅταν as a translation of Heb. ‫ ב‬+ inf. cs. (50% of occurrences in G-Num; cf. Num 4:5; 8:2; 9:19; 10:7; 11:9; 15:19; 18:30; 28:26; and 35:19). Levine, Numbers 21–36, 205 reads, “who will survive from the northland”, whereas Vetter, Seherspruch, 49–50 reads, “islands will assemble from the north”. He based this on ancient Egyptian data from the time of the supposed exodus; cf. Vetter, Seherspruch, 55. 285

2.5 The Character of the Textual Traditions

119

24:4: read ‫ עיניו‬with G (and Smr?) for M’s ‫עינים‬. 24:6: read ‫ נטה‬with Smr, G, and V for M’s ‫נטע‬. 24:7: read with G for the first five words: ‫יאזל איש מזרעו ומשל בעמים‬. 24:8: replace M’s ‫ מוציאו‬with Smr’s and G’s ‫נחהו‬. 24:8: emend the verb ‫ ימחץ‬to ‫ ימחצו‬so that the subject (universally plural) and verb match. 24:13: add ‫ אלהי‬after ‫ יהוה‬with several Hebrew manuscripts and V (supported by Smr and G). 24:14: read ‫( למקומי‬with G) for M’s ‫לעמי‬. 24:16: read ‫ עיניו‬with G (and Smr?) for M’s ‫עינים‬. 24:17: delete the ‫ ו‬in ‫ אשורנו‬with G. 24:17: replace M’s ‫ דרך‬with ‫יזרח‬, as in G, S, and V. 24:17: replace M’s ‫ שבט‬with ‫ איש‬or ‫ אדם‬with G (supported by others). 24:17: replace M’s ‫ פאתי‬with G’s ‫( ראשי‬supported by S, T, TJ, and V). 24:17: replace M’s ‫ קרקר‬with Smr’s ‫( קדקד‬found also in Jer 48:45M). 24:18: read ‫ עשו‬with Smr and G for M’s ‫שעיר‬. 24:18: delete ‫ איביו‬from the end of 18a (cf. Ps 68:22). 24:19: change ‫ וירד‬to ‫ וער‬with G. 24:20: M’s ‫ עדי אבד‬incorrectly divided the words as found in Smr: ‫( עד יאבד‬cf. G). 24:22: read ‫ ואם‬with G for M’s ‫כי אם‬. 24:22: remove the ‫ י‬in M’s ‫ קין‬with G in 22b. 24:22: G reads ‫ערמה‬, which is favored, instead of M’s ‫עד מה‬. 24:23: Add ‫ וירא את עג‬to the beginning of 23 with G. 24:24: read (combined from G and Smr) ‫ ויצא‬for M’s ‫וצים‬. 24:24: read ‫ וענה‬for M’s ‫ וענו‬in 24. 24:24: correct M’s ‫ הוא‬to ‫המה‬, cf. G. 24:24: correct M’s ‫ עדי אבד‬to ‫ יחדו יאבד‬based on G and Smr.

2.5 The Character of the Textual Traditions 2.5 The Character of the Textual Traditions

At this stage some reflections on the character of each textual tradition based on the observations of the aforementioned section are in order. The results here concentrate on the (proto-)Masoretic, (pre-)Samaritan, and Septuagint traditions – in reverse order – and their variation from the reconstructed older state of Numbers 22–24.

2.5.1 The Septuagint Tradition A number of interpretative elements have found their way into the Greek textual tradition known under the rubric “Septuagint”. Yet, few of the differences identified in the Septuagint attest a consonantal Hebrew Vorlage distinct from that of the proto-Masoretic tradition. At the same time, it cannot be ignored that some of the idiosyncrasies of G indicate a Hebrew Vorlage distinct from M and Smr. Both of these matters will be briefly addressed in the following comments.

120

The Text of Numbers 22–24

Several differences between G and the reconstructed older text demonstrate a tendency toward consistency, both within these chapters and in broader biblical contexts.286 Examples of this can be identified: 1) in 22:18, in which G attests “officers” as found in many other verses of the pericope; 2) in 22:35, which harmonizes the text with 23:12 (“you shall guard to say”); 3) the plurality of offerings (cf. 23:1) and the mention of the spirit of God (cf. 24:2) in 23:6; and 4) the verb “return” in 23:17, which matches that in 23:6. One theological harmonization should be mentioned: 23:19 presents a more harmonized image of God when considered in the light of texts like Gen 6:6 and Jonah 3:10. In these few ways, G presents a text more consistent than the reconstructed older version of the text. A few differences appear to have resulted from scribal errors. In 22:32b, G reads οὐκ αστεία, which appears to have resulted from an incorrect reading (‫הרע‬ as in Smr instead of ‫ ירט‬as in the reconstructed text). The translator of G also misunderstood the referent ‫ קרית חצות‬in 22:39 and translated it as “cities of the encampments” (πόλεις ἐπαύλεων), apparently confusing ‫ חצות‬and ‫חצרות‬, an understandable error.287 The number of these errors could be readily increased. Some idiosyncrasies in G came about due to minor reinterpretations of the text. The reference to the πρέσβεις in 22:5 may be an attempt to integrate the two groups of elders (γερουσία) attested in v. 7, as Wevers suggested. 288 In terms of style, G presents a better parallel structure in 23:10b than in the other witnesses or in the reconstructed text. A few elements in G should also be viewed as clarifying glosses, such as the added subject Balaam in 23:4 and 15. Other distinctive elements reflect a refocusing of the text in other ways. Probably the most interesting category of differences can be found in interpretations of the text. For example, Balak’s use of the plural in 22:6 G transposes the conflict away from a personal level: no longer does Balak alone feel inferior to Israel; rather, even the whole of his people, the Moabites, could not have overwhelmed the Israelites. Such a minor distinction presents a major interpretive difference: Israel is made to look even stronger in G than in the other versions; the conflict is one of peoples and less of personalities. A similar understanding lay presumably behind the curious – and honestly, inaccurate – translation of “dust” with “seed” in 23:10. Such an interpretation can only be understood with a text like Gen 13:16 in its background. The common understanding of the vastness of the people of the promise should not be overlooked. Theologically, it should be noted that G does not permit Balaam to refer to YHWH as “my God” in 22:18 and 24:13. This quiescence regarding his personal

286

Cf. Zacharias Frankel, Einfluss, 174–75. Cf. Wevers, Notes on the Greek Text of Numbers, 382. 288 Cf. Wevers, Notes on the Greek Text of Numbers, 361. Rösel and Schlund, “Arithmoi,” 481 followed this understanding. 287

2.5 The Character of the Textual Traditions

121

conviction distances the God of Israel from this foreigner to a degree not attested in, e.g., M. One should note G’s reference to the “honor” of God, reflected exemplarily by the unicorn, in 23:22 and 24:8. Finally, G could evince a somewhat more apparent eschatological understanding in its reading of “Gog” in 24:7: the person who comes from Jacob/Israel will be greater than the kingdom of Gog. This understanding does not mesh entirely with Ezekiel 38–39, which explicitly pronounces that God will be the one who dominates Gog, whether in M or in G. Though it should not be entirely excluded that the reading in G resulted from an error (‫ גג‬instead of ‫אגג‬, as in M), G suggests that a person will succeed above and beyond the figure Gog. However, it would be going too far to suggest that G as a whole in Numbers 24 attests a more messianic text than M. Consider the following: “There are significant differences between the Greek and the Hebrew [of Num 24,7.17]; these cannot be entirely explained in textual terms or in terms of the methods of translation followed by the translator; the introduction of the name ‘Gog’ already gives v. 7 an eschatological thrust; the ‘man’ has a kingdom (v. 7) and acts as a military leader (v. 17) and thus is presented as a messianic ruler; and there is clear evidence in the [Dead Sea] Scrolls of a more or a less contemporary messianic interpretation of v. 17. Thus Num 24,17 is interpreted in a messianic way in CD VII.18–21 and in 4QTestimonia, although it shou1d be noted that in 1QM XI.6–7, part of a reworked quotation of Num 24,17–19, the passage is interpreted with reference to the people of Israel.”289

This ignores that 1) we have a Hebrew reading of Gog in Smr, thus suggesting that this cannot be an interpolation from the translator; and 2) that the notices in the Dead Sea Scrolls attest the version of M and thus cannot be used as evidence for a messianic understanding in the Greek version. Rather, one should be more cautious in identifying messianic tendencies in one textual tradition over another, a lesson taught particularly through the discovery of the texts from Qumran. In sum, one notes that G (or, in some cases, possibly its Vorlage to some degree) demonstrates a tendency to clarify the text through additions and variations. At other times, G demonstrates ideological and theological variance from the reconstructed text (and the other known versions). It sometimes uses terms that attest the translator’s interpretive ability, such as the reference to the honor of the unicorn and the translation of “dust” with “seed”. Some of these variants suggest an attempt at inner-biblical exegesis. And yet, some of the distinctions of G appear reminiscent of simple errors in the textual tradition or ideological interpretation.

289

Michael A. Knibb, “The Septuagint and Messianism. Problems and Issues,” in The Septuagint and Messianism, ed. Michael A. Knibb (Leuven: University Press, 2006), 18.

122

The Text of Numbers 22–24

2.5.2 The (Pre-)Samaritan Tradition Like G, Smr demonstrates a number of differences from the oldest reconstructable text. Many of these differences follow patterns similar to those recognized in G, albeit often in other manners. Generally, it remains unclear whether these differences present an innovation or error in Smr or pre-Smr. Some examples will be presented in the following. Consistency appears to be an important factor in the Smr tradition of these chapters. For example, Smr presents more consistency about the deity’s interaction with Balaam, in that Smr more often attests the divine “messenger” than the other textual traditions; cf. Num 22:20; 23:4–5 and 16. Like G, Smr is more consistent in its attesting multiple altars in 23:6, presumably again under the influence of 23:1. On a syntactical level, Smr demonstrates consistency in Balak’s bringing Balaam to the various locations. For example, in 22:39 Balak brings him to Kiriath-huzoth, whereas in M and G they travel together. The singular verb with an object suffix better matches the other movements of these characters with each other; cf. 22:41; 23:14, and 23:28 (although this last example relies on an unattached object and not an object suffix). A similar remark could be made about the opening gloss of 24:16, which sets it parallel to 24:4. Other attempts at consistency go beyond the context of Numbers 22–24 or even the whole of Numbers, such as the reference to Balaam as an Ammonite in 22:5 (which suggests conformity with Deut 23:4–7) or the use of ‫ שׁבט‬in Num 24:17, just as in Gen 49:10. Balaam’s command in 23:1, that Balak “make” (‫ )עשׂה‬an altar (as opposed to “building” [√‫ ]בנה‬one), reminds the reader more of the language of Covenant Code (cf. Exod 20:24–25) rather than, e.g., the primeval history or patriarchal narratives (cf. Gen 8:20; 12:7–8; 13:18 [n.b., however, ‫ עשׂה‬in 13:4]; 22:9; 26:25; and 35:37). This could also be an attempt to maintain consistent terminology after the explication of the Covenant Code. Thus, like G, Smr demonstrates an affinity for consistency, though somewhat more pronounced than in G.290 At the same time, Smr presents evidence for a variety of errors in the transmission of its text. Several elements have been overlooked: ‫ לאמר‬in 22:10, the notice that the people “is dwelling across” from Balak in 22:11; ‫ בלבי‬in 22:18; the lengthy material in 23:3; the shorter reading in 23:15; the missing object suffix in 24:4; and the missing reference to Og in 24:23. Other errors in Smr include the plural as a referent to “the people” in 22:5, the mistaking of ‫ י‬for ‫ה‬ in 22:9, the incorrect verb form in 22:31, the mistaken readings 22:32b and 23:10, the metathesis in 23:21 and 24:14, and the mistakes in 24:24 (although these last instances may have resulted from a corrupted text). This brief list demonstrates that Smr is hardly free from mistakes in its textual tradition.

290

Cf. Tov, Textual Criticism, 80–83.

2.5 The Character of the Textual Traditions

123

Yet, again like G, some of the differences within Smr – both in regards to the reconstructed text and the other textual traditions – evince interpretive undertakings. For example, Smr implies that Balaam might have been cheaper in 22:18: it would not take silver and gold and everything in Balak’s palace to bribe him; any one of these things might have sufficed (should anything have sufficed at all). Like G, Smr refused to have Balaam refer to YHWH as “my God” in 22:18 and 24:13, a clear theological reinterpretation of the text. To this category one might also reckon the more consistent recourse than in the other witnesses to the divine messenger as an agent in Smr. The reinforced usage of this character increases the distance between Balaam and the deity and limits their direct interaction. In Smr, the only real interaction occurs in Balaam’s nocturnal encounters, which could easily – though not explicitly – be understood as dreams. Samaritanus adds “Balak and Balaam” as the subject in 23:2, making it explicit that both of these actors undertook the offering, whereas it remains somewhat ambiguous in G and the reconstructed text. The reference to Gog in 24:7 might also attest a theological understanding, in this case – and contrary to G – not referring to a person who will defeat Gog. Rather, like M, this could be a reference to God (the most apparent antecedent to “his” in 24:7 [M and Smr] syntactically speaking is YHWH in v. 6). Again, like G, one notices that Smr developed some interpretive elements over the course of transmission, some of them even in common with G (and M). The character of the Smr text reflects tendencies similar to those recognized in G, yet sometimes taking matters in distinct directions. Both G and Smr increased the amount of consistency in their text, particularly when considered over and against the older reconstructed text. Like G, Smr demonstrates a moderately adapted theological agenda, yet Smr increases the distance between God and Balaam more than could be found in G (or M, for that matter). Finally, Smr transmits a number of errors that crept in over the course of transmission, which led to some abbreviations in its text vis-à-vis the older reconstructed text.

2.5.3 The (Proto-)Masoretic Tradition Having covered both G and Smr, M must be examined regarding its distinctions from the older version as well as regarding its idiosyncrasies. Here as with Smr it should be noted that it remains essentially impossible to distinguish in each case whether a difference in the text of M stems from M or proto-M. Therefore, the two will be addressed as the same, just as Smr and pre-Smr and G and its Vorlage were generally regarded as identical. Again, M demonstrates a number of errors and the development of the older text in a distinct manner, with its own unique emphasis. Again, one cannot overlook the tendency toward consistency in M, just as in G and Smr.

124

The Text of Numbers 22–24

A number of cases demonstrate that elements were lost over the course of transmission in the M tradition. Examples of missing words and particles have been identified in Num 22:4, 10–11, 18, 32b, 33, 23:3, 15; 24:1, 4, and 7. Most of these appear to be errors in the text. Some errors resulted not in shorter, but rather distinct readings, like ‫ ירט‬for ‫ ירע‬in 22:32b, the misreading of ‫ אולי‬for ‫ ולולי‬in 22:33, the erroneous reading “quarter” in 23:10, the confusion of ‫ ו‬for ‫ה‬, the incorrect pointing of 23:18, the metathesis in 23:21, and the various errors in 24:20–23. Other changes appear to have been intentional, though they often change the meaning little: the addition of the article in 22:11, the consistency of the verb ‫ נכה‬in 22:33 with 22:29, the harmonization of the name ‫ במת בעל‬in 22:41 with Numbers 21:19–20, the addition of the subject Balak and Balaam in 23:2, the addition of “thus” in 23:15, and even the very late addition in 24:4, as well as the consistent text of 24:6, which reflects the root ‫ נטע‬and the noun “aloes” in M, making the reference better match its botanical context. Some of M’s distinctions focus on the consistency of the text, whether within the pericope, within Numbers, or within the Bible. The aforementioned differences in 24:4, 22:41 and 22:33 could all be included under this rubric. Even the metathesis in 23:21 could have been inspired by the newly created relationship to Amos 2:2, rather than a mere accident. 291 Additionally, one should note the consistency between 24:8 and 23:22 in M, the replacement of ‫ מקום‬with ‫ עמי‬in 24:14, the additional phrase in 24:16, the common reading in 23:9 and 24:17, the use of ‫ שׁבט‬in 24:17 and Gen 49:10, the relationship to Psalms (particularly 68 and 110, but also 72) in Num 24:18–19, and the relationship to Dan 11:30 in Num 24:24. The absence of the character Og better matches the context following Numbers 21 in M than it does in G. How should the reader understand Balaam seeing him in Num 24:23, when he was killed in Bashan in Numbers 21? M resolves this problem. Such differences seem hardly coincidental and probably indicate intentional textual editing in a more systematic sense. Conspicuously, none of these similarities present themselves in G.292 Further evidence of editing can be found in ideological and theological differences to the other witnesses. Only M maintains Balaam’s reference to “my God” when referring to YHWH (22:18), but the other case is missing in M (24:13). The Masoretic Text (like Smr) includes Balaam with Balak in its first mentioning of the sacrifices before Balaam encounters God (Num 23:2). This may be a negative reflection on the character of Balaam, since he takes part in an offering that the audience – especially in later periods – must have regarded 291 Cf. Bogaert, “Heshbon,” 47–48: “Parmi les explications possibles, la plus raisemblable est qu’il s’agit d’un hybride | créé à partir de l’oracle sur Moab en Am 2,2, où le mot ‘Sha’on’ est utilisé, et de la phrase parallèle de Nb 24,17.” 292 Samaritanus can barely help in this case, since the only significant reinterpretation that occurs in the Torah appears at Gen 49:10. There, however, Smr contains the same terms as M in contrast to G.

2.6 Conclusions

125

as invalid. In 23:20 M precludes the understanding that it is in fact Balaam who carries out the blessing of Israel. Rather, God blesses. This transformation certainly contributes to the partial detriment of Balaam’s character. The use of ‫דלי‬ in 24:7 may present an oblique theological reference to Second Isaiah (Isa 40:15). In 24:17 a number of distinctions have been made: the coming destroyer of Moab has been depersonalized and the use of the verb ‫ דרך‬can be found in a number of soteriologically loaded texts. It can hardly be denied that the similarities to Dan 11:30 and Psalm 68 reflect a theological intention. Bearing these factors in mind, we note that M also follows a particular theological and ideological mind-set, moving some of the attention away from individual human agents and focusing on either the tribe (24:17) or God (24:7). This may have been a reaction to some developing distressed or disappointed messianic groups or hopes sometime after the translation of G, but there is no way to determine that with any certainty. With these data, it becomes apparent that M, like the other traditions, has its own approach to the text.293 Its approach demonstrates a unique ideology and theology within M, as well as indicating that M apparently had a particular interest in developing a consistent text, not only within the immediate context, but within the developing Tanakh. However, M demonstrates many errors and misunderstandings as well. Some of its verses also appear to have been corrupted over time, such as 24:6–7 and 24. Nonetheless, in spite of these factors, M presents a remarkably similar consonantal text to the older reconstructed text, being about one percent shorter and having been changed about forty times. Most of these changes are superficial, with substantial differences only in 23:3; 24:7, 17 and 24.

2.6 Conclusions 2.6 Conclusions

A number of conclusions must be drawn from the preceding discussion. First, no single witness generally proffers the oldest reconstructable text. Rather a more ancient ancestor stands in the background of the various traditions, which is why text-critical decisions must be made on a case by case basis. This textual ancestor can be reconstructed cautiously with a relative degree of certainty by considering the evidence available from the various witnesses. As these witnesses generally demonstrate very limited differences from each other, it is possible to both develop a plausible textual ancestor and reflect on the kinds of changes made within each textual tradition. This, quite frankly, speaks to the

293

This supports the findings in Adrian Schenker, “Der Ursprung des massoretischen Textes im Licht der literarischen Varianten im Bibeltext,” Textus 23 (2007): 51–67.

126

The Text of Numbers 22–24

great stability in the textual tradition in general. Nevertheless, it remains probable that other errors and changes remain in the reconstructed text that can no longer be identified due to the paucity of other ancient witnesses. That being said, the various traditions demonstrate several common tendencies. No tradition is free from errors. Each tradition has a tendency to move toward a more consistent text, whether in the immediate or macro context. This trend toward consistency seems most pronounced and broad in M, but can also be found in G and Smr. This tendency towards consistency reflects intertextual considerations within the traditions. That is, reading Numbers 22–24 within different textual corpora sometimes resulted in new readings or terminology. Occasionally these distinctions appear theologically motivated. At the same time, each textual tradition demonstrates corruptions and misunderstandings, leading to some differences in the Hebrew witnesses and especially in the Greek translation. The translators of G generally attempted to present their Vorlage text, but on some occasions, applied a limited amount of freedom to their translational technique. In M, it seems quite possible that the fourth oracle underwent very late recensional editing. The most prominent evidence for this can be found in the unique relationship to other biblical texts in M. Some of these texts demonstrate particular theologies, like the kingship of God in Psalm 68. Other elements focus on foreign peoples and Israel’s interaction with them. Such texts can be found particularly in Dan 11:30, Amos 2:2 and Jer 48:45 M. These texts remain important because their relationship is unique to M and not attested elsewhere. For example, Jer 48:45 does not even exist in G, meaning that Numbers 24 G could hardly have been edited toward it. This late recension would have been influenced by or would have also edited other texts in the HB, suggesting it could be one of the latest elements to have worked on the developing Tanakh. With some caution an older version of the Balaam story has been reconstructed. From this version of the text the other historical critical steps follow. Its reconstruction represents the mandatory preparation for the successive literary-critical undertakings and tradition-historical analysis in the following chapters. At the same time, recognizing and describing the attested changes in the versions provides a limited canon of processes for how texts may have changed over time. We have noted some of the ways that texts actually vary in the manuscripts and translations and how editors engaged with them. The question then becomes whether we can postulate and identify similar processes behind the texts we have and whether we can reconstruct sources and/or redactional diachrony behind these known texts. The recensional characteristics of some text-historical distinctions commend reflecting particularly on redactional-historical processes. Yet, these must be more securely demonstrated and – where possible – corroborated, not merely postulated. These considerations

2.6 Conclusions

127

lead automatically to the literary-critical investigation, to which we can now turn our attention.

Chapter 3

Literary Criticism Having considered the evidence for reconstructing the oldest Balaam textual tradition, we can turn our attention to the diachronic development of that story. As described in the introduction in Chapter One, several explanations for the existence of the Balaam narrative remain principally within the realm of theoretical possibility. The narrative could be a unity that contains no additions. It could be the product of the compilation of multiple sources. The development of the text could have started as a basic core that was expanded through one or several redactional undertakings. This core or these sources could have existed independently or as part of a longer composition or may have been composed for an extant literary context. These matters will be addressed in the following discussion, beginning first with a brief structural analysis. Then, some conspicuous problems that commend literary-critical engagement will be offered. Finally, I will conclude with a literary-critical analysis of Numbers 22–24 and identify, catalogue, and organize the text’s fractures before offering a relative chronology of the different elements identified in the text. This will ultimately contribute to the subsequent discussion on the place of these chapters in the redaction and literary history of the Bible. For the undertaking here, it will become clear that all of the literary-critical models advanced for appreciating Numbers 22–24 have some element of truth in them, but that none of them has satisfactorily addressed all of the problems. Numbers 22–24 stems from at least one primary source, with redactional elements added at several different stages over time.

3.1 Structural Analysis 3.1 Structural Analysis

The tale in Numbers 22–24 can be divided into several episodes and speeches.1 1) The Israelites’ arrival in Moab (22:1) 1 Contrast this organizational scheme with the simpler, albeit less convincing “two times three” scheme of Hans-Christoph Schmitt, “Der heidnische Mantiker als eschatologischer Jahweprophet. Zum Verständnis Bileams in der Endgestalt von Num 22–24,” in “Wer ist wie Du, HERR, unter den Göttern?” Studien zur Theologie und Religionsgeschichte Israels für Otto Kaiser zum 70. Geburtstag, ed. Ingo Kottsieper, et al. (Göttingen: Vandenhoeck & Ruprecht, 1994), 182–83.

3.1 Structural Analysis

2) 3) 4) 5)

6)

7)

8) 9) 10)

11)

2

129

Balak witnesses Israel destroying the Amorites (22:2) Moab fears the people (22:3a) Moab loathes the Israelites and complains to King Balak (22:3b–4) Balak’s first emissaries to Balaam (22:5–14) a. Balaam’s introduction and Balak’s declaration of intent (22:5–6) b. Arrival of the message and Balaam’s response (22:7–8) c. Balaam’s first encounter with God (22:9–12) d. Balaam’s refusal to go along (22:13) e. The return of Balak’s first emissaries (22:14) Balak’s second emissaries to Balaam (22:15–21) a. Balak’s declaration of intent and promise of fiduciary compensation (22:15–17) b. Balaam’s response and preliminary refusal of compensation (22:18–19) c. Balaam’s second encounter with God (22:20) d. Balaam’s departure (22:21) Balaam and his donkey (22:22–35a)2 a. Exposition: God’s anger and messenger (22:22) b. The donkey steps aside; strike one (22:23) c. The donkey steps aside; strike two (22:24–25) d. The donkey stops; strike three (22:26–27) e. First discussion: The donkey and Balaam (22:28–30) f. Second discussion: The messenger and Balaam (22:31–35a) Balaam’s journey to Balak (22:35b) Balak and Balaam’s first encounter and journey together (22:36–40) Balaam’s first oracle (22:41–23:12) a. Relocation (22:41) b. Preparation (23:1–3) c. Balaam’s encounter with God and return to Balak (23:4–6) d. Balaam pronounces his first oracle (23:7–10) e. Balak responds (23:11) f. Balaam responds (23:12) Balaam’s second oracle (23:13–26) a. Relocation (23:13–14a) b. Preparation (23:14b–15) c. Balaam’s encounter with God and return to Balak (23:16–17) d. Balaam pronounces his second oracle (23:18–24) e. Balak responds (23:25) f. Balaam responds (23:26)

Contrast this structural analysis with Kenneth C. Way, Donkeys in the Biblical World. Ceremony and Symbol, HACL (Winona Lake, IN: Eisenbrauns, 2011), 184–86.

130

Literary Criticism

12) Balaam’s third oracle (23:27–24:14a) a. Relocation (23:27–28) b. Preparation (23:29–30) c. Balaam remains and is overcome with the spirit of God (24:1–2) d. Balaam pronounces his third oracle (24:3–9) e. Balak responds (24:10–11) f. Balaam responds (24:12–14a) 13) Balaam’s final oracles (24:14b–24) a. Balaam advises Balak about Israel (24:14b) b. Introduction (24:15–16) c. Oracle against Moab and Edom (24:17–19) d. Oracle against Amalek (24:20) e. Oracle against Cain (24:21–22) f. Oracle about Kittim, Ashur, and Eber[-Hanahar (?)] (24:23– 24) 14) Balaam and Balak part ways (24:25) A cursory examination of the structure demonstrates several factors. Narrative material dominates the first portion of the pericope, while Balaam’s speeches dominate the latter half. In general the narrative operates through the recounting of various, mostly parallel-structured episodes. However, the story opens with several expositions and concludes abruptly following a series of brief oracles that is structurally, thematically, and theologically distinct from the rest. Particularly the narrative material about Balak’s emissaries and the framework of the first three oracles attest parallel structures. Even the most superficial reading confirms this structural conformity, but this must be nuanced to some degree as well. Some of the differences within the structure provide grounds for the application of diachronic explanations, whereas others appear stylistic or narratological. These will be addressed below. The lexical interdependence of much of the text demonstrates strong cohesion among the major portions of Number 22–24. Repetition of several phrases throughout the episodes suggests one compositional hand responsible for most of the text. For example, the phrase ‫ואך את־הדבר אשׁר־אדבר אליך אתו תעשׂה‬ with moderate variation, appear in 22:38b; 23:12b, and 26b. The concluding conversation between Balaam and Balak in 24:11–13 – before the intonation of Balaam’s final oracles – picks up elements from Balak’s second message to Balaam in 22:17aα (‫ )כבד אכבדך‬and 18aβb ( ‫לא אוכל לעבר את־פי יהוה אלהי‬ ‫)לעשׂות קטנה או גדולה‬. Such phrases structure the narrative and suggest a common background for much of the text. On a larger scale, the similar construction of episodes strongly suggests their narrative cohesion, as demonstrated in the Table 11: The Structure of Balak’s Missions to Balaam.

3.2 Literary-Critical Remarks and the Unity of the Composition V. 22:5aα 22:7b 22:8aα 22:8aβ

22:9a 22:12aα 22:13aα

Balak’s First Mission ‫וישׁלח מלאכים אל־בלעם בן־‬ ‫בעור‬ ‫ויבאו אל־בלעם וידברו אליו‬ ‫דברי בלק‬ ‫ויאמר אליהם‬ ‫לינו פה הלילה והשׁבתי אתכם‬ ‫דבר כאשׁר ידבר יהוה אלי‬ ‫ויבא אלהים אל־בלעם‬ ‫ויאמר אלהים אל־בלעם‬ ‫ויקם בלעם בבקר‬

Balak’s Second Mission ‫ויסף עוד בלק שׁלח שׂרים‬

V. 22:15abα

‫ויבאו אל־בלעם ויאמרו לו כה‬ ‫אמר בלק בן־צפור‬ ‫ויען בלעם ויאמר אל־עבדי בלק‬ ‫ועתה שׁבו נא בזה גם־אתם‬ ‫הלילה ואדעה מה־יסף יהוה‬ ‫דבר עמי‬ ‫ויבא אלהים אל־בלעם לילה‬ ‫ויאמר לו‬ ‫ויקם בלעם בבקר‬

22:16abα

131

22:18aα 22:19

22:aα1 22:aα2 22:21aα

Table 11: The Structure of Balak’s Missions to Balaam

However, other matters suggest that some elements must be regarded as later interpolations or editorial remarks on the text, which will be dealt with below.

3.2 Literary-Critical Remarks and the Unity of the Composition 3.2 Literary-Critical Remarks and the Unity of the Composition

Some tensions and contradictions cannot be overlooked in Numbers 22–24, demonstrating that it is probably not the exclusive product of a single hand.3 Models for reconstructing the compositional history of Numbers 22–24 have recognized diverse issues within the text that could evince the disparate heritage of some elements in the story. The most substantial can be covered briefly here. On a most general level, one quickly recognizes the repetitive style of the narrative: Numbers 22 recounts two Moabite missions to retrieve Balaam; Numbers 23 and 24 each attest a pair of oracles, making four in total; multiple offerings occur before the deliverance of these oracles. Noticing these repetitions, many (cf. Chapter One) have considered the possibility that more than one source or editorial hand stands behind these chapters. Often, to aid in this source- or redactional-critical appreciation, they recognized the variance in the nomenclature for the divinity: ‫ אלהים‬and ‫יהוה‬. Those observations present the most global, but a number of specific issues could also buttress the claim that multiple hands (in whatever fashion, for now) contributed to the composition of Numbers 22–24. 1)

3

The story opens on several occasions. Numbers 22:1 presents a solid exposition to the tale. Numbers 22:3 even presents two: Moab feared the people and Moab was

A more detailed refutation of such positions follows below.

132

2)

3) 4) 5)

6) 7) 8) 9)

10)

11) 12)

13) 14)

15) 16)

17)

Literary Criticism disgusted by the Israelites. The repetition of the subject “Moab” in v. 3 is particularly suspicious and syntactically superfluous. The multiple expositions could evince distinct hands in some fashion. The narrative mentions Balak in 22:2, failing to introduce him sufficiently until v. 4. The mention in v. 2 references an event outside the immediate context of Num 22:2, looking back to Num 21:32, the last time the Amorites appeared. In chapter 22, verses 4 and 7, the elders of Midian appear and just as quickly disappear. They do not fit the surrounding context in any fashion. Balaam’s heritage is described both as “Pethor on the river” and “in the land of the sons of his people” in 22:5. Balak’s message to Balaam in v. 6 remains distinct from Balaam’s quoting it in v. 11 (‫ קבב‬instead of ‫ארר‬, missing the element of blessing). However, God’s response to Balaam in v. 12 implicitly suggests it had been reported as in v. 6. To the Midianite elders in v. 7, cf. to v. 4 above (#3). Numbers 22:22 presents a marked, surprising contradiction to Num 22:20. In tone and content, the story of the donkey in Num 22:22–35 certainly differs from the rest of Numbers 22–24. Numbers 22:36 presents Balak going to meet Balaam, having heard that he has come. To some degree this could stand in tension with the surrounding material. This has led some to argue that Balaam never went to Balak in one version of the story, but that Balak came to him. This would mean that 22:36a would be redactional. Balaam’s noting his construction of altars in 23:4 stands in tension with v. 1–2. Balaam’s command to build the altars and sacrifice is absent in vv. 13–14, but appears again in v. 29. The repetition – roughly speaking – of 23:22 in 24:8 could indicate editorial copying. The interaction between Balaam and Balak in 23:25–26 reads like a satisfactory conclusion when succeeded immediately by 24:25, suggesting that 23:27–24:24 may have been from (an)other hand(s). The repetition of the subject “Balaam” in 24:1–2 without syntactical necessity is not without problems. Balaam’s lengthy self-introduction in 24:3b–4 has led some to regard this oracle as originally divorced from a context that contained preceding oracles. According to this logic, Balaam should be familiar both to the audience in the text and implied by the text, thus precluding the existence of any preceding oracles. Numbers 24:14b presents a marked break from anything that might have been readily anticipated after 24:14a. The repetition of material in the introductions of the third and fourth oracles could either imply a common compositional background against the first two oracles or scribal copying of one from the other. The Masoretic tradition strengthened this common character (cf. the note to 4a in the BHS). The introductions to other oracles in 24:20, 21, and 23 could indicate later expansions.

Not all exegetes recognize all of these matters as relevant; nor do I. But many agree on some or all of them as problematic for regarding Numbers 22–24 in its entirety as a compositional unity, as do I. These issues will be dealt with in

3.3 The Narrative’s Exposition(s): Numbers 22:1–4

133

the following pages, which focus on literary-historical observations from Numbers 22–24. At the same time, the volume of these contradictory elements and tensions should not be exaggerated. Only limited material in Numbers 22–24 can be identified as demonstrating tension with or standing in contradiction to its surrounding material. Such paucity of tension and contradiction implies only limited editorial activity, much like that identified in the previous chapter reflecting on the history of the text’s transmission. It remains improbable that the story arose from the combination of several initially disparate narrative versions that were part of extensive sources woven together to form the whole of the Pentateuch.4 Rather – with the exception of the lengthy insertion of Num 22:21–35a and some updates to the narrative embedding at the beginning and end of the story – Numbers 22–24 should be regarded as a generally systematically composed narrative unit. This position will be argued in detail in the following examination; beginning with the opening of chapter 22 and continuing following the narrative, relevant literary-critical problems will be dealt with generally as they occur in the text.

3.3 The Narrative’s Exposition(s): Numbers 22:1–4 3.3 The Narrative’s Exposition(s): Numbers 22:1–4

The opening verses of Numbers 22 present the exegete with a variety of expositions. Numbers 22:1, the currently most conspicuous beginning, reads like the conclusion to the preceding material in 21:32–35;5 the verse serves to bring the Israelites into the geographical context presumed by the Balaam narrative, i.e., with the people of the exodus encamped in or near Moab. The excursion to Bashan in 21:33–35, or even that to Jazer in 21:32, demands this return to Moab.6 That suggests that 21:32 and/or 33–35 and 22:1 belong together on a compositional level, or that Num 22:1 perhaps serves to afford the copying and insertion of Deut 3:1–3 into Numbers 21 (cf. Deut 2:26–3:11 in which the conquest of Og follows the conquest of Sihon). At the same time, Num 22:1 relies on the preceding material to identify where the Israelites are located, as only context dictates whence they set out (cf., e.g., Num 10:12; 20:22; 21:4, and 10– 13). Otherwise the notice remains incomplete. With this in mind, the only possibility available is that some editor must have inked Num 22:1 to create an 4

This will be addressed more expansively below. N.b. the Setumah in the Jewish reading tradition. To the difficulties reflected in the itineraries about the journey through or around Moab, cf., .e.g., J. Maxwell Miller, “The Israelite Journey Through (Around) Moab and Moabite Toponymy,” JBL 108, no. 4 (1989): 577–87. Some of these issues may stem from a long-standing misidentification of the Wadi Zered; cf. Yoel Elitzur, “Naḥal Zered in the Bible and the Baraita de-Teḥumin,” PEQ 145, no. 2 (2013): 108–18. 6 Only Isa 16:–89 // Jer 48:32 suggest Jazer (and Heshbon) belonged to the Moabites. 5

134

Literary Criticism

appropriate narrative context for an anecdote about Israelite occupation of and dominion over lands north of the Amorites. 7 Without Num 21:32–35, Num 22:1 remains superfluous. The group undertaking the exodus had already arrived three times in the region neighboring Moab (Num 21:11, 13, and 20). The combination of the verbs ‫ נסע‬and ‫ חנה‬commends its belonging to the same ideological milieu as Exod 13:20; 17:1; 19:2, and – more importantly for the context at hand – Num 21:10–13*, as well as Numbers 33.8 The reference to the exodus generation in 22:1 as the “children of Israel” distinguishes it from the material that follows; Numbers 22–24 otherwise attests this nomenclature only in Num 22:3b (see below).9 The previous usage can be found in the opening of the itinerary in Num 21:10; following the Balaam pericope one encounters it again in 25:6. These texts apparently stem from a common author or editor of Numbers, the Enneateuch, Hexateuch, or Pentateuch. Similarly, verse 22:2 references content found in 21:21–31*, implying that these materials may also stem from the same compositional level.10 The narrative mentions Balak for the first time in 22:2, preempting his appearance in 22:4b. Thus, it seems most likely that 22:2 and 22:4b come from different hands;11 at the very least one should note that they provide two introductions for the character Balak, a duplication that could be diachronically relevant. That 22:2 presupposes the existence of 22:4b can be seen in the missing data about Balak in 22:2, which only mentions his patronymic and fails to clarify to 7 Ulrich Fistill, Israel und das Ostjordanland, ÖBS, vol. 30 (Frankfurt a.M.: Peter Lang, 2007), 74–76 argued that Num 22:1 presents a redactional addition to the text in line with the redaction responsible for the composition of Num 22:21–35. 8 Since the details of this itinerary fail to match that of Deuteronomy 1–3, it seems likely that the author responsible for Deuteronomy 1–3 did not know them. That in turn commends regarding this itinerary in Numbers as younger than that in Deuteronomy 1–3, or at least compositionally distinct from it. 9 The first author I have found to suggest distinct provenance for pieces of Numbers 22–24 based on the nomenclature for Israel is Gall, “Zusammensetzung und Herkunft,” 4–5, though he relied on them for source-critical undertakings. Gross, Bileam, 86–88 made similar observations to mine about the terms used to describe the people undertaking the exodus. His literary-critical findings regarding the terms “Israel” and “Jacob”, excepting the oracles, generally matches that in this chapter. However, he permitted the terminology for the exodus party only a supporting role. Similarly, Levin, Jahwist, 381 noted that “Israelites” was editorial and “the people” belonged to the older source: “Schon der Ausdruck ‘Israeliten’ (‫ )בני ישׂראל‬weist auf den Jahwisten als Verfasser (vgl. Ex 1,9.12; 12,35; 14,10). Die Quelle dagegen spricht von dem Volk.” It is possible to determine the literary uniformity of material based on how it names “Israel”. 10 This observation contradicts the position of Fistill, Israel und das Ostjordanland, 73–74, who opined that Num 21:21–35 stemmed from a single author’s hand. 11 N.b., for example, how even Albright, “Home of Balaam,” 387 regarded verse 2 as “editorial” when contrasted with verse 4.

3.3 The Narrative’s Exposition(s): Numbers 22:1–4

135

the reader why he or she should care who Balak is. Numbers 22:2 thus presupposes familiarity with the character Balak. Again, as in 22:1, the terminology applied to those undertaking the exodus remains distinct in its context: Num 22:2 identifies them simply as “Israel”. The simple term Israel matches the usage in Num 21:21–31*, strengthening the argument for their belonging to the same diachronic narrative level or source. References to Israel in the oracles, and – more likely – in Num 24:1–2 could also belong to this layer, though this requires more clarification; see below. Based on these observations, the relative chronology of these elements becomes immediately apparent. Numbers 21:32–22:1* interrupt the narrative logic of Num 21:21–31* + 22:2. Therefore Num 21:32–22:1* must have been inserted into an extant literary context and thus be younger in that context than Num 21:21–31* + 22:2. Verse 22:1 only becomes necessary after the Israelites’ excursion to Jazer in Num 21:32 or to Bashan in Num 21:33–35. Without this insertion at the end of Numbers 21, one would have a logical, consistent narrative transition from Num 21:21–31* to Num 22:2. The alternative, that Num 22:2 is later than Num 22:1, seems, at best, implausible and most likely impossible. Verse 22:2 serves as a good transition from 21:21–31* to 22:3*, but becomes superfluous once 22:1 precedes it. In all probability, the narrative logic of the tale indicates that Num 22:2 is older than 22:1. Numbers 22:2 remains at the same time, however, younger than Num 22:4b. The repetition of “Moab” as the subject of both clauses in verse 3 raises suspicions.12 Moab reacts to the arrival of the exodus group in their territory in two distinct, yet remarkably similar ways.13 Each expression of negativity contrasts with that offered in v. 2: whereas Israel is dangerous in v. 2, there are many of them in 3a, and they are disgusting in 3b.14 The motif ‫ קוץ‬connects to Exod 1:12, which Achenbach identified with P in his discussion of Numbers 22–24.15 That would mean that Num 22:3b was either P (whether Pg or Ps) or post-P, but in the P tradition.16 Two different manners identify the object of 12

John Sturdy, Numbers, Cambridge Bible (NEB) (New York: Cambridge University Press, 1976), 160 suggested that this is emphatic. Budd, Numbers, 265 followed him in this view. That we are dealing with a duplication seems more likely based on the surrounding material. 13 For these reasons, Gross, Bileam, 65 regarded this as one of the few duplications in Numbers 22–24 that is literary-critically relevant. Ashley, Numbers, 444, on the other hand, for spurious reasons, did not. 14 Cf. Greßmann, Mose und seine Zeit, 321–22. Achenbach, Vollendung, 395 identified both ‫ רב‬and ‫ עצום‬as Deuteronomistic. Such an ascription could aid in the identification of one or both of these terms to that tradition-historical background. 15 Cf. Achenbach, Vollendung, 395. 16 It remains outside the scope of this study to consider the layering of the priestly literature, particularly as elements that legitimately can be identified as priestly hardly impacted Numbers 22–24 in any appreciable way. Cf. already Nöldeke, Untersuchungen zur Kritik des Alten Testaments, 87: “Dass die Geschichte Bileam’s Num, 22, 2 – 24, 25 mit der Grundschrift Nichts zu thun hat, bedarf keiner Darlegung.” In this study, when talking about “P” or “Priestly”, I use

136

Literary Criticism

Moab’s ire: “the people” in 3a and “the children of Israel” in 3b. The last time that “the people” was used to speak of Israel during its exodus from Egypt in Numbers was in 21:4–9, and 16.17 The term “the people” is more consistent with the terminology applied to the exodus group in Numbers 22–24: cf. 22:6, 11–12, 17, 41; and 24:14, in addition to just “a people” in 22:5; 23:9 and 24. As the usage of “sons of Israel” matches the clearly tertiary literary parenthesis in 22:1, it seems probable that 3b belongs to the same redactional level as that verse.18 In principle, this implies that Num 22:3a presents the original Balaam narrative’s exposition, should it have existed as an independent narrative. This matter will be addressed further below. However, it remains possible that Numbers 3a could be the product of two hands, a matter to which we will return later. The elders of Midian are out of place in their current setting in v. 4 (and also in v. 7). Their sudden appearance and subsequent conspicuously vanishing suggest that they are the product of a redactional addition in this context.19 Presumably they represent an interpolation pointing toward Numbers 25, which features (like Num 22:4 and 7) a conglomeration of Moabites and Midianites, and ultimately to Numbers 31.20 The addition of the Midianites into Num 22:4 and 7 prepares the reader for their being together in Numbers 25 and ultimately for Balaam’s dying amongst Midianite rulers in Numbers 31.21 this term in a very broad sense, including essentially everything that demonstrates affinity with something like classical iterations of the Priesterschrift, including any post-P elements that reflect a similar vocabulary or ideology. 17 The attestations of “people” (‫ )עם‬in 21:23, 29, and 33–36 reference non-Israelites. The reference in 18 is plural (i.e., also not Israel) in G. 18 Rudolph, Der “Elohist”, 107 suggested that Num 22:3b was initially a “Randglosse” inspired by Exod 1:12. According to him, this marginal note then erroneously made its way into the transmitted text through a copyist error. While an interesting and legitimate possibility, currently I favor its belonging to a consistent editorial undertaking. 19 Cf. Baentsch, Exodus, Levitikus, Numeri, 592–95. The assertion of Budd, Numbers, 263, that a later editor would have included more about the Midianites had they been a later insertion, is spurious and without merit. The insertion of the Midianites in this chapter probably stems from the incorporation of a fragment or another piece of extant textual tradition into this context. 20 Cf. Achenbach, Vollendung, 395: “V. 4 indes nimmt die Midianiterthematik mit hinein, die erst in Kombination von Num 25,1–5 und 25,6ff. plus Num 31 überhaupt einen Sinn macht.” 21 This position is contra Theodor Nöldeke, Ueber die Amalekiter und einige andere Nachbarvölker der Israeliten (Göttingen: Dietrich, 1864), 8, who regarded the Moabites in Numbers 22–24 as a younger replacement of the more ancient Midianites that was not fully implemented. The appearance of Moabites and Midianites together in Num 22:7 precludes this reconstruction. According to Achenbach, Vollendung, 394, the Pentateuch redaction (theologische Bearbeitung) found an earlier Hexateuch version of the story (which included Israel’s prostituting themselves to the Moabites in Numbers 25) and took that as an opportunity to kill Balaam with the Midianites. That is, Achenbach regarded the incorporation of the Midianites

3.3 The Narrative’s Exposition(s): Numbers 22:1–4

137

Two possible resolutions exist for these circumstances: 1) the elders of Midian were inserted alone into this context; or 2) the elders of Midian were inserted with other material into this context. Starting with v. 4, one can remove the words ‫ אל־זקני מדין‬and leave the rest of the text intact, which reconstructs a plausible, albeit somewhat unclear reading. To whom does Moab speak? And why does Balak, should he have heard the formulation of their complaint, fail to consider it as the dangerous issue for his people (v. 5–6)? The reference to the bull devouring that which is green in the field sounds like the fear of starvation (cf. v. 4). Balak considers the danger to be of a military nature (cf. v. 6). Balaam echoes only this military threat as the impetus for his summoning in his reporting to God (cf. v. 11). Both of these verses apparently lack familiarity with Moab’s complaint in v. 4a, suggesting it may have been added in its entirety. The question then remains what the basis of Moab’s fearing the people entailed. At least one other case in the Hebrew Bible uses ‫ גור‬in a similar fashion; cf. 1 Sam 18:15.22 The fear of the size of “the people” in 3a could match Balak’s insecurity in his ability to defeat them militarily. Admittedly, the great size could additionally exacerbate any fears about famine, in this case presupposing it as a result of overpopulation. Returning to our consideration of ‫קוץ‬ in v. 3b above and remembering the apparent connection to Exod 1:12, it becomes likely that the phrase ‫ כי רב־הוא‬was added to strengthen the connection to Exod 1:12.23 There one also finds the motif of size (‫)רבב‬. However, the narrative logic does not necessitate placing the people’s great size on the same diachronic level as Moab’s fear of starvation. Particularly if one considers 22:1 to be of one piece with 22:3b, it seems likely that a later editor added 3a* in order to more intimately connect these two stories. But the great size could also reflect an addition in the vein of Num 22:2, which focused on Israel’s military might. That might would presumably be more threatening with greater size. Either way, in the oldest version of the narrative, Moab probably merely feared the people without a stated reason in v. 3a to which their king Balak, introduced in v. 4b, reacted in v. 5. Thus 4a probably belongs as a whole to another redactional layer. However, one can read it together with 3b sensibly, as Achenbach noted, though perhaps not as a parallel version of the Balaam story, as he suggested.24 into the Balaam narrative in Num 22:4 and 7 as a later interpolation. This common observation seems likely, and we can support this finding. 22 The other uses of ‫ גור‬with the meaning “fear” include Deut 1:17; 18:22; 32:27; Job 19:29; 41:17; Ps 22:24; and 33:8. These instances often, but not always, base the fear on something concrete. There is thus no need for Num 22:3a to supply a reason for Moab to be afraid. 23 Cf. Alan W. Jenks, The Elohist and North Israelite Traditions, SBLMS, vol. 22 (Missoula, MT: Scholars Press, 1977), 57, who regarded ‫ כי רב הוא‬as a gloss. 24 Cf. Achenbach, Vollendung, 395–96. This comment should not preclude it have a sourcelike character. Rather, at this point, it seems appropriate merely to reiterate that neither Balaam

138

Literary Criticism

With this in mind, it seems most likely that at least three or four hands composed elements of 22:1–4.25 The oldest narrative began in 3a*. If this narrative existed independently, 3a* must have been its opening.26 It then continued in 4b (i.e., without Moab and its complaint to the Midianite elders) and then onto v. 5. The phrase ‫ בעת ההוא‬in 4b refers to the time of Moab’s fear and is not based on 4a.27 Simultaneously, this phrase demonstrates that the oldest Balaam narrative was told from its inception as a report of an earlier period. That is, the responsible scribe telling the story regarded it as history in some fashion. The first addition to the story occurred with the supplement of v. 2, which incorporated the Balaam narrative into a position following the destruction of the Amorites under Sihon’s hegemony, and perhaps the notice of the people’s great size in 3a*. A second layer of additions can be identified in vv. 1 and 3b. This second narrative embedding returned the Israelites to Moab in Num 22:1 after the insertion of Num 21:32 or the story of Og of Bashan between the story about the Amorites and the Balaam narrative, presumably from Deut 3:1–3.28 The journeys to Jazer and/or Bashan mandated the recontextualization of the story, as narrated in v. 1. Verse 3b reminds the reader of the racist sentiments against the Israelites mentioned in Exod 1:12, strengthening the narrative context within the exodus event attested elsewhere in Numbers 22–24. Verse 4a follows well immediately on 3b, suggesting they may be of one piece.29 Finally, either in combination with one of these two additions or independently from them, someone added the motif of great size in 3a*. This could have been added at the same time as v. 2, focusing the ascription of Israel’s might on their nor Balak appear in these verses, demonstrating that they were not part of an alternative “Balaam story”. 25 Contra Fistill, Israel und das Ostjordanland, 84, who regarded 2–4a as stemming from a single hand and interrupting the original union of Num 22:1 and 4b. 26 This position is contra Schmidt, “Bileam,” 343: “Gegen die These, dass mit V. 3 eine selbständige Bileam-Erzählung begann, spricht der Begriff ‘das Volk’ in V. 3a. Er setzt ‘Israel’ in V. 2 voraus, da nur dann klar ist, welches Volk mit ihm meint ist. V. 3a war also schon ursprünglich die Fortsetzung von V. 2. Da V. 2 an den Grundbestand der Sihon-Erzählung in 21,21ff.* anknüpft, war die elohistische Bileam-Erzählung schon immer Bestandteil eines größeren literarischen Werks. Mit V. 3a bereitete E 23,10a in dem ersten Spruch vor, wo Bileam die außerordentliche Größe von Jakob/Israel rühmt.” While it would be premature based on the observations presented thus far to determine that the Balaam narrative in Numbers 22–24 existed as an independent unit, if it did, it must have begun in 22:3a*. 27 Contra Schmidt, “Bileam,” 342. 28 Thus already Carl Steuernagel, Lehrbuch der Einleitung in das Alte Testament mit einem Anhang über die Apokryphen und Pseudepigraphen, Sammlung theologischer Lehrbücher (Tübingen: Mohr Siebeck, 1912), 269–70 and many others. 29 For Achenbach, the arrival at ‫ ערבות מואב‬settles Israel at the point where the Pentateuch redaction places Moses’ death. Connected to this are 3b–4a, in which Moab turns to the elders of Midian; cf. Achenbach, Vollendung, 395. That suggests that Num 22:1 + 3b–4a belong to the same diachronic level.

3.4 The Main Narrative Body: Numbers 22:5–23:6, 11–17, 25–30

139

size. Alternatively, it could have been added at the time of the incorporation of 1 + 3b–4a or afterward, strengthening the connection to Exodus 1 (cf. vv. 9 and 12). Syntactical considerations preclude it being part of 1 + 3b–4a. Thus, at least three to four hands were active in the pericope of Numbers 22–24 based solely on the evidence from the first four verses. The question remains at this point whether they represent redactional layers, individual sources, or some combination of the two.

3.4 The Main Narrative Body: Numbers 22:5–23:6, 11–17, 25–30 3.4 The Main Narrative Body: Numbers 22:5–23:6, 11–17, 25–30

Numbers 22:5, which informs the reader where Balaam was when he was summoned is curiously phrased, text-critically problematic, and stands somewhat in tension with Balaam’s account of his summoning found in Num 23:7. For these reasons, the data about Balaam’s origin raise suspicion of their being editorial. The location of Pethor remains unclear, but the author responsible apparently envisioned it by a river, here unnamed.30 The curious phrase “in the land of his people” (‫ )בארץ עמו‬remains both unclear and conspicuously superfluous, factors which ultimately led to its revision in some later witnesses.31 The focus on non-Israelite peoples comes about mostly because of the material in 24:14b–24, which will be shown to be secondary (and tertiary) to the original narrative.32 In this capacity, the phrase ‫ בארץ בני־עמו‬bears the most similarity to 24:14b, and will therefore be attributed to the same hand as that insertion (see below).33 Alternatively, and quite speculative, should it have one time read

30 It could be the Euphrates, but nothing in the text indicates that. The identification of the river with the Euphrates seems to be an attempt to read Numbers 22:5 through the lens of Deut 23:5, in which Balaam appears to be Mesopotamian. Cf. Greßmann, Mose und seine Zeit, 320, who proposed this conjecture. The conjecture of Sigmund Mowinckel, “Der Ursprung der Bilʿamsage,” ZAW 48 (1930): 235–36, that Pethor should be “Edom”, is baseless speculation mandated by his presumption of the supposed E itinerary he recreated. It should be rejected outright. 31 Cf. the discussion to this verse in Chapter Two. 32 See below. 33 Cf. already Mowinckel, “Ursprung,” 236–37, who identified this phrase as a “redactional corrective” from Num 24:14. His correction to an older ‫ בני־קדם‬is fanciful. Contra Fistill, Israel und das Ostjordanland, 82–83, who regarded an Ammonite heritage an impossible later ascription when contrasted with a Mesopotamian heritage. Since neither of these appeared initially in Num 22:5, we should reject this position outright. Rudolf Smend Sr., Die Erzählung des Hexateuch auf ihre Quellen untersucht (Göttingen: Georg Reimer, 1912), 230 and Rudolph, Der “Elohist”, 99, e.g., regarded the attribution of Balaam’s coming from Pethor by the river (= Euphrates) as a secondary harmonization with Deuteronomy 23. This possibility should not

140

Literary Criticism

‫בארץ בני־עמנו‬, it could have followed the conclusion of 4a: “and the elders of Moab said to the elders of Midian, ‘Now this assembly will lick up all that surrounds us, just like the bull licks up the green of the field, in the land of the sons of our people(s)’”. This rereading of the term ‫ עמו‬as ‫ עמנו‬would have the added benefit of adding further evidence to explain the divisions in the textual transmission that read either ‫( עמו‬haplography) or ‫( עמון‬metathesis). While that could present a relatively cogent text, it remains quite speculative. The duplication of the term ‫ הנה‬in v. 5 presents another curious case. The repeating of this term could lead to one phrase being regarded as redactional, since the phrases associated with it essentially introduce the people twice. On the one hand, Balak notes that they have come from Egypt. Whence he received this information remains a mystery. On the other hand, he notes that they cover the “eye of the land” (‫)את־עין הארץ‬. This curious phrase only occurs one other time outside of the Balaam story: in Exod 10:5, where it is used to describe the coming plague of locusts.34 “Israel is clearly compared with a locust plague, a common simile for an invading army (cf. Judg. 6:5; 7:12).”35 Should one have to choose which one of these phrases should be regarded as secondary, the second (‫ )הנה כסה את־עין נארץ‬seems most likely in that its subject relies on the foregoing clause and it disturbs the syntax slightly; 36 the transition from ‫ ממצרים‬to ‫ והוא‬certainly reads smoother than the current transition. The other usage of this phrase in Exod 10:5 in combination with locusts suggests regarding it as oblique reference to famine. That would place it firmly on an editorial or compositional layer consistent with 4a and thus remove it as part of the oldest narrative. In turn, that would imply that ‫ ויכס את־עין הארץ‬in Num 22:11aβ also belongs to this or an even later layer. Syntactically, the phrase ‫והנה כסה‬ ‫ את־עין הארץ‬appends well with the complaint of the Moabite elders in 22:4a, suggesting that it could have immediately followed it.37 The slightly different phrasing in 22:11 and the lack of connecting material between v. 5* and v. 11* at a common diachronic level (see below), suggest that the phrase ‫ויכס את־עין‬ ‫ הארץ‬in 22:11 represents a later editorial addition.

be entirely precluded, as both the Numbers and Deuteronomy traditions about Balaam underwent harmonization even into the recorded period of textual transmission; cf. the discussion to Deuteronomy 23 in Chapter Four, on the Bible’s Balaam. 34 Some have reckoned this phrase in Exodus to the “alte Exodus-Erzählung”; cf. Helmut Utzschneider and Wolfgang Oswald, Exodus 1–15, IEKAT (Stuttgart: Kohlhammer, 2013), 230 = Helmut Utzschneider and Wolfgang Oswald, Philip Sumpter, trans., Exodus 1– 15, IECOT (Stuttgart: Kohlhammer, 2015), 221. 35 Milgrom, Numbers, 186. 36 Contra Kratz, Composition, 299 = Kratz, Komposition, 295–96. 37 It could have even followed the speculative phrase ‫בארץ בני־עמנו‬, should it ever have stood there.

3.4 The Main Narrative Body: Numbers 22:5–23:6, 11–17, 25–30

141

Some elements of 22:6 awaken a feeling of their secondary status. The words ‫ כי־עצום הוא ממני‬perhaps represent an addition.38 When stating the messengers’ purpose in 22:11, Balaam does not include this datum. Further, the phrase ‫ כי־עצום הוא ממני‬expounds on the people’s might, as opposed to merely their size (cf. v. 3a).39 Both of these facts could imply that ‫ כי־עצום הוא ממני‬is secondary in its context. On the other hand, the oracles – particularly in their oldest form (see below) – tend to focus on the people’s might and not its size. The exception to this appears in 23:10a, which also alludes to Israel’s great size and makes a secondary impression (see below). This could then imply that ‫ כי־עצום הוא ממני‬in 22:6 is original, in spite of Balaam’s not repeating it in 22:11. In turn, this could affirm our identification of the phrase ‫ כי רב־הוא‬as a secondary interpolation in 3a.40 That would, again, in turn affirm that 3a did not originally mention the reason for Moab’s fear. That could suggest that Num 22:6 delivered the reason: their might. Nonetheless, it cannot be overlooked that the might in 22:6 readily reflects Balak’s fear in the editorial addition 22:2, suggesting they stem from the same hand. As was seen above, the size also did not originally play a role. Rather, the original version never mentioned why the Moabites feared “the people”, only that Balak sought to drive them out. The whole of 22:6b could be regarded as redactional as well. It reads somewhat peculiarly in the logic of the narrative and seems to presuppose and anticipate the coming undertakings, in which Balak contrasts Balaam’s repeated blessing with the desired and commanded cursing. To some degree it also reminds the reader of Gen 12:3 and 27:29, which could commend its attribution to an editorial expansion to afford it a better integration into the developing Pentateuch (or Hexateuch, etc.). However, at first glance, no internal criteria immediately mandate its omission; it neither contradicts nor stands in tension to its surroundings. The terminology does raise an issue however, and this affects several verses in these chapters. The ‫ ארר‬appears in Num 22:6 (3x), 12b; 23:7b; and 24:9b (2x). In most of these cases, it serves directly to contrast ‫ברך‬ (22:6b [2x], 12b; and 24:9). That is, it looks like a motif. These usages also follow a pattern: bless–curse (22:6b); curse–bless (22:12); bless–curse (24:9b). In the first case, Balak says what he knows about Balaam. The second case presents God telling Balaam about Israel. Finally, Balaam repeats God’s sentiment from 22:12 to Balak in the order that Balak reported what he knows about 38

Cf. Witte, “Segen Bileams,” 205. Achenbach, Vollendung, 395 nonetheless ascribed these two pieces to a common tradition-historical background, a background which he then recognized as combined with Exod 1:7 and 9. However, Numbers 22–24 fails to combine these terms, suggesting that they could each attest from distinct diachronic provenience. 40 This is contra Van Seters, Life of Moses, 416: “The theme of the people being numerous (rb), v. 3b, and powerful (ʿṣm), v. 6, also takes up the familiar theme of J in Ex. 1:7, 9, 12, 20… Since this serves as a basic theme of the whole Balaam story, it cannot be relegated to a series of redactional glosses.” Without further comment, Van Seters blends these two motifs. 39

142

Literary Criticism

Balaam. This creates a marked irony around the figure Balak: he basically announces a self-fulfilling prophecy in 22:6b. These usages take on the appearance of later interpretations about the story, strengthening the motif of blessing and cursing, but that still leaves the appearances of ‫ ארר‬in 22:6a and 23:7b. These are different and probably present transformations in the terminology for consistency. The form ‫ ארה‬appears only in these two cases in the whole HB. That is suspicious. And when contrasted with the surrounding context, it becomes apparent what happened. When Balaam reports Balak’s words to God in 22:11, he uses ‫קבב‬. When Balak’s ambassadors repeat his first message in his second (22:17), they also use ‫קבב‬. The same root appears consistently otherwise in every interaction between Balak and Balaam (23:11, 13, 25, 27; and 24:10). That is, ‫ קבב‬is the older, standard root for cursing in Numbers 22–24. The parallels between 22:6a, 11, and 17 demonstrate that someone over the course of transmission changed ‫ קבב‬to ‫ ארר‬after the insertion of 22:6b, 12b; and 24:9. The same is true of ‫ ארה‬in 23:7: the parallel between 23:7b and 8a demonstrates unequivocally that someone changed the term from ‫ קבב‬to ‫ ארר‬there. The ‫ ארר‬presents an interpolation (and gloss) in Numbers 22–24. It could be indicative of Deuteronomistic theology; cf. Deut 27:15–26 and 28:16–19. Since 12b also contains this contrast with these roots, it presumably belongs to the same editorial hand.41 Originally, God only forbade Balaam’s travel. A later editor expanded this with a reason. That is why in the original Balaam also names no reason why he cannot go with Balak’s messengers in 22:13: none was given to him. By leaving this open, the narrator created an opportunity for Balaam, and opportunity that he is able to exploit. God seems to be almost testing Balaam in the development of the story, once the knowledge of the people’s blessedness is removed from the story. When the second group of messengers come, they appeal to Balaam financially, but he rejects this and states that he can only do what God says (22:16–18). This establishes the motif of the word in Balaam’s mouth, an important thread throughout the rest of the story (22:18, 38; 23:5, 12, 16; and 24:10). Once God has witnessed Balaam’s testimony, God permits Balaam to go with them. Now the reader knows that Balaam will become God’s instrument in the further development of the story. Initially, in the oldest version, Balaam even made himself an instrument willingly. Verse 22:7 contains some elements that appear secondary. Most conspicuous are the “elders of Midian” found here again, as in v. 4. However, they do not stand alone. The elders of Moab also appear to be a secondary insertion; only in this case are the “messengers” of v. 5 identified as “elders”, whereas 41 Based on the surrounding context, the editor relied on the term ‫ עם‬to refer to the exodus group in favor of referring to it as “Israel”. Since neither Balak (22:5–6) nor Balaam (22:11) called it “Israel”, it would make no sense for God to do so here.

3.4 The Main Narrative Body: Numbers 22:5–23:6, 11–17, 25–30

143

the more standard identification for them is “officers” (‫)שׂרים‬. 42 Perhaps at some point they were known in 22:7 as “officers” and were intentionally or unintentionally adapted to match the secondarily inserted “elders of Midian”, but this is impossible to verify or falsify. With the subjects most likely secondarily added, the verb becomes superfluous and is thus secondary. Note the phrasing used in the recounting of the second mission to Balaam in 22:16, which does not include ‫ וילכו‬before ‫ויבאו‬. Thus, all of ‫וילכו זקני מואב וזקני מדין‬ presents a later interpolation. With that whole sentence presumably secondary, the verbless clause ‫וקסמים‬ ‫ בידם‬must be an addition, as it would otherwise remain without a sensible context. The unclear nature of this phrase could support its identification as a later insertion. While it remains opaque as to what precisely this should refer,43 the pejorative understanding of the root ‫ קסם‬in biblical texts should lead none to doubt that it should reflect negatively on Balaam.44 In terms of tone and context it should not be regarded as original. However, it does match the rest of Num 22:7a and therefore does not have to be separated from that. Therefore, the entirety of 22:7a does not belong to the oldest narrative about Balaam.45 The elements common to 22:7a and 4a commend ascribing them to the same hand. Thus, we can identify a fragmentary narrative in addition to the primary narrative about Balaam. The original Balaam narrative thus began with 22:3a*, 4b, 5a*, 5b*, 6a*, 6b, and 7b. Moab feared the people, so the king of Moab, Balak, sent messengers to Balaam ben Beor in Pethor on the river to request assistance. This appears to be a first source in Numbers 22–24 (S1). A first editor expanded this introduction, adding 22:2, perhaps 3a*, and 6*, contextualizing Israel’s arrival in Moab after their defeating the Amorites and emphasizing Israel’s military might (R1). To this, someone incorporated a parallel story describing the Israelites’ arriving in Moab, consisting of vv. 1, 3b–4a, 5b*, and 7a (S2). Because of Moab’s disgust over the Israelites, they sent some of their elders with elders of Midian with some magical implements in their hands. To all of this, perhaps 42

Cf. 22:8, 13–15, 18G, 21, 35, 40. It could refer to either the prophecies themselves or the devices used to divine them. Cf. R. Walter L. Moberly, “On Learning to Be a True Prophet. The Story of Balaam and His Ass,” in New Heaven and New Earth. Prophecy and the Millennium. FS Anthony Gelstron, ed. Peter J. Harland and Robert Hayward (Leiden: Brill, 1999), 3–4 and the discussion of the root ‫קסם‬ in the chapter on Tradition History. Milgrom, Numbers, 187 understood this verse as demonstrating the mantic ability of the travelling elders, “…that is, they themselves were diviners (see Ezra 7:25) and might, therefore, prevent him from backing out on the pretext that the time was unpropitious.” 44 Cf., e.g., Achenbach, Vollendung, 397 and the more lengthy discussion to this lexeme in the chapter on tradition history. 45 Cf. Greßmann, Mose und seine Zeit, 319, who attributed this half-verse to J and divorced it from an E context, seeing it as a contradiction to v. 5. 43

144

Literary Criticism

someone later added a comment on the Israelites’ size in 3a* and the reference to the land of the sons of his people in v. 5. In the first case, 3a*, the common connection it shares with 3b to Exod 1:12 commends regarding 3a* as the product of a second redactor (R2), who incorporated S2 into S1 + R1. To R2 we can also ascribe the insertion of 11a, which picks up elements of S2 currently found in v. 5. The sons of the people belonged either to S2, if my conjecture about the form ‫ עמנו‬in its background is correct, or more likely the earlier editor (R1), also responsible for the addition of 24:14b–19a (see below). The first editor probably also added 6b in order to strengthen this contrast between these two persons; with the addition of 6b the editor prepared the reader for the implied blessedness of Balaam and Balak’s implied cursedness, the second of which matches the ideology of 24:14b–19a and therefore belongs to that edition. From this reconstruction, then, it appears we can identify four hands in the first seven verses of Numbers 22: S1, R1, S2, R2. This great disparity and diversity does not continue through the whole of chapters 22–24, however. Rather, Numbers 22 appears to be an important seam, on which several people sewed. Following v. 7 and aside from the issues already noted in vv. 11–12, the story generally reads uniformly until 22:21.46 Balaam’s encountering God and God’s inquiry about the visitors does not stand in tension to the rest.47 Balaam’s briefer transmission of God’s message in 22:13 does not indicate that someone expanded 23:12 regarding Balaam’s inability to curse the people.48 The repetitive encounter between Balak’s messengers and Balaam does not sufficiently merit dividing the story into two layers. Rather, the second encounter builds upon and presumes the existence of the first encounter, in which Balaam refused to go with Balak’s officials. The duplication of these episodes serves a compositional function by increasing the tension in the narrative and developing the characters of Balak and Balaam. Recourse to the second message is made in 24:11–13, strengthening my claim of the uniform compositional intention behind this duplication.

46

As noted already in, e.g., Bewer, “Literary Problems,” 241–42. Cf. Van Seters, Life of Moses, 417: “In the divine dialogue, the question posed by the deity, v. 9, is not for the purpose of information but to allow Balaam the chance to state the request of the visitors. It is a device used elsewhere, especially in the primeval history (Gen. 3:9, 11, 13; 4:9–10; cf. 18:9), and often suggests a testing of the person questioned.” 48 Milgrom, Numbers, 188 understood the matter pragmatically in terms of narrative logic: “Balaam omits the full reason – that cursing Israel is an exercise in futility – not because he hoped later to change God’s mind but because it would have brought the story to an end had the emissaries reported this reason to Balak.” 47

3.4 The Main Narrative Body: Numbers 22:5–23:6, 11–17, 25–30

145

In 22:21aβ + 22–35 we find the largest insertion into the Balaam story of Numbers 22–24.49 God’s disposition in this material contradicts God’s commanding Balaam to go with Balak’s emissaries.50 For the first time in the story, God uses a medium to interact with Balaam and does not engage him directly.51 This new manner of interaction distinguishes the story of the donkey not only from the preceding material, but also from what follows it. That is, when considering Numbers 22–24 as a whole, 22:21aβ + 22–35 attests a unique theology.52 Furthermore, the narrative attitude toward Balaam is much more negative than in the remaining material. Finally, Num 22:21a + 22–35 remains

49

While this has generally been accepted in scholarship, some attempt to read the story of the donkey as part of the original; cf., e.g., Greßmann, Mose und seine Zeit, 319 and Rudolph, Der “Elohist”, 108. More recently, Edgar Kellenberger, “Jahwes unerwarteter Widerstand gegen seinen Beauftragten. Erwägungen zur Episode von Bileams Eselin (Num 22,22–35),” ThZ 45 (1989): 69–72 regarded this and other stories demonstrating God’s surprising anger (Gen 32:23–33 and Exod 4:24–26) as literarily uniform. Rather than regard the texts as products of some literary-critical undertaking, he ascribes their arcane narratives to an oral background. Such an assertion hardly convinces in this case. Clinton J. Moyer, “Who is the Prophet, and Who the Ass? Role-Reversing Interludes and the Unity of the Balaam Narrative (Numbers 22– 24),” JSOT 37 (2012): 167–83 described this tale as part of a whole in order to develop a more multifaceted characterization of Balaam. While observant and poignant on many points, he did not consider whether a redactor might be responsible for this multifaceted characterization, particularly as the only new aspect apparent in this addition stems from Balaam contravening the divine will, even if Balaam ultimately acknowledges that and seeks to make amends. A competent editor could just have easily added this aspect. 50 Cf., e.g., Baentsch, Exodus, Levitikus, Numeri, 598–99 and Löhr, “Bileam,” 88. Contrary to what Mary Douglas, “Balaam’s Place in the Book of Numbers,” Man (N.S.) 28, no. 3 (September 1993): 411–20 argued, the divine command and the divine anger clearly stand in tension to one another. Her suggestion about the whole story uniformly serving as a satire roots on the translation of Num 22:20aβ as “if the men come to you again, rise up and go with them”. However, that translation remains inaccurate. The text clearly does not say “again”, but reads, “if the men have come (perfect = past tense!) to call to you, arise, go with them”. Balaam was, thus, not deceitful from the story’s inception, as Douglas suggested. 51 Erhard S. Gerstenberger, “Boten, Engel, Hypostasen: Die Kommunikation Gottes mit den Menschen,” in Gott und Mensch im Dialog. Festschrift für Otto Kaiser zum 80. Geburtstag, ed. Markus Witte (Berlin: de Gruyter, 2004), 148 described this medium (traditionally called “angel”) as, “den Widerstand leistenden Schutzgeist”. 52 Cf. Achenbach, Vollendung, 403: “Die Veränderung [zwischen Num 22,20 und 35] besteht darin, daß Bileam nach v. 20b gewiesen ist, zu tun, was das Wort Jahwes gebietet, nach v. 35a indes zu reden, was der Engel Jahwes ihm mitteilt. Durch diese Nuance verrät der Interpolator der Eselinnen-Episode sein Interesse an der Theologie des Wortes.” If Achenbach is correct in this identification, then the interpolation of the donkey story would have necessitated the emendation of 22:38 and 24:13 to reflect the same idea, assuming of course they are not from the same hand as Num 22:22–35.

146

Literary Criticism

distinct from the rest of Numbers 22–24 in terms of genre.53 Whereas in general Numbers 22–24 can be identified as a kind of hagiography about Balaam, Num 22:21aβ + 22–35a presents the exegete with a satirical fable.54 In terms of narrative logic, the story has not progressed any further by the end of Num 22:35 than it was in Num 22:21. That could also commend regarding it as an interpolation or expansion.55 The boundaries of this insertion are somewhat rough, but most clearly defined in vv. 22 and 35.56 However, since 21aβ mentions Balaam’s donkey for the first time – a necessary prerequisite for the following episode – it probably also belonged to that layer. Since 35b proffers Balaam’s name again as the subject of the sentence, which would be superfluous and cumbersome immediately after 21aα, 35b probably represents the conclusion of the addition, 53

Greßmann, Mose und seine Zeit, 324–27 identified it as a saga with fairy-tale elements. Achenbach, Vollendung, 403 called it a “burlesk ausgemalte Groteske”. Milgrom, Numbers, 191 identified several satirical and ironic elements: 1) YHWH opens the donkey’s mouth before Balaam’s; 2) there is a sword nearby while Balaam wishes he had one; and 3) the donkey’s attitude presents a justifiable response to Balaam’s irrational and irritated behavior. At Qumran they apparently understood 22:21 as a new passage’s beginning. Formal evidence for their evaluation of a possibly secondary character of this narrative in its context could be identified via the red ink used for this verse in 4QNumb, visually distinguishing it from the surrounding material. Alan R. Millard, “Epigraphic Notes, Aramaic and Hebrew,” PEQ 110– 111 (1978–79): 24–25 and Émile Puech, “Balaʿam and Deir ʿAlla,” in The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity and Islam, ed. George H. van Kooten and Jacques van Ruiten, TBN, vol. 11 (Leiden; Boston: Brill, 2008), 28–29 recognized this at Deir ʿAlla as an Egyptian method for distinguishing a text’s beginning. At such a late stage of transmission, the veracity of such a claim cannot be verified, however, merely postulated. 54 Jonathan D. Safren, “Balaam and Abraham,” VT 38, no. 1 (1988): 105–13 described the story of Balaam’s ass as a burlesque, a kind of mirror-image of the binding of Isaac (Genesis 22). To this end, he described a few similar elements in the stories: 1) Setting (a trip by donkey), Characters (angel, prophet ), donkey (dumb vs. smart); 2) stylistic devices: Num 22:21a // Gen 22:3aαβ. (vocabulary dissimilar!); Num 22:22b // Gen 22:3aαβ (dissimilar vocabulary!); 3) Vocabulary: ‫ראה‬, the sword (two distinct terms!). These similarities remain generally superficial (note the distinct vocabulary), which generally precludes a common compositional interest and background. However, Safren did accurately recognize the negative characterization of Balaam implied through the addition of the donkey story. David Marcus, From Balaam to Jonah: AntiProphetic Satire in the Hebrew Bible, BJS, vol. 301 (Atlanta: Scholars Press, 1995), 29–41 excellently and lucidly demonstrated the satirical characteristics of the passage. 55 While this could represent a negative characterization of Balaam (he has not developed over the course of this story), that does not imply that the donkey narrative is of one piece with the surrounding material, as Amos Frisch, “The Story of Balaam’s She-Ass (Numbers 22:21– 35): A New Literary Insight,” HS 56 (2015): 103–13 has suggested. The evidence that Frisch cited in favor of the unity of this passage with the rest all belongs to editorial additions. That is, any negative characterization of Balaam implied by this passage or the rest stems from a redactor. 56 Gross, Bileam, 66 observed that v. 35 returns the reader to the narrative point reached in 22:20–21 and that 22:22–34 stands in tension to 22:20–21.

3.4 The Main Narrative Body: Numbers 22:5–23:6, 11–17, 25–30

147

which may have copied it from 21b. This created what appears like a virtual Wiederaufnahme, and it should be regarded as the older conclusion of the donkey episode and not that of Balak’s messengers visiting Balaam before the insertion of the story about Balaam’s donkey.57 The combination of 21aα (‫)קום‬ and 21b (‫ )הלך‬literally fulfills what God has commanded Balaam in 22:20 and thus supports this reconstruction: “get up” (‫ )קום‬and “go” (‫)הלך‬. The same is not true for 22:35b, which does precisely match the command in 35a. Since the insertion of 22–35 lacks a proper introduction for the character Balaam, identifying him merely by his name and not his patronymic or any other data, we should regard this material as having been composed for the literary context in which it currently stands. The tale of the donkey presumes that the reader is already familiar with Balaam. This familiarity and the inability to identify any other material that must have preceded this story in an earlier version indicate that in 22:21aβ + 22–35 we are dealing with a Fortschreibung, a text composed for its context that expounds on and reinterprets the older material.58 The otherwise unbound inclusion of the donkey in 21a confirms this suggestion’s accuracy. No conspicuous need to remove vv. 28–30 exists, as Holzinger proposed, other than seeking to remove any fabulous elements from the story.59 The tale of Balaam’s interaction with his donkey is best understood as a unit. It is a unit that reflects negatively on Balaam, in marked contrast to the preceding and following story and oracles: “In truth, Balaam is depicted on a level lower than his ass: more unseeing in his inability to detect the angel, more stupid in being defeated verbally by his ass, and more beastly in subduing it with his stick whereas it responds with tempered speech. […] The lampooning of Balaam, then, serves the purpose of downgrading his reputation. It aims to demonstrate that this heathen seer, who was intent on cursing Israel without God’s consent, is in reality a fool, a caricature of a seer, one outwitted even by his dumb beast.”60

57 Cf. similarly Barré, “Portrait,” 260, who did not make some redactor responsible for the insertion. 58 Cf. Hans-Christoph Schmitt, “Heidnischer Mantiker,” 193: “Insofern ist nicht unbedingt auszuschließen, daß es sich bei Num 22,22–35 um eine von Num 22* abhängige redaktionelle Bindung handeln kann. Dafür könnte auch sprechen, daß 22,35 eine direkte Wiederaufnahme von 22,20f. darstellt.” and Achenbach, Vollendung, 403, who commented that the story of the donkey is a fragment that does not demonstrate any linguistic or literary-critically relevant connection to other sources. Rüdiger Bartelmus, “Von Eselinnen mit Durchblick und blinden Sehern. Numeri 22,20–35 als Musterbeispiel narrativer Theologie im Alten Testament,” TZ 61, no. 1 (2005): 27–43 argued convincingly that the donkey story was not an older independent narrative and was familiar with the older story. Contra Ashley, Numbers, 435, who regarded it as a separate source integrated into the whole. 59 Cf. Holzinger, Numeri, 114. 60 Milgrom, Numbers, 469.

148

Literary Criticism

Balaam’s negative characterization in this passage remains clear and generally indisputable.61 Nonetheless, it is a unit that does not fit into the original narrative, nor the version edited to include the connection to the Amorites, nor the fragment including the elders of Midian. That is, this was the product of distinct editors from those, and therefore probably later than them (R3+). At this stage we can note that another Mediterranean text features a talking beast of burden. “The story of Balaam’s talking jenny is not the only text from the ancient Mediterranean world that attests to divinely endowed equid speech. An additional example of this scenario is attested in Homer’s Iliad (XIX:404– 24), where Achilles converses with his horsed named Xanthus.”62 Rouillard presented three convergences between these two tales: 1) the divine origin of animal’s speech; 2) the animal’s speech coincides with the rider’s blindness; and 3) the rider becomes angry at the animal.63 To these three, Way added a fourth: the dialogues feature death as a major theme.64 “Perhaps this Homeric parallel to Numbers 22 can be regarded as evidence for the existence of a typescene that was employed in the literature of the ancient Mediterranean world.”65 Alternatively, one could consider the impact of Greek literature on the author of the donkey story as an explanation. That could have repercussions for dating this passage. A number of phrases reflect how Balaam should, will, and did engage with the word of God or that which was set in his mouth. Such texts appear in 22:18b, 20b, 35aβ, 38b; 23:5, 12b, 16, 26b; and 24:13aβγδb. Since the phrase in 35aβ appears to be part of an interpolation, we must consider the possibility that some or all of the other such phrases are as well. The most similar phrase to that in 22:35aβ appears in 22:20b. The only differences are the word for “only” (‫ אפס‬vs. ‫ )אך‬and the concluding verb (‫ דבר‬vs. ‫)עשׂה‬. These are understandable distinctions that could have developed over the course of transmission. Both 22:20b and 35aβ are lacking a common motif in most of the subsequent statements, namely the “mouth” (‫)פה‬. Taken together, these suggest that the editor responsible for the addition of the donkey story or an even later editor added 22:20b for the sake of harmonization. When read together with 22:21aβ, these two elements better incorporate Num 22:22–35. In the other cases, one recognizes a pattern. The final reference to Balaam’s speaking, in Num 24:13, represents essentially the compilation of two previous statements: 22:18b and 23:26b. These represent Balaam’s statement before he set 61 Gershon Hepner, “The Mockery of Kings and Prophets. The Balaam Narrative Contains an Implied Critique of Moses,” RB 118 (April 2011): 180–85 read this addition in contrast to Exodus 10 and interpreted it as a negative commentary on Moses. The evidence for this remains thin, particularly since Moses plays no role anywhere in Numbers 22–24. 62 Way, Donkeys, 189. 63 Cf. Rouillard, Balaam, 118. 64 Cf. Way, Donkeys, 189. 65 Way, Donkeys, 189.

3.4 The Main Narrative Body: Numbers 22:5–23:6, 11–17, 25–30

149

out to encounter Balak and his statement before he set out with Balak on the final leg of their journey. In between these two statements, we find Balaam’s affirmations of what he will say before the first and second oracles (22:38b and 23:12), each with a corresponding note about how the deity interacted with Balaam (23:5 and 16). Again, the minor variations between these pieces could have come about from processes in the transmission of the text similar to those described in Chapter Two. In fact, many of these verses attest variants precisely for that reason. Be that as it may, we nevertheless recognize a pattern in these statements. Before Balaam sets out, he states that he remains unable to transgress the “mouth” of God (22:18b). When he meets Balak and still before his first oracle, he tells him that he will say the word that God puts in his mouth (22:38b). God then puts a word in his mouth (23:5), which Balaam apparently repeats. Balaam repeats his interaction with Balak and God before reciting the second oracle (23:12 and 16). Then before the third oracle, Balaam changes his statement, just as the form of the third oracle is distinct from those that preceded it (23:26). This prepares the audience for what is coming. No longer does Balaam claim that he will say what God puts in his mouth. Rather, he will do what God says. God will instrumentalize him and apparently speak through him. This is precisely what the third oracle presents, noting for the first time that the spirit of God was upon him (24:2). No longer does God put the word in Balaam’s mouth. Balaam becomes the literal divine mouthpiece. After that conclusion, when Balak dismisses him, Balaam repeats both messages, his first (22:18b) and his last (23:26), as a conglomeration and the fulfillment of what just happened (24:13aβγδ). This presents quite a remarkable literary structure, fractured by the insertion of the material in 22:20b and 35aβ. Read in this, the text presents Balaam as an upstanding figure. When Balak’s officers return to him, Balaam tells them that he will not – cannot – transgress God’s message. This statement prepares Balaam for the journey that God can now send him on. The story as a whole develops this theme in the oldest version, rounding it out before the original conclusion in 24:14a + 25 (see below). With that, we can regard 22:20b and 35aβ as later interpolations, but the other pieces about Balaam, his mouth, and God’s word as belonging to the oldest layer.66 Just as previously mentioned, regarding the two encounters between Balak’s messengers and Balaam, the duplicate structures in Num 22:36–23:26 cannot a priori be regarded as coming from different sources or editors. However, as early as Wellhausen, some have suggested tension, particularly in Num 22:36– 41. Wellhausen himself regarded vv. 37 and 39 as a fragment of the original continuation of the story from 22:21–35*.67 In this recounting, then, Balaam returned after his encounter with the messenger of YHWH, a note that has gone 66 67

Contra Achenbach, Vollendung, 406–7. Cf. Wellhausen, Composition des Hexateuchs, 109.

150

Literary Criticism

missing or was deleted in the combination of the sources. Therefore Balak went to Balaam, explaining the circumstances behind vv. 37 and 39. Since Wellhausen proposed this, some have followed him.68 Others, however, have suggested other possibilities and rejected Wellhausen’s reconstruction.69 Bewer, for example, argued for a pragmatic understanding of Balak’s rhetorical question, “did I not send for you?” Rather than understand it literally, Bewer recognized that this question demonstrates the character Balak’s frustration in the episode.70 He is exasperated with Balaam and thus dramatically questions him. This deals better with the text as it stands. Proponents of dividing the text in Num 22:36–41, like Wellhausen, must make recourse to a missing portion of text. While that is possible, it remains more secure to work with evidence we do have to explain the situation than to postulate evidence that might have been in order to divide the text. Without 22:36, which Wellhausen and his adherents place in a different source than 37 and 39, Balak and Balaam would not be present at the same location. That remains the fatal flaw in their reconstruction. Further, there is no tension between the twofold structures such as the journey that Balak and Balaam undertake together from Ir-moab to Kiriath-huzoth and further to Bamath-baal and finally to the top of Pisgah.71 The second leg of their journey together (from Kiriath-huzoth to Bamath-baal) builds on the first leg of their journey together (from Ir-moab to Kiriath-huzoth). The same must be true of Balaam’s first two oracles and the episodes in which they are embedded: the second builds on the first. It appears that the composer of this story tends to use two-fold structures: two missions from Balak to Balaam; God’s twofold nocturnal visitation to Balaam; two legs of Balak’s and Balaam’s journey together. This usage of pairs, however, breaks after the second oracle, demonstrating the special status of the third oracle in the original composition. For the first time in the oldest composition, a third journey begins. Balak and Balaam undertake a third leg on their journey together, and Balak makes a third offering. However, Balaam no longer goes out to receive his third oracle. He no longer says that the word will be set in his mouth. Something has changed, and the narrator implies this with the recounting of a third encounter. What follows and (originally) concluded the story presents new terrain. 68

Cf., e.g., Greßmann, Mose und seine Zeit, 319 and Graupner, Elohist, 161. Cf. already Abraham Kuenen, “Bijdragen tot de critiek van Pentateuch en Jozua. X. Bileam,” ThT 18 (1884): 535. 70 Cf. Bewer, “Literary Problems,” 248–50. 71 One does wonder if the duplicate data in Num 23:13 regarding the pinnacle of Pisgah // the field of the Zophim anticipates Moses’ viewing of the land in Deut 34:1. Perhaps the “field of the watchers” was added to connect to this later narrated event. Cf. Ernst Axel Knauf, “Welches ‘Israel’ Bileam sah,” in Wort und Stein. Studien zur Theologie und Archäologie. Festschrift für Udo Worschech, ed. Friedbert Ninow (Frankfurt am Main; Berlin; Bern; Brussels; New York; Oxford; Vienna: Peter Lang, 2003), 184, who recognized this connection, but reversed the relationship suggested here. 69

3.4 The Main Narrative Body: Numbers 22:5–23:6, 11–17, 25–30

151

Before considering the relationship of the oracles to the narrative material, one substantial element presents a problem or at least awakens a sense of its secondary status in the context. Verse 23:4b probably was not originally part of the story. Beyond its moderately disturbing of the context, two major factors suggests this verse’s editorial background. First, v. 23:4b seeks to paint Balaam in a more negative manner by reiterating his (illegitimate and inefficacious) sacrifice. Secondly, comparing the first oracle’s prelude (23:3–5) with the prelude to the second oracle (23:15–16) suggests that Balaam’s explicit mention of the sacrifices presents a secondary addition before the first oracle.72 4a 4b 5a 5b

Num 23:4–5 ‫ויקר אלהים אל־בלעם‬ ‫ויאמר אליו את־שׁבעת המזבחת ערכתי‬ ‫ואעל פר ואיל במזבח׃‬ ‫וישׂם יהוה דבר בפי בלעם‬ ‫ויאמר שׁוב אל־בלק וכה תדבר‬

Num 23:16 ‫ויקר יהוה אל־בלעם‬

16aα

‫וישׂם דבר בפיו‬ ‫ויאמר שׁוב אל־בלק וכה תדבר׃‬

16aβ 16b

Τable 12: Parallel Structure in Num 23:4–5 + 16.

The curious location of this phrase and the fact that it presents Balaam as the one sacrificing (cf. 23:2 ML) in contrast to the rest of the narrative suggest its secondary status. With suspicions raised about that singular phrase, one should consider the other material focused on the motif of Balaam’s culpability for Balak’s offerings. The parallel structure of the narrative elements surrounding the first three oracles could imply that the material about Balaam’s encouraging Balak to sacrifice is secondary. But this must be reflected in detail.

72 Contra Van Seters, Life of Moses, 416–17: “…the instructions to construct altars and offer whole burnt offerings as a preparation for the revelations cannot be dissociated with the practice of divination through which the deity offers Israel his blessing.”

‫‪152‬‬

‫‪Literary Criticism‬‬ ‫*‪Num 23:27–24:11a‬‬ ‫ויאמר בלק אל־בלעם לכה־נא‬ ‫אקחך אל־מקום אחר אולי‬ ‫יישׁר בעיני האלהים וקבתו לי‬ ‫משׁם‬ ‫ויקח בלק את־בלעם ראשׁ‬ ‫הפעור הנשׁקף על־פני הישׁימן‬ ‫ויאמר בלעם אל־בלק בנה־לי‬ ‫בזה שׁבעה מזבחת והכן לי בזה‬ ‫שׁבעה פרים ושׁבעה אילים‬ ‫ויעשׂ בלק כאשׁר אמר בלעם‬ ‫ויעל פר ואיל במזבח‬ ‫וירא בלעם כי טוב בעיני יהוה‬ ‫לברך את־ישׂראל ולא־הלך‬ ‫כפעם־בפעם לקראת הנחשׁים‬ ‫וישׁת אל־המדבר פניו‬ ‫וישׂא בלעם את־עיניו וירא את־‬ ‫ישׂראל שׁכן לשׁבטיו‬

‫ותהי עליו רוח אלהים‬

‫וישׂא משׁלו ויאמר‪...‬‬ ‫ויחר־אף בלק אל־בלעם ויספק‬ ‫את־כפיו‬ ‫ויאמר בלק אל־בלעם לקב איבי‬ ‫קראתיך והנה ברכת ברך זה‬ ‫שׁלשׁ פעמים‬ ‫עתה ברח־לך אל־מקומך‬ ‫אמרתי כבד אכבדך והנה מנעך‬ ‫יהוה מכבוד‬

‫*‪Num 23:13–25‬‬ ‫ויאמר אליו בלק לכה־נא אתי‬ ‫אל־מקום אחר אשׁר תראנו‬ ‫משׁם אפס קצהו תראה וכלו‬ ‫לא תראה וקבנו־לי משׁם‬ ‫ויקחהו שׂדה צפים אל־ראשׁ‬ ‫הפסגה‬

‫ויבן שׁבעה מזבחת‬ ‫ויעל פר ואיל במזבח‬ ‫ויאמר אל־בלק התיצב כה על־‬ ‫עלתך ואנכי אקרה כה‬

‫ויקר יהוה אל־בלעם‬

‫וישׂם דבר בפיו‬ ‫ויאמר שׁוב אל־בלק וכה תדבר‬ ‫ויבא אליו והנה נצב על־עלתו‬ ‫ושׂרי מואב אתו‬ ‫ויאמר לו בלק מה־דבר יהוה‬ ‫וישׂא משׁלו ויאמר‪...‬‬

‫ויאמר בלק אל־בלעם גם־קב‬ ‫לא תקבנו גם־ברך לא תברכנו‬

‫‪Num 22:41–23:11‬‬

‫ויקח בלק את־בלעם ויעלהו‬ ‫במת בעל‬ ‫וירא משׁם קצה העם‬ ‫ויאמר בלעם אל־בלק בנה־לי‬ ‫בזה שׁבעה מזבחת והכן לי בזה‬ ‫שׁבעה פרים ושׁבעה אילים‬ ‫ויעשׂ בלק כאשׁר דבר בלעם‬ ‫ויעל פר ואיל במזבח‬ ‫ויאמר בלעם לבלק התיצב על־‬ ‫עלתך ואלכה‬

‫אולי יקרה יהוה לקראתי ודבר‬ ‫מה־יראני והגדתי לך‬ ‫ויתיצב בלק על עלתו ובלעם‬ ‫נקרה אל אלהים וילך‬ ‫וילך שׁפי‬ ‫ויקר אלהים אל־בלעם‬ ‫את־שׁבעת‬ ‫אליו‬ ‫ויאמר‬ ‫המזבחת ערכתי ואעל פר ואיל‬ ‫במזבח‬ ‫וישׂם יהוה דבר בפי בלעם‬ ‫ויאמר שׁוב אל־בלק וכה תדבר‬ ‫וישׁב אליו והנה נצב על־עלתו‬ ‫הוא וכל־שׂרי מואב‬ ‫וישׂא משׁלו ויאמר‪...‬‬

‫ויאמר בלק אל־בלעם מה‬ ‫עשׂית לי לקב איבי לקחתיך‬ ‫והנה ברכת ברך‬

‫‪Table 13: Travelling and Building Altars.‬‬

‫‪The first episode does not begin with Balak declaring that he will take Balaam‬‬ ‫‪elsewhere as in the second and third episodes. There are two reasons for this:‬‬ ‫‪1) as Balaam has not yet failed to curse the people in the first episode, there’s‬‬ ‫‪no reason for Balak to demand a relocation or explain his logic to Balaam (and‬‬

3.4 The Main Narrative Body: Numbers 22:5–23:6, 11–17, 25–30

153

the reader); 2) Balak’s brief speeches before the second and third oracles consider the element of sight (‫)ראה‬, already noted after the relocation in the first episode. Without these explicatory speeches in the second and third cases, the story as a whole would become more incomprehensible. For these reasons 22:41 does not present a perfect parallel to 23:13–14a and 27–28. Nonetheless, all of these elements appear original to the story. The narrative logic demands the distinct form from the opening of the first oracle. A problem with a different solution appears in Balaam’s demanding that Balak build altars and carry out sacrifices. The narratives are again inconsistent in this regard. The second episode conspicuously lacks Balaam’s command to offer sacrifices or build altars. The first and third episodes contain such information and that Balak did as Balaam told him (23:1–2 and 29–30). The second episode only notes that Balak built seven altars and sacrificed (14b). For that matter, Balaam repeatedly identifies the offerings as explicitly Balak’s (23:3a, 15), as does the narrator (23:3b [in the restored reading of G], 6, and 17). Balaam does not ever mention “our” sacrifice, only “your” sacrifice. The narrator never mentions “their” sacrifice”, only “his” (i.e., Balak’s) sacrifice. Thus it seems possible that Balaam’s demanding that Balak build the altars and sacrifice there may be secondary.73 Initially the offerings may have represented Balak’s attempt to influence the events.74 We notice that Balaam has had no need for them in his encounters with God in Numbers 22 and that Balak sacrifices in 22:40 without being told to do so.75 Balaam’s leaving in the first two cases even might have presented his attempt to divorce himself from Balak’s undertaking in an older version. Within the narrative introduction of the various sacrifices, tension remains in each case as to how many altars there were. Balaam always wants seven, but Balak only ever actually sacrifices on one at each location (cf. 23:2b, 14b, and 30b).76 Balak’s building the altars in 23:14 presents a minor problem for the

73 Gross, Bileam, 67 regarded primarily the recounting in 23:29–30 as suspicious since it returns to a longer version than that found in 14b. Mostly he seems required to argue the point in this manner to maintain his model of the original version ending in 23:25: nothing in the older tale could follow that verse, meaning 23:29–30 must be ejected. 74 Cf., e.g., Achenbach, Vollendung, 407; Achenbach presumed that Balak sacrifices to Balaam’s god in Num 22:40 in order to make him amenable, but the text makes no such suggestion or implication. 75 Albeit reflecting a different verb (√‫ )זבח‬and thus perhaps presuming a distinct kind of offering. The offering in 22:40 does serve a different purpose from those in Numbers 23, precluding the distinct terms presenting any kind of tension, contra Noth, Numeri, 159, who regarded 22:40 as a doublet to 23:1–2, based also on the distinct kinds of offerings. Cf. Van Seters, Life of Moses, 420. 76 Regarding the numbers “seven”: “Midrashic tradition attributes the seven altars to the need to recall the altars erected previously by seven righteous men: Adam, Abel, Noah, Abra-

154

Literary Criticism

theory of an older version in which the command to build and sacrifice stems from Balaam, yet perhaps this geographical location should be understood as the only one without an altar when they arrived.77 In such a case, only the number seven should be removed from the text there. This minor ascription to a later hand would also mandate the changing of ‫ מזבחת‬to ‫ מזבח‬in 23:14b, the removal of a single, similar-looking letter.78 To this evidence, one can add the distinct terminology for Balak’s undertaking: ‫ בנה‬in 23:14b vs. ‫ עשׂה‬in 23:2 and 30. By removing Balaam’s imperatives to build and sacrifice in 23:1–2a and 29–30a, as well as the word ‫ שׁבעה‬in 23:14b, we achieve a smoother text, and a text that does not raise suspicions about Balaam’s motives or abilities. Balaam has no need to sacrifice to receive the messages he desires and Balak initially sacrificed without Balaam’s command to do so. That is, later additions make Balaam culpable for his part in these sacrifices.79 Following the offerings, the texts remain distinct to a certain degree. The second episode presents the shortest version, relying implicitly on the more expansive first episode, and yet their striking similarities cannot be overlooked. The third episode takes the narrative in an entirely new direction and thus departs more radically from the formula of the first two. For example, God no longer encountered Balaam and put a word in his mouth. Rather, the spirit of God was upon him, precluding Balaam’s needing God’s word placed in his mouth. Balaam does not depart and, thus, must not return; Balak must not inquire – as in the more terse second episode – what YHWH said, as he has already heard Balaam proclaim it in front of him. The climactic nature of the encounter after the third oracle entirely explains the third episode’s distinction from the other two at its denouement. Balak attempted no more. Frustrated, he dismissed Balaam. Yet, Balaam took a parting shot, reminding Balak that Balaam remained steadfast, neither wavering nor offering himself for sale, before returning to his people. Compositional reasoning explains the unique conclusion of the third round of Balaam’s receiving oracles. ham, Isaac, Jacob, and Moses [Num. R. 20:8]. Sefer ha-Mivḥar, in seeking an astrological explanation (i.e., seven altars for the seven planets), is close to the mark – it was part of Balaam’s divinatory technique.”( Milgrom, Numbers, 194). 77 The names of the locations might commend this reading. Particularly in the first case, both elements of the name of the location of Balak’s sacrifice suggest some kind of cultic activity: ‫במת בעל‬. While the last case is not as certain, apparently biblical authors were familiar with some religious overtones around the name ‫ ;פעור‬cf. Num 25:3, 5; Deut 4:3; Ps 106:28; and Hos 9:10. Such an understanding may be in the background of Num 23:28, though it is not explicit. 78 The proposed literary-critical resolution makes good sense and is not strictly a text-critical matter, since no manuscripts preserved the proposed reading. 79 Cf. already Gall, “Zusammensetzung und Herkunft,” 10, who suggested one source for this material (E) that only presented Balak offering. Rost, “Fragen um Bileam,” 378 came close to this position, when he noted that the ritual with its many altars remains inconsistent with later Israelite religious mores. I.e., it should reflect negatively on Balaam.

3.4 The Main Narrative Body: Numbers 22:5–23:6, 11–17, 25–30

155

Bearing the narrative structure in mind, it seems most likely that any identification of Balaam as responsible for the offerings is secondary and should presumably cast him in a more negative light.80 This contradicts the opinion of several earlier exegetes, who often regard the offerings and the oracles as intimately related.81 By encouraging Balak to build altars and offer sacrifices in Num 23:1 and 29, Balaam explicitly acts contrary to 22:20b: by demanding that Balak sacrifice, Balaam undertakes an action not expressly mandated by God, whereupon Balak did “exactly as Balaam said” (‫)ויעשׂ בלק כאשׁר דבר בלעם‬ in 23:2a and 30a. This ascription makes Balaam at least partly responsible for Balak’s action. Particularly the plurality of altars must have presented a problem for later readers; obviously, this is never regarded positively in the Bible. But the construction of multiple altars at one location is associated with some of the most disreputable biblical characters, as well as with Israel’s enemies, and is often a theme of prophetic injunctions.82 Only rarely does anyone build (‫ )בנה‬an altar and then sacrifice (‫ )עלה‬on it without an explicit divine command to do so: Noah in Gen 8:20; Aaron and the people (Israel) in Exod 32:5–6; and Uriah and Ahaz in 2 Kings 16.83 From this list, the Bible affords only Noah a positive image regarding the described incidents. In this way, the redacted narrator suggests to the reader somewhat subversively why Balaam was ultimately bad and must be defeated. The introduction of God’s demand that Balaam only do what he says in 22:20b facilitated this negative image. This defamation of Balaam’s character matches the milieu of the preparation of his downfall in Numbers 31; that is, Num 23:1–2a and 29–30a cannot be earlier than the incorporation of Num 22:4a and 7a and probably belong to the even later additions. They cannot be regarded as part of the narrative material consisting of Num 22:1 + 3b–4a,

80 This interpolation could simultaneously be regarded as polemic against sacrificial cults, specifically non-Israelite cults. Nothing that Balak and Balaam sacrifice is able to change the mind of God, making foreign offerings – therefore – futile. 81 Cf., e.g., Rost, “Fragen um Bileam,” 380–82; Van Seters, Life of Moses, 412–13; and Achenbach, Vollendung, 412: “Die Nähe zwischen Divination und Opfer ist Grundlegend für die Erzählung.” The only evidence Achenbach and the others offered for this assertion refers to the similar structure of Xenophon’s Anabasis. That hardly presents a literarily cogent or necessary support for the biblical text at hand. 82 Cf., e.g., Deut 7:5; 12:3; Judg 2:2; 2 Kgs 16:1–16; 21:3–5; Isa 17:8; JerM 11:13; 17:1–2; Ezek 6:4–6, 13; Hos 8:11; 10:1–2, 8; 12:12; and Amos 3:14. 83 Other builders who sacrifice, all of whom were commanded to do so, include Abraham (Gen 22:9–14), Moses with the young Israelites (Exod 24:4–5), Joshua (Josh 8:30; cf. Deut 27:5–7), and Gideon (Judg 6:24–27). Some of the commands to build also come indirectly from God, just as for Jeshua, Zerubbabel, and the priests in Ezra 3:1–3. Ezra 3:1–3 cites scripture and presumes a context following Ezra 1, in which Cyrus has commanded the rebuilding of the temple. The same could be said of Gad reporting God’s command to David in 2 Sam 24 // 1 Chr 21.

156

Literary Criticism

5*, and 7a (S2), since the addition of the altars is familiar with Balak and Balaam. Neither of these characters figured in the restored version of this fragment, precluding the identification of Balaam’s encouraging Balak’s building with that composition, should it have existed as an independent unit. That is, either the editors responsible for adding it (R2) or some later editors (R3+) must have been the first to include it. Balaam’s relationship to the altars does not append to the donkey episode, so it clearly was not part of that narratologically, though it could have been part of the same Fortschreibung. Most likely one should regard it as stemming from the same hand as 22:20b, as these two matters compliment each other well.

3.5 The Oracles: Num 23:7–10, 18–24; 24:3–9, and 15–24 3.5 The Oracles: Num 23:7–10, 18–24; 24:3–9, and 15–24

A fundamental matter remains to be discussed: How are the oracles related to the narrative material? There are four possible answers to this question: 1) the oracles are older than the narrative;84 2) the narrative is older than the oracles;85 3) the oracles and the narrative were composed together from the outset;86 and 4) some portion of the oracles and narrative were originally composed together, with the rest being added later.87 Each possibility should be considered in turn. Divorced from the context of the story, the oracles by themselves remain barely comprehensible. Many oracular verses explicitly relate to the narrative context.88 One must note that the first two oracles each begin with a reference to the circumstances at hand by mentioning Balak (23:7 and 18), while the second and third oracles also reference the narrative background of the story with their recourse to the exodus from Egypt (23:22 and 24:8). Without the narrative the oracles would make little sense, and particularly the first oracle would remain devoid of content. Conspicuously, the oracles build up to a climax in the third proclamation, with the fourth oracle reversing the peoples’ fortunes ironically. This datum suggests compositional insight in constructing

84 Cf., e.g., Albright, “Oracles of Balaam”; Vetter, Seherspruch; Rost, “Fragen um Bileam”; Gross, Bileam; and Levine, Numbers 21–36, 41. 85 Cf., e.g., Gall, “Zusammensetzung und Herkunft”. 86 Eissfeldt, “Komposition der Bileam-Erzählung”. N.b. the comment of Van Seters, Life of Moses, 411, albeit in the context of the relationship between Numbers 22–24 and DAPT: “Consequently, the content of DAT does not take us very far in the specific interpretation of Numbers 22–24, although it could be used to argue that the prose and poetic oracles should be taken together as part of an organic whole.” 87 Often exegetes have considered the independence of the oracles in Numbers 24, with those in 23 always bound to their context, as a particular iteration of this theory. Cf. Mowinckel, “Ursprung”; Rudolph, Der “Elohist”, 97–127; and Schmidt, “Bileamüberlieferung”. 88 Cf. Num 23:7, 9a, 18, 20, 22–23; 24:3–8, and 15.

3.5 The Oracles: Num 23:7–10, 18–24; 24:3–9, and 15–24

157

the order of the first three oracles, with a redactional turn responsible for the fourth (see below).89 But is it possible that the narrative existed in some fashion without the oracles? Again such a circumstance seems most implausible. Removing the oracles leaves a rump of a tale that makes no sense. What does Balaam do when he leaves that angers Balak repeatedly? If no oracles or pronouncements stood there, the story would have no drama, no purpose, no sense. Reading the narrative as, e.g., Num 23:6 or even 7a leading directly to v. 11 leaves an implausible tale devoid of clarity. The same is true of 23:17 or 18a segueing into 23:25 or 24:2 or 3a into 24:10. Of course it might be possible that something else stood where the oracles now stand and that this was replaced when the oracles were incorporated, but no extant evidence supports this. Any attempt at reconstructing it must of needs remain fanciful. Something must have stood between Balaam’s reporting and Balak’s reactions, since his reactions otherwise would occur unprovoked and not follow from the preceding material. In this way, we can agree with the assessment of Zobel: “Doch auch so ist klar, daß die beiden ersten Lieder vom Erzähler für seine Erzählung gedichtet worden sind, die Erzählung nicht ohne die Lieder und diese nicht ohne die Erzählung existieren können.”90 The caveat being, of course, that the third oracle should also be included. “The poetry was composed for the sake of the prose. Without the narrative, the poetic oracles would make no sense, and all their allusions to personalities, nations, and events would be incomprehensible.”91 The simplest solution is to maintain the material currently found there as opposed to postulating something else for which no evidence exists. Yet, one exception can be found here again: the fourth oracle can be readily removed from its context – a context that it currently disturbs – without doing any violence to the story. Verse 24:25 can immediately follow 24:14a and even makes more sense when it does. Having noted that the first two possibilities remain most improbable, one of the other two must present the answer to the question about the relationship of the oracles to the narrative. Thus, either all of the oracles were originally found in their current context connected with the narrative, or only some of them are original. Based on the data discussed above, the last solution suggests the most 89

Cf. already László M. von Pákozdy, “Theologische Redaktionsarbeit in der Bileam-Perikope,” in Von Ugarit nach Qumran: FS Otto Eissfeldt, ed. J. Hempel and L. Rost (Berlin: Töpelmann, 1958), 173: “Mögen die Sprüche einen Ursprung haben, welchen sie wollen, sie sind im gegenwärtigen Zusammenhang keine sinnlosen Wiederholungen aus parallelen Quellen, sondern sie fügen sich klimaktisch zusammen unter der Hand des Redaktors.” 90 Hans-Jürgen Zobel, “Bileam-Lieder und Bileam-Erzählung,” in Die hebräische Bibel und ihre zweifache Nachgeschichte: Festschrift für Rolf Rendtorff zum 65. Geburtstag, ed. Erhard Blum, Christian Macholz, and Ekkehard Stegemann (Neukirchen-Vluyn: Neukirchener Verlag, 1990), 150. 91 Milgrom, Numbers, 467.

158

Literary Criticism

obvious answer: some of the oracles are original and integral parts of the narrative, whereas at least one presents a later addition and perhaps some elements of others are later additions as well. At first glance it appears possible that an older version of the Balaam narrative might have concluded in 23:26 (as Balak has forbidden Balaam to either bless or curse Israel in 23:25) or that 23:26 immediately preceded 24:25 in a theoretical “first draft” of the story. Such a transition would have been abrupt, but not entirely impossible. However, for narratological reasons I find this improbable. After the third oracle, Balak angrily dismisses Balaam, presenting an otherwise absent climax in their relationship. The discussion between Balak and Balaam ends their relationship with a bang, whereas an ending after the second oracle would have allowed it to simply fizzle out. The conclusion in chapter 24 seems more fitting, as 24:11–12 directly refers back to 22:17–18, Balak’s second mission to Balaam. This strong narrative link suggests that an older story of Balaam in Num 22–24* continued until 24:14a and then closed with 24:25. The more likely older conclusion can rather be found in 24:14a + 24:25 than in 23:25–26 + 24:25. Within Balaam’s first two speeches, there is no readily available evidence of extensive secondary reworking or literary expansion. That is, no obvious internal contradictions or tension stand out especially. The stichs of the oracles are presented in pairs, and, while their presence in the speeches does not always appear in a fashion one might expect (cf. the unclear context of 23:10b), no particular basis for removing any element can be identified.92 While the first speech is about 50% shorter than the second (and third), this by itself should not be used as a criterion for the diminution of the second (and third) oracle(s). One element in the third oracle stands out and should presumably be removed: the last two words of 24:8 (‫)וחציו ימחץ‬. They disrupt the context and break the consistent twofold structure of the oracle, demonstrating their secondary nature. Numbers 24:8 should end with “and their bones he will break”, which was originally followed by the – appropriate in this restored context – lion metaphor. Redactors perhaps added this element (‫ )וחציו ימחץ‬to create a pun via alliteration between the name “Egypt” (‫ )מצרים‬and the words for “piercing” (‫ )ימחץ‬and “enemies” (‫)צריו‬, as Friedman has suggested.93

92 For this reason, Angelo Tosato, “The Literary Structure of the First Two Poems of Balaam (Num. XXIII 7–10, 18–24),” VT 29, no. 1 (January 1979): 98–106 regarded them as a unity and reconstructed a common chiastic structure for the two oracles, when taken together as part of one composition. However, other considerations preclude this interpretation, as will become apparent below. 93 Cf. Friedman, Commentary on the Torah, 510.

3.5 The Oracles: Num 23:7–10, 18–24; 24:3–9, and 15–24

159

Having considered the oracles’ narrative embedding and relationship to the story in general, we can focus our attention on the oracles themselves.94 From a first glance, a paucity of internal criteria precludes dissecting the oracles too much in a literary-critical abattoir. That is, one does not immediately stumble across tensions and contradictions within each oracle or among all of them. Still, there are some elements – particularly when evaluated and contrasted with external criteria – that raise some suspicion. These can be dealt with here. The first oracle (Num 23:7b–10) consists of a total of 16 stichs, organized in pairs: two introductory stichs (23:7bα), followed by two stichs that summarize Balak’s message to Balaam from 22:6 (7bβ), and then six pairs (8a, 8b, 9a, 9b, 10a, and 10b).95 None of these pairs appears particularly distinct from its context, and none of them appears secondary based solely on internal criteria; i.e., they evince no obvious tension or contradiction to their surroundings. However, based on the external identification of a secondary layer in the narrative that addressed the people undertaking the exodus with the moniker “Israel” (as in 22:2), both 7bβ and 10a appear problematic. Removing 7bβ from its context is impossible, however, since without it the oracle neither has real content nor makes grammatical sense: 8a could not follow easily on 7bα and the subject of the rest of the oracle bases on the identification of the object in 7bβ. Therefore, the following options seem possible, that 1) the oracle in its entirety was added later; 2) the oracle was edited from a form that can no longer be reconstructed; 3) the oracle did not originally attest any explicit direct objects in 7bβ; or 4) later editors changed the original terms in 7bβ from something else into ‫ יעקב‬and ‫ישׂראל‬.96 94 For an alternative version of the editorial development of the oracles, cf. Witte, “Segen Bileams,” 197–202. I do not follow his proposals; cf. the discussion of Witte in the Introduction in Chapter One. 95 Witte, “Segen Bileams,” 201 identified two stanzas of three bicola each, organized in parallel. One could follow him regarding the stanzas, but the parallel structure is only obvious regarding the rhetorical questions in 8 and 10a. Contra Gall, “Zusammensetzung und Herkunft,” 25–26, who saw two stanzas of three diptychs with an additional gloss in 10b. He identified these stanzas with cursing (first stanza) and blessing (second stanza). Cf. Holzinger, Numeri, 116, who followed him in this ascription. Regarding Num 23:10b, Noth, Numeri, 161 commented, “Immerhin ist es wahrscheinlicher, daß ein späterer Ergänzer die unpassende Schlußzeile hinzugefügt hat, als daß der ursprüngliche Spruch diesen merkwürdigen Ausklang gehabt hätte.” However, that is not a literary-critical argument, so much as a disparaging evaluation of the content. 96 Already Gall, “Zusammensetzung und Herkunft,” 19–22 suggested an exilic or later date for the parallelization of Jacob//Israel. More recent scholars have followed this; cf., e.g., Achenbach, Vollendung, 410: “Der Parallelismus Jakob/Israel setzt die Umbenennungslegende von Gen 32,29 und von P, Gen 35,9ff. voraus und erscheint ansonsten als Stilmittel vorwiegend im Deuterojesajabuch.” Since the oldest layer may be older than that, it remains possible or probable that one would have to recognize this parallel identification as a later interpolation, unless the whole stich should be regarded as later.

160

Literary Criticism

I tend to favor the latter two options, with a preference for the third. I find no obvious reason that anyone should have expanded the narrative so expansively by introducing a new first oracle (to the first option), particularly one that does not add much actual content to what is said in the other oracles. Really, what would that add to the story other than mostly fluff? The core meaning of the oracle is simply that the people is not cursed, nothing more. For some structural reasons that will become apparent below, I also disregard the addition of the whole oracle at a later time. It remains impossible to clearly define the boundaries of what would have had to been added in terms of narrative framing to afford such an addition, at least within the confines of literary criticism. To the second option, this is purely speculative, but admittedly entirely possible and plausible. It nonetheless remains the option of last resort, in my opinion, if the evidence provides for an alternative. That only leaves the third or fourth options: the oracle contained no names in 23:7bβ or the names were changed in 7bβ to reflect the parallelization of Jacob and Israel. This probably would have had to occurr at the time of the addition of v. 10a and the fourth oracle; cf. 24:17bαβ and 18b–19a and the discussion to the fourth oracle below. Should it have actually been the case that the names were editorially adjusted, I would commend the nomenclature from 22:6: ‫ ;העם הזה‬perhaps ‫אתו‬ stood as a parallel to this phrase (cf. 22:11bα). These terms parallel easily (as they already do in 22:11), identify the same object as the current terms, and are inherent to the oldest narrative level of the story. Note particularly the similar style and vocabulary in 22:6aα and 23:7bβ; Num 22:17b also presents a solid similar, though somewhat distinct, parallel. That makes them at least a plausible and sensible option. The similarities between 23:7bβ and 22:6aα, and 12bα cannot be overlooked. Somewhat more distinct, yet nonetheless relevant are probably 11bα and 17bα, which – other than the synonym ‫ קבב‬for ‫ – ארר‬are essentially identical. The common elements favor recovering the postulated older readings ‫ העם הזה‬and ‫ אתו‬in this oracle. Otherwise, it remains possible – though clunky – to reconstruct the oracle without an object in these phrases. Either of these alternatives seems preferable to inventing other plausible, though fanciful readings. Bearing these factors in mind, I suggest that the first oracle originally consisted of seven pairs of stichs in a slightly altered form, with the names ‫יעקב‬ and ‫ ישׂראל‬either inserted or having replaced ‫ העם הזה‬and ‫אתו‬. Were that the case, they presumably would have had to stand in that order. At a later date, probably in tandem with the addition of the fourth oracle, a later scribe added 10a, bringing the total to eight pairs or sixteen stichs. This addition may have matched the first redactional layer identified in Numbers 22 as reflected in its preference for the term “Israel” for the group undertaking the exodus. Continuing with the second oracle (Num 23:18b–24), the final form of the poem consists of eleven pairs of stichs, beginning with a reference to the situation into which the poem was spoken (18b). After this, Balaam continues,

3.5 The Oracles: Num 23:7–10, 18–24; 24:3–9, and 15–24

161

delivering ten more pairs (19a, 19b, 20, 21a, 21b, 22, 23a, 23b, 24a, and 24b). As in the first oracle, no internal criteria really mandate the removal of elements from this poem, whether as tensions or contradictions. Again however, as in the first oracle, some of these lines feature the parallel identification of Jacob / Israel: 21a, 23a, and 23b. That would, should it also have been the case in the first oracle and in Num 22:2, imply that these additions were secondary in their context. To this one can consider ‫ נבט‬in 21a as having a somewhat Deuteronomistic flavor;97 cf. Gen 15:5; Exod 3:6; 33:8; Num 12:8; and several texts in Samuel and Kings. One other element could also be redactional as well, though this is somewhat more difficult to assign with any certainty: 23:22 contains essentially the same material as 24:8a, suggesting that some relationship, compositional or redactional, may exist between these two elements. We will consider this more below, but for now, it makes the impression that this phrase was copied into 23:22 in order to develop a closer relationship between the second and third oracles. Removing the elements that feature Jacob / Israel (21a and 23) or that were potentially copied from the third oracle (22) would thus leave the second oracle with seven pairs of stichs and the whole beginning with a reference to the situation at hand. This structure matches that of the first oracle in its older version. On the other hand, it could be argued – again based solely on external criteria – that 23:19aβ attests influence from 1 Sam 15:29aβb. Such an appreciation would again imply its redactional status in Numbers 22–24. However, I find this hardly likely. Notice how the verses do not match. First Samuel 15:29aβb reads ‫לא ישׁקר ולא ינחם לא אדם הוא להנחם‬, “He [the luster of Israel, i.e., God; cf., however, G] will not recant and he will not relent, for not a human is God that he might relent”. The verse in Samuel uses the same root as in Num 23:19a, ‫נחם‬, but whereas Samuel has it in the Niphal, Numbers records it in the Hithpaʿel. For that matter, the usage of the Hithpaʿel as in Numbers remains unique with this meaning; cf. Gen 27:42; 37:35aβ; Deut 32:36; Isa 66:13 (1QIsaa); Ezek 5:13; Ps 119:52; and 135:14. The metaphor remains distinct as well, as Samuel does not reference “lying”, ‫כזב‬, as Numbers does. Instead, Samuel reports ‫שׁקר‬, “to deal falsely”. Finally, reference in Samuel is to an ‫אדם‬, but in Numbers to a ‫בן־אדם‬. So, while the texts are thematically linked, even to a degree theologically linked, they are hardly necessarily the product of a single hand. The differences certainly outweigh their similarities, making

97 Though the form of ‫ נבט‬does not precisely match in Num 23:21a, one cannot overlook the potential connection to Jeroboam son of Nebat, the primary enemy of the Deuteronomist in Kings.

162

Literary Criticism

even the copying of one text into the other rather outside the realm of probability. The meaning of the phrase in Num 23:19–20 is that Balaam’s word should not remain unfulfilled.98 Turning, then, to the third oracle (Num 24:3b–9), we recognize in its current form eleven pairs (3b, 4b, 5, 6a, 6b, 7a, 7b, 8a, 8b, 9a, and 9b) plus two independent stichs (4a [only in M] and 8bβ* [‫)]וחציו ימחץ‬. Disregarding the textcritically suspicious addition of 4a in M,99 that leaves us with a total of 11.5 pairs, slightly longer than the second oracle. However, some of the elements here again arouse suspicion of perhaps being secondary in their context. For example, the parallel structure of blessing and cursing, as in 9b, appears editorial, just like the addition of 22:6b (see above). Regarding its syntax, one notes that for the first time in oracles, here Balaam addresses his object in the second person (together with v. 5; see below). Thematically, this half-verse, 24:9b, more than any other, anticipates the point of 24:17: the people who cursed Israel will in turn be cursed. That would commend regarding it as belonging to the same editorial level as the (oldest layer of the) fourth oracle (see below) and the insertion in 22:5 that developed the contrast between the peoples ( ‫בארץ‬ ‫)בני־עמו‬, i.e. R1. The phrase in 24:9b establishes that Balaam will be blessed for his blessing Israel and Balak will be cursed for his cursing Israel. This cursed status presumably will affect Moab as well, since, in the edited version of R1 they also carry responsibility for their fearing the people. Moab’s cursedness presents precisely the point of the original state of the fourth oracle (see below). The relationship between 24:9b and 22:6b commends regarding 22:6b as part of R1 as well. In terms of form, 24:6 appears distinct from all of the other stichs in the oracles identified thus far.100 Its idyllic presentation of Israel also falls somewhat outside of the framework of the rest (and presumes the occupation of the land in a kind of a simile of paradise). Verse 6 also, strictly speaking, lacks the parallel structure of the rest. The more expansive introduction of the third oracle remains conspicuous when contrasted to the two preceding oracles, implying that some of it may have been editorial as well (n.b. it is even longer in M, suggesting that this lengthening process continued well into the transmission of the text). I would suggest that it initially consisted of 24:3bα and 4bα: “The oracle of Balaam ben Beor, who sees the vision of the Shaddai”. Later editors added 3bβ and 4bβ to focus more attention on Balaam’s insight. Finally, one element remains suspicious for its transmitting the parallel of Jacob / Israel: v. 5. As suggested above, this presents something of a problem should it have belonged to the oldest layer. However, it remains possible that 98 Cf. Michael Fishbane, Biblical Interpretation in Ancient Israel (Oxford: Clarendon, 1985), 468. 99 Cf. the information to this verse in Chapter Two. 100 This is contra Witte, “Segen Bileams,” 201, who regarded v. 6 as older than v. 7.

3.5 The Oracles: Num 23:7–10, 18–24; 24:3–9, and 15–24

163

the parallel here presents a later addition (just as it may have been in the first oracle). We note, to this end, that the form of the rhetorical question in 24:5 does not follow the pattern for all of the other rhetorical questions in the oracles in Numbers 23–24 in which the question is built as a complete parallel: 23:8, 10a, 19b (but cf. 24:23b, which stands alone). The lack of the pure parallel structure appears even more pronounced when one considers the other Jacob / Israel parallels, which are all perfectly structured. That could imply that this phrase might not have belonged originally to the phrases that feature the parallelization of Jacob / Israel. As we suggested in the first oracle, that might imply that something else stood here as well, but it would be difficult to ascertain what. Perhaps it followed the structure proposed above for 23:7bβ, though this must remain speculation, and either proffered no explicit vocatives or included something like ‫ אתו‬and ‫העם הזה‬. Neither of these terms particularly fits the syntax, however. Like its precursor in 23:7bβ, 24:5 remains an integral part of the oracle, which would remain grammatically and structurally difficult, were one to remove it without replacing it. To what would the repeated “his”, “him”, and “he” in verses 6–9a then refer? At the same time, one must recognize the difficulty in maintaining its second-person addressing of Jacob / Israel. That all could imply that something substantially different may have stood here in an older version. Maintaining as much as possible of the current text and reflecting the postulated reconstruction in 23:5 above, I would suggest something like ‫מה טבו אהלי יעקב ומשׁכנת העם הזה‬. Yet, I admit that this is quite speculative, if not fanciful. I see, nonetheless, no better resolution. Considering those additions and problems, we can reconstruct a possible older layer of the third oracle that again consisted of seven pairs (24:3bα, 4bα, 5*, 7–9a),101 the first of which introduced the oracle. This was later expanded to consist of eleven pairs and then ultimately to the awkward 11.5 of Smr and G or 11 + 1+ 1 that we now find in M. We will consider this further below, when we address the structure of all of the oracles together. Finally, the fourth oracle complex (Num 24:15b–24) consists of several disparate elements, was probably composed later than the other three (see below), and was presumably expanded on one or more occasions.102 The current form of the oracle consists of some 13 pairs of stichs (15b, 16a, 16b, 17a, 17bαβ, 17bγδ, 18a*, 18b–19a, 20b, 21b, 22, 23b–24a, and 24b), one independent stich (19b), and three renewed introductions (20a, 21a, and 23a). The position of 101

Contra Witte, “Segen Bileams,” 201, who suggested that 6 and 9b were part of the oldest

layer. 102 Cf. Zobel, “Bileam-Lieder,” 145–46: The fourth oracle was originally independent as it is introduced with a ‫נאם‬, which when applied to human subjects always represents an independent text-tradition (cf. 2 Sam 23:1 and Prov 30:1). While this is possible, it remains by no means necessary. He regarded the fourth oracle as the oldest piece of the whole, with the third oracle and an early story composed to include it. From this, another scribe created the first two oracles and the rest of the story; cf. Zobel, “Bileam-Lieder,” 152–54. This is highly unlikely.

164

Literary Criticism

Schmitt, that these oracles develop in a chiasm, does not satisfactorily explain the structure.103 Rather, a more complicated structure evinces itself. The material from 24:14b–24 appears not only secondary in its context, but even internally inconsistent as an oracle that could have been expanded in phases. Evidence for such a theory can be readily found in the different introductions recorded in the biblical text: Balaam begins oracles in 24:15, 20, 21, and 23. Such a construction indicates editorial expansion. At this point, I reiterate only that 24:14b–24 looks secondary in its context, and it appears most likely that it was lengthened on perhaps a couple of occasions; initially someone may have added 24:14b–19a to Numbers 22–24* before a later editor added 24:20, and then other editors added 24:21–22 and 24:23–24 (and 19b); see above. However, these problems will be expounded further below. The odd number of stichs in the fourth speech may be an indication that a single stich was added somewhere, most likely 24:19b, which alone does not have a corresponding compliment. 104 The otherwise parallel structure of 24:18–19a strengthens this suggestion: Edom will be a possession // Esau will be a possession; Israel will be mighty // from Jacob someone will rule his enemies. The half-verse 19b clearly falls outside of this framework. Nor does it match the following material in 20, which again has a parallel structure. Perhaps the addition of 19b occurred at the same time as the addition of the story about Og added in 21:33–35, as this presents the only other occurrence of ‫ שׂריד‬in Numbers. It could just have easily been added later, under the influence of Num 21:33–35.105 Even in the most superficial engagement with the text, several factors become clear. First, Balaam’s self introduction is significantly longer than in the third oracle (three pairs of stichs: 15b, 16a, and 16b), but adapts and expands on elements from that oracle. Second, after eight pairs there is a new introduction to a new oracle (20a), consisting of a single pair of stichs. On this follow two more introduced oracles (21a and 23a), each consisting of two pairs of stichs (21b–22 and 23b–24). That is, this complex fourth oracle can be divided into four oracles: 15b–19a, 20b, 21b–22, and 23b–24.106 In turn, that means

103 Cf. Hans-Christoph Schmitt, “Heidnischer Mantiker,” 184. One of his key observations in this regard, that the outer oracles begin distinctly from the inner oracles, ignored the textcritical evidence of G (probably the older reading in this case) and thus remains incorrect. 104 The suggestion of Albright, “Oracles of Balaam,” 220–21 to transpose elements of 19a and 19b into 24:17 and 18a does not alleviate the issue and is highly speculative. A redactioncritical approach better resolves the tensions. 105 Obad 14 and 18 also present conspicuous usages of this root in oracles against Edom. Consequently, this reading could attest an attempt at harmonization within the developing Tanakh. 106 Cf. already Sayce, “Balaam’s Prophecy,” 1, who noted that “…the introductions in verses 20, 21 and 23 suggest either that successive prophecies have been attached one by one

3.5 The Oracles: Num 23:7–10, 18–24; 24:3–9, and 15–24

165

that we are dealing no longer with four oracles in the final form of Numbers 22–24, but with seven oracles, each of which is defined by a distinct introduction, with the form of the last three being the most similar. That could have compositional relevance (see below). Third, the fourth oracle (Num 24:15b– 19a) consists of eight pairs, matching the structure of the first oracle in its redacted form. The last two oracles also consist of a total of eight pairs, broken up by two introductions. Between these two sets of eight pairs of stichs, the fifth oracle consists only of a single pair of stichs. That is, the material from 15b–24 attests a concentric structure around the fifth oracle, that against Amalek in 24:20b: 8:2:8. Finally, it cannot be overlooked that in its final form the final series of oracles is substantially longer than those that preceded it. Above and beyond that, in their final form the oracles attest a structure in which they get progressively longer: eight pairs, eleven pairs, eleven pairs plus a single stich, and ultimately 13 pairs plus one independent stich. But what does this mean? Consider, for a moment, the short-form fourth oracle, i.e., 15b–19a. It attests some of the features that we identified with redactional material in the preceding oracles. The most poignant evidence is, of course, the parallel structures featuring Jacob / Israel. We find these in 17bα and 18b–19a. What’s more, they attest the parallels in reverse order: Jacob / Israel // Israel / Jacob. The midpoint of this oracle, then, and that which is surrounded by the Jacob–Israel parallels, consists of the oracles against Moab (17bβ) and Edom (18a*).107 These verses, 15b–19a, probably attest the oldest core of this well-planned and tightly constructed oracle.108 Someone probably composed it at the same time as the additions to the other oracles that feature Jacob / Israel and the increasing contrast between blessing and cursing (‫)ארר‬, i.e. Num 22:6b and 12b, in addition to editing 23:8b and 24:5a to contain this parallel. In order to make the fourth oracle somewhat of a good fit, the composer copied and expanded opening elements from the preceding oracles. For example, as with all of the other oracles, the fourth begins with the phrase ‫וישׂא משׁלו‬ ‫ויאמר‬. From there, the composer added the opening of the third oracle, which had itself presented a departure and transition from the first two: ‫נאם בלעם בנו‬ to the original prophecy in verses 17–19, or else that the passages they introduce have been taken from other documents of various age and ancestry.” 107 It remains unclear to what ‫ שׁת‬might have referred. Achenbach, Vollendung, 423 suggested the mention of ‫ שׁת‬in Num 24:17bδ only makes sense if one combines it with the mentioning of the elders of Midian in 22:4a, and he saw them as an exegesis of 2 Sam 8:12 (cf. 1 Chr 18:11). This connection is hardly evident and rather it might have something to do with Amalek and not Midian. According to Gen 36:12, however, Amalek belonged to Esau’s progeny and – unlike Midian – does stem not directly from Abraham’s (cf. Gen 25:1–4). The most obvious understanding would be a parallel to “Moab”. 108 Contra Zobel, “Bileam-Lieder,” 145, who regarded vv. 18–19 as an expansion on vv. 15 and 17.

166

Literary Criticism

‫בער ונאם הגבר שׁתם העין‬. Then, he or she went one step further and provided even more background for Balaam (‫)נאם שׁמע אמרי־אל וידע דעת עליון‬,109 before returning to the opening of the third oracle (‫)מחזה שׁדי יחזה נפל וגלוי עינים‬. Only then did he or she begin the final oracle, which in its current form is by far the longest if one includes everything through 24:24. Even should one include material only as far as v. 19a, it is still longer than the first oracle: it has 16 stichs to the first oracle’s 14, while the second and third oracles each have 22. To date this addition more concretely, we can consider Achenbach’s observation. He noted that astral metaphors have their roots in the late Babylonian and Persian periods; cf. Isa 14:5ff. and Ezek 32:7–8. They were widely accepted and explicated in the Hellenistic period, as in Dan 8:9ff.110 The passage in Isaiah to which he probably referred is Isa 14:12–15. But Num 24:17a has precisely the opposite function as Isa 14:12–15: the astral image in Isaiah is one of the king regarding himself as mighty, but falling from that high station. Precisely the same is true of Ezek 32:7–8 and Dan 8:9–10: astral imagery is not positively applied to any regent in the Bible outside of Numbers 24. So, should there be any relationship between these other biblical texts and Numbers 24, it is an inverse relationship. One side presumably reacted to the other and since the negative side seems to have come to dominate, it remains more likely that the positive imagery must have been older. That would place this editorial addition pretty firmly in the Neo-Babylonian period. That this material should reflect on David seems most logical.111 However, this redaction apparently already reckoned with this oracle stemming from some narratological period before David, as demonstrated by v. 17a.112 That is, the text points toward the time of David from some preceding period, at least in terms of narrative logic. The established setting of the Balaam narrative during the exodus probably mandated this, as seen in, e.g., 22:4b–5. Staying with the material often identified with the fourth oracle for a moment, but which here has been described as the fifth (24:20b), sixth (21b–22), and seventh oracles (23b–24), we notice some more structural matters. First, the final two oracles consist of a number of stichs equal to the number in the fourth. That is, the sixth and seventh oracles together (21b–24) can be understood as a parallel of the fourth oracle (15b–19a) and probably stem from a later period. The fifth oracle may have been added at the same time as the sixth and the seventh to serve as the center of the concentric structure, but was, perhaps more likely, later inserted to divide these two oracles and serve as a kind 109

The Masoretic attestation of ‫ נאם שׁמע אמרי־אל‬in 24:4 represents a late harmonizing gloss. Cf. Smr, G, and the preceding chapter on textual criticism. 110 Cf. Achenbach, Vollendung, 422. 111 Cf. Klaus Seybold, “Das Herrscherbild des Bileamorakels Num. 24,15–19,” TZ 29 (1973): 1–19. Whereas Seybold sought to date it to the time of David, I prefer to focus on its literary connection to the figure David. 112 Cf. Seybold, “Herrscherbild,” 10–11.

3.5 The Oracles: Num 23:7–10, 18–24; 24:3–9, and 15–24

167

of lyrical and ideological climax of the final oracular composition in 15b–24.113 To that end, it stands in the middle of the concentric structure. The insertion of 19b could have occurred in conjunction with either of these later expansions (or even later).114 Returning to the larger structure of the oracles, one notes distinct structures apparent in each version. The oldest layer consisted of three oracles, each of which began with one pair of introductory stichs and reported seven pairs of stichs in total. The structure was 7:7:7 with a climax about the anticipated king, who would be greater than Agag (24:7), metaphorically devouring the nations of his enemies.115 Notarius’ discursive analysis of the oracles supports this appreciation, even if that was not her intent. She recognized a development in the addressees (Balak in the first two oracles and Israel in the third).116 A similar development occurs over the course of the oracles regarding the speech acts, from a report to an argument to a blessing.117 The oracles’ aspects feature this development as well, from simultaneity, to atemporality, to anticipation.118 the examples from Notarius demonstrate the development of these oracles from a logical perspective. The first redaction added the parallels about Jacob / Israel and explicated the contrast between Balak’s desired cursing and Balaam’s bespoken blessing to the end that those who bless Israel would be blessed and those who cursed Israel would be cursed (24:9b). This, of course, naturally climaxed in the newly composed fourth oracle (24:15b–19a), anticipated by the new redactional conclusion of the third oracle (24:9b). The oracular structure of this new “edition” was 8:11:11:8 pairs of stichs, featuring lengthier introductions in the third and fourth oracles. The climax of this version was the anticipated crushing of Moab and Edom. Notarius’ discursive analysis again supports the secondary status of

113 This would be contra Smend, Erzählung des Hexateuch, 230, who regarded Num 24:23– 24 as younger than 20–22. 114 Achenbach, Vollendung, 423 suggested, however, that it was part of his HexRed, i.e., his oldest reconstructable layer. In order to afford it that status, he undertook an unnecessary conjectural textual emendation, just as BHS did. 115 No need to regard this eschatologically exists, as Hans-Christoph Schmitt, “Heidnischer Mantiker,” 187–88 suggested. Rather than referring to Saul, Hans-Christoph Schmitt thinks the mention of Agag should be understood as a reference to Haman. However, Schmitt overlooked the text-critical issues in Esther, in which the G text fails to identify Haman as ‫האגגי‬. So, while it is possible that someone understood Num 24:7 in this way, it most likely after Numbers had been translated into Greek (n.b. also the distinctions in Num 24:7 to this effect). This reference to Haman indicates, thus Schmitt, an eschatological understanding for even Balaam’s third oracle. 116 Cf. Tania Notarius, “Poetic Discourse and the Problem of Verbal Tenses in the Oracles of Balaam,” HS 49 (2008): 63–64. 117 Cf. Notarius, “Poetic Discourse,” 65–66. 118 Cf. Notarius, “Poetic Discourse,” 66–69.

168

Literary Criticism

the fourth oracle, in that she noticed that it varies from the others in terms of addressee (unspecified), speech act (prophetic vision), and aspect (future).119 Finally, later editors revised the structure on one or more occasions, leaving the whole with an increasingly long oracular structure: 8:11:11.5:13.5. This version has more than one climax: with the oracle against Moab and Edom still featuring prominently, but being figuratively overshadowed by the oracles against Amalek and the mysterious conclusion of the whole, featuring the downfall of empires.120 It should not be overlooked that the final version (not counting the text-historically difficult 24:4a) consists of seven oracles in 44 stichs plus six quasi-stichs in the introductions to the oracles (Num 23:7a, 18a; 24:3a; 15a, 20a, 21a, and 23a) for a total of 50 pairs, 100 stichs. That hardly seems coincidental. With this, we can offer a plausible, though by no means certain reconstruction for the diachronic development of the Balaam oracles within their narrative context. Originally the Balaam story contained three oracles, each consisting of seven pairs of stichs, which can be reconstructed with more or less certainty: 1) 23:7b*, 8–9, 10b; 2) 23:18b–20, 21b, 24; 3) 24:3bα, 4bα, 5*, 7–9. Later scribes probably edited the verses 23:7b and 24:5 to include the parallels Jacob and Israel, which may have replaced some older elements (the most likely candidates would be some permutations ‫ העם הזה‬and ‫ אתו‬or ‫יעקב‬, though this is quite speculative). This editing probably occurred already in the first major redactional reworking of Numbers 22–24, as evinced by the verses 23:10a, 21a, 23, and – more importantly – one of the key texts of the first redaction: 24:17–19a. Each oracle originally began with a single pair of stichs that served as a kind of introduction: 23:7bα, 18b, and 24:3bα + 4bα. The first oracle continues its introduction with an oblique reference to the narrative material in 22:4–6*, embedding this oracle in the presumed historical background. (Numbers 24:8 serves the same function.) Its opening combines the two characters and focuses on their interaction: Balak has summoned Balaam. The second and third oracles each hone in on one of these characters: the second oracle begins explicitly addressed to Balak, whereas Balaam explicitly ascribes the third to himself. This concentration on the characters demonstrates a logical progression, a dramatic development towards a climax. The reference to the “mountains” in 23:9 as part of the first oracle strengthens this observation, in that Balaam recites second and third oracles on the mountains Pisgah and Peor, respectively. In terms of content, the first oracle, delivered at Bamath, Balaam notes that the people is not cursed, but unique. This oracle’s content is both the most open 119

Cf. Notarius, “Poetic Discourse,” 63–70. The Masoretic edition clearly stands in a striking relationship to Dan 11:30, thus placing a similar historical proclamation just before Israel’s entrance into the land and before a description of the end of time. 120

3.5 The Oracles: Num 23:7–10, 18–24; 24:3–9, and 15–24

169

and the most insecure in terms of reconstruction. At its conclusion, he wishes himself well, presumably in anticipation of Balak’s negative reaction(s) to the messages he delivers. The second oracle, at Pisgah, supersedes the first in content, noting that God will do as God has said and that Balaam (in the original version) will bless the people. This people is like a lion and will devour its enemies. The metaphor of the lion expands in the third oracle, delivered at Peor. Who will raise this lion, and thus risk endangerment? Balaam looks ahead to the people’s mighty king, who will be greater than Agag. This oracle’s connection to the exodus event is noted explicitly and understood as evincing God’s might, proximity to, and affinity for the people. The systematic structure becomes obvious. “This [third] oracle is the climatic one: In the first, only God determines blessing and curse (23:8); in the second, God’s blessing cannot be revoked (23:20); in this, the third, those who bless or curse Israel will themselves be blessed or cursed.”121 Though it must be noted that an editor appended the explication of blessedness and cursedness for the subjects cursing or blessing Israel. Originally the third oracle only implied the danger for the people’s future enemies. The first redaction rebuilt the 7:7:7 pattern of three oracles by expanding each of them and adding a fourth. This reworking created a structure of 8:11:11:8, advanced the motif of blessing contrary to cursing, and incorporated the parallels Jacob and Israel. This refocusing occurred through 1) the additions of a) 23:10a; b) 23:21a and 22–23; c) 24:3bβ, 4bβ, 5*, 6, and 9b; and 2) the composition of the fourth oracle, 24:15b–19a (including the framing elements in 14b–15a). This version’s zenith appears in the oracles against Moab and Edom in 17bβ–19a, essentially the conclusion of the whole. It consists of parallel structures in a kind of chiasm: Jacob / Israel – Moab – Edom – Israel / Jacob. Including the introductions to the four oracles, we could count a total of forty pairs of stichs in the total structure, again, hardly an accident. Later redactions expanded this text on at least one occasion, but perhaps on more. These redactions 1) left the first oracle untouched; 2) left the second oracle untouched; 3) added 24:8bβ* to the third oracle, probably incidentally and late (cf. Ps 68:22); and 4) added 24:19b–24 to the fourth oracle.122 These additions, in total, created a system of increasingly long speeches (8:11:11.5:13.5) that now consist of seven formally distinct oracles.123 Including the narrative openings, the total number of pairs in this comes to 50 (= 100 lines). These additions and this restructuring framed the fourth oracle as the midpoint (three precede and three more follow), presumably maintaining or even increasing its 121

Milgrom, Numbers, 202. Hans-Christoph Schmitt regarded this “little apocalypse” as stemming from the period during Alexander the Great’s military entanglements with Persia, cf. Hans-Christoph Schmitt, “Heidnischer Mantiker,” 184–85. 123 Cf. René Vuilleleumier, “Bileam Zwischen Bibel und Deir ʿAllā,” ThZ 52 (1996): 154. 122

170

Literary Criticism

– metaphorical – centrality. Probably for this reason, more than any other, this oracle became the one with the largest reception history (cf. Qumran and the Gospel of Matthew). Determining the number of redactions necessary or probable for the addition of 24:19b–24 remains particularly difficult. It could be as few as one or as many as four. For the maximal number: a distinct redactor may have added each new oracle, meaning (in the order in which they appear in the text) one added 19b, one Amalek, one the Cainites, and one the empires. The simplest explanation would be that one added them all, but that remains uncertain, particularly regarding the disparate nature and lengths of the additions.124 I tend to favor a middle position. Based on the structure, one could argue that one editor added the oracles against Amalek and the Cainites, consisting of two introductions and three pairs of oracular verses, i.e., a somewhat parallel structure to the oracles against Moab and Edom, which also had four pairs (not counting the lengthy introduction). To round out this structure, a later editor added the oracle about Og and the empires in the vv. 23–24. That pushed the oracle about Amalek to be the midpoint of these concluding oracles: four pairs (Moab and Edom), two pairs (Amalek), four pairs (Cainites and Empires). A final editor added the phrase in 24:19b, finishing the whole thing off and increasing its similarity to Num 21:35, Deut 2:34; and 3:3, albeit regarding Edom and not Sihon or Og.125 Realistically and honestly, this final editor could have been identical with the one who added 24:23–24. Only in the fourth oracle does internal evidence for expansion really appear conspicuously, here in the renewed introductions to oracular speech in 24:20a, 21, and 23a. Otherwise only external criteria aid in this reconstruction. Nonetheless, it becomes possible to reconstruct diachronic layers of different size and quality within the oracles. The original layer, which was always part of the oldest layer identified in the narrative portions of Numbers 22–24, consisted of three oracles. Thereupon followed two to four redactions, permitting the following layering: 1) Original layer (S1): 23:7b*, 8–9, 10b; 18b–20, 21b, 24; 24:3bα, 4bα, 5*, and 7–9, with their narrative embedding in 23:7a, 18a; and 24:3a. 2) The first redaction (R1): 23:10a, 7bβ*; 21a, 22–23; 24:3bβ, 4bβ, 5*, 6, 9b, and 15b–19a. 3) A second redaction (R3+): 24:20–22. 124 According to Hans-Christoph Schmitt, “Heidnischer Mantiker,” 185, the tight linguistic connections between Num 24:20 and 23, as well as 20 and 21–22, suggest that at least the materials in 24:20–24 were composed at the same time. This remains possible. 125 Alfred Marx, “A propos de Nombres XXIV 19b,” VT 37, no. 1 (January 1987): 100–104 also recognized 19b as a late addition (though with his description “la periode exilique ou le debut de la periode postexilique”, not nearly as late as I am suggesting) and noted that it could have been added as an implied reference to Gen 27:40. That could even be added to the proposed list of biblical allusions behind the additions in this pericope.

3.6 The Narrative’s Conclusion: Numbers 24:1–2, 10–14, and 25

171

4) Another redaction: 24:23–24.126 5) Final (and/or incidental) additions: 24:8bβ* and 19b. These redactions appear generally consistent with the editorial additions identified otherwise in the narrative material in Numbers 22–24, though with some distinction. While the oracular material identified as the first redactional layer in the oracles is consistent and complimentary to the first redaction identified in Numbers 22, nothing in the material in 24:19b–24 must have belonged to the tertiary material identified in Numbers 22. That is, nothing from the oracles appears to match the context immediately subsequent to Num 22:1 + 3b– 4a, 5*, and 7a. Nothing connects these elements. Syntax and narratology indeed seem to preclude their connection. That would mean that latest oracular material (24:23–24 and 24:8bβ* and 19b) stems from some editor after the one who combined that fragmentary text with the redacted oldest source at the earliest. With that, we can return to discussing the narrative material, now with chapter 24.

3.6 The Narrative’s Conclusion: Numbers 24:1–2, 10–14, and 25 3.6 The Narrative’s Conclusion: Numbers 24:1–2, 10–14, and 25

Much of the opening verses of chapter 24 may be secondary. The reference to Balaam’s having sought out “omens” (‫ )נחשׁים‬in 24:1 seems most suspicious, as this term had not been applied to his undertakings in the preceding chapters.127 Some or all of 24:1aγ probably consequently should be identified as secondary.128 Taking a minimalist approach, I would suggest the words ‫לקראת‬ ‫נחשׁים‬, but admittedly, it could be all of 24:1aγ. The phrase (‫)לקראת נחשׁים‬ presumably attempts to caste Balaam in a somewhat negative (or at least ambivalent) light.129 It does not match the image of Balaam that we find in the oldest layer, nor that of the first redaction. That means it must be from a later editor or part of the fragment assembled in 22:1 + 3b–4a, 5*, and 7a. Appending the phrase ‫ לקראת הנחשׁים‬to 22:7a does present a legible text: “so the elders of Moab and the elders of Midian went and oracular instruments were in their hand to summon hexes”. That is either an incredible coincidence or a sign of remarkable editorial finesse. Therefore, it seems likely that one should attribute 126 Some have argued for a Seleucid or similar date for these verses; cf., e.g. Wellhausen, Composition des Hexateuchs, 351 and Gall, “Zusammensetzung und Herkunft,” 42–43. 127 However, Barré, “Portrait,” 262–63, saw ‫ נחשׁ‬in this verse as referencing some mantic practice on the previous occasions that the narrator did not describe. This assumption seems unlikely. 128 Cf. Gross, Bileam, 138–40. Gross, however, pled for a common background for this addition and the one in 22:7, which seems unlikely based on the appearance of the Midianites in 22:7. 129 Cf., e.g., Noth, Numeri, 165 and Achenbach, Vollendung, 418–19.

172

Literary Criticism

the phrase ‫ לקראת נחשׁים‬to the same material found in Num 22:1, 3b–4a, 5*, and 7a. Taken together, this piece presents the exposition of a story in which Moab and Midian sought to curse the Israelites while they were in Moab. In and of itself, it has nothing to do with Balaam or Balak thus far. The mention of “Israel” as the object of Balaam’s blessing raises suspicion in 24:1aβ, as such references have thus far been identified as editorial. That would imply that at least ‫ את־ישׂראל‬is redactional. More likely, the narrative context should be regarded as stemming from the same hand, meaning that all of 24:1aβ (‫ )כי טוב בעיני יהוה לברך את־ישׂראל‬is secondary. The desert or wilderness (‫ )מדבר‬has not played a role in the original narrative, implying that the reference in 24:1b is secondary. Since there is no way to excise only that element and leave a sensible phrase, that whole sentence must be removed. That the exodus generation was not “in the desert” at this point in the oldest reconstructed narrative of Numbers 22–24 will be affirmed in section 3.9.130 Removing these editorial elements – which do not all stem from the same redactor – means 24:1* in the oldest version consisted only of ‫וירא בלעם ולא־הלך כפעם־‬ ‫בפעם‬.131 Balaam looked, but did not go out as the previous times. Turning to 24:2, the mention of “Israel” in 2aβ raises the same suspicion as in 24:1. Here it is probably secondary as well. Likewise, “dwelling according to his tribes” (‫ )שׁכן לשׁבטיו‬in 24:2aγ raises suspicion since no mention of tribes has been made up to this point in Numbers 22–24. These words probably present secondary interpolations into the text. The phrase (‫ )שׁכן לשׁבטיו‬includes an element familiar from, and somewhat common to late priestly language, particularly when compared to other texts in Numbers. The term ‫ שׁכן‬occurs several times in Numbers, usually referring to the cloud/God dwelling in the midst of the Israelites (‫)בני ישׂראל‬.132 That could imply that it belongs to either S2 or R2. It remains most familiar in this through its representing the tabernacle, the ‫משׁכן‬. However, the root ‫ שׁכן‬sometimes refers to others than God dwelling, to something other than the ‫משׁכן‬. In Numbers, such is the case for Joshua and Caleb, Israel among the nations, and the case here at 24:2.133 Taken together, that could indicate that the usage here is neither original to the story, nor the product of (post-)priestly vernacular (see below), and therefore more likely R1. 130

Rather, the oldest strand of Numbers 21 apparently placed “the people” in the plain of Moab at the pinnacle of Pisgah overlooking the wasteland (Num 21:20). The first redaction apparently brought them to the desert (Num 21:13). 131 This only partially affirms the position of Achenbach, Vollendung, 418, in which he opined that 24:1aα marks the distinct characterization of the third oracle scene. On the other hand, cf. Achenbach, Vollendung, 419: “Die besondere Geistbegabung ist Ausdruck der Steigerung und kann, muß aber nicht zwingend als sekundäres Element beurteilt werden.” 132 Cf. Num 5:2–3; 9:17–18, 22; 10:12; and 35:34. 133 Cf. Num 14:30 (Joshua and Caleb shall live in the land) and 23:9 (Israel dwelling among the nations).

3.6 The Narrative’s Conclusion: Numbers 24:1–2, 10–14, and 25

173

The term ‫ שׁבט‬occurs six times in Numbers, but generally refers to the Israelites or some priestly group.134 This connection to ‫ בני ישׂראל‬perhaps commends a common background with Num 22:1 and 3b (= S2 or R2). On the other hand, the camping according to tribes here awakens an expectation of larger groups than the half-tribes or clans that this term references in the material in Numbers that demonstrate priestly – in the broadest sense of the term – affinity. Again, this would suggest that, while foreign to the original story, this phrase did not belong to the same diachronic level as Num 22:1 and 3b (see below). With these data in mind, the material of 24:2 that remains in the oldest version is 24:2aαb: ‫וישׂא בלעם את־עיניו ותהי עליו רוח אלהים‬. As has been noted, this introduction to the third oracle transforms Balaam in some degree, making him an explicit participant in the spirit of God and not merely the recipient of God’s word in his mouth.135 Apparently intended to stress its importance, Balaam’s third oracle begins differently than, and remains structurally distinct from the two that preceded it. This third oracle, in which Balaam does not leave Balak’s presence, represents the pinnacle of an older version of the story (see above to the oracles). Now the spirit of God adheres to Balaam and he introduces the oracle in a previously unseen manner. No longer does he begin with Balak, rather he notes his own capability. This turn in manner matches the turn in scope: he no longer concludes with relatively open statements of the paucity of the people’s cursedness, but rather notes that their king will dominate their enemies. This literary climax suggests that a man will come from Israel and rule over many peoples – possibly even Moab, though this is, at most, implicit in the third oracle. Based on the concluding conversation between Balak and Balaam in 24:10– 14a (again, one should note the climactic manner of Balak’s reaction, which now even includes a physical manifestation of his anger, seen in the clapping of his hands),136 one should consider this the most likely conclusion of an earlier “edition” of the Balaam story that proceeded immediately to 24:25. The only material in 10–14a that raises suspicion of being secondary is v. 10b in which Balak expressly contrasts the blessing with the anticipated cursing. This 134 Cf. Num 4:18 (Kohath) and 18:2 (Levi) for the priestly groups and 36:3 for “Israelites”. The exceptional cases are 32:33 (the half-tribe of Manasseh), here at 24:1 and the scepter in 24:17 M. 135 Cf. Pákozdy, “Theologische Redaktionsarbeit,” 175: The transition to the third oracle, the raising of his eyes, transitions Balaam’s activities from augury to visionary. He now possesses the spirit of prophecy (‫)רוח אלהים‬. 136 Cf. Milgrom, Numbers, 205: “Note the progression: (1) Balak mildly rebukes Balaam (23:11); (2) his anger surfaces (23:25); (3) his anger bursts forth in words and gestures.” Since there is no accompanying comment after the fourth oracle that would develop this theme, one wonders if Num 24:10 presents the episode’s original narrative climax. The clapping of hands is, “apparently a sign of contempt or derision (cf. Lam 2:15; Job 27:23).” (Budd, Numbers, 269).

174

Literary Criticism

explication presents a developed theme of the first redaction identified in Numbers 22–24. However, the terminological distinction (‫ קבב‬instead of the editorial ‫ )ארר‬demonstrates that 24:10b belongs to the oldest narrative level. “Thus the climax parallels the second scene in the story, binding the narration, or at least the series of vignettes, into a significant unit. Moreover, the final speech highlights the key motif as static. It does not change from the beginning to the end. Balaam can speak only what God gives him to speak. Whatever that is, whether blessing or curse, Balaam will report.”137

Further, the removal of this piece as secondary tends to blunt the point of the story, making it far less poignant. That affirms its status as part of the original. The sudden return to Balaam’s presenting an oracle in 24:14b after he had already announced his departure disturbs the narrative flow and interrupts the development of the story. This disruption demonstrates its redactional character. Originally, 24:14a concluded in 24:25. Additional support for this can be found in G’s reading “my place” (εἰς τὸν τόπον μου = ‫ )למקמי‬in 24:14a, which anticipates the conclusion in 24:25. The Masoretic Text’s “my people” (‫)לעמי‬ reflects the addition of Balaam’s people in 22:5 and – more significantly – focuses on the distinction between Israel and Balak’s and Balaam’s peoples for ideological reasons.138 Presumably M’s reading was also influenced by the two occurrences of ‫ עם‬in 24:14b. The reference to the later days (‫)באחרים הימים‬ cannot exclusively indicate a late dating of this addition.139

3.7 Summary of Literary-Critical Results 3.7 Summary of Literary-Critical Results

Summarizing the results of this brief literary-critical overview, we arrive at the following list of material suspected of having been secondarily introduced into Numbers 22–24: 22:1, 2, 3a* (‫)כי רב־הוא‬, 3b, 4a, 5aβ* (‫)בארץ בני־עמו‬, 5bβ* (‫)והנה כסה את־עין הארץ‬, 6* (‫)כי־עצום הוא‬, 6b, 7a, 11a, 12b, 20b, 21aβ, 22–35; 23:1–2a, 4b, 7bβγ* (‫ ישׂראל‬+ ‫)יעקב‬, 10a, 14aα* (‫)שׂדה צפים‬, 14bα* ( + ‫שׁבעה‬ ‫)ת‬, 21a, 22–23, 29–30a; 24:1aβγ* (‫כי טוב בעיני יהוה לברך את־ישׂראל וישׁת אל־‬ ‫ המדבר פניו‬and ‫)לקראת חנחשׁים‬, 2aβγ (‫)וירא את־ישׂראל שׁכן לשׁבטיו‬, 3bβ, 4bβ, 5* (‫ ישׂראל‬+ ‫)יעקב‬, 8bβ* (‫)וחציו ימחץ‬, 9b, and 14b–24. This material should be

137 George W. Coats, “Balaam: Sinner or Saint?” in Saga, Legend, Tale, Novella, Fable. Narrative Forms in Old Testament Literature, vol. 35, ed. George W. Coats, JSOTSup, vol. 35 (Sheffield: JSOT Press, 1985), 60, originally published as George W. Coats, “Balaam: Sinner or Saint?” BR 18 (1973): 1–9. 138 This idea will be developed further below. It is logical that M would focus on the people after the introduction of the fourth oracle, which distinguishes various peoples from Israel, indirectly even Balaam’s people. The emphasis here is on the different types of foreigners: those who curse Israel (Balak) and those who don’t (Balaam). 139 Cf. Lipiński, “‫ באחים הימים‬Dans Les Textes Préexiliques,” 448.

3.7 Summary of Literary-Critical Results

175

considered foreign interpolations or secondary additions to an older Balaam narrative, though it was apparently added in distinct versions. The shape of the oldest reconstructable Balaam story began with Moab’s fear of “the people” within the context of their journey toward the promised land and covered their being blessed by an outsider. Three times the reader encounters speeches about Israel’s greatness, beginning with Balaam’s inability to curse the people (23:7b–10a). The next episode continues by affirming God’s special relationship with the people and unwillingness to withdraw this blessing (23:18b–20, 21b, 24). This oldest version climaxes with the currently penultimate episode, reflecting on the people’s anticipated king, his might, and his ability to destroy the people’s enemies (24:5* + 7–9a). The story ended with the final exchange between Balak and Balaam, with the outsider Balaam returning to his home having frustrated the plans of the king who summoned him to his aid. In this first version, Balaam appears in an entirely positive light as an unexpected ally who affirms God’s exceptional relationship to the people undertaking the exodus. The secondary elements add nuance to this image of Balaam, even reversing it in some cases. Under the second rubric, reversal of attitude, one should primarily include the lengthy additions of the story of the donkey (22:21aβ + 22– 35) and the addition of Balaam’s mandating the offerings; under the first, one can consider the final oracles (24:14b–24). We will look at these first. Beginning with the oracles, verses 24:15–19a reflect on Moab’s destruction at the hands of Israel’s coming king. This oracle fully expounds the third oracle’s redactionally appended concluding anticipation (from 24:9b: those who curse Israel are cursed; cf. the ironically placed 22:6b) and focuses this condemnation on Moab, the people who tried to curse Israel under Balak’s leadership. Balak will not be able to defeat them, as he desired (22:6); rather, his fortune has been entirely reversed: someone from Israel will destroy Moab. This oracle continues against another neighboring people, the Edomites, noting that Israel will possess and rule them. Presumably this reflects on Edom’s denying passage to Israel on their exodus from Egypt (Num 20:14–21*). Thus, this oracle suggests a context familiar with the story of the Edomites in Num 20:14–21.140 Both of these elements still apparently reflect positively on the character Balaam. To these two, one might consider the addition of 24:20, which may reflect and expound upon 24:7. To this end, one could consider 2 Sam 8:11–12 // 1 Chr 18:11, which include the Moabites, Edomites, and Amalekites (among others) in a list of peoples whom David subdued and plundered. This presents the only case in which these people are listed together. While the context of Num 24:17– 19a makes it clear that Israel’s coming ruler will dominate Edom and Moab, 140

One might also note the consistent use of “Israel” in some duplicated material of Num 20:14–21 (14–18 + 21) as opposed to the use of “children of Israel” in the rest version (19–20).

176

Literary Criticism

Num 24:20 does not specify who would be responsible for Amalek’s downfall. Perhaps this reticence should be explained by insecurity about the destruction of Amalek in the combined traditions within the developing book of Samuel. Did Saul (1 Sam 14:48), Samuel (1 Sam 15:32–33), or David (2 Sam 8:11–12) ultimately destroy Amalek? The book of Samuel does not entirely resolve this issue. Nonetheless, it remains clear that the book of Samuel envisioned the Amalekites’ end, as they no longer appear later in Samuel or Kings or any narrative accounts of later periods. Numbers 24:20 may reflect this openness about Amalek’s demise. Because of its reticence about precisely who would destroy Amalek, v. 20 presented a suitable verse for the later transition to the successive imperial forces that dominate vv. 21–24. The downfalls noted in the final verses of Numbers 24 also do not explicate who will cause the destruction. This can still be regarded as a positive reflection on Balaam’s character. Yet, other additions reflect negatively on Balaam and prepare the reader for his ultimate demise as recounted in Num 31:8 and 16 (and Josh 13:22).141 Of particular interest in this light is the story of Balaam’s interaction with the donkey (22:21aβ + 22–35) on the way to meet Balak. This fable clearly mocks the famous seer, who cannot interpret the situation as accurately as his mount. Both its tone and its genre distinguish it from the surrounding material. The tale of Balaam’s donkey presents both the lengthiest and most conspicuous addition in this category, but there are other, smaller additions that demonstrate negativity towards Balaam as well. In this vein, one notes the addition of the Midianites in 22:4a and 7a, which both point toward a context that includes Numbers 25 and 31. They prepare the reader for Balaam’s demise, while at the same time anticipating the justification of his death as found in 31:16. The use of ‫ קסם‬in 22:7a lends unambiguous support to this interpretation. Other minor additions – at least in terms of length – that reflect negatively on Balaam can be identified in Num 23:1–2a, 4b, 29–30a; and 24:1aβ. They tend to cast Balaam in a somewhat negative light, reinforcing the idea that he personally mandated sacrifices (presumably viewed as illegitimate by later audiences) to influence God’s proclamation in chapter 23 and suggesting that he practiced some kind of divination in chapter 24. Both of these ideas tend to place Balaam in a less than upstanding light, at least from the perspective of later biblical tradents. Additionally, and more readily apparent, some literary expansions recontextualize the narrative into larger literary structures. The new contexts can be 141

This observation is contra Ulrike Sals, “Bileam – der lächerliche Falschprophet? Eine Widerlegung,” in Der Bibelkanon in der Bibelauslegung. Methodenreflexionen und Beispielexegesen, ed. Egbert Ballhorn and Georg Steins (Stuttgart: Kohlhammer, 2007), 163–88, who unconvincingly used a canon-critical approach to determine that Balaam must be viewed positively. The reasoning for this essentially roots in the (implicit) comparison of Balaam to Jacob and Joseph (among others) as prophets (or other upstanding men). That neither of these men, nor Balaam for that matter, are ever called prophets does not particularly strengthen her thesis.

3.7 Summary of Literary-Critical Results

177

recognized most easily in the opening verses of Numbers 22. On two occasions the story was embedded in new contexts: the older embedding in 22:2 made recourse to the encounter with the Amorites in 21:21–31, whereas the younger interpolation can be found in 22:1 + 3b. This younger insertion brings the Israelites back from the more northern city of Jazer or perhaps even the further reaches of Bashan. It remains essentially impossible from the perspective of narrative logic that these two additions occurred in the opposite order; after the addition of 22:1, there would be no need to genuflect on the Amorites after the “children of Israel” arrived in Moab from Jazer or Bashan, particularly as Numbers does not otherwise regard Og of Bashan as an Amorite.142 A chronological element should not be overlooked: why did Balak witness the downfall of the Amorites only after the Israelites return from Bashan? While the additions in 22:1–3 look to the material that precedes the Balaam narrative in Numbers 22– 24, the incorporation of the references to the Midianites in 22:4a and 7a anticipate the episodes that follow, connecting Balaam and the Moabites with the Midianites in preparation for Numbers 25 and 31. Other elements, e.g., 24:8bβ*, are more difficult to place in any particular redactional scheme. It is entirely possible that this was added at the same time as 24:14b–19a or 20, since both 24:8bβ* and 24:14b–19a + 20 possess vocabulary and ideas found in Ps 68:22. At the same time, one notes the similarity in the verbs between Num 24:8 and 2 Sam 22:39 M. Only in these two cases can one find ‫ אכל‬and ‫גרם‬. This could imply a focus on the figure David, who supposedly composed 2 Sam 22:39; Num 24:8* would thus anticipate his coming, as one could also suggest for Num 24:14b–19a + 20. The difficulty here comes in explaining the location of the addition in Num 24:8: shouldn’t it have been added into the later oracle if the same hand was responsible? I see no concrete resolution to this problem, but the simplest option appears to be ascribing 24:8bβ to the same hand as either 24:14b–19a or 20. More likely however, one could consider the refocusing on David implicit in 24:8bβ* and the parallel with Ps 68:22 as an indication that 24:8bβ represents a late addition into the Torah. It represents an attempt to make it more consistent with the developing Tanakh. That would push its date much later than that of the first redaction. Within the category of elements difficult to assign to any particular redactional hand is the notice about Israel camping according to its tribes in 24:2aβ. Only rarely does Numbers refer to the tribes of Israel as ‫ שׁבט‬as in 24:2. In the cases where it does, it generally refers to specific groups, like the family of Kohath (4:18), the Levites (18:2), and the half-tribe of Manasseh (32:33). Most often Numbers refers to the tribes of Israel with the noun ‫מטה‬, roughly 85 142 This distinguishes Numbers from the redacted text of Deuteronomy; cf. Eckart Otto, Deuteronomium 1–11, Teilband 1: 1,1–4,43, HThKAT (Freiburg; Basel, Vienna: Herder, 2012), 336.

178

Literary Criticism

times. In the aforementioned case of 18:2, it uses both of these terms. Also, the case of 32:33 matches usage in Deut 3:13, implying an innerbiblical allusion. The case of Num 36:3 uses the term most similarly to Num 24:2, yet referring to the “children of Israel” instead of just “Israel”. It is technically possible that a relationship exists between these texts. Further aid might be found in ‫שׁכן‬, which Numbers only rarely attests, often in verses that mention the “Israelites”: 5:2–3; 9:17–18, 22; 10:12; and 35:34. The only instances that do not match this usage can be found in 14:30 and 23:9. Taken with the aforementioned, this datum could suggest that the phrases ‫שׁכן‬ ‫ לשׁבטיו‬in 24:2 can be attributed to the same hand as 22:1 + 3b. However, the issue is more complicated than that. The texts in Numbers which rely on ‫ שׁכן‬as a verb for dwelling and reflect the usage of ‫ בני ישׂראל‬all use ‫ שׁכן‬exclusively for God’s dwelling among the Israelites. The usage here in Num 24:2 does not reflect this and thus remains inconsistent with most other attestations in Numbers. Add to that the usage of ‫שׁבט‬, which does not appear to be the favored term for “tribe” in the layer featuring the ‫בני ישׂראל‬, and the picture becomes clearer. The layer that relied on ‫ בני ישׂראל‬as the term for the exodus group favored ‫ מטה‬for “tribe”. A glance at the concordance permits the guidelines for the division of these terms into something like schools of thought still attested in the Bible. The term ‫ שׁבט‬occurs almost 100 times as a term for “tribe” in those books classically subsumed under the rubric of the Deuteronomistic History (Deuteronomy–Kings). Contrast this with ‫מטה‬, which – in the same corpus in M – never appears in Deuteronomy, Judges, or Samuel, and only twice in Kings, but almost 60 times in Joshua. Four points about the occurrences in Joshua merit mentioning: 1) G attests fewer of them than M;143 2) Joshua 21, the Levitical cities of refuge, accounts for almost half (27) of the attestations in that book; 3) many of the occurrences reflect the Transjordanian tribes; 144 and 4) some usage occurs within materials that reference the ‫בני ישׂראל‬.145 Taken together, these points commend regarding ‫מטה‬, and not ‫שׁבט‬, as the favored term for “tribe” for what has been identified with the ‫ בני ישׂראל‬material in Numbers 22–24. The ‫שׁכן‬, appearing immediately adjacent to ‫ שׁבט‬in Num 24:2 and there not referring to God, commends regarding the insertion of 24:2aγ as belonging to our first redactional layer. Thus, it was appended at the same time as 24:2aβ. To the last levels of redaction, one can add the insertion of 24:19b. As noted above, this element demonstrates some minor lexical affinity with the material now found in 21:33–35. Possibly the oracles in 24:20 and 21–22 come from the same hand as 14b–19a, but most likely not. More likely that they were added to 14b–19a before or after or together with 19b. The last option, the 143

G does not attest ‫ מטה‬in its Vorlage in Josh 7:18; 13:24; 18:21; 19:1, 24, and 40. Cf. Josh 13:15, 24, 29; 14:3–4; and 22:1. 145 Cf. Josh 7:1; 14:1–5; 19:51; 21:1; and 22:14. To these we can add 1 Kgs 8:1. 144

3.7 Summary of Literary-Critical Results

179

simultaneous appending of 19b + 20–21, would at least more easily explain why 19b was added to the end of the proclamation against Edom. But it still remains within the realm of possibility that 24:20 and 24:21–22 were added in two separate phases. The brevity of the comments against Amalek complicate this matter. Why does Amalek only receive two stichs (20b), whereas the other speeches all consist of four?146 Perhaps it serves a compositional function, as it is the oracle in the middle and presents the transition from Israel’s domination to foreigners dominating each other. Only in the case of the Amalekites does Balaam not specify who the destroyer will be. Should the proclamation against Amalek really serve this transitional function, it most likely should be regarded as coming from the same hand as 24:23–24. The same would then also have to be said of 21–22. However, it seems more likely that the Amalekites were simply not regarded as being as significant as the other peoples. Genesis 36:12 and 16 regard the Amalekites as subordinate to the Edomites. Such an idea may be reflected here as well, but it is by no means certain.147 But to which redactional layer can one ascribe Num 24:23–24? The evidence remains most unclear. The text has been corrupted to some degree and was apparently even updated at a very late date, i.e., even after its translation into Greek, to accord with Dan 11:30. The verb ‫ענה‬, “to oppress or afflict” is rarely used without both a subject and an object in the narrative books of the Bible. Some exceptions can be identified, but they are easily recognizable as distinct from the cases here: either they are reflexive (Niphal, Hithpaʿel, or Puʿal) or they attest the object with a preposition.148 In only two other cases is the object marked by neither the particle ‫ את‬nor an object suffix.149 More significantly perhaps, the peoples mentioned in Num 24:21–24 remain the most distant from Israel in biblical genealogical understanding, making them distinct from the preceding Moabites, Edomites, and Amalekites. The peoples named in 24:24 even reflect post-diluvian divisions of humanity, suggesting a different hand from the one that added the Amalekites and others, who are not as distantly related in Genesis’ conception. The Kenites of 24:21–22 perhaps cast an even wider net, going back to a genealogical differentiation in the antediluvian period, even to humanity’s second generation.150 The name Assyria conjoins the 146

Moab: 17b; Edom: 18–19a; Kenites: 21b–22; Kittim: 23b–24. Perhaps the opposite is true and the Amalekites should be emphasized. They are the center of a chiastic structure, after all, and present the only people in these oracles that is completely wiped out in biblical literature. Cf. Exod 17:8–16. 148 Reflexive: Gen 16:9; Exod 10:3; Lev 23:29; and Ezra 8:21; with Preposition: Ruth 1:21. 149 Exod 22:21 and Num 30:14. 150 Alternatively, could this be an oblique reference to the Midianites? N.b. Moses’ fatherin-law, regardless of his name, is often identified as a Midianite (e.g., Exod 3:1), but occasionally as a Kenite (e.g., Judg 1:16). It could have also been a reference to Midianites from a corrupted form of ‫ ;קנת‬cf. Num 32:42 and Judg 8:11. Cf. S. Abramski, “The Qenites [Hebrew],” EI 3 (April 1954): 116–24, as well as Milgrom, Numbers, 209, who identified the 147

180

Literary Criticism

oracles in 21–22 and 23–24. These last oracles place Israel in a substantially larger ethnic and geographical context, going far beyond the preceding material’s horizon. This broader perspective makes them generally distinct from the rest of the material in Numbers 22–24, which has no relation to the wider populations outside of the southern Levant. Considering the evidence, we can thus identify several distinct textual levels within Numbers 22–24, from the initial story to the final product now found in the Pentateuch. The stages can be summarized thus (cf. also Table 13 at the conclusion of this section): 1) The oldest Balaam narrative, the primary source (S1): Num 22:3a*, 4b, 5*, 6a*, 7b–11aαb, 12a, 13–20a, 21aαb, 36–41; 23:2b–4a; 5–9* (without ‫ יעקב‬and ‫ ישׂראל‬in v. 7), 10b–20 (without ‫ שׂדה צפים‬and ‫ שׁבעה‬in v. 14), 21b, 24–28, 30b; 24:1aαγ*b, 2aαb, 3abα, 4b, 5* (without ‫יעקב‬ and ‫)ישׂראל‬, 7–9a, 10–14a, and 25 2) The first redaction (R1): Num 22:2, 5* (‫)בארץ בני־עמו‬, 6a* ( ‫כי־עצום‬ ‫)הוא‬, 6b, 12b; 23:7bβγ* (‫ יעקב‬and ‫)ישׂראל‬, 10a, 21a, 22–23; 24:1aβb, 2aβ, 5* (‫ יעקב‬and ‫)ישׂראל‬, 6, 9b, and 14b–19a 3) A second, fragmentary narrative (S2): Num 22:1, 3b–4a; 5bα* ( ‫הנה‬ ‫)כסה את־עין הארץ‬, 7a; and 24:1aβγ* (‫)לקראת הנחשׁים‬ 4) The second redaction (R2), which incorporated the second fragment into R1: Num 22:3a* (‫)כי רב הוא‬, 11aβ 5) The third redaction (R3): 24:20–22 6) The fourth redaction (R4): 24:23–24 7) The negative redaction (R5): Num 22:20b, 21aβ, 22–35; 23:1–2a, 4b, 14bα* (‫ שׁדה צפים‬and ‫ ת‬+ ‫)שׁבעת‬, 29–30a; 24:3bβ, 4bβ 8) Glosses added for consistency within the developing Tanakh: 24:4a, 8bβ* (‫ )וחצו ימחצו‬and 19b. It is quite possible that the numbers 5–7 belonged to a single redactional undertaking or to other editorial processes listed here, but in order to present the maximum possible number, I have distinguished several layers within the final oracle complex. Honestly, these brief additional orations read more like contextual Fortschreibung than like portions of some larger editorial undertaking. There is hardly evidence to affirm points 5–7 as part of a consistent redaction that edited Numbers, the Pentateuch, the Hexateuch, or the Enneateuch. Such an appreciation of these elements would require further elucidation, a matter that goes beyond the focus of this chapter and even this study more generally (but cf. the discussion below). For that reason, the verses Num 24:20–24 will be regarded as elements of one redactional layer for the subsequent discussion Kenites as a “…nomadic group that attached itself to Midian, Amalek, and Israel and ranged from the Sinai Peninsula to the Galilee […] In Judges 1:16, 4:11 and 1 Samuel 15:6–7, the Kenites who settled in Canaan are represented as allies of the Israelites, suggesting that the Kenites of the following oracle are of an earlier period.”

3.7 Summary of Literary-Critical Results

181

in this chapter and the rest of this monograph (R3+). Readers should also understand this rubric (R3+) as containing the glosses and other materials on boundary between textual and literary criticism, such as number 8 in the list above. From this list four matters become apparent: 1) the vast majority of the text belongs to a single, oldest narrative level; 2) redactional insertions were – with the singular exception of the donkey story and, to a lesser extent, the creation of the fourth oracle – minor and specific; 3) the resultant literary-critical analysis precludes the atomization of the text into infinitesimal units stemming from a plethora of distinct redactional undertakings; 4) the addition of incidental material reflecting the developing consistency of Tanakh evinces a process that continued into the attested transmission of manuscripts and versions, as obviously apparent in M’s refining Num 24:24 to be more like Dan 11:30. Each version has a profile distinct from the others. The initial story covers Balaam’s interaction with Moab’s king, Balak. Balaam delivers three oracles to Balak, each expounding the message of the previous. First, Balaam pronounces his inability to curse Israel, as Balak had requested (23:7–8). In the second oracle, Balaam states that he will bless Israel (23:20, reconstructed with G, Smr, and T). These two oracles belong together and also with the third. These oracles, delivered with the divine name ‫יהוה‬, among others, cannot – contra Van Seters – have the Deuteronomic law about prophecy in mind.151 The Deuteronomic law in Deut 18:18 contradicts Balaam being regarded as a true prophet both for its future perspective (‫ )אקים‬and for his ethnicity ( ‫מקרב‬ ‫)אחיחם‬.152 The use of the future tense in Balaam’s stating of his coming blessing in the second oracle (Num 23:20) anticipates the coming of the third oracle, in which the nature of the people’s blessing is sketched in terms of the circumstances of their livelihood, well-being, and the coming of their king. The connection to Agag could imply the anticipation of Saul, as known from 1 Samuel 15.153 The third oracle fills Israel’s blessedness with substance, demonstrating 151

Cf. Van Seters, Life of Moses, 418: “For the first two oracles Yahweh is said to place the words in his mouth. It would appear, therefore, that the Balaam story has the specific model of the true prophet of Deut. 18:18–20 in mind in composing this account (cf. also Jer. 1:17; 1 Kings 22:14).” 152 Here, Van Seters contradicted himself. He regarded the Balaam account as later than Deuteronomy 18, since Balaam could not have been the idealized prophet; cf. Van Seters, Life of Moses, 419–20: “That Balaam, a foreign prophet, should be so like Moses, the model prophet, to whom Yahweh speaks without dream or vision, could hardly be envisaged or intended by Deut. 18:18. The Balaam account must be later.” That would mean that Balaam would be deceitful and that Israel was not blessed. Is that really the moral of the story in Numbers 22–24? I hardly think so. 153 Contra Hans-Christoph Schmitt, “Heidnischer Mantiker,” 187, who argued that even the third Balaam oracle should be understood eschatologically since there is nothing historical in its contents. He rejected the connection to Saul in the third oracle, since its introduction is closely related to that of the fourth oracle. This fourth oracle, particularly the metaphor of the

182

Literary Criticism

its current and future state of bliss, which will culminate in the monarchy. At this point in the original narrative, Balak angrily dismisses Balaam and each goes his own way. The historical background for this story reflects most likely a period in which Moab and Israel were in conflict.154 The most likely scenario for this is from the Northern Kingdom in the eighth or ninth centuries BCE.155 Considering the origin of this story, a few matters stand out. The oldest narrative’s geography suggests something about the story’s provenience. We recognize the importance of Moabites and an Aramean, i.e., someone from the region around Damascus. Each of these characterizations commends the text’s northern heritage. That is, the oldest biblical Balaam narrative in Numbers 22– 24* stems from the Northern Kingdom, Israel. Linguistic observations may support this analysis as well. Rendsburg has argued that Numbers 22–24, as one of only a few such texts, demonstrates dialectic characteristics similar to Aramaic, indicative of its northern provenience. It does this in order to accentuate the characterization of Balaam as an Aramean, a phenomenon he refers to as “style-switching”. Specifically, he identified Aramaic-like evidence in my reconstructed oldest narrative in the following points: 1) ‫ הררי‬in Num 23:7; 2) ‫ מלכת‬in 24:7; 3) the hitpaʿel as a passive for ‫ יתחשׁב‬in 23:9; 4) ‫ נאם‬in 24:3; 5) ‫( צרים‬cf. Aramaic ‫ )טורים‬for “mountains” in 23:9; 6) ‫ מות ישׁרים‬in 23:10; 7) the ‫ גרם‬in 24:8; 8) the ending ‫ יהם‬appended to a plural noun ending in ‫ות‬ in 24:8; and 9) the use of ‫ עדי‬in 23:18.156 Rendsburg’s observations generally preclude the characterization of the Pentateuch as a whole as a product of the Persian period, since the paucity of Aramaic influence over broad portions of text preclude its composition during that time.157 Rather, the few times that one star, certainly references David and his supposed conquest of Moab and Edom according to Schmitt. The fourth oracle’s oblique reference to David thus precludes identifying Saul in the third oracle. This, of course, presumes based solely on the introductions of the oracles that they belong to the same diachronic level. That this probably was not the case remains evident even in the text history of the passage in which 24:4 more closely matches 24:16 in M than in Smr and G. That in turn suggests that scribes long continued to emend texts for consistency, as even witnessed within the distinctions among the versions. 154 Cf. Mowinckel, “Ursprung,” 239–41. 155 Cf., e.g., Galo W. Vera Chamaza, Die Rolle Moabs in der neuassyrischen Expansionspolitik, AOAT, vol. 321 (Münster: Ugarit-Verlag, 2005), 29–60. 156 Cf. Gary A. Rendsburg, “Aramaic-Like Features in the Pentateuch,” HS 47 (2006): 169– 71. In contrast, he only recognized five such elements in material that I regard as redactional, some of which are even textual errors: 1) the noun ‫ רבע‬in 23:10 (scribal error); 2) the noun ‫נחשׁ‬ in 23:23 and 24:1; 3) the form ‫ נטיו‬in 24:6; 4) ‫ נאם‬in 24:4 and 15–16 (copied from 24:3); and 5) the noun ‫ נחלים‬in 24:6, which he ultimates related to the Arabic cognate, questioning whether it might have existed in Old Aramaic. Disregarding the error (23:10) and editorial glosses (24:4 and 15–16), the other material all belongs to the first redactional layer (23:23; 24:6) or could be based on it (24:1). 157 Cf. the few scattered examples in Rendsburg, “Aramaic-Like Features in the Pentateuch,” 172–75.

3.7 Summary of Literary-Critical Results

183

finds Aramaic influence it is either for stylistic or geographical reasons. In this case, it seems to be both: Northerners describing an Aramean with Aramaic (sounding) linguistic features. This commends regarding the core of Numbers 22–24 as a product of the north. The supposed tension with Moab and the connection of Transjordan and Aram-Damascus presumed in the story’s background commend a dating for the core narrative in the ninth or eighth centuries BCE.158 The reference to Agag could refer to some earlier version of the Saul tradition in 1 Samuel 15 that still regarded Saul as a hero. The first redaction edited this older version by placing it in a new, presumably larger narrative context and adding the logical progression of Balaam’s third oracle: Israel will now be responsible for the demise of Moab and other neighboring peoples. This outcome presumes and expounds upon the conclusion of Balaam’s last statement in his third oracle. In an ironic twist, the new version of the third oracle ends with the explicit reversal of Balak’s expected outcome; rather than everyone being blessed or cursed depending on Balaam’s whim (22:6b), Balaam transmits from God that peoples are blessed or cursed depending on their regard for God’s people (24:9b). By trying to curse Israel, Balak and – because of him – Moab have become objects of a curse. The opening lines of the fourth oracle explain precisely what this curse entails: Israel’s coming monarch will dominate Moab and even the neighboring kingdom of Edom. This first editor maintained a more localized focus; the mention of Edom could be indicative of a southern, Judean perspective. The dominated peoples named suggest a particular interest in the figure David as opposed to Saul.159 By culminating Balaam’s oracles against Judah’s enemies, this redaction generally maintains the largely positive tone toward Balaam. He became an explicit ally of God’s people against their foes, present and future. The second redaction changed the context preceding the Balaam narrative by including the story the Moabites and Midianites going out to curse the Israelites. This fragmentary story, and thus the minor redaction that incorporated it into the Balaam narrative, anticipated Numbers 25* and 31. Unique among the other editorial undertakings identified here, the few elements included in this second layer identified at the outset of Numbers 22 can be read together to

158

The discussion of the inscription from Deir ʿAlla in Chapter Five will develop this proposal further. 159 I reject any eschatological or messianic understanding of this verse stemming from the time of its composition. That later scribes and scholars understood it this way stands without question, but initially this probably only referred to the literary figure David as known in Samuel. My position is, thus, contra Hans-Christoph Schmitt, “Heidnischer Mantiker,” 186, who argued that the anticipation of David presumed by Num 24:17 assumed an eschatological character, much like Zech 9:9, from its inception. This eschatological understanding is also inherent to the text beginning in Num 24:15. It refers, according to Schmitt, to a new David coming to restores Israel’s control over its neighbors.

184

Literary Criticism

form the exposition of a separate episode. Combining Num 22:1, 3b, 4a, 5bα* and 24:1aγ* creates a cogent beginning of a narrative: “And the Israelites set out and camped in steppes of Moab across the Jordan from Jericho. And Moab was disgusted by the Israels, so Moab spoke to the elders of Midian: ‘Now this assembly will devour our whole surrounding, just as the bull devours the green of the field. And Behold! It covers the visible landscape!’ And the elders of Moab and the elders of Midian went (and magical items were in their hand to conjure spells)”.

This reads like the opening of a conflict between the Israelites, Moabites, and Midianites, precisely the circumstances we find in Numbers 25 and 31. That could support a common compositional background. The next chapter will address this possibility briefly. The possibility should not be discounted from the outset that the material identified as a second interpolation above, could in fact present at least a fragment of a second source. This case of the conflict between the Israelites, Moabites, and Midianites, however, makes no mention of Balaam. As a brief aside, it would even be possible to include the editorial gloss in Num 22:5aβ* (‫ )בארץ בני־עמו‬in this source, particularly with a minor modification. If the term ‫ עמו‬originally read ‫עמנו‬, in a conjectural variant that no longer exists, this phrase would fit perfectly in the context: “Just as the bull licks up the green of the field in the land of our people(s). Behold!” While this is a tenuous conjecture, it remains astounding how well it fits the context and could simultaneously explain the variant readings of ‫ עמו‬and ‫ עמון‬in the traditions. This addition permitted the increasingly negative tone toward Balaam in later traditions.160 The final redactions cast Balaam in a more negative light, making him more foolish than his donkey and implicating him as complicit in sponsoring Balak’s inappropriate offerings. Whereas most of the work of the first two redactions and Fortschreibungen was undertaken at the edges of the story in Numbers 22– 24, these final editors included expansive elements in the middle as well. They added more material into the last two oracles, leaving the first two essentially untouched. Picking up on the inclusion of Og from Num 21:33–35, a later redaction or Fortschreibung incorporated the oracle now found in Num 24:23– 24, which also includes Og in an older version attested by G. The audience appreciates Balaam presenting these final oracular statements seeing the defeated army and dead body of Og. For this reason, the final oracles open with his exasperated rhetorical question: who can survive when God does this? This editor takes a much larger vision of the relationships in Israel’s future. No longer are only smaller people groups involved, even the great empires have been added, making the perspective more universal. This editorial expansion should probably be dated to the end of the Persian period or the beginning of 160 Contra Robert R. Wilson, Prophecy and Society in Ancient Israel (Philadelphia: Fortress Press, 1980), 150, who identified already the Elohist as the negative editor of the Balaam material.

3.7 Summary of Literary-Critical Results

185

the Hellenistic period at the earliest, as suggested by the reference to the Greeks.161 With the conclusion of the preceding survey, we can return to our introductory comments in Chapter One. The various compositional models proffered for Numbers 22–24 all find some justification in the study here. The core of Numbers 22–24, both narrative and oracular, consists of a uniform story with a beginning, a middle, and an end, presumably from the hand of some individual, anonymous scribe. A fragment of a second source may have been edited into this earliest source. However, neither of these sources looks anything like J or E as in the iterations of the Documentary Hypothesis. At the same time, clear evidence of redactional work cannot be ignored, whether in the narrative or in the oracles. Several hands have seen fit to add material to the oldest version of Numbers 22–24, presumably in several phases throughout history (though this will be discussed briefly in the next sections). None of these editors look anything like a J or an E editor in redaction-historical constructions that develop such.162 Yet not all of this redactional work can conspicuously be ascribed to larger recensional undertakings within Numbers, the Pentateuch, the Hexateuch, or the Enneateuch. Two varieties an be appreciated: those that look like Fortschreibungen and elements that, when combined, read like a second source has been incorporated into the narrative. Each of these will also be described in overview in the next sections. The main thrust of the oldest version of the Balaam story transmitted in Numbers 22–24* sought to affirm the power of Israel over and against Moab. One cannot overlook the lack of military prowess anticipated in the text, at 161 I favor a dating in the Persian or Hellenistic Period as opposed to the Roman period, as suggested already by Gall, “Zusammensetzung und Herkunft,” 43–47. Von Gall’s suggestion that G and Smr were constantly updated to afford changes to the proto-M seems less likely than their stemming from a common Vorlage before the Roman Era. 162 Could one potentially identify a ‫ יהוה‬redaction in Numbers 22–24? Discounting the usages in ‫ מלאך יהוה‬in 22:22–35 (i.e., 22–27, 32, 34–35), the Tetragrammaton appears in 22:8a, 13b (not G), 18, 19b, 28, 31; 23:5 (not G), 8, 12 (not G), 16a (not G), 17b (cf. 23:6–7), 21b, 26 (not G); 24:1aβ, 6bβα, 11b, and 13. Of these, 22:28, 31; 24:1aβ and 6 clearly present editorial interpolations. Of the remaining thirteen appearances, G lacks five, precluding them as part of some “Yahwistic redaction”. Of the remaining eight, 23:8 and 21b are firmly anchored in the original oracles and cannot be removed as interpolations. The narrative’s logic requires the remaining six occurrences (22:8aγ, 18–19; 23:17b; 24:11b, and 13). The removal of 22:8aγ leaves Balaam’s demand that the officers spend the night baseless; 13b presents the same problem. Removing these elements and 18–19 would completely remove the motif of the night and would have God appear not only to Balaam, but to the officers, and the dismissal in chapter 24 reads unusually without the references to YHWH. Really, no reason appears to regard any of this as data for a Yahwistic redaction. If there had been one, it would have been minute outside of the addition of the donkey story. More likely, it seems like the variance in the divine names could represent recensional undertaking in either direction, toward YHWH or the more generic Elohim.

186

Literary Criticism

least on the side of Israel. Israel does not set out against Moab with columns of soldiers, but arrives and survives only with the support of its God. Even in the oldest version of this tale it appears that Balaam recognized his inability to undermine YHWH’s relationship with Israel. That is, an Aramean mantic called to assist Moab cannot overcome Israel. Rather, he recognizes that Israel will rise into a mighty kingdom. This could speak to the historical situation behind the composition of this oldest version. Later editors reinterpreted this story for their own contexts, updating the form, content, and understanding of the story. However, it must still be explained what process or processes could have been responsible for the later incorporation of this material into Numbers 22– 24*. The earliest version and each of the expansions must be placed, where possible and necessary, in broader narrative and more specific historical contexts. These tasks – and how the redactions within Numbers 22–24 relate to redactions within the rest of the Pentateuch and other larger narrative works found in the Hebrew Bible – will be undertaken immediately. R3+

R2

‫כי רב־‬ ‫הוא‬

S2

R1 Numbers 22 ‫ויסעו בני‬ ‫ישׂראל ויחנו‬ ‫בערבות מואב‬ ‫מעבר לירדן‬ :‫ירחו‬ ‫וירא בלק בן־‬ ‫צפור את כל־‬ ‫אשׁר־עשׂה‬ :‫ישׂראל לאמרי‬ ‫ויקץ מואב‬ ‫מפני בני‬ :‫ישׂראל‬ ‫ויאמר מואב‬ ‫אל־זקני מדין‬ ‫עתה ילחכו‬ ‫הקהל הזה את־‬ ‫כל־סביבתינו‬ ‫כלחך השׁור‬ ‫את ירק השׂדה‬

‫והנה כסה את־‬ ‫עין הארץ‬

‫בארץ בני־עמו‬

S1

V. 1

2

‫ויגר מואב מפני העם‬ ‫מאד‬

3

4a

‫ובלק בן־צפור מלך‬ :‫למואב בעת ההוא‬ ‫וישׁלח מלאכים אל־‬ ‫בלעם בן־בעור‬ ‫פתורה אשׁר על־‬ ‫הנהר לקרא־לו לאמר‬ ‫הנה עם יצא ממצרים‬ :‫והוא ישׁב ממלי‬

4b 5

‫‪187‬‬

‫‪3.7 Summary of Literary-Critical Results‬‬

‫‪6‬‬

‫ועתה לכה־נא ארה־‬ ‫לי את־העם הזה ממני‬ ‫אולי אוכל נכה־בו‬ ‫ואגרשׁנו מן־הארץ‬

‫‪7‬‬

‫ויבאו אל־בלעם‬ ‫וידברו אליו דברי‬ ‫בלק‪:‬‬

‫‪8‬‬

‫ויאמר אליהם לינו פה‬ ‫הלילה והשׁבתי‬ ‫אתכם דבר כאשׁר‬ ‫ידבר יהוה אלי וישׁבו‬ ‫שׂרי־מואב עם־בלעם‪:‬‬ ‫ויבא אלהים אל־‬ ‫בלעם ויאמר מי‬ ‫האנשׁים האלה עמך‪:‬‬ ‫ויאמר בלעם אל־‬ ‫האלהים בלק בן־צפר‬ ‫מלך מואב שׁלח אלי‬ ‫לאמר‪:‬‬ ‫הנה עם יצא ממצרים‬ ‫וישׁב ממלי עתה לכה‬ ‫קבה־לי אתו אולי‬ ‫אוכל להלחם בו‬ ‫וגרשׁתיו‪:‬‬ ‫ויאמר אלהים אל־‬ ‫בלעם לא תלך עמהם‬

‫‪9‬‬

‫‪10‬‬

‫‪11‬‬

‫‪12‬‬

‫‪13‬‬

‫‪14‬‬

‫‪15‬‬

‫‪16‬‬

‫‪17‬‬

‫ויקם בלעם בבקר‬ ‫ויאמר אל־שׂרי בלק‬ ‫לכו אל־אדנכם כי‬ ‫מאן יהוה לתתי להלך‬ ‫עמכם‪:‬‬ ‫ויקומו שׂרי מואב‬ ‫ויבאו אל־בלק ויאמרו‬ ‫מאן בלעם הלך עמנו‪:‬‬ ‫ויסף עוד בלק שׁלח‬ ‫שׂרים רבים ונכבדים‬ ‫מאלה‪:‬‬ ‫ויבאו אל־בלעם‬ ‫ויאמרו לו כה אמר‬ ‫בלק בן־צפור אל־נא‬ ‫תמנע מהלך אלי‪:‬‬ ‫כי־כבד אכבדך מאד‬ ‫וכל אשׁר־תאמר אלי‬ ‫אעשׂה ולכה־נא‬

‫כי־עצום הוא‬ ‫כי ידעתי את‬ ‫אשׁר־תברך‬ ‫מברך ואשׁר‬ ‫תאר יואר‪:‬‬ ‫וילכו זקני‬ ‫מואב וזקני‬ ‫מדין וקסמיהם‬ ‫בידם‬

‫ויכס‬ ‫את־‬ ‫עין‬ ‫הארץ‬ ‫לא תאר את־‬ ‫העם כי ברוך‬ ‫הוא‪:‬‬

‫‪Literary Criticism‬‬

‫‪18‬‬

‫‪19‬‬

‫‪20‬‬

‫‪21‬‬ ‫‪22‬‬

‫‪23‬‬

‫‪24‬‬

‫‪25‬‬

‫‪26‬‬

‫קבה־לי את העם‬ ‫הזה‪:‬‬ ‫ויען בלעם ויאמר אל־‬ ‫עבדי בלק אם־יתן־לי‬ ‫בלק מלא ביתו כסף‬ ‫וזהב לא אוכל לעבר‬ ‫את־פי יהוה אלהי‬ ‫לעשׂות קטנה או‬ ‫גדולה בלבי‪:‬‬ ‫ועתה שׁבו נא בזה‬ ‫גם־אתם הלילה‬ ‫ואדעה מה־יסף יהוה‬ ‫דבר עמי‪:‬‬ ‫ויבא אלהים אל־‬ ‫בלעם לילה ויאמר לו‬ ‫אם־לקרא לך באו‬ ‫האנשׁים קום לך אתם‬ ‫ויקם בלעם בבקר‬ ‫וילך עם־שׂרי מואב‪:‬‬

‫‪188‬‬

‫ואך את־הדבר‬ ‫אשׁר־אדבר אליך‬ ‫אתו תעשׂה‪:‬‬ ‫ויחבשׁ את־אתנו‬ ‫ויחר־אף אלהים‬ ‫כי־הולך הוא‬ ‫ויתיצב מלאך‬ ‫יהוה בדרך לשׂטן‬ ‫לו והוא רכב על־‬ ‫אתנו ושׁני נעריו‬ ‫עמו‪:‬‬ ‫ותרא האתון את־‬ ‫מלאך יהוה נצב‬ ‫בדרך וחרבו‬ ‫שׁלופה בידו ותט‬ ‫האתון מן־הדרך‬ ‫ותלך בשׂדה ויך‬ ‫בלעם את־האתון‬ ‫להטתה הדרך‪:‬‬ ‫ויעמד מלאך יהוה‬ ‫במשׁעול הכרמים‬ ‫גדר מזה וגדר‬ ‫מזה‪:‬‬ ‫ותרא האתון את־‬ ‫מלאך יהוה‬ ‫ותלחץ אל־הקיר‬ ‫ותלחץ את־רגל‬ ‫בלעם אל־הקיר‬ ‫ויסף להכתה‪:‬‬ ‫ויוסף מלאך־יהוה‬ ‫עבור ויעמד‬ ‫במקום צר אשׁר‬

‫‪189‬‬

‫‪27‬‬

‫‪28‬‬

‫‪29‬‬

‫‪30‬‬

‫‪31‬‬

‫‪32‬‬

‫‪33‬‬

‫‪3.7 Summary of Literary-Critical Results‬‬ ‫אין־דרך לנטות‬ ‫ימין ושׂמאול‪:‬‬ ‫ותרא האתון את־‬ ‫מלאך יהוה‬ ‫ותרבץ תחת‬ ‫בלעם ויחר־אף‬ ‫בלעם ויך את־‬ ‫האתון במקל‪:‬‬ ‫ויפתח יהוה את־‬ ‫פי האתון ותאמר‬ ‫לבלעם מה־‬ ‫עשׂיתי לך כי‬ ‫הכיתני זה שׁלשׁ‬ ‫רגלים‪:‬‬ ‫ויאמר בלעם‬ ‫לאתון כי‬ ‫התעללת בי לו‬ ‫ישׁ־חרב בידי כי‬ ‫עתה הרגתיך‪:‬‬ ‫ותאמר האתון‬ ‫אל־בלעם הלוא‬ ‫אנכי אתנך אשׁר־‬ ‫רכבת עלי מעודך‬ ‫עד־היום הזה‬ ‫ההסכן הסכנתי‬ ‫לעשׂות לך כה‬ ‫ויאמר לא‪:‬‬ ‫ויגל יהוה את־עיני‬ ‫בלעם וירא את־‬ ‫מלאך יהוה נצב‬ ‫בדרך וחרבו‬ ‫שׁלפה בידו ויקד‬ ‫וישׁתחו לאפיו‪:‬‬ ‫ויאמר אליו מלאך‬ ‫יהוה על־מה‬ ‫הכית את־אתנך‬ ‫זה שׁלושׁ רגלים‬ ‫הנה אנכי יצאתי‬ ‫לשׂטנך כי־ירע‬ ‫הדרכך לנגדי‪:‬‬ ‫ותראני האתון‬ ‫ותט מלפני זה‬ ‫שׁלשׁ רגלים ולולי‬ ‫נטתה מפני כי‬ ‫עתה גם־אתכה‬ ‫הכיתי ואתה‬ ‫החייתי‪:‬‬

‫‪Literary Criticism‬‬ ‫‪34‬‬

‫ויאמר בלעם אל־‬ ‫מלאך יהוה‬ ‫חטאתי כי לא‬ ‫ידעתי כי אתה‬ ‫נצב לקראתי‬ ‫בדרך ועתה אם־‬ ‫רע בעיניך‬ ‫אשׁובה לי‪:‬‬ ‫ויאמר מלאך‬ ‫יהוה אל־בלעם‬ ‫לך עם־האנשׁים‬ ‫ואפס את־הדבר‬ ‫אשׁר־אדבר אליך‬ ‫אתו תדבר וילך‬ ‫בלעם עם־שׂרי‬ ‫בלק‪:‬‬

‫‪35‬‬

‫‪36‬‬

‫‪37‬‬

‫‪38‬‬

‫‪39‬‬ ‫‪40‬‬

‫‪41‬‬

‫‪190‬‬

‫וישׁמע בלק כי בא‬ ‫בלעם ויצא לקראתו‬ ‫אל־עיר מואב אשׁר‬ ‫על־גבול ארנן אשׁר‬ ‫בקצה הגבול‪:‬‬ ‫ויאמר בלק אל־בלעם‬ ‫הלא שׁלח שׁלחתי‬ ‫אליך לקרא־לך למה‬ ‫לא־הלכת אלי‬ ‫האמנם לא אוכל‬ ‫כבדך‪:‬‬ ‫ויאמר בלעם אל־בלק‬ ‫הנה־באתי אליך עתה‬ ‫היכול אוכל דבר‬ ‫מאומה הדבר אשׁר‬ ‫ישׂים אלהים בפי אתו‬ ‫אדבר‪:‬‬ ‫וילך בלעם עם־בלק‬ ‫ויבאו קרית חצות‪:‬‬ ‫ויזבח בלק בקר וצאן‬ ‫וישׁלח לבלעם‬ ‫ולשׂרים אשׁר אתו‪:‬‬ ‫ויהי בבקר ויקח בלק‬ ‫את־בלעם ויעלהו‬ ‫במת בעל וירא משׁם‬ ‫קצה העם‪:‬‬ ‫‪Numbers 23‬‬

‫‪1‬‬

‫ויאמר בלעם אל־‬ ‫בלק בנה־לי בזה‬ ‫שׁבעה מזבחת‬ ‫והכן לי בזה‬

‫‪191‬‬

‫‪3.7 Summary of Literary-Critical Results‬‬

‫‪2‬‬

‫ויעל פר ואיל במזבח‪:‬‬

‫‪3‬‬

‫ויאמר בלעם לבלק‬ ‫התיצב על־עלתך‬ ‫ואלכה אולי יקרה‬ ‫אלהים לקראתי ודבר‬ ‫מה־יראני והגדתי לך‬ ‫וילך ויתיצב בלק על‬ ‫עלתו ובלעם נקרה‬ ‫אל אלהים וילך שׁפי‪:‬‬ ‫ויקר אלהים אל־‬ ‫בלעם‬

‫‪5‬‬

‫וישׂם יהוה דבר בפי‬ ‫בלעם ויאמר שׁוב‬ ‫אל־בלק וכה תדבר‪:‬‬ ‫וישׁב אליו והנה נצב‬ ‫על־עלתו הוא וכל־‬ ‫שׂרי מואב‪:‬‬ ‫וישׂא משׁלו ויאמר‬ ‫מן־ארם ינחני בלק‬ ‫מלך־מואב מהררי־‬ ‫קדם לכה ארה־לי‬ ‫ולכה זעמה‪:‬‬ ‫מה אקב לא קבה אל‬ ‫ומה אזעם לא זעם‬ ‫יהוה‪:‬‬ ‫כי־מראשׁ צרים‬ ‫אראנו ומגבעות‬ ‫אשׁורנו הן־עם לבדד‬ ‫ישׁכן ובגוים לא‬ ‫יתחשׁב‪:‬‬ ‫תמת נפשׁי מות‬ ‫ישׁרים ותהי אחריתי‬ ‫כמהו‪:‬‬ ‫ויאמר בלק אל־בלעם‬ ‫מה עשׂית לי לקב‬ ‫איבי לקחתיך והנה‬ ‫ברכת ברך‪:‬‬ ‫ויען ויאמר הלא את‬ ‫אשׁר ישׂים יהוה בפי‬ ‫אתו אשׁמר לדבר‪:‬‬ ‫ויאמר אליו בלק‬ ‫לכה־נא אתי אל־‬ ‫מקום אחר אשׁר‬

‫‪4‬‬

‫‪6‬‬

‫‪7‬‬

‫‪8‬‬

‫‪9‬‬

‫‪10‬‬

‫‪11‬‬

‫‪12‬‬

‫‪13‬‬

‫שׁבעה פרים‬ ‫ושׁבעה אילים‪:‬‬ ‫ויעשׂ בלק כאשׁר‬ ‫דבר בלעם‬

‫ויאמר אליו את־‬ ‫שׁבעת המזבחת‬ ‫ערכתי ואעל פר‬ ‫ואיל במזבח‪:‬‬

‫יעקב‬ ‫ישׂראל‬

‫מי מנה עפר‬ ‫יעקב ומי ספר‬ ‫משׁפחת ישׂראל‬

‫‪192‬‬

‫‪Literary Criticism‬‬

‫‪14‬‬

‫‪15‬‬

‫‪16‬‬

‫‪17‬‬

‫‪18‬‬

‫‪19‬‬

‫‪20‬‬ ‫‪21‬‬

‫תראנו משׁם אפס‬ ‫קצהו תראה וכלו לא‬ ‫תראה וקבנו־לי‬ ‫משׁם‪:‬‬ ‫ויקחהו אל־ראשׁ‬ ‫הפסגה ויבן מזבח‬ ‫ויעל פר ואיל במזבח‪:‬‬ ‫ויאמר אל־בלק‬ ‫התיצב על־עלתך‬ ‫ואנכי אקרה כה‬ ‫אלהים‪:‬‬ ‫ויקר יהוה אל־בלעם‬ ‫וישׂם דבר בפיו ויאמר‬ ‫שׁוב אל־בלק וכה‬ ‫תדבר‪:‬‬ ‫ויבא אליו והנה נצב‬ ‫על־עלתו ושׂרי מואב‬ ‫אתו ויאמר לו בלק‬ ‫מה־דבר יהוה‪:‬‬ ‫וישׂא משׁלו ויאמר‬ ‫קום בלק ושׁמע‬ ‫האזינה עדי בנו צפר‪:‬‬ ‫לא אישׁ אל ויכזב‬ ‫ובן־אדם ויתנחם‬ ‫ההוא אמר ולא יעשׂה‬ ‫דבר ולא יקימנה‪:‬‬ ‫הנה לברך לקחתי‬ ‫אברך ולא אשׁיבנה‪:‬‬ ‫יהוה אלהיו עמו‬ ‫ותורעת מלך בו‪:‬‬

‫‪22‬‬

‫‪23‬‬

‫‪24‬‬

‫‪25‬‬

‫הן־עם כלביא יקום‬ ‫וכארי יתנשׂא לא‬ ‫ישׁכב עד־יאכל טרף‬ ‫ודם־חללים ישׁתה‪:‬‬ ‫ויאמר בלק אל־בלעם‬ ‫גם־קב לא תקבנו גם־‬ ‫ברך לא תברכנו‪:‬‬

‫שׂדה צפים‬ ‫שׁבעה ‪ +‬ת‬

‫לא־אביט און‬ ‫ביעקב ולא־ראה‬ ‫עמל בישׂראל‬ ‫אל מוציאו‬ ‫ממצרים‬ ‫כתועפת ראם‬ ‫לו‪:‬‬ ‫כי לא־נחשׁ‬ ‫ביעקב ולא־קסם‬ ‫בישׂראל כעת‬ ‫יאמר ליעקב‬ ‫ולישׂראל מה־‬ ‫פעל אל‪:‬‬

‫‪193‬‬ ‫‪26‬‬

‫‪27‬‬

‫‪28‬‬

‫‪3.7 Summary of Literary-Critical Results‬‬ ‫ויען בלעם ויאמר אל־‬ ‫בלק הלא דברתי‬ ‫אליך לאמר כל דבר‬ ‫אשׁר־ידבר יהוה אתו‬ ‫אעשׂה‪:‬‬ ‫ויאמר בלק אל־בלעם‬ ‫לכה־נא אקחך אל־‬ ‫מקום אחר אולי יישׁר‬ ‫בעיני האלהים וקבתו‬ ‫לי משׁם‪:‬‬ ‫ויקח בלק את־בלעם‬ ‫ראשׁ הפעור הנשׁקף‬ ‫על־פני הישׁימן‪:‬‬

‫‪29‬‬

‫‪30‬‬

‫ויעל פר ואיל במזבח‪:‬‬

‫‪1‬‬

‫וירא בלעם ולא־הלך‬ ‫כפעם־בפעם‬

‫‪2‬‬

‫וישׂא בלעם את־עיניו‬ ‫ותהי עליו רוח‬ ‫אלהים‪:‬‬ ‫וישׂא משׁלו ויאמר‬ ‫נאם בלעם בנו בער‬ ‫אשׁר מחזה שׁדי יחזה‬ ‫מה־טבו אהליך‬ ‫משׁכנתיך‪:‬‬

‫‪3‬‬ ‫‪4‬‬ ‫‪5‬‬ ‫‪6‬‬

‫‪7‬‬

‫‪8‬‬

‫יאזל־אישׁ ‪ /‬אדם‬ ‫מזרעו ומשׁל בעמים‬ ‫רבים וירם מאגג‬ ‫מלכו ותנשׂא מלכתו‪:‬‬ ‫אל נחהו ממצרים‬ ‫כתועפת ראם לו‬ ‫יאכל גוים צריו‬ ‫ועצמתיהם יגרם‬

‫ויאמר בלעם אל־‬ ‫בלק בנה־לי בזה‬ ‫שׁבעה מזבחת‬ ‫והכן לי בזה‬ ‫שׁבעה פרים‬ ‫ושׁבעה אילים‪:‬‬ ‫ויעשׂ בלק כאשׁר‬ ‫אמר בלעם‬ ‫‪Numbers 24‬‬ ‫לקראת‬ ‫כי טוב בעיני‬ ‫הנחשׁים‬ ‫יהוה לברך את־‬ ‫ישׂראל וישׁת‬ ‫אל־המדבר פניו‪:‬‬ ‫וירא את־ישׂראל‬ ‫שׁכן לשׁבטיו‬ ‫ונאם הגבר שׁתם‬ ‫העין‪:‬‬ ‫נפל וגלוי עיניו‪:‬‬ ‫יעקב‬ ‫ישׂראל‬ ‫כנחלים נטיו‬ ‫כגנת עלי נהר‬ ‫כאהלים נטה‬ ‫יהוה כארזים‬ ‫עלי־מים‪:‬‬

‫וחצו ימחצו‪:‬‬

‫‪Literary Criticism‬‬ ‫‪9‬‬ ‫‪10‬‬

‫‪11‬‬

‫‪12‬‬

‫‪13‬‬

‫‪14‬‬

‫‪15‬‬

‫‪16‬‬

‫‪17‬‬

‫‪18‬‬

‫כרע שׁכב כארי‬ ‫וכלביא מי יקימנו‬ ‫ויחר־אף בלק אל־‬ ‫בלעם ויספק את־כפיו‬ ‫ויאמר בלק אל־בלעם‬ ‫לקב איבי קראתיך‬ ‫והנה ברכת ברך זה‬ ‫שׁלשׁ פעמים‪:‬‬ ‫ועתה ברח־לך אל־‬ ‫מקומך אמרתי כבד‬ ‫אכבדך והנה מנעך‬ ‫יהוה מכבוד‪:‬‬ ‫ויאמר בלעם אל־בלק‬ ‫הלא גם אל־מלאכיך‬ ‫אשׁר־שׁלחת אלי‬ ‫דברתי לאמר‪:‬‬ ‫אם־יתן־לי בלק מלא‬ ‫ביתו כסף וזהב לא‬ ‫אוכל לעבר את־פי‬ ‫יהוה לעשׂות טובה או‬ ‫רעה מלבי אשׁר־ידבר‬ ‫יהוה אלהי אתו‬ ‫אדבר‪:‬‬ ‫ועתה הנני הולך‬ ‫למקומי‬

‫מברכיך ברוך‬ ‫וארריך ארור‪:‬‬

‫לכה איעצך‬ ‫אשׁר יעשׂה העם‬ ‫הזה לעמך‬ ‫באחרית הימים‪:‬‬ ‫וישׂא משׁלו‬ ‫ויאמר נאם‬ ‫בלעם בנו בער‬ ‫ונאם הגבר שׁתם‬ ‫העין‪:‬‬ ‫שׁמע אמרי־אל‬ ‫וידע דעת עליון‬ ‫מחזה שׁדי יחזה‬ ‫נפל וגלוי עיניו‪:‬‬ ‫אראנו ולא עתה‬ ‫אשׁרנו ולא קרוב‬ ‫יזרח כוכב‬ ‫מיעקב וקם אישׁ‬ ‫‪ /‬אדם מישׂראל‬ ‫ומחץ ראשׁי‬ ‫מואב וקדקד‬ ‫כל־בני־שׁת‪:‬‬ ‫והיה אדום ירשׁה‬ ‫והיה ירשׁה עשׂו‬ ‫וישׂראל עשׂה‬ ‫חיל‪:‬‬

‫‪194‬‬

195

3.8 Redaction History ‫והאביד שׂריד‬ :‫מעיר‬ ‫וירא את־עמלק‬ ‫וישׂא משׁלו‬ ‫ויאמר ראשׁית‬ ‫גוים עמלק‬ ‫ואחריתו עד‬ :‫יאבד‬ ‫וירא את־הקיני‬ ‫וישׂא משׁלו‬ ‫ויאמר איתן‬ ‫מושׁבך ושׂים‬ :‫בסלע קנך‬ ‫ואם־יהיה לבער‬ ‫קן ערמה אשׁור‬ :‫תשׁבך‬ ‫וירא את־עג וישׂא‬ ‫משׁלו ויאמר אוי‬ ‫מי יחיה משׂמו‬ :‫אל‬ ‫ויצא מיד כתים‬ ‫וענה אשׁור וענה־‬ ‫עבר הנהר וגם־‬ :‫המה יחדו יאבד‬

19

‫וער מיעקב‬

20

21

22

23

24

‫ויקם בלעם וילך וישׁב‬ ‫למקמו וגם־בלק הלך‬ :‫לדרכו‬

25

Table 14: The Division of Numbers 22–24 into Sources and Redactional Layers

3.8 Redaction History 3.8 Redaction History

The literary-critical overview of Numbers 22–24 permitted the reconstruction of several diachronic levels behind the final edition of the text. Two sources, one of which appears quite fragmentary, stand in the background of the whole, as do several layers of editorial reworking. Nonetheless, several matters demand further consideration: Was the primary source (S1) behind Numbers 22– 24 part of a longer literary work or did it exist independently? How about the material identified as the second, fragmentary source? How are the different layers to be understood as part of the larger contexts in which they are contained? And when was the oldest version composed and during what periods was it edited? To these questions we now turn briefly, with a particular focus, for obvious reasons, on the place of the Balaam story in the book of Numbers. Since the full development of the redactional history of Numbers (and the other compositions identified in the Hebrew Bible) goes beyond the bounds of a

196

Literary Criticism

study on the figure Balaam, these reflections must be regarded as preliminary; they must, unfortunately, remain a desideratum for continued study at the conclusion of this survey, but I hope to develop these ideas further in the future. First, we must consider the status of S1 as an independent or interdependent narrative. What prerequisites did this version have and do they mandate literary links to other material? What criteria can be legitimately applied to the text in order to determine such potential narratological links? One need only develop a few as guidelines to distinguish related elements within the text. The most obvious is of course consistency, whether narrative, theological, ideological, or historical. The paradigms addressed in the introductory chapter all failed in some capacity or another to sufficiently cover this issue. Were that not the case, all scholars would have reached a consensus and this study would be irrelevant and superfluous. In addition to the various matters of consistency, one can consider cogency: can the reconstructed oldest layer be read? Can each new redactional creation be understood in its new context, both narratologically and historically? If not, they must be rejected as implausible. These original and renewed narratological contexts, of course, will aid in the ascription of a date to each reconstructed version. A third criterion could be something like vocabulary and terminology. Since the development of the ältere Urkundenhypothese, the fascination with the distinct names for God has dominated much discourse about any source or sources; indeed, based on the transitions in Genesis, some still speak of J and E, or recognize that P favored ‫אלהים‬. The text-historical and literary-critical evaluation of Numbers 22–24 demonstrated that these terms present a moot point: it is no longer feasible to reconstruct sources or redactional material based solely, or even initially, on divine nomenclature. The witnesses demonstrate how inconsistently this was maintained through the process of textual transmission. In Numbers 22–24, we noted distinct terms for the group undertaking the exodus: ‫עם‬, ‫ישׂראל‬, and ‫בני ישׂראל‬. Less significantly for the purposes of Numbers 22–24, the term ‫ קהל‬appeared as well, though in Numbers 22–24 it demonstrates unusual, even unique characteristics. Perhaps it will continue to be fruitful to follow this line of logic for determining diachronic layers within Numbers. That presents the working thesis here. One could object at this point based on my refusal to recognize the various names of God as literary-critically relevant, that the terminology for “the Israelites” etc. is just as problematic. This objection would be merited and accurate, if only the terminology for “the people” was as theologically or ideologically loaded as the terms for God or was as text-critically insecure as the divine names. Neither of these is the case. The survey of Numbers 22–24 failed to evince a single case in which one of these terms changed over the course of transmission. Only in one case did one editor copy the terminology of the

3.9 The Story of Balaam in the Redacted Version(s)

197

source he or she edited.163 These data suggest that the terminology referencing the group undertaking the exodus may be more literary-historically reliable. Barring that objection, we can return to the reconstruction. Turning to the first question, regarding the oldest version’s independence or interdependence, requires reflecting on the text’s prerequisites. The exposition in Num 22:3a and 5, as well as Balaam’s report in v. 11, mandate that “the people” here are either in or adjacent to Moabite territory. Verse 22:5, 11; and 24:8 recognize that this people had departed from Egypt; that is, the Balaam story and its oracles presumed the connection to the exodus tradition from their first iteration. Fundamentally, “the people” probably requires further explication, since otherwise no audience who know who this text is about. That is, Num 22:3a does not present an acceptable narrative exposition for the subsequent material. Finally, the note in 22:4b (‫ )בעת ההוא‬demonstrates that the narrator recognized this story as such past occurrence and not as part of the immediate past. Thus, based on the preceding literary-critical analysis, three possibilities remain for the oldest core of Numbers 22–24: 1) it was an independent narrative about the period of the exodus of which the exposition has been lost; 2) it was part of a longer work of which the rest has been lost; 3) it was part of a longer work of which other elements remain. Logically, one must proceed with an examination of material that could potentially have belonged to the same work as the oldest narrative and its oracles. Only once that has been eliminated as a possibility would one need to address the issue of the independent status of the Balaam narrative, whether it stood alone or functioned as a single element in a longer work. That will be the first step here. To that end, we must consider what else might have belonged to such a source.

3.9 The Story of Balaam in the Redacted Version(s) 3.9 The Story of Balaam in the Redacted Version(s)

The story of Balaam and his oracles did not remain untouched. The first editor incorporated the Balaam’s story into a series of stories about confrontations with other peoples. Of particular contrast to the circumstances regarding the Moabites as found in Numbers 22–24 is the now preceding story of the Amorites under Sihon. The editor placed the Balaam story behind this tale, now found in Numbers 21, providing a new context for the Balaam story and more poignantly justifying the Moabite fear of Israel. The addition of this story before the Balaam narrative presumes an attitude toward the Moabites consistent with their defeat or disappearance, without claiming that Israel was responsible for it. That could imply that Moab no longer played a substantial role in Israel’s 163 For narratological reasons, the first editor once used the term ‫ עם‬when adding a component to the text in Num 22:12. Otherwise this redactor referred only to the exodus group as “Israel”.

198

Literary Criticism

geopolitical relations. The focus contrasting the peoples according to their blessing or cursing Israel, apparent both in the new conclusion of Balaam’s third oracle and the reference to Balaam coming from “the land of the sons of his people”. Now he is the generic foreigner, suggesting that Aram-Damascus had ceased to play an important geopolitical role by the time of this redaction. Both of these political relationships commend the late seventh or the first half of the sixth century as a probable context for this editing. This redaction presumably reworked the older composition in portions of text outside of the Balaam pericope as well. One finds the usage of “Israel” as a term for the group undertaking the exodus in a few cases. Particularly in Numbers 20–21, one notices Israel’s engagement with Edom in 20:14–21* (with the interpolation of the “Israelites” in v. 19, as well as some duplications in vv. 18–20), the Canaanites in 21:1–3, and the Amorites in 21:21–31*, as well as Israel’s song in 21:17–18a. This observation again commends the literary-critical decisions of the previous chapter. Other than these stories, however, little material in Numbers conspicuously stems from this redaction. The term “Israel” without ‫ בני‬occurs only sporadically, often still in contexts that unequivocally belong to a later redactional context.164 Perhaps some other elements belonged to this diachronic layer, such as the anecdote about Hobab in 10:29–32, some institution of the elders in Numbers 11*, and the incident with the Baal Peor in Num 25:1–5*. However, since none of these texts particularly impact our understanding of Balaam in his context, we will save the discussion of the extent of this editing for another occasion. In terms of literary connectivity, looking backwards in the Pentateuch, the additions in Num 22:6b and 24:9b commend a connection to the Abram story reported in Gen 12:1–3 and the Jacob story in 27:29.165 So, whereas the oldest layer of Numbers 22–24 evinced no connection to Genesis, the first redaction appears to connect to the Abram and Jacob cycles. When regarding the peoples 164 “Israel” appears by itself in Num 1:3, 16 (with the congregation), 20 (here, it references the man, not the group), 44 (Princes); 3:13; 4:46 (Princes); 7:2 (Princes), 84 (Princes; cf. G); 10:4 (thousands), 29, 36 (thousands); 11:16, 30; 16:9 (God of Israel and congregation of Israel), 25 (elders), 34; 18:14 (cf. G), 21; 19:13; 25:1–5; 26:5 (like 1:20); 31:4–5 (military); 32:4 (congregation; cf. G), 13–14, and 22. Many references are to the military, firmly embedded in the context of the ‫בני־ישׂראל‬, and, therefore, not R1 (Num 1:3, 16, 44; 4.46; 7:2, 84; 10:4, 36; 26:5; and 31:4–5). The usages in Numbers 18–19 demonstrate a (Post-)Priestly connection, as do the references to the “congregation” in 16:9 and 32:4. Numbers 11 refers to the “elders of Israel” in vv. 16 and 30, which could stem from R1, at least theoretically. The remaining uses that could belong to R1: 10:29 (Hobab); 16:25 and 34; 25:1–5; and 32:13–14 and 22. It remains outside the bounds of this study to demonstrate where these verses belong diachronically. 165 For Josef Forsling, Composite Artistry in the Book of Numbers. A Study in Biblical Narrative Conventions, Studia Theologica Holmiensia, vol. 22 (Åbo, Finland: Åbo Akademis förlag, 2013), 134–36, the promises to the patriarchs presented a central aspect to the subsequent conquest. That would fit well with the perspective of some form of Deuteronomistic composition.

3.9 The Story of Balaam in the Redacted Version(s)

199

mentioned in the first redaction, one especially notices general overlap between the peoples mentioned in the closing oracles and the genealogical data gleaned from Genesis. The generational coincidences are too strong here to be accidental. According to Genesis, Moab (Gen 11:27; 19:37) and Edom/Esau (24:25–26) come from the same generation as Jacob/Israel.166 In other words, the contrasts developed in Num 24:17–19a are all great-grandsons of Terah in Genesis’ conception. The first redaction of the Balaam story thus focuses on Jacob’s generation (the twelfth after the flood = the second after Abraham). Import comes to those groups most closely related to Israel in Genesis’ narrative, and nearest in geographical terms. The focus moves in a clockwise fashion geographically from the east (Moab) to the southeast (Edom) of Israel and Judah. The later additions incorporate distinct geographical or geopolitical data, as will become apparent below. Looking towards the material that must have followed Numbers 22–24*, the first redaction implies a connection with the David traditions found in Samuel. Taken together, these data would suggest something like a Deuteronomistic Enneateuch composition as the redactional context associated with this layer. Theologically, one recognizes this in the conception of just desserts in this editorial layer: the one who curses merits his reciprocated accursed state, just as the one who blesses deserves the blessedness he receives. This is consistent with Deuteronomistic theology, which connects fate – even national fate – with action; German scholars call this the “Tun-Ergehen-Zusammenhang” (the connection of deed and consequence). Precisely for this reason, this edition reckons with Moab’s destruction as a result of their desiring Israel’s destruction, whereas Balaam is allowed to leave in peace. Each gets what he deserves. At the same time, the impact of Deuteronomistic vernacular on the redacted version of the Balaam story remains rather underwhelming. This edition’s favored term for cursing, ‫ארר‬, tends to appear more in Deuteronomistic texts.167 The phrase ‫( טוב בעיני יהוה‬or its opposite) appears often in Deuteronomistic texts.168 The first editor of the Balaam story added it in Num 24:1, which could affiliate this addition with such an editor. As suggested in the previous chapter,

166 The mention of the “sons of Seth” in Num 24:17 probably does not recognize Seth as a son of Adam. N.b., in this verse in Numbers the sons of Seth are paralleled by Moab, suggesting an understanding of Seth distinct from that of Genesis. 167 Of 63 attestations, eighteen appear in Deuteronomy, ten appear in Joshua–2 Kings, and six appear in Jeremiah. That is, more than half appear in texts often affiliated with Deuteronomic / Deuteronomistic theology and phraseology. 168 ‫ בעיני יהוה‬appears nine times in Deuteronomy, eight times in Judges, five times in Samuel, and 42 times in Kings. It also appears once in Jeremiah. Taken together, these attestations easily present a majority of the 93 usages.

200

Literary Criticism

the term for tribe in Num 24:2 is more consistent with Deuteronomistic usage.169 The identification of the Moab and Edom as possession (‫ )ירשׁה‬is consistent with Deuteronomistic terminology.170 So, while there may only be a paucity of supposed Deuteronomistic linguistic characteristics in these chapters, enough terminological evidence is there to postulate a connection to that school of thought. Add to that the theological considerations, and it appears likely that the first editing of the Balaam narrative took place in that milieu. At the same time, one notices that the Deuteronomistic character of Numbers 22– 24 remains limited, affirming the generally uniform characterization of the older reconstructed version. That is, the Deuteronomistic editing of this chapter did not recast the whole narrative with a distinct mold. It nonetheless did leave a fine imprint on the text as passed through the generations. The dating of this layer could have been any time during the late monarchic period in Judah, through the exile, or perhaps even later. The second editor took the text in another direction. By the inclusion of the material about the Moabite and Midianite elders seeking to undermine the Israelites due to Moab’s disgust, this editor prepared the audience for the stories in Numbers 25* and 31*. This new addition creates a new subplot and prepares the reader for the Israelites’ military engagement with the Midianites, justifying this interaction.171 By inserting this material into Balaam’s story this editor began through implication and innuendo the process of negatively recasting the Balaam figure. Now his affiliation with the Moabites turned into a liabilty, rather than his contrast with the Moabites serving to his beatification as had been the case in the earlier version. The connection of the passages in Numbers 25* and 31* to priestly literature – regardless of what stage of priestly literature – commends regarding this expansion of Numbers 22–24 as part of that school of thought. Through the addition of the Midianites and the connections to the opening of the book of Exodus, Numbers 22–24 officially became part of a priestly composition subsequent to its incorporation into a Deuteronomistic composition. Most of the book of Numbers presumably stems from the second and later redactions, as apparent through the extensive use of ‫ בני־ישׂראל‬as the name for the group undertaking the exodus. Even later editors appear to have included information about the ‫ עדה‬and the ‫קהל‬, though these remain beyond the bounds 169 Of the 191 usages in the HB, eighteen are in Deuteronomy, 33 are in Joshua, sixteen are in Judges, eighteen are in Samuel, and thirteen are in Kings, only two appear in Jeremiah, but there are an additional eight in four Deuteronomistically redacted prophets of the Book of the Twelve. As with ‫ארר‬, more than half of the appearances of ‫ שׁבט‬are in Deuteronomistically edited or affiliated literature. 170 Of the sixteen times this term appears in the HB, only two are not in Deuteronomistic literature. Deuteronomistic literature also broadly attests the ‫ ירשׁ‬. 171 Cf. Forsling, Composite Artistry, 129–30, who regarded the connections between Numbers 22–24; 25; and 31 as a significant subplot of the Numbers narrative.

3.9 The Story of Balaam in the Redacted Version(s)

201

of the discussion here, as we can identify no impact of these additions on the Balaam story in Numbers 22–24. However, they do play a role in the expansion of information about Balaam in Numbers 31. The next chapter addresses this. Other than the incorporation of a fragmented source, of which the continuation presumably stands behind Numbers 25* and 31*, R2 barely expanded the text of Numbers 22–24. The paucity of priestly language demonstrates the limited interaction this theological school had on shaping the tradition. Like the integration into the Deuteronomistic composition, this expansion left only a slight mark, though it had lasting repercussions for Balaam’s characterization. The most likely dating for this new edition would be exilic or later, depending on how one dates the Deuteronomistic version. The narrative arc of this version could have spanned the creation account in Genesis 1 or the beginning of the priestly exodus account to the death of Moses in Deuteronomy 34 or even further. Nothing in particular within this diachronic layer in Numbers 22–24 permits the identification of this priestly composition’s beginning or end. The later additions to Numbers 22–24 do not necessarily demonstrate any particular affinity with any larger redactional undertakings in the book or elsewhere. Should the story of the donkey be understood as a parallel to Genesis 2–3,172 one could consider these pieces of one diachronic layer. On the other hand, the common material between these tales, beyond the speaking animal, is of a more general nature. Other minor additions for consistency within the developing biblical literature are ‫ וחצו ימחצו‬in Num 24:8 (cf. Ps 68:22) and ‫ וחאביד שׂריד מעיר‬in Num 24:19b (cf. 21:35). Each of these phrases evinces a connection to other literature and could be identified as part of a larger tendency towards consistency. However, it would be impossible to ascribe these additions to one or several redactional undertakings. The rest of the additions in Numbers 22–24 cannot immediately and unequivocally be ascribed to any larger editorial activity within the book, the confines of the Pentateuch, Hexateuch, Enneateuch, or elsewhere. Scribes presumably added these pieces over the course of time, most likely into the Hellenistic period (cf. Num 24:24). Some of these final elements may be among the youngest in the Torah. Returning to our considerations of the peoples mentioned in the expanded final oracles, we recognized that the Deuteronomistic composition focused on Israel’s and Judah’s neighbors Edom and Moab. They all stem from the same generation in Genesis’s conception (see above). The Amalekites present something of a crux in this conception. Amalek (Gen 36:12) comes from the generation of Esau’s and Jacob’s grandsons, i.e., the same generation as Perez and Zerah (Gen 38:27–30), as well as Ephraim and Manasseh (Gen 41:50–52). The Fortschreibung with Amalek thus concentrates on the fourteenth generation after the flood, the fourth after Abraham, according to Genesis. Geographically 172

George W. Savran, “Beastly Speech: Intertextuality, Balaam’s Ass and the Garden of Eden,” JSOT 64 (1994): 33–55 suggested reading these texts together.

202

Literary Criticism

they could also present a continuation of the east–southeast progression in Num 24:17–19a, in that the Amalekites can be understood as residing to the south of Judah.173 This suggests a proximity between the Moab–Edom oracles and the Amalek Fortschreibung, but also some common redactional impetus between Genesis and Numbers at this stage. Exodus 17:8–16 might have inspired this oracular addition and could attest a tendency like that of R1 in Numbers 22– 24. N.b. it describes the exodus generation as “Israel”, consistent with the first redaction identified here. In Num 24:20 the Amalekites could thus reflect this subsequent generation and present a further step beyond that attested in the Deuteronomistic version of Numbers 22–24, a Fortschreibung with recourse to Genesis 36 and Exod 17:8–16.174 The later Fortschreibungen in the final oracles in Numbers 24 have a much broader interest, covering groups that go back to humanity’s second generation in Genesis’ conception, the Kenites presumably being the descendents of Cain (Genesis 4) or a group affiliated with the Amalekites.175 Following this, the identification of the other peoples are post-diluvian, but the divisions still go back to the earliest time following the flood. Kittim (Cyprus) was the greatgrandson of Noah via Japhet’s line according to Gen 10:1–4. From the line of Shem, Assyria (Ashur) was the grandson of Noah according to Gen 10:1 + 21– 22. Should the Kenites be identified with a group south of Judah, the geographical orientation of the peoples would differ from that of the preceding oracles, moving from the south to the northeast and then the northwest.

173

No certain information about the location of the Amalekites exists and some biblical data contradict each other. While the Amalekites sometimes appear in Ephraim (Judg 5:14), they can generally be localized at the northeastern end of the Sinai Peninsula in the Bible’s conception. Cf. already Nöldeke, Ueber die Amalekiter, 11–19. 174 The datum in Gen 36:12 could be an editorial interpolation, just as seems to be the case in Num 24:20. N.b. Amalek’s status as a bastard and his falling outside of the twelve descendants of Esau; cf. John Skinner, A Critical and Exegetical Commentary on Genesis. Second Edition, reprint, 1930, ICC (Edinburgh: T. & T. Clark, 1980), 431. Perhaps the anti-Amalekite tradition of Exodus 17 inspired both of these additions. These observations remain, nonetheless, beyond the focus of this study. 175 The latter seems less likely: The Kenites might, of course, reference the group of people affiliated with Moses’ father-in-law in several texts. While this would place them in a similar geographical context to the Amalekites, somewhere in southern Palestine (Cf. Gen 15:19; Judg 1:16; 1 Sam 15:6; and 1 Chr 2:55), the contrast of this group with the others presents a problem, in that the Kenites are particularly insignificant when compared to Assyria and Greece, the other peoples in this oracle; contra Nöldeke, Ueber die Amalekiter, 19–21, who regarded the Kenites here as a tribe within the Amalekites. Perhaps ‫ קין‬references the Midianites? The identification of Moses’ father-in-law as both a Kenite and a Midianite could indicate that biblical authors regarded these groups as identical, whether at different times or in different places. Should this reference to ‫ הקיני‬actually be to the Midianites, this would encompass another group under the rubric of Abraham in Genesis’ genealogical concept; cf. Gen 25:1–2.

3.9 The Story of Balaam in the Redacted Version(s)

203

The account in Genesis affords about ten literary generations between the peoples mentioned in the later Fortschreibungen and those encountered in the first redaction and earlier Fortschreibung of Numbers 22–24. This contrast suggests the possibility of identifying increasingly larger narrative contexts both behind and common to the expanding Balaam text. That is, later versions of the Balaam pericope may reflect knowledge of the primeval history, while the earlier versions only seem to know about the patriarchs and the exodus. Interestingly, the number of generations implied behind the peoples mentioned in the Balaam story suggest a compositional intention in both directions, preceding and succeeding the story. The conception presents an idea stemming from the combination of P and non-P. According to Genesis 5 (P), Noah is the ninth generation conceived after the creation of the world.176 Jacob, his brother (Esau), and his cousins (Moab and Ammon) present the twelfth generation after Noah.177 Considering the genealogical information in Genesis, Ruth, Samuel, and Kings, Solomon is the twelfth generation after Jacob.178 From Solomon, the twelfth generation is Hezekiah according to Kings.179 And finally, according to Kings, Ezra 3:2 (cf. Hag 1:1, etc.), and 1 Chr 3:16–7, Zerubbabel was the seventh generation from, and the ninth ruler after Hezekiah.180 Thus, these data divide the history of the world up to the conclusion of the exile schematically: Creation – nine generations – twelve generations – twelve generations – twelve generations – nine rulers (from seven generations) – restitution. This systematic, chiastic portrayal can hardly be accidental. A: From Creation through the Flood (9) B: From the Flood through Jacob/Israel (12) C: From Jacob/Israel through the United Monarchy (12)181 B’: From the United Monarchy to Judah Alone (12) A’: From Judah Alone through the Return from the Exile (9) At the center of this chiasm, one clearly sees the culmination in the period of the so-called United Monarchy. The conclusion in the generation returning from the exile also places special emphasis on this event. 176

Seth – Enosh – Kenan – Mehalalel – Jared – Enoch – Methuselah – Lamech – Noah. Shem – Arpachshad – Shelah – Eber – Peleg – Reu – Serug – Nahor – Terah – Abram – Isaac – Jacob. 178 Judah – Perez – Hezron – Ram – Amminadab – Nahshon – Salmon – Boaz – Obed – Jesse – David – Solomon. 179 Rehoboam – Abijam – Asa – Jehoshaphat – Jehoram – Ahaziah – Joash – Amaziah – Azariah – Jotham – Ahaz – Hezekiah. 180 Manasseh – Amon – Josiah – Jehoahaz – Jehoiakim – Jehoiachin – Zedekiah – Shealtiel – Zerubbabel. 181 N.b. this historical construct does not afford time to the so-called period of the judges, suggesting that it may be a later insertion into its context. Cf. also 1 Kgs 6:1: the 480 years between the exodus and the construction of the temple = 12 generations à 40 years. 177

204

Literary Criticism

Yet, the original Balaam story did not reflect these understandings; they were the product of later additions, particularly the concluding oracles. The first redaction reflected elements from the Jacob–Solomon era. This observation again commends recognizing the first redaction of Numbers 22–24 as part of a Deuteronomistic composition. The original fourth oracle, composed and appended by the primary editor of the older Balaam composition, covered precisely these peoples, Moabites and Edomites, and focused on the period between the patriarchs and the united monarchy. The later redactions added the rest, completing this scheme and systematizing it with data in the rest of the Bible. To this end, the final oracles demonstrate the increasingly consistent incorporation of genealogical material in Genesis and elsewhere as part of a priestly composition and then as part of the Tanakh. The objective of determining compositional dates for the original Balaam narrative and oracles, their greater context, and the various redactional undertakings has thus far remained unfulfilled. To this matter we can now turn our attention, beginning with the latest layers of the text and working our way backward. The edition of the text present in M certainly contains some features from the Hellenistic or even Roman periods, as addressed in the chapter on text history. These latest textual transformations shine through most apparently in Num 24:23–24. The text of 4QNumb, which shares many similarities with Smr, G, and M, demonstrates that a version similar to that found in M (and the other relevant witnesses) was certainly known by the first century BCE. Unfortunately, the text recovered from Qumran does not attest Num 24:23–24, meaning that we cannot recognize whether its version followed M, G, Smr, or something entirely distinct. At the very least we can ascertain that all redactional historical processes that shaped Numbers 22–24 must have been finalized by that time, leaving only the possibility of minimal changes and/or errors in the transmission as responsible for distinctions in the versions.182 The redactional elements identified with the terminology of ‫ קהל‬or ‫עדה‬ seem to have been the latest larger additions and glosses to the book of Numbers. But is it possible to date these? They clearly presuppose texts that belonged to P in whatever form.183 That means that we should regard them as

182

For more on this, consult the conclusion of the chapter on textual history. For problems and issues regarding the identification, or even the description of P, cf. Baruch J. Schwartz, “Introduction: The Strata of the Priestly Writings and the Revised Relative Dating of P and H,” in The Strata of the Priestly Writings: Contemporary Debate and Future Directions, Sarah Shectman and Joel S. Baden (Zürich: Theologischer Verlag Zürich, 2009), 1– 12; Joel S. Baden, “Identifying the Original Stratum of P: Theoretical and Practical Considerations,” in The Strata of the Priestly Writings: Contemporary Debate and Future Directions, Sarah Shectman and Joel S. Baden (Zürich: Theologischer Verlag Zürich, 2009), 13–29; and 183

3.9 The Story of Balaam in the Redacted Version(s)

205

post-P additions into the book of Numbers. In terms of dating, they could stem from any period after the exile. The final additions in Num 24:23–24, for example, may even come from the Hellenistic period. Certainly nothing suggests that particularly the material in the final oracles stems from the premonarchic period.184 The second redaction looks like P, most likely Ps. That would mean that the source fragment it adapted would be Pg. However, should the source have been part of Ps, that would imply that the redaction must also be regarded as post-P in some form.185 For our purposes here, it only matters that this R2 expanded the Balaam story in Numbers 22–24 as part of some priestly composition. If the source fragment was exilic, then the redaction was postexilic. They may both have been postexilic. Nothing about either piece suggests that it might be preexilic. The first redaction adds elements looking forward to developments in the Deuteronomistic history. Therefore, they evince a background with some Deuteronomistic composition, though clearly not precisely in the form Blum identified. These editors expanded Balaam’s oracles to anticipate David as presented in the Deuteronomistic version of the book of Samuel and his domination of Moab and Edom. This layer cannot have existed by itself (i.e., it was not a source), but relied upon and expanded on some earlier source about Balaam. This composition seems likely to have been late preexilic or, more likely, exilic. It probably cannot be dated earlier than Josiah, and more likely belongs to the height of the Neo-Babylonian period of the exile. Erhard Blum, “Issues and Problems in the Contemporary Debate Regarding the Priestly Writings,” in The Strata of the Priestly Writings: Contemporary Debate and Future Directions, Sarah Shectman and Joel S. Baden (Zürich: Theologischer Verlag Zürich, 2009), 31–44. 184 Contra Seebass, “Vertrauenswürdige Nachrichten”, who argued that this anticipates David. While that may be true, these oracles seem to anticipate the literary figure David and not necessarily any such historical figure. That pushes their dating down to the era when these texts about him were formed and edited, probably later in the monarchic period and thereafter. 185 All of the ruminations on the (barely) priestly characteristics of these elements of Numbers 22–24 only impact the discussion here peripherally. Ultimately, it all comes down to where one identifies the end of any Priesterschrift as a source, in Exodus 40 (e.g., Thomas Pola, Die Ursprüngliche Priesterschrift. Beobachtungen Zur Literarkritik und Traditionsgeschichte von Pg, WMANT, vol. 70 [Neukirchen-Vluyn: Neukirchener Verlag, 1995]), in Leviticus 16 (e.g., Christophe Nihan, From Priestly Torah to Pentateuch: A Study in the Composition of the Book of Leviticus, Forschungen Zum Alten Testament. 2. Reihe, vol. 25 [Tübingen: Mohr Siebeck, 2007]), somewhere in Numbers (e.g., Jean Louis Ska, “Le récit sacerdotal: une ‘histoire sans fin’?” in The Books of Leviticus and Numbers, ed. Thomas Römer, BETL, vol. 215 [Leuven: Peeters, 2008], 631–53 and Jean Louis Ska, “Old and New in the Book of Numbers,” Bib 95, no. 1 [2014]: 102–16), or even later in the narrative (e.g., Christian Frevel, Mit Blick auf das Land die Schöpfung erinnern. Zum Ende der Priestergrundschrift, Herders Biblische Studien, vol. 23 [Freiburg im Breisgau: Herder, 2000]). These matters are beyond the scope of this study and will hopefully dealt with elsewhere.

206

Literary Criticism

The oldest material recoverable in Numbers 22–24 already presumes some exodus story, but also anticipates some form of the conquest. From its inception, then, the tale in Numbers 22–24 stood in the crux of the composition that became Numbers, the transition from Egypt to Israel.186 This oldest Balaam story could have theoretically been part of a larger narrative, but was a quality of independence about it. Its interaction with other texts could commend it being part of a larger literary collection of material about the exodus as a time before the monarchy. To this end, the earliest stratum of Balaam’s tale in Numbers noted that Israel had come from Egypt. Yet, the oldest version also awaited Saul’s kingship and the establishment of a state, at least in a literary form. The conflict with Moab presumed in the text commends a monarchic dating for this layer. In particular, the conflict with Moab and probably the identification of Balaam as an Aramean commend a date during the Omride or Jehuide dynasties. Such dating would also fit with the dating of the other later layers, which range from late preexilic (at the earliest) to Persian or even Hellenistic or Hasmonean, should one include the text history.

186 Cf. the description of Numbers as a composition in Rolf P. Knierim, “The Book of Numbers,” in Die Hebräische Bibel und ihre zweifache Nachgeschichte: Festschrift für Rolf Rendtorff zum 65. Geburtstag, ed. Erhard Blum, Christian Macholz, and Ekkehard Stegemann (Neukirchen-Vluyn: Neukirchener Verlag, 1990), 160: “The Book of Numbers has indeed two parts. the conceptual aspects of the preparartory [sic] organization and execution of the campaign are the two constitutive elements of its structure. These two aspects complement each other as the two basic elements of a report or narrative about a campaign. Numbers is a conceptual unity in its own right – it is the saga of a campaign. And since this campaign belongs to the type of migration from one territory to another rather than to the type of campaign which after completion returns to its point of departure, it should specifically be characterized as the saga of a migratory campaign.”

Chapter 4

The Bible’s Balaam beyond Numbers 22–24 Outside of the long Balaam narrative in Numbers 22–24, one encounters a variety of biblical traditions about this character and his sponsor, Balak. This chapter will briefly consider these passages in terms of content and any potential relationship that they might have to one another. The various references to Balaam and Balak in the Hebrew Bible will be expounded in the order in which they appear in the Hebrew canon before a brief consideration of Balaam at Qumran and in the New Testament. Numbers 22–24 presents Balaam’s first appearance, but one other text before that may have some relationship to the figure Balaam known from other texts, namely Gen 36:32 // 1 Chr 1:43–44, to which we will turn first. Balaam’s first appearance in the Bible subsequent to Numbers 22–24 occurs in Numbers 31, the text to which we will subsequently return. The other texts to be discussed here are Deut 23:4–6; Josh 13:21–22; 24:9–10; Judg 11:25; Mic 6:5; Neh 13:1–3; and the New Testament texts 2 Pet 2:15–16; Jude 11; and Rev 2:14.

4.1 Genesis 36:32–33 // 1 Chronicles 1:43–44 4.1 Genesis 36:32–33 // 1 Chronicles 1:43–44

While not quite identical to the name Balaam son of Beor (‫)בלעם בן־בעור‬, a man called Bela ben Beor (‫ )בלע בן־בעור‬appears in Gen 36:32–33. The difference consists of a single Hebrew letter. This text in Genesis identifies Bela as a king of Edom from the pre-monarchic period in Israel, noting also that “his city”, perhaps his capital, was Dinhabah. Context commends regarding Bela son of Beor as the first king of Edom. Finally, the text mentions his death and his successor, a man from another family and another city (Jobab son of Zera from Bozrah). A parallel version of this information appears in 1 Chr 1:43–44.1 Otherwise, this figure Bela and even his city are absent from other biblical

1 Kurt Galling, Die Bücher der Chronik, Esra, Nehemia übersetzt und erklärt, ATD, vol. 12 (Berlin: Evangelische Verlagsanstalt, 1958), 20 regarded the list of Edom’s kings as a later insertion into its context, copied from Genesis 36. He provided no basis for this.

208

The Bible’s Balaam beyond Numbers 22–24

texts. 2 Is there some relationship between the figure Balaam and the figure Bela, both named son of Beor? The possibility, of course, cannot be precluded from the outset. The similarity of the names and the common patronymic would at least permit the attribution of both to some common background tradition, a tradition that developed in diverging directions: one a king, the other a diviner; one an Edomite, the other an Aramean.3 These differences, however, suggest that distinct traditions stand behind these literary figures. Of the manifold traditions about the seer Balaam (see below and also in the chapter about Deir ʿAlla), none of them suggests, nor even goes so far as to imply anything like a royal background for him. Nothing in them would commend his geographical background being as far south as Edom either.4 If a common background for these two figures of tradition could be identified, the most likely circumstances would be that the famous tradition of the seer Balaam served as the basis for the literary creation of this Edomite king.5 The representation of this figure in the extrabiblical Tell Deir ʿAlla inscription affirms this possibility by demonstrating that a ninth or eighth-century tradition of a seer named Balaam existed, while such a tradition about a king remains otherwise unknown. Any other alternative than the seer as the background or the independent development of each seems unlikely. Therefore, it seems best to regard Bela ben Beor in Gen 36:32–33 as manufactured from the traditional figure Balaam ben Beor.6 Should Balak ben Beor have been the older reading in Genesis (as in G), this creation must have been

2

First Chronicles does mention two other Belas in 1 Chr 5:8 and 7:6–7. However, the only known “sons of Beor” in the Bible are the Bela in Gen 36:32–33 and Balaam in Numbers 22– 24 et al. 3 Cf., e.g., already Nöldeke, Untersuchungen Zur Kritik Des Alten Testaments, 87–88, who recognized a common legendary tradition for both figures. Others have followed this, such as Holzinger, Numeri, 105–6; Greßmann, Mose und seine Zeit, 318–20; Skinner, Genesis, 435; and Gray, Numbers, 324. Rudolph, Der “Elohist”, 100 remained cautious, noting that the author of Numbers hardly could have been familiar with the tradition in Genesis. 4 Contra Mowinckel, “Ursprung”, who regarded him as an Edomite. 5 Perhaps G serves as an oblique confirmation of this suggestion. Rather than report the name ‫ בלע‬in Genesis, G records the first king of Edom’s name as Βαλακ, i.e., the Moabite king from Numbers 22–24 as opposed to the Aramean diviner from that same story. That suggests that even early translators or later recensors recognized a connection between these figures in Genesis 36 and Numbers 22–24 and their primary narrative, regardless of which version (G or M) of the Genesis text was older. N.b. Chronicles transmits precisely the same variant as Genesis in this case. This position is, of course, contra, e.g., Gerhard von Rad, Das erste Buch Mose. Genesis, übersetzt und erklärt, ATD (Göttingen: Vandenhoeck & Ruprecht, 1967), 302– 3, who recognized this list in Genesis as both old and trustworthy regarding its data about Edom. 6 This position is contrary to Ephraim Avigdor Speiser, Genesis, AB, vol. 1 (Garden City, NY: Doubleday, 1982), 280, who regarded the similarity as “coincidental”.

4.2 Numbers 31

209

quite late in transmission of the biblical texts, i.e., after the translation of Genesis into Greek.7

4.2 Numbers 31 4.2 Numbers 31

Two brief notices about Balaam appear in Numbers 31: one in verse 8 and one in verse 16. Contextually, Balaam appears here in a description about a military conflict between the Israelites (‫ )בני ישׂראל‬and the Midianites. The story’s obvious goal is the narrative destruction of the Midianites at the hands of the second generation after the exodus from Egypt.8 In this context, Balaam’s first appearance in this chapter describes his demise at the Israelites’ hands. Verse 8 includes him in a list of five Midianite kings killed by the Israelites: ‫ואת־‬ ‫מלכי מדין הרגו על חלליהם את־אוי ואת־רקם ואת־צור ואת־חור ואת־רבע המשׁת מלכי‬ ‫מדין ואת בלעם בן־בעור הרגו בחרב‬. Several of this notice’s features stand out. First, the whole of this verse fits at best loosely in its context.9 The preceding verse concludes with a notice that the Israelites killed all males in Midian: ‫( ויהרגו כל־זכר‬31:7b). After this general notice, in which the kings – as males – presumably should have been included, the two notices in verse 8 follow. Whereas verse 9 continues with the syntax (and arguably the content) of verse 7 (‫)וישׁבו בני ישׂראל את־נשׁי מדין‬, verse 8 reports its data via inverted sentences (with the object preceding the verb) and changes from the narrative form to the simple perfect. This is true for both elements of verse 8, both for the notice about and the list of the kings, as well as for the notice that they killed Balaam with the sword. Second, the text seems to mention Balaam almost as an afterthought.10 That is, even within this loosely attached verse, Balaam’s appearance serves as kind of a non sequitur. Why is he here? At any rate, based on the last mention of Balaam (Num 24:25), his death here remains at least contextually unanticipated. 7

One should also note that in the Septuagint tradition, Balak became the father of Job. According to the postscript of Job in the Greek version, Job’s name had previously been Jobab and he was the son of Balak ben Beor, an Edomite; cf. JobG 42:17b–d; Martina Kepper and Markus Witte, “Job / Das Buch Ijob / Hiob,” in Septuaginta Deutsch: Erläuterungen und Kommentare Band II. Psalmen bis Daniel, ed. Martin Karrer and Wolfgang Kraus (Stuttgart: Deutsche Bibelgesellschaft, 2011), 2048–50, and the literature cited there. 8 The narrative function should be stressed here. N.b. already Gray’s appropriate comments about this chapter: “This is not history, but Midrash [italics in original]…” and “The unhistorical character of the narrative is so obvious that it need not be proved at length…” in Gray, Numbers, 418. 9 Cf., e.g., Noth, Numeri, 199, who regarded 31:8 as a late addition in a late context. 10 Cf., again, Noth, Numeri, 199, who recognized that Balaam does not fit the context here and was probably added to v. 8 under the influence of v. 16.

210

The Bible’s Balaam beyond Numbers 22–24

To this end – third – it remains surprising that Balaam should be included in a list of apparent royalty; nothing in the data about him in the HB suggests he should be subsumed into such a class.11 Yet, the incorporation of his patronymic demands that this Balaam surely be understood as the Balaam known from Numbers 22–24 and thus not some other royal figure in Midian. Fourth, the identification of Zur as one of the kings of Midian surprises the reader. His only appearance thus far – in Num 25:15 – presented him parenthetically thus: ‫“( ראשׁ אמות בית אב במדין הוא‬A head of the tribes of the house of his father in Midian was he”; cf., however, 25:18: ‫)נשׂיא‬. For this reason, one cannot consider Numbers 31 without making appropriate consideration of Numbers 25 and their interrelationship. Nothing in Numbers 25 suggests that he should be considered a king. This minor, yet poignant distinction could be relevant for redaction-historical reconstructions, particularly when one notes the contrast between 25:15 and 18. To the fourth point, Numbers 25:15 and 18 are not free of text-historical problems. Most importantly, it appears that G may have read a text distinct from M in 25:15: ἄρχοντος ἔθνους Ομμωθ presents a longer reading than M. Either one term is missing in M or G translated one term twice.12 To the former, G provided ἔθνος exclusively as a translation of either ‫( עם‬Num 13:28, 31; 21:18; 23:9; and 24:7 [reconstructed; cf. textual criticism to this verse in Chapter Two]) or ‫( גוי‬14:12, 15;. 24:8, and 20). That could imply that one of these terms stood in the G Vorlage. Of the options available, ‫ עמים‬probably attests the most likely candidate. Both other usages of ‫א ָמה‬, ֻ Gen 25:16 and Ps 117:1 could aid our reconstruction as well. It appears from Ps 117:1 that either ‫אמים‬ there presents an error for ‫( עמים‬cf. G) or that ‫ ֻא ָמה‬could also take a masculine plural. That would make both a potential confusion (‫ אמים‬for ‫ )עמים‬and subsequent replacement (‫ אמות‬for ‫ )אמים‬in Num 25:15 more likely. The person who translated Genesis into Greek apparently thought that ἔθνος served as an adequate translation for ‫אמה‬, permitting the possibility that the G translator of Numbers simply translated the term and then transliterated it for clarity.13 At the same time, it remains possible that the G Vorlage had one more element in the text and that it read ‫ נשׂיא‬instead of ‫ראשׁ‬, since the G of Numbers relied on ἄρχοντος as a standard equivalent for both of these terms. 11

The only possible indication that Balaam might have been royal can be found in Gen 36:32, which identifies ‫ בלע בן בעור‬as the king of Edom. However, since only the patronymic matches, but the personal name and geographical or ethnic ascription to Edom differ from everything else known about Balaam, the datum should be regarded as irrelevant to the case at hand; cf. the preceding discussion of this verse. 12 Wevers, Notes on the Greek Text of Numbers, 428 regarded G as both a translation and a transliteration. 13 Cf., however, Dorival, Nombres, 103: “…en réalité les choses sont plus compliquées, puisque ’ummōt est d’abord traduit par le mot éthnos, ‘nation’, au singulier, puis ensuite transcrit au pluriel. Quoi qu’il en soit, il y a dans la LXX un nom de peuplade absent du TM.”

4.2 Numbers 31

211

Finally, Num 31:8 must be appreciated from a text-historical standpoint. The Greek tradition of this text contains several variants from M that could be literarily relevant.14 The surplus of a few words at the end of the verse presents the most substantial distinction. The verse as a whole reads αἳ τοὺς βασιλεῖς Μαδιαν ἀπέκτειναν ἅμα τοῖς τραυματίαις αὐτῶν καὶ τὸν Ευιν καὶ τὸν Σουρ καὶ τὸν Ροκομ καὶ τὸν Ουρ καὶ τὸν Ροβοκ πέντε βασιλεῖς Μαδιαν καὶ τὸν Βαλααμ υἱὸν Βεωρ ἀπέκτειναν ἐν ῥομφαίᾳ σὺν τοῖς τραυματίαις αὐτῶν. In the Greek tradition, Num 31:8 ends with the phrase σὺν τοῖς τραυματίαις αὐτῶν, which presumably reflected the Hebrew ‫ את־חלליהם‬or something similar. Several factors commend this reconstructed reading behind the Greek. First, ML reads ‫על־‬ ‫ חלליהם‬in Num 31:8, for which G attests ἇμα τοῖς τραυματίαις αὐτῶν. In G, ἇμα would present an unusual translation of ‫על‬: generally ἇμα reflects ‫יחד‬, ‫יחדו‬, or the prepositions ‫ את‬or ‫עם‬. This variance suggests that ‫ על‬may not have been the reading of the G-Vorlage. The Masoretes responsible for ML identified this usage of the preposition ‫ על‬as singular with a note in the Masorah parva that the phrase ‫ על־חלליהם‬is hapax legomenon. This observation supports the possibility of another preposition or word originally standing here. Finally, the parallel in Josh 13:22 records ‫ אל־חלליהם‬at the conclusion of the notice about Balaam in M, but does not reflect the first occurrence ‫ על־חלליהם‬as found in Num 31:8.15 The distinction in the preposition in Joshua may be indicative of an older version of Numbers, which changed through the course of textual transmission, and many of the previous observations can be made in this case as well (e.g., the Masoretes also identify the phrase ‫ אל־חלליהם‬as hapax legomenon). Perhaps there is an older corrupted tradition behind all of the readings. The most likely candidate would be the preposition ‫את‬. It presents a good Vorlage for G’s ἇμα and σὺν, as well as the background for Joshua’s ‫ אל‬in 13:22. Presumably it also stood in Num 31:8. Later scribes must have changed it in Num 31:8 M to reflect the preposition ‫ על‬and removed the second occurrence of the phrase ‫ את־חלליהם‬over the course of transmission. Thus, Num 31:8 in Greek contains a virtually duplicated reading not attested in M, but which may reflect an older reading that was lost in the proto-Masoretic tradition.16 The reading of

14

BHS notes none of these, however. That the phrase ‫ אל־חלליהם‬at the end of Josh 13:21–22 is not reflected in G presents no problem, contrary to the suggestion of Dorival, Nombres, 520, who regarded the duplication in Num 31:8 as a harmonization with the beginning of the verse. The text of G Joshua was probably damaged at this point; cf. the discussion to Josh 13:21–22 below. Contrary to Gray, Numbers, 420–21 I do not regard the version in Joshua and the version in Numbers as each developing distinctly from “a tradition anterior to either”. Probably one copied the other. 16 Contra Wevers, Notes on the Greek Text of Numbers, 508; Dorival, Nombres, 520; and Rösel and Schlund, “Arithmoi,” 507, who viewed G as a harmonization even within the verse Num 31:8 itself. However, none of them effectively demonstrate how such an inharmonious 15

212

The Bible’s Balaam beyond Numbers 22–24

the verse could be reconstructed thus: ‫ואת־מלכי מדין הרגו את־חלליהם את־אוי‬ ‫ואת־רקם ואת־צור ואת־חור ואת־רבע חמשׁת מלכי מדין ואת בלעם בן־בעור הרגו בחרב‬ ‫את־חלליהם‬. The variant translations of ‫ את‬before ‫ חלליהם‬in G resulted from context. This duplication suggests that the phrase between the duplicated words should be identified as a later insertion. The insertion remains marked by the Wiederaufnahme that frames it. The presence of the conjunction καί in the Greek before the list of kings could also support this; it implies that the list began ‫ ואת־אוי‬in the G-Vorlage, and the translation of Joshua reflects this as well. Thus, ‫)ו(את־אוי ואת־רקם ואת־צור ואת־חור ואת־רבע חמשׁת מלכי מדין ואת‬ ‫ בלעם בן־בעור הרגו בחרב‬should be removed as a later insertion into this narrative context. Perhaps it was taken from Josh 13:21–22, but the presence of ‫אל־‬ ‫ חלליהם‬at the end of Josh 13:22 M makes this improbable unless the G reading there is actually older; see below. At any rate, it seems quite likely that the information about Balaam here presents a later insertion.17 An older version of Numbers 31 included neither the names of these monarchs nor any mention of Balaam. Numbers 31:16 again provides the reader with new data about Balaam, and – again – like Num 31:8 it evinces significant text-critical problems. This case remains more opaque and difficult to resolve, at least text-historically. ‫הן הנה היו לבני ישׂראל בדבר בלעם למסר־מעל‬ ‫ביהוה על־דבר־פעור ותהי המגפה בעדת יהוה‬

αὗται γὰρ ἦσαν τοῖς υἱοῖς Ισραηλ κατὰ τὸ ῥῆμα Βαλααμ τοῦ ἀποστῆσαι καὶ ὑπεριδεῖν τὸ ῥῆμα κυρίου ἕνεκεν Φογωρ καὶ ἐγένετο ἡ πληγὴ ἐν τῇ συναγωγῇ κυρίου

Table 15: Numbers 31:16 in M and G.

The Septuagint’s reading remains distinct from that in M, but it is unclear to what extent. Making the issue more difficult is M’s unclear text, which can be translated (barely) as, “Behold! They were [the ones] for the Israelites, by Balaam’s word, to instruct treachery against YHWH because of the Peor affair. And there was a plague against the congregation of YHWH.” This unclear text has led some to consider the Greek version a reformulation and improvement upon the Masoretic text,18 but the possibility should not be dismissed out of hand that G attests an older reading and that M is simply corrupt. note should be viewed as a harmonization (e.g., the plural cannot refer to Balaam and the “addition” of such a note would be superfluous and nonsensical). Probably (proto-)M removed the duplicate material to make the sentence read more smoothly. 17 Contra, e.g., Levine, Numbers 21–36, 445, who regarded the whole of Numbers 31 as a unit attributable to P. 18 Cf., e.g., Rösel and Schlund, “Arithmoi,” 508. Dorival, Nombres, 523 remarked: “…loin d’être une glose, la traduction de la LXX s’efforce de rendre le texte hébreu qui lui servait de modèle en tenant compte de traditions d’interpretation”.

4.2 Numbers 31

213

The most significant text-critical issue revolves around the phrase recorded in M as ‫למסר־מעל ביהוה‬.19 Some have suggested correcting M with a conjecture, such as ‫ למעל‬for the elements ‫למסר מעל‬.20 For M’s ‫ למסר־מעל ביהוה‬G proffers τοῦ ἀποστῆσαι καὶ ὑπεριδεῖν τὸ ῥῆμα κυριοῦ. No Masoretic pendant exists for the Greek τὸ ῥῆμα before κυριοῦ, but the most likely candidate – should it have stood in the Vorlage – would have been ‫דבר‬.21 It is possible that this element was lost due to haplography. Notice the two other appearances of ‫ דבר‬in this verse. As to the other differences, it remains unclear what should be a translation of which element from the Hebrew, an issue stemming from the variant syntax; whereas M attests an (apparent) infinitive followed by a substantive, G reads two infinitives conjoined with the conjunction καί. At the same time, it remains uncertain how G interpreted the term ‫למסר‬, assuming of course that this is what stood there.22 The most likely reconstruction suggests some relation to ‫ ;סור‬cf. Num 12:10 and 14:9 in which ‫ סור‬has been translated with a form of ἀφίστημι. However, no such known form of this biblical Hebrew verb exists; it would have to reflect a Hiphil masculine single participle written defectively with an attached preposition ‫ל‬. Perhaps some scribe accidentally introduced the ‫ מ‬or the ‫ ל‬under the influence of the preceding ‫בלעם‬ or the following ‫)ו(מעל‬. To the following term, it remains certainly within the realm of plausibility that the ‫ ו‬was lost under the influence of the preceding ‫ר‬. The Septuagint also certainly understood this root as a verb and not as a substantive. Perhaps it also had an affixed ‫ ל‬that was lost in transmission, but this remains speculative. The most likely scenario for the loss of the ‫ ל‬together with the ‫ ו‬would be haplography of both letters before the ‫מ‬. Possibly the combination ‫ ול‬resembled a ‫מ‬ enough that both letters could have been lost in one scribal error. Perhaps as well, some error in this way led to the insertion of a ‫ מ‬in the preceding word. To me, it certainly seems possible that the reading ‫ לסר ולמעל‬or ‫ לסור ולמעל‬was corrupted over the course of transmission to read ‫למסר למעל‬. Reconstructing a missing ‫ דבר‬as well then, the Vorlage of G may have read ‫הן הנה היו לבני ישׂראל‬ ‫בדבר בלעם לסר ולמעל בדבר יהוה על־דבר־פעור ותהי המגפה בעדת יהוה‬. The verb ‫ מסר‬in combination with the noun ‫ דבר‬might have led some scribes – especially 19 Levine, e.g., saw no need to emend the text here, for which he offered the somewhat fanciful translation “to array forces in sacrilegious rebellion”; cf. Levine, Numbers 21–36, 455– 56. 20 Cf. Dillmann, Numeri – Deuteronomium und Josua, 190, who based this conjecture on T. 21 LXX.E seems more certain about the Vorlage matching M than I am; cf. Rösel and Schlund, “Arithmoi,” 508. I hardly find that plausible. More likely, the text of (proto-)M was emended. At the same time, Seebass, Numeri III, 287–88 accepted that the witnesses against M probably attest a distinct Vorlage, but he favored M nonetheless, even though (or because, actually, in his argumentation) it is the narratological lectio facilior. That hardly seems likely. 22 Gray, Numbers, 423 regarded M’s ‫ למסר‬as a corruption. Cf., however, Dorival, Nombres, 523, who regarded it as an interpretation based on other traditions.

214

The Bible’s Balaam beyond Numbers 22–24

perhaps scribes in a tradition identified as “Masoretic”23 from √‫ – מסר‬to remove “the word” so that Balaam no long “instructed [them] in transgression against the word of YHWH”, but rather only “instructed [them] in transgression against YHWH”. Further, Numbers 25 – the “Peor Affair” – makes no explicit mention that anyone transgressed against the word of YHWH. That is, M maintains a higher level of consistency than G does. Once the errors reflected by the distinctions in the verbs crept in, perhaps the ‫ דבר‬was intentionally removed so that it would not be possible to understand Balaam as having in any way corrupted the text of scripture. This interpretation and reconstruction are by no means certain, however. The content of this verse again varies from anything encountered thus far and provides unique information about Balaam. The incident referred to in Num 31:16 must be some or all of Numbers 25, the last time the words ‫מגפה‬ and ‫ פעור‬were used in Numbers. Likewise, Num 31:16 presumes the conflation of the Moabites and Midianites attested in Numbers 25; this chapter notes that the Moabite women collectively invited Israel to their offerings, but a single Israelite man brought a single Midianite women into the assembly. Numbers 31:16 appears to have understood the text such that the Midianites and Moabites were identical.24 Otherwise the “vengeance” (‫ )נקם‬in Numbers 31 would have no basis, at least not in any known text. To what other incident might it refer? The phrase ‫ על־דבר־פעור‬in Num 31:16 and twice in Num 25:18 make this innerbiblical reference a virtual certainty. But Balaam appears nowhere in Numbers 25. For the first time, then, in Num 31:16 the reader encounters the suggestion that Balaam encouraged the Midianite women to lead the Israelites astray. The only interaction that Balaam might have had with Midianites in the text of Numbers comes from the secondary notices in Num 22:4a and 7a, in which Midianite elders were among the Moabite elders. “Der Redaktor bringt diese [die Ältesten Midians] wie durch eine Hintertür ins Spiel, indem er die Ältesten Moabs sich an die Ältesten Midians wenden läßt, sie also praktisch herbeiruft. Dadurch daß Moab den Ältesten Midians seine Sorgen mitteilt, werden schließlich auch diese in den Kreis derer mit einbezogen, die sich vor Israel fürchten und es zugrunde richten wollen. Dies ist der erste Schritt, um die Midaniter als potentielle Gefahr (vgl. Num 25,6– 15) und als Israels Feinde einzuführen (vgl. Num 25,16–18; 31). “In einem zweiten Schritt werden die Ältesten Midians in Kontakt mit Bileam gebracht (22,7). Mehr braucht es nicht, damit die Erzähllogik für die nachfolgenden Kapitel Num 25 und Num 31 einigermaßen stimmt. In dem Augenblick, wo die Ältesten Midians auf Bileam treffen, haben diese die Möglichkeit, von ihm den listigen Ratschlag einzuholen, der die Israeliten zum Verderben führen kann (vgl. Num 31,16).”25

23

I am thinking here, of course, of the ‫ ;בעלי המסרה‬cf. Tov, Textual Criticism, 26. Cf. Seebass, Numeri III, 292 and the literature cited there. 25 Fistill, Israel und das Ostjordanland, 79–80. 24

4.2 Numbers 31

215

The secondary introduction of the Midianites into Numbers 22–24, the combination of the Moabites and the Midianites in Numbers 25, and the conflation of both peoples while referencing both pericopes in Numbers 22–24 and 25 could indicate a common ideological background or historical understanding for these elements.26 Certainly it seems likely that Numbers 31:16 must have been familiar with the combination of Numbers 22–24 with Numbers 25.27 This combined text, Numbers 22–25, may have served as the background for the introduction of Balaam into Num 31:16, making him responsible for the Israelite’s apostasy under Midianite influence.28 Smend described the process thus: “Man kombinierte daher c 22–24 mit der Erzählung von c 25 dahin, daß Bileam ein böser Zauberer (vgl. Jos 13,22 ‫ הקוסם‬mit Num 23,23 ‫ )קסם‬war, der Israel zum Götzendienst und zur Unzucht mit den midianitischen Weibern verführte (Num 31, 8.16; Jos 13,22). Diese Auffassung kommt 22,4.7 in den deutlich eingetragenen Worten ‫ אל זקני מדין‬und ‫וזקני מדין‬ ‫ וקסמים בידם‬zum Ausdruck.”29

Based on the literary-critical analysis of Numbers 22–24 in Chapter Three, the editor likely responsible for this combination was R2. He or she presumably added this verse in Numbers 31 at the same time that he included the story of the Moabite and Midianite elders who went out to hex the Israelites in Numbers 22–24. Perhaps this indicates that the fragment of S2 identified in Numbers 22–24 continued with an earlier, abbreviated version of Numbers 25 and 31. To this end, one notes the fairly consistent use of ‫ בני ישׂראל‬for the exodus group in these texts, commending a common diachronic and ideological, tradition-historical background. Once the moment reported in Num 31:16 became integrated into the text, ideological necessity – a sense of justice – probably motivated someone to incorporate the notice that the Israelites killed Balaam in Num 31:8, a datum that reads like a foreign interpolation. Theoretically both of these elements in Numbers 31 could have been composed at the same time, but the glossy features of Num 31:8 commend understanding it as a later insertion to resolve Balaam’s outcome.30 This outcome, by the way, and Numbers 31 in general consider Balaam in a geographical context inconsistent with Num 24:25, which presupposes Balaam’s returning to Aram. Such a relocation precludes his presence among the Midianites. The tension thus created suggests a status for the notices in Numbers 31 distinct from – and probably secondary or tertiary to – the story 26

Cf. Levine, Numbers 21–36, 144–45 and 445, who regarded all of them as stemming from

P. 27

Cf., e.g., Noth, Numeri, 200. N.b. Seebass’s identification of Balaam as a Horrorfigur in this verse; cf. Seebass, Numeri III, 292. 29 Smend, Erzählung des Hexateuch, 230. 30 This contradicts, e.g., Seebass, Numeri III, 292–99 and Levine, Numbers 21–36, 445, who regarded Num 31:8 and 16 as part of the same priestly layer of Numbers. 28

216

The Bible’s Balaam beyond Numbers 22–24

in Numbers 22–24. That is, we notice similar tendencies in Numbers 25 and 31 to those we identified in Numbers 22–24.31 Regarding Numbers 25, one can divide the text along the same diachronic lines as Numbers 22–24 with 25:1–5 belonging to the first redactional layer identified in Numbers 22–24 (here, the term ‫ העם‬represents a military organization, just as in, e.g., Deut 20:1 and 5 and in the core of Numbers 31 [see below], and not the whole exodus group as in the oldest identified layer in Numbers). The second redactional layer appears in 6aαβ, 7a, 8, and 10–13. Verse 9 either belonged to this second redactional layer or the source it apparently incorporated into Numbers. The later redactional material consists of 6aγδb, 7b, 14–19. Turning back to Numbers 31, the text does not appear unified. Rather, some layers may be discerned.32 “The people” (‫ )העם‬appears in 31:2–3 and 32. Verse 2 uses ‫ העם‬in a euphemism for death. In v. 3 it refers to the military and not the exodus generation. The term in 31:32 is an error and the text should be emended to reflect G.33 “The Israelites” (‫ )בני ישׂראל‬dominate the chapter and attest its general status as part of the secondary redaction identified in Numbers 22–24. Yet, some material awakens a tertiary or later impression. For practical reasons, we need not cover the whole chapter, but v. 16 impresses as a portion of the second redaction through its identification of the exodus group with the term ‫בני ישׂראל‬. It describes the scenario in Num 25:6, which also features the ‫בני ישׂראל‬. That is, at least the core of verse 16 belonged to an earlier version of Numbers 31 than that which we now possess.34 Most relevant to the discussion at hand: this editor added Balaam, identifying and condemning him as the party responsible for the Israelites’ treachery against YHWH. The reference to

31 Cf. Thomas B. Dozeman, “The Midianites in the Formation of the Book of Numbers,” in The Books of Leviticus and Numbers, Thomas Römer (Leuven: Peeters, 2008), 278: “The web of literary relationships indicates that the introduction of the Midianites into the story of Balaam sets the stage for a polemical interpretation, which is also directed against Balaam. The negative interpretation of the Midianites cannot be harmonized with their liminal role in the non-Priestly narrative, where they function positively as foreigner and as religious insider.” and Dozeman, “Midianites,” 283: “The insertion of the Midianites into the story of Balaam lays the foundation for a rejection of the positive role of non-Israelites in the cult of YHWH. The focus on cultic origins at Mount Sinai, in conjunction with the genealogy of Phineas as the authentic priest of YHWH, provide the basis for exterminating Balaam and the Midianites as a danger to the purity of the cult of YHWH.” 32 Contra Levine, Numbers 21–36, 445, who suggested that the whole chapter is “entirely attributable to P”. Others had already questioned the unity of this chapter; cf. e.g., Noth, Numeri, 198–99. 33 Οἱ ἄνδρες = ‫אנשׁי‬. 34 Cf. Noth, Numeri, 200, who distinguished 14–18 from the later additions in 13 and 19– 20.

4.2 Numbers 31

217

the ‫ עדה‬in 16b commend regarding it as a later addition, from R3+.35 Similarly, the identification of ‫ על־דבר־פעור‬appears to be a tertiary interpolation, as will become apparent presently. The second redactor, who included Balaam into Numbers 31, thus demonstrates a familiarity with the composite text of Numbers 22–25 including the conglomeration of Moabites and Midianites, Balaam, and an Israelite transgression. This texual amalgamation began the downward trajectory of Balaam’s characterization in the Bible. The third redactor, or someone even later, must have added the Midianite monarchs and Balaam to v. 8. It interrupts material from the second redaction (vv. 7 and 9). Numbers 31:8 connects explicitly to Num 25:15 and 18, which read like interpolations in their context. The mention of ‫ על־דבר־פעור‬in Num 25:18, thus apparently from R3, commends regarding this phrase as R3 in Num 31:8 as well. This editor, the one responsible for the comment on Balaam’s demise, thus completed the transformation of Balaam’s characterization to one of villainy. Having familiarity with Num 31:16, this author did not shy from condemning Balaam, but permitted his death at the hand of the Israelites for what he had done. From this point on, there was no more hope for Balaam ben Beor in the Bible. In sum, then, Numbers 31 contributes new information about Balaam, but information not directly motivated by his image in Numbers 22–24. As will become apparent below, this distinguishes Numbers 31 somewhat from the others texts addressed here. Through no fault of his own, the character Balaam became affiliated with negative circumstances in the text of Numbers over the course of its redaction history. Through the introduction of the Midianites to Numbers 22–24 and the editorial appending of Numbers 25, Balaam transformed from one of Israel’s allies in Numbers 22–24 to one of its great nemeses just a few chapters later. Since neither the fragment about Midian in Numbers 22–24, nor the episode at Peor in Numbers 25, nor even the war against the Midianites in Numbers 31 initially had anything to do with Balaam, we must recognize that only editors in the book of Numbers were responsible for this later negative characterization of this figure within Numbers. For Greene, who regarded this editorial phase as P, Balaam in this reconstructed P represents a typos, the pinnacle or culmination of foreign religious practices. That is why the combination of texts now found in Numbers 25 and 31 demanded that he die for his activities.36 As opposed to texts like Deuteronomy 23 and Joshua 24, Numbers 31 fails to report the motif of blessing.37 However, they were not

35

Cf. already Dillmann, Numeri – Deuteronomium und Josua, 190, who identified it as editorial in a P context. 36 Cf. John T. Greene, Balaam and his interpreters. A hermeneutical history of the Balaam traditions, BJS, vol. 244 (Atlanta: Scholars Press, 1992), 69–74. 37 Cf. Schmidt, “Bileamüberlieferung,” 260.

218

The Bible’s Balaam beyond Numbers 22–24

the only authors still found in the Pentateuch to ascribe negative actions to Balaam in the Torah.

4.3 Deuteronomy 23:4–638 4.3 Deuteronomy 23:4–6

The so-called Qahal-legislation in Deuteronomy 23:2–9 designates who can and cannot belong to the “assembly of the lord”, to borrow its vernacular. Roughly in the middle of this list, one encounters the figure Balaam ben Beor as the basis for the exclusion of some peoples from the assembly. This line of argumentation is surprising, since Balaam appears nowhere else in Deuteronomy. Reading from the beginning of the Pentateuch, Balaam has only appeared thus far in Numbers 22–24 and 31. But Deuteronomy’s retelling of the incident with Balaam does not particularly match that found in Numbers. That suggests in turn that Deuteronomy 23 must be basing its arguments on other traditions or interpreting Numbers in a way not immanent to Numbers 22–24 or even 31. This section reflects on how Deuteronomy’s version of Balaam’s tale relates to other data about this figure in the Hebrew Bible and addresses how Balaam came to find his place in the legal collection of the Qahal-legislation. The argumentation will proceed in three steps: 1) Deuteronomy 23:4–7 in its context; 2) Text- and literary-historical issues in Deuteronomy 23:2–9, with particular emphasis on vv. 4–7; 3) the relationship of Deuteronomy 23:4–7 to other biblical Balaam traditions. First, let us turn to Deuteronomy 23:4–7 within its context in Deuteronomy 23:2–9 in the Masoretic version. The Qahal-legislation excludes four types of people (vv. 2–7) from the assembly and admits only two (vv. 8–9). The first two disenfranchised groups appear to have some kind of sexual irregularity or impurity, either as men with mutilated genitals or as a ‫מ ְמזֵ ר‬. ַ These proscriptions apply to individuals, apparently irrespective of their culture. Following that, Deuteronomy 23 outlines some permissible and inadmissible ethnic groups. Verses 4–7 preclude two peoples, the Ammonites and Moabites, on historical grounds. Verses 8–9 permit two peoples based on their filiation (Edomites) or historically founded reciprocity (Egyptian). That is, the text defines the Edomites as brothers and admits Egyptians since the assembly had previously lived in Egypt as ‫גרים‬. At first glance, these explanations are simple enough, but the text more forcefully explicates Moabite and Ammonite proscription.

38 The section on Deuteronomy 23 represents a reworking of a paper I presented at the International SBL / EABS Meeting in Berlin in 2017. A preliminary version appeared in German as Jonathan Miles Robker, “Der fremde Prophet Bileam im Qahal-Gesetz des Deuteronomiums,” ZABR 23 (2017), 137–149.

4.3 Deuteronomy 23:4–6

219

The Ammonites and Moabites, despite arguably being related to Israel, cannot join. Deuteronomy 23:4bα precludes their admission to the assembly for ten generations. Simultaneously, 23:4bβ refuses them eternally. One half-verse sets two distinct time limits for their inadmissibility. Verse 5 explains why they remain inadmissible, namely because they failed to bring the Israelites sustenance (5a), and they hired Balaam ben Beor, who subsequently sought to curse Israel (5b). That is, the hostility toward Moabite and Ammonite participation develops argumentatively in two stages: 1) they were inhospitable, failing to come out with bread and water; 2) they were antagonistic, going so far as to hire a professional from Mesopotamia to curse the assembly. Yet, Balaam’s cursing remained inefficacious through YHWH’s intervention, or rather due to YHWH’s unwillingness to listen to Balaam: YHWH transformed Balaam’s attempted curse into a blessing. This transformation occurred because of YHWH’s love for the assembly. Many matters obviously distinguish this recounting of Balaam’s interaction with the group undertaking the exodus from that attested in Numbers 22–24. However, before considering that, we must address the textual transmission attested by the various witnesses to reconstruct the oldest possible version of the Deuteronomy text and trace its development through recorded transmission history. Deuteronomy bases its refusal to allow Ammonites and Moabites into the community on Balaam’s attempt to curse Israel. Yet, the presupposed attempt remains distinct from the material found in Numbers 22–24 in a number of matters. In contrast to the episode in Numbers 31, however, only sparse textcritical matters deserve attention here and they are more easily dealt with. Nonetheless, a number of distinctions between Numbers and Deuteronomy exist, and they apparently are literarily and redaction-historically relevant. First, G apparently translated the ‫ גם‬found in Deut 23:4 M with (καὶ) ἕως, which remains an appropriate translation of this particle – even if not the most apparent – especially considering its context. The same could be said for the καὶ ἕως εἰς at the end of the verse as a translation for ‫ ;עד‬while it is possible that the Vorlage differed in these cases, it is hardly necessary.39 However, it is unclear why G does not reflect the suffixed preposition ‫ להם‬in this verse. It does not appear in V or G, but GO and T attest it. The resolution to this remains opaque, but the European languages have no need of this particle, and thus their unedited translations may have simply ignored it.40 The plural of ‫ שׂכר‬probably represents a harmonization based on context;41 cf. the plural ‫ קדמו‬earlier in 23:5. On the other hand, it could just as easily be 39

Cf. the discussions in John William Wevers, Notes on the Greek Text of Deuteronomy, SBLSCS (Atlanta, Ga.: Scholars Press, 1995), 364–65 and McCarthy, Deuteronomy, 115*. 40 Cf. Wevers, Notes on the Greek Text of Deuteronomy, 365, who identified the Hebrew ‫“ להם‬largely untranslatable” in this instance. 41 The editors of BHQ favor this explanation; cf. McCarthy, Deuteronomy, 66.

220

The Bible’s Balaam beyond Numbers 22–24

an oversight, the ‫ ו‬having been overlooked after the ‫ר‬. Neither alternative can be granted certain priority over the other, yet the meaning remains clear regardless of the form. More importantly, the identification of Balaam’s land of origin attests some distinctions worth noting. The Masoretic text, α’, S, and T all reflect the same reading: ‫מפתור ארם נהרים‬. Syntactically, this presents the most well-integrated reading, which probably indicates its later status. The Samaritan Pentateuch could thus be indicative of an older phase in the text history: the specific location in Smr matches that found in Num 22:5 (‫)פתורה‬. Presumably this similarity, which makes no grammatical sense here (the directive ‫ה‬, while sensible in Num 22:5 remains superfluous in Deut 23:5), attests an older phase in the harmonization of the data between Numbers 22–24 and Deuteronomy 23.42 The Septuagint, here supported still by V (de Mesopotamia Syriae), attests what probably presents the oldest reconstructable reading: ἐκ τῆς Μεσοποταμίας, most likely from a Vorlage that read ‫)מ(ארם נהרים‬.43 While Numbers 22–24 merely permits the possibility that Balaam comes from Mesopotamia, Deuteronomy 23 makes it explicit, even in the oldest version. Over the course of transmission, these two versions were presumably harmonized, as witnessed by the variants at both locations.44 Thus, e.g., Deut 23:5 was expanded from its original reading (‫ ]מ[ארם נהרים‬with G + V) to incorporate the specific city in the form attested in Num 22:5 (‫ פתורה‬with Smr), before this was syntactically ironed out in the proto-M textual tradition, leading to the reading found in Deut 23:5 M (‫)מפתור ארם נהרים‬.45 But to what degree is such an addition relevant? There are two important answers to this question. First, we noticed a continued tendency toward intertextual harmonization in the Smr and M traditions, just as noted about the textual history of Numbers 22–24 in Chapter Two. Second, it is possible to identify two distinct backgrounds for Balaam. Whereas Numbers describes him as an Aramean in the oldest reconstructable form, the text of Deuteronomy apparently regarded him as a Mesopotamian, not even from the city of Pethor, in its

42 Cf. Richard D. Nelson, Deuteronomy. A Commentary, OTL (Louisville; London: Westminster John Knox Press, 2002), 276. Contra Eckart Otto, Deuteronomium 12,1–23,15, HThKAT (Freiburg; Basel, Vienna: Herder, 2016), 1735. 43 Contra Wevers, Notes on the Greek Text of Deuteronomy, 365–66, Otto, Deuteronomium 12,1–23,15, 1735, and McCarthy, Deuteronomy, 115*, who favored M in this instance, disregarding its harmonizing aspect. 44 Cf. the text-historical and literary-critical discussion about Num 22:5 in chapters 2 and 3. 45 My reconstruction here opposes the opinion offered in BHQ; cf. McCarthy, Deuteronomy, 115*. Cf. already Albright, “Home of Balaam,” 388, who identified ‫ פתור‬in Deut 23:5 as a “post-Septuagintal gloss”. Rouillard, Balaam, 211–12 identified Deut 23:5 as a later interpretation of Num 22:5 and 23:7, a plausible explanation for Balaam’s later Mesopotamian heritage.

4.3 Deuteronomy 23:4–6

221

oldest version of this verse.46 Therefore, this minor distinction could present evidence of distinct Balaam traditions within the Bible: one found in Numbers 22–24 and one found in Deuteronomy 23. Alternatively, and more likely, Deuteronomy 23 could represent an interpretive amalgamation of data from Num 22:5 (‫ פתר‬and ‫ )נהר‬and 23:7 (‫)ארם‬. The variants in Deut 23:6 can be covered and explicated quickly. The Septuagint and S do not reflect M’s ‫ לך‬and neither does Neh 13:2 M; the shorter reading should be preferred, the longer reading (M, Smr, GO [behind the asterisk], and V [but affixed at another point]) having resulted from some form of dittography from the preceding ‫אלהיך‬.47 The plurals in G et al. for blessings and curses should be understood as collectives and may reflect Balaam’s multiple attempts to curse the people in Numbers 22–24. Thus, they should be regarded as reflecting a harmonizing tendency and ignored as secondary.48 Clearly a harmonizing tendency exists between Numbers 22–24 and Deuteronomy 23, as attested in the various witnesses. Once these tendentious elements have been identified and removed, as above, the following reconstruction of an older version of Deut 23:4–6 seems probable: ‫לא־יבא עמוני ומואבי‬ ‫בקהל יהוה גם דור עשׂירי לא־יבא להם בקהל יהוה עד־עולם על־דבר אשׁר לא־קדמו‬ ‫אתכם בלחם ובמים בדרך בצאתכם ממצרים ואשׁר שׂכר עליך את־בלעם בן־בעור מארם‬ ‫נהרים לקללך ולא־אבה יהוה אלהיך לשׁמע אל־בלעם ויהפך יהוה אלהיך את־הקללה‬ ‫לברכה כי אהבך יהוה אלהיך‬. Several features suggest that this text developed uniquely and only secondarily became increasingly affiliated with Numbers 22–24. The only connecting features between the two texts are names and the motif of cursing and blessing. Otherwise both the background implied by and semantics used for the circumstances surrounding Balaam’s interaction with Israel are distinct. First, Numbers 22–24 fails to connect the Ammonites with the Moabites as in Deuteronomy 23.49 The conjunction of these groups is hardly insignificant in that the tradition in Deuteronomy regards both of these peoples as neglecting Israel during its exodus and – even going beyond that – together summoning 46

In his magisterial history of Arameans, Younger too hastily combined the terms in Num 23:7 and Deut 23:5 regarding both Aram and Aram-Naharaim as the same place; cf. K. Lawson Younger, Jr., A Political History of the Arameans. From Their Origins to the End of Their Polities, ABS, vol. 13 (Atlanta: SBL Press, 2016), 96. They refer instead to two different places. Cf. the discussion below in Chapter Six. 47 Contra Wevers, Notes on the Greek Text of Deuteronomy, 366; cf. McCarthy, Deuteronomy, 115*. 48 Cf. Wevers, Notes on the Greek Text of Deuteronomy, 366 and McCarthy, Deuteronomy, 115*. 49 Cf. Samuel Rolles Driver, A Critical and Exegetical Commentary on Deuteronomy, ICC (Edinburgh: T&T Clark, 1986), 261: “…the ‘Ammonites are not mentioned in connexion [sic] with Balaam”, apart from the harmonization in 22:5 in some manuscripts, Smr, S, and V, of course…

222

The Bible’s Balaam beyond Numbers 22–24

Balaam to curse Israel. Nothing in Numbers merits such an interpretation. There, only the Moabites, specifically their king, Balak, play any role in summoning him. Later editors did add the Midianites, but not the Ammonites. Further, Deut 23:5 implies that Balaam received remuneration for his undertaking (‫)שׂכר‬, which specifically contradicts Num 24:11.50 Second, the Numbers text conspicuously and particularly precludes any attempt Balaam might have made to curse the people. Rather, from the beginning of the story – and in every redactional iteration until perhaps the last – he only ever sets out under the premise that he can only do that which the deity permits. Numbers 22–24 notes from essentially the outset that the people Balaam should curse is blessed (Num 22:12) and neither Balaam nor the narrator ever questions this blessedness. Balaam’s willingness to curse “the community of YHWH” (‫ )קהל יהוה‬in Deuteronomy 23 directly contradicts the story in Numbers 22–24. Indeed, Deut 23:6 presumes that Balaam indeed uttered a curse, which God refused to hear.51 For that matter, only after such a curse was spoken would it have been possible for God to transform it. This demonstrates Deuteronomy’s negative attitude toward Balaam.52 For that matter, the nomenclature for the group he should curse is distinct in Deuteronomy (‫)קהל יהוה‬ when compared to Numbers (‫בני ישׂראל‬, ‫ישׂראל‬, and ‫)]ה[עם‬.53 Third, YHWH’s willingness to listen to Balaam and YHWH’s love for Israel – the twin foci of Deuteronomy 23:4–6 – play no role at all in Numbers 22–24. The roots √‫ אבה‬and √‫אהב‬, the keywords and primary theological motivation behind Deut 23:6 and the episode it reflects, do not even appear in Numbers 22–24. Fourth, as noted in the preceding text-critical reflections, Balaam apparently came from two different places in the traditions: Mesopotamia in Deuteronomy and Pethor (or “Seertown”) in Aram, i.e., near Damascus, in Numbers. Finally, the primary motif of the whole story – Balaam cursing – relies on distinct verbs for cursing in Deuteronomy 23 vis-à-vis Numbers 22–24. In spite of Deuteronomy’s often using the root √‫ ארר‬for cursing,54 it did not use this root in its recounting of Balaam’s tale, but used √‫ קלל‬instead.55 And Deuteronomy uses only √‫ קלל‬for cursing in this story. Numbers never uses √‫קלל‬, but

50

Cf. David Frankel, “The Deuteronomic Protrayal of Balaam,” VT 46, no. 1 (January 1996): 32. 51 Cf. David Frankel, “The Deuteronomic Protrayal of Balaam,” 31–32. 52 Curiously, David Frankel, “The Deuteronomic Protrayal of Balaam,” 38 did not regard Balaam cursing Israel as a negative reflection on him. Frankel thinks Balaam is only doing his job, something which biblical authors would not hold against him. Based on content and context, this position seems spurious. 53 But, n.b., the pejorative ‫ הקהל הזה‬in the Moabites’ mouths in Num 22:4. 54 Sixteen times total in Deuteronomy, all in chapters 27–28. 55 Eleven times total in Deuteronomy.

4.3 Deuteronomy 23:4–6

223

attests no fewer than three other roots for cursing: √‫ארר‬, √‫קבב‬, and √‫זעם‬.56 This circumstance largely precludes Deuteronomy copying from Numbers 22– 24, as well as Numbers copying from Deuteronomy. Had that been the case, it seems particularly unlikely that the two texts would have relied on distinct roots for the most poignant motif in the story and narrated the undertakings in such disparate versions. Going beyond superficial textual matters, one must also regard the distinct form of Deuteronomy 23 when contrasted with Numbers 22–24 (and even Numbers 31 and the other biblical Balaam texts, for that matter). Deuteronomy 23 relies on Balaam as indicative of Israel’s enemies’ negative behavior toward them during the exodus. In other words, Balaam appears in a context defining Israel’s social mores regarding admission into the community (‫)קהל‬. In particular, that is, even the form of Deuteronomy’s sole reference to Balaam appears distinct from other texts about this figure. This distinction should not be easily overlooked or pushed aside. Turning to Balaam’s literary-historical embedding in Deuteronomy 23, one notes that he stands out as the basis for legislation. Singularly the refutation of admission for Ammonites and Moabites merited a lengthy and repetitive basis in this passage in Deuteronomy.57 The admission criteria for the ‫ קהל‬begins in Deut 23:2, noting that no emasculated person should be admitted. Following this, 23:3 (for which the shorter reading in G should again be preferred, and 23:3b deleted as recensional;58 cf. 23:4 and 9) precludes the admission of any ‫ממזר‬, probably a bastard or child of some other illicit union, such as incest.59 Neither of these first two exclusions present a reason; cf. G. Thereupon follows the prohibition of Ammonites and Moabites, with – for the first time – a historical basis of the proscription. One notes the two chronological limitations: ten generations and forever. The refutation of Moabite and Ammonite membership in the ‫ קהל‬continues through verse 7, whereupon vv. 8–9 permit the admission of Edomites and Egyptians in the third generation. Verse 8bβ does 56 While noting the general lexical distinctions regarding cursing between Deuteronomy 23 and Numbers 22–24, David Frankel, “The Deuteronomic Protrayal of Balaam,” 39 overlooked some of the terms Numbers attests for cursing. 57 Primarily for stylistic reasons, Gerhard von Rad, Das fünfte Buch Mose. Deuteronomium, übersetzt und erklärt, ATD (Göttingen: Vandenhoeck & Ruprecht, 1968), 104 regarded the material about Balaam as a later interpolation. In this we can readily follow him, as, e.g., Ruth Ebach, Das Fremde und das Eigene. Die Fremdendarstellungen des Deuteronomiums im Kontext israelitischer Identitätskonstruktionen, BZAW, vol. 471 (Berlin; New York: de Gruyter, 2014), 69–104. Alternatively, Nelson, Deuteronomy, 278 considered Balaam as part of a historical expansion of the text, also a possibility. Contra Otto, Deuteronomium 12,1– 23,15, 1743–48, who argued for the unity of Deut 23:2–9. 58 Contra Rösel and Schlund, “Arithmoi,” 579 and McCarthy, Deuteronomy, 114*-115*. 59 Cf. the brief discussions in Driver, Deuteronomy, 260–61 and von Rad, Deuteronomium, 105 in addition to the dictionaries for literature for the seldom-attested lexeme.

224

The Bible’s Balaam beyond Numbers 22–24

proffer some historical basis for the admission of Egyptians. Regarding Edomites, v. 7aβ presents not a historical basis to permit their admission, rather stating instead that, “he is your brother”. This unclear situation could suggest that scribes secondarily appended v. 8bβ as a historical basis for the admission of Egyptians. This historical recourse links v. 8 with vv. 5–7, which thus may also attest later editing. With the notice in v. 9, the ‫ קהל‬legislation ends. The embedding and organization of this ‫ קהל‬material are, at first glance, quite murky.60 The ‫ קהל‬jurisprudence begins with implied sexual issues, either personally or in one’s heritage, that preclude admission. Neither or these exclusions finds an explanation; cf. G. Sexuality could reflect a motif association between the ‫ קהל‬and the sexual legislation in Deut 22:13–23:1. Someone apparently found this position appropriate for the addition of the Qahal material, whether as an insertion or a Fortschreibung. The middle focuses on the disavowed ethnic groups (Moabites and Ammonites). The conclusion concentrates on permitted ethnic groups (Edomite and Egyptian), before returning to considerations about the enemies (23:10; ‫)איביך‬. The Moabites and Ammonites present the crux and combination of the sexual and ethnic elements, in my opinion.61 Moving from the notice about men with mutilated genitals and the ‫ממזר‬, probably products of illicit affairs and presumably including children of incestuous relationships, 62 the reference to the Ammonites and Moabites makes some sense. This, of course, is the case, provided one is familiar with a text like Gen 19:30–38, which narrates the incestuous origins of the Ammonites and Moabites. They thus present ethnic groups precluded for sexual reasons in the first instance. No need existed to include any reference to Balak and Balaam or the exodus in an older, perhaps earliest version of this legislation; Deut 23:4 in an earlier, briefer version may have only implied the incestuous reason for forbidding Ammonites and Moabites.63 But for how long? The variant lengths of disenfranchisement evince later editorial expansion to include new information. Based on the form of vv. 2–3 and assuming that G is older for the ‫מ ְמזֵ ר‬, ַ a general, permanent exclusion of Ammonites and Moabites seems originally likely.64 Thus, the core of the legislation against Moab-

60

Cf., however, the synchronic analysis in Otto, Deuteronomium 12,1–23,15, 1736–38. Cf. also already Driver, Deuteronomy, 261: “The ‘Ammonite and the Moabite are to be placed on the same footing as the bastard, on account of their ancestors’ unfriendly treatment of Israel at the time of the Exodus.” 62 Cf., e.g., already Driver, Deuteronomy, 260 and the literature cited there. 63 Cf. already Dillmann, Numeri – Deuteronomium und Josua, 348: “Stünde V. 4 für sich, so könnte [italics in original] man immerhin als Grund dieses Verbots die Rücksicht auf die blutschänderischen Ursprünge und Sitten dieser Völker (s. Gen. 19, 30 ff.) annehmen.” 64 The addition of the number of generations for admission of the offspring of a ‫ ממזר‬reflects the same coordinating tendency described below. If one group had it, the others should as well. 61

4.3 Deuteronomy 23:4–6

225

ite and Ammonite inclusion initially consisted of only 23:4a, which stood parallel to 23:2 + 3a, but which all contrasted with 8aαbα. This oldest form was a simple apodictic list describing who could or could not join the assembly, and it probably rooted in social mores about these groups, presumably with eternal consequences.65 I would suggest that editors reduced this general proscription against Ammonites and Moabites to only ten generations when including the reference to the inhospitable Ammonite and Moabite reception of the Israelites during the exodus. To this end, they added 4b* + 5a. This renewed version stood roughly parallel to the adapted Edomite and Egyptian legislation in vv. 8–9 including the elements in 8aβbβ + 9, with Ammonites and Moabites excluded for ten generations, but with Egyptians and Edomites being permitted after only three. The final editors reintroduced an eternal exclusion of Moabites and Ammonites. Their addition of 4bβ* + 5b–7 created a roughly circular structure, placing the historical basis for Moabite and Ammonite disenfranchisement (vv. 5– 6) between two notices that their partaking in the assembly was eternally precluded (vv. 4bβ and 7). The implausibility of vv. 4bβ* + 5b–7 existing in an independent form or as a fragment of some other source commends its identification as a contextual Fortschreibung composed for this setting. For our purposes here, it suffices to notice that 1) Balaam probably presents the latest interpolation into Deuteronomy 23:2–966 and that 2) this interpolation does not conform to Balaam’s image as presented by Numbers 22–24*, particularly as it existed in its older forms.67

Of course, there is no need to append the number of generations of a castrated or sexually mutilated man, since he presumably cannot produce his own offspring. 65 This is contra Dillmann, Numeri – Deuteronomium und Josua, 348, who continued from the previous citation in note 63, “Aber daraus, dass man dies könnte [italics in original], folgt nicht, dass man es muss, oder gar dass V. 5–7 interpoliert sein werden.” 66 Cf. Donner, “Balaam pseudopropheta,” 113, who regarded Deut 23:5b–6 as a later addition to Deuteronomy 23. 67 This simultaneously precludes the possibility that Deuteronomy 23 served as the basis for Numbers 22–24. E.g., Graeme A. Auld, “Samuel, Numbers, and the Yahwist-Question,” in Abschied vom Jahwisten. Die Komposition des Hexateuch in der jüngsten Diskussion, ed. Jan Christian Gertz, Konrad Schmid, and Markus Witte (Berlin; New York: de Gruyter, 2002), 245: “The brief mention in Deuteronomy of Balaam hired by Balak to curse Israel will relate to the extended stories in Numbers likes it abstract to many a conference paper or its proposal to many a book – not as summary but as programme: written long before the final version, and superseded by the extended text in important aspects.” He presumes quite a bit here, even going so far as to insert Balak into Deuteronomy, a text that was apparently unfamiliar with him! This suggestion assumes that a direct line of tradition exists between these two versions, though this cannot be postulated with anything like certainty.

226

The Bible’s Balaam beyond Numbers 22–24

Later Expansion

Context in Deuteronomy

‫עד־עולם׃‬ ‫ואשׁר שׂכר עליך את־בלעם בן־בעור‬ ‫מארם נהרים לקללך׃‬

‫גם דור עשׂירי לא־יבא להם‬ ‫בקהל יהוה‬ ‫על־דבר אשׁר לא־קדמו‬ ‫אתכם בלחם ובמים בדרך‬ ‫בצאתכם ממצרים‬

Oldest Layer ‫לא־יבא פצוע־דכא‬ ‫וכרות שׁפכה בקהל‬ ‫יהוה׃‬ ‫ממזר‬ ‫לא־יבא‬ ‫בקהל יהוה׃‬ ‫עמוני‬ ‫לא־יבא‬ ‫ומואבי בקהל יהוה‬

V. 2

3 4 5

6

‫ולא־אבה יהוה אלהיך לשׁמע אל־בלעם‬ ‫ויהפך יהוה אלהיך את־הקללה לברכה‬ ‫כי אהבך יהוה אלהיך׃‬ ‫לא־תדרשׁ שׁלמם וטבתם כל־ימיך‬ ‫לעולם׃‬

7 ‫כי אחיך הוא‬ ‫כי־גר היית בארצו׃‬ ‫בנים אשׁר־יולדו להם דור‬ ‫שׁלישׁי יבא להם בקהל יהוה׃‬

‫לא־תתעב אדמי‬ ‫לא־תתעב מצרי‬

8a 8b 9

Table 16: Proposed Layering of Deuteronomy 23:2–9.

Thus the question: whence Deuteronomy’s version of Balaam and of the presumed historical circumstances more generally? The most likely answer would be from Joshua 24, which also describes a scenario closer to Deuteronomy’s prerequisites and terminology (see also the following discussion to this chapter). Three factors commend this: 1) most importantly, YHWH does “not desire to listen”, ‫ אבה‬+ ‫שׁמע‬, in both Deut 23:6 and Josh 24:10, an important motif to both versions absent in Numbers.68 2) A certain wordplay exists between the two versions, in that in Deut 23:5, the Moabites and Ammonites did not greet the assembly “with bread” (‫)בּלֶּ ֶחם‬ ַ and in Josh 24:9, Balak “battled” (‫)וַ יִּ לָּ ֶחם‬ against Israel. Again, Numbers does not reflect either of these elements substantially, though admittedly it does peripherally.69 3) Both versions, Joshua 24 and Deuteronomy 23, relate the same verb for cursing, ‫קלל‬. These three aspects commend relating these versions to each other. On the other hand, the Joshua version appears to have some familiarity with the Numbers text; one notes particularly the precise repetition of ‫ שׁלח‬and ‫ קרא‬in Num 22:5 and Josh 24:9b. This would suggest that Joshua 24 adapted the version in Numbers 22–24* and that Deuteronomy represents a further development on Joshua 24, more distantly removed on a continuum from Numbers 22–24. Clear evidence 68 The element of listening, ‫שׁמע‬, may have been added over the course of transmission into the Joshua text from Deuteronomy 23. Cf. the following discussion to this verse below. 69 Balak does desire to strike (‫ )נכה‬the people in Num 22:6 and Balaam reports his desire to battle them in 22:7 (‫)לחם‬.

4.3 Deuteronomy 23:4–6

227

of Deuteronomy’s developing Joshua 24 appears in the addition of the motifs of YHWH’s “transforming” (‫ )הפך‬the curse into a blessing and loving Israel, elements yet foreign to Joshua. Perhaps the reference to Balaam in a context combining Ammonites and Moabites also reflects familiarity with some version of the Jephthah story of Judges 11, in which one finds the same combination, albeit with Balak instead of Balaam. What inspired this Fortschreibung in Deuteronomy 23? The answer probably relates to the historically based permission to admit Egyptians and Edomites in Deut 23:8–9. The text grants permission to each of these groups for different reasons after three generations. The basis for the Egyptians is apropos: in the narrative of Deuteronomy Israel is in its liminal phase between leaving Egypt and arriving in the land. If this historical circumstance permits Egyptians, one wonders why Deuteronomy allows the Edomites admission. Deuteronomy grounds this in the identification of Edomites as relatives, as narratively embedded in Genesis. Genesis also recounts that Moab and Ammon are also related to Israel, but Deuteronomy has excluded them. The text thus must base this distinction on something, so it first chose their inhospitality as the reason for Ammonite and Moabite exclusion. Later editors expanded this with the Balaam episode. The first addition probably occurred in tandem with the historical argumentation for Egyptian admission. Should the ‫ קהל‬legislation have existed in an independent form, a solution I currently favor, these elements were presumably added when it was incorporated into the narrative context of Deuteronomy as part of the exodus. The Balaam episode must have later served as a more marked contrast to the basis for Egyptian and Edomite admittance and forbade the Ammonites and Moabites forever. Yet, is it possible to reconstruct when editors inserted Balaam into Deuteronomy 23? The Qahal legislation may itself be a late interpolation into Deuteronomy, but the addition of Balaam into the text could not have happened too late. Nehemiah 13:1–3 knows and pretty much copies the text of Deuteronomy 23:4–6 into its recounting (cf. the discussion to this chapter below). At the same time, the author of Nehemiah 13:1–3 harmonized the text of Deuteronomy, removing the mention of the ten generations. For Nehemiah, the focus is on the later version of the text including Balaam, excluding Ammonites and Moabites unequivocally and indefinitely. One other reception of Balaam from Deuteronomy 23 merits mention. In the Samaritan version of Numbers 22:5 (as well as in the Vulgate and the Peshitta), Balaam comes from “the land of the sons of Ammon” (‫)ארץ בני־אמון‬. This identification presumably resulted from the influence of Deuteronomy 23, which implies that the Ammonites must have engaged in some transgression in the matter of Balaam and his seeking to curse Israel. The most obvious way to account for that was the simple appending of a ‫ נ‬to the curious phrase “land of the sons of his people” (‫)ארץ בני־אמו‬. Thus, Deuteronomy 23 transformed the

228

The Bible’s Balaam beyond Numbers 22–24

Moabite sin of Joshua 24 into a Moabite and Ammonite sin through the insertion of Balaam into this context. Ultimately, this addition in Deuteronomy led to the Samaritan (and subsequent traditions’) metamorphosis of Balaam into an Ammonite. Deuteronomy 23’s reference to Balaam presents a later reinterpretation of the ‫ קהל‬legislation after its insertion into an exodus or Pentateuch context. Here, he demonstrates the low point of Moabite and Ammonite depravity, leading to their eternal exclusion from the assembly. Thus, in Deuteronomy 23 1) Balaam probably presents a later interpolation, and 2) this interpolation about Balaam in Deuteronomy remains generally distinct from Balaam in Numbers. For example, whereas Deuteronomy regarded him as a Mesopotamian, Numbers regarded him as an Aramean. Nevertheless, Deuteronomy borrows from the Joshua tradition about Balaam, itself a rereading of the Numbers version. By including Balaam, later scribes sought to emend Deuteronomy such that it eternally precluded Ammonite and Moabite admission to the assembly instead of allowing it after some ten generations as in a previous edition. This transformation appears to have somewhat ironically restored the earliest eternal prohibition against Moabite and Ammonite participation. Finally, this adapted text of Deuteronomy served as the basis and model for the proscription in Nehemiah 13 and the Samaritan, Syriac, and Latin identification of Balaam as an Ammonite. In these ways, the image of Balaam in Deuteronomy 23 stands in the crux of the literary and reception histories of the biblical versions of this figure. In conclusion, really the only elements common to the stories of Balaam in Numbers and Deuteronomy are the name (Balaam ben Beor) and the motif – but not the semantics – of cursing. Everything else in the stories varies. The Moabites are with the Ammonites in Deuteronomy, but – at most – with the Midianites in Numbers 22–24. These groups both hire Balaam together in Deuteronomy, whereas Balak (who remains conspicuously absent in Deuteronomy) orders Balaam to come because of his people’s fear. The name of the group to be cursed varies in Deuteronomy from those in Numbers, again in spite of the plethora of names available for the scribe to copy from Numbers 22–24. Context demands that the group referenced in Deuteronomy 23 be identified as the ‫קהל יהוה‬, a term most similar to one of later redactions identified thus far in Numbers (and in the Pentateuch, more generally).70 Numbers never describes Balaam as desiring to curse the people, whereas Deuteronomy requires this as the background for the entire anecdote. The motifs of willingness and love – both present in Deuteronomy 23 – are absent in Numbers 22–24. The genres of 70 For this reason, one can agree to some degree with Otto’s dating this text after Joshua 24 and after the post-priestly Hexateuch, though one need not identify this with a Pentateuch composition; cf. Otto, Deuteronomium 12,1–23,15, 1747. Important for the position here is the dating of the text after both Numbers 22–24 and Joshua 24.

4.4 Joshua 13:21–22

229

the tales are absolutely distinct, Numbers 22–24 recounting something like a Heilsgeschichte, whereas Deuteronomy 23 embeds this motif firmly within the confines of jurisprudence in leges sacrae. Deuteronomy thus apparently lacks direct knowledge of some of the most important material in Numbers 22–24, while adding other characteristics to the figure Balaam. In Deuteronomy, Balaam served two other foreign groups against the assembly during the exodus. He sought to curse them, and apparently may have been successful if God hadn’t intervened. In Numbers 22–24, he apparently was successful in blessing Israel, at least based on Balak’s reactions (Num 23:11, 25; and 24:10). Balaam undertook this endeavor for financial gain in Deuteronomy 23. Numbers 22– 24 lacks these aspects of his character. Ultimately, any direct, unfiltered connection between the two traditions is at best tenuous and at worst impossible, at least beyond a common tradition-historical background around the figure Balaam. Essentially no evidence commends reconstructing a common literary heritage behind both, nor that one story relied exclusively on the other.

4.4 Joshua 13:21–22 4.4 Joshua 13:21–22

The tradition about Balaam in Josh 13:21–22 remains intimately and inextricably intertwined with that in Num 31:8 (see above). Yet even their striking similarity makes certain distinctions apparent.71 The first conspicuous distinction is the context of the repeated data: whereas in Num 31:8 the list of conquered kings comes in the context of the Israelites’ “revenge” against Midian, Josh 13:21–22 reports the demise of the Midianite leaders within a report about the division of Transjordan. The distinct nomenclature for the rulers in the previous sentence was not accidental; while Num 31:8 lists Midianite “kings” ( ‫מלכי‬ ‫)מדין‬, Josh 13:21 references “Midianite princes” (‫ )נשׂיאי מדין‬and “Sihon’s leaders” (‫)נסיכי סיחון‬.72 The record in Josh 13:21–22 combines the killing of the Midianite rulers with Sihon’s downfall; that is, the death of the Midianite princes/rulers occurred at the time of the Amorites’ destruction; cf. Num 21:21–32* + 31:8. That means that the author either did not know or did not care to report the information about Moab and Midian found in Numbers 22– 25. The notice about Balaam remains distinctly syntactically appended to the Joshua text and contains the extra datum that he was “the diviner” (‫)הקוסם‬. An

71 Cf., e.g., J. Alberto Soggin, Joshua. A Commentary, OTL (London: SCM Press Ltd., 1972), 158 who recognized the obvious harmonizations between this text and Numbers 22 and 31–32 (which in turn had been Deuteronomistically edited) without citing the specific textual interactions. 72 To the text-critical issues of this phrase, cf. Leonard J. Greenspoon, Textual Studies in the Book of Joshua, HSM, vol. 28 (Chico, CA: Scholars Press, 1983), 138–40.

230

The Bible’s Balaam beyond Numbers 22–24

important similarity between the two notices in Num 31:8 and Josh 13:21–22 are the text-critical difficulties. Num 31:8 G: καὶ τοὺς βασιλεῖς Μαδιαν ἀπέκτειναν ἅμα τοῖς τραυματίαις αὐτῶν καὶ τὸν Ευιν καὶ τὸν Σουρ καὶ τὸν Ροκομ καὶ τὸν Ουρ καὶ τὸν Ροβοκ πέντε βασιλεῖς Μαδιαν καὶ τὸν Βαλααμ υἱὸν Βεωρ ἀπέκτειναν ἐν ῥομφαίᾳ σὺν τοῖς τραυματίαις αὐτῶν

Num 31:8 M: ‫ואת־מלכי מדין הרגו‬ ‫על־חלליהם את־אוי‬ ‫ואת־רקם ואת־צור‬ ‫ואת־חור ואת־רבע‬ ‫חמשׁת מלכי מדין‬ ‫ואת בלעם בן־בעור‬ ‫הרגו בחרב‬

Josh 13:21b–22 M: ‫ואת־נשׂיאי מדין את־‬ ‫אוי ואת־רקם ואת־צור‬ ‫ואת־חור ואת־רבע‬ ‫נסיכי סיחון ישׁבי הארץ‬ ‫ואת־בלעם בן־בעור‬ ‫הקוסם הרגו בני־‬ ‫ישׂראל בחרב אל־‬ ‫חלליהם‬

Jos 13:21b–22 G: καὶ τοὺς ἡγουμένους Μαδιαμ καὶ τὸν Ευι καὶ τὸν Ροκομ καὶ τὸν Σουρ καὶ τὸν Ουρ καὶ τὸν Ροβε ἄρχοντας παρὰ Σηων καὶ τοὺς κατοικοῦντας τὴν γῆν καὶ τὸν Βαλααμ τὸν τοῦ Βεωρ τὸν μάντιν ἀπέκτειναν ἐν τῇ ῥοπῇ

Table 17: Numbers 31:8 // Joshua 13:21b–22

In general, one notes that G proffers a more succinct reading than M. This brevity can be readily seen in the elements absent from G but present in M in Josh 13:21–22: ‫ אשׁר מלך בחשׁבון‬in v. 21 and ‫ בני ישׂראל‬and ‫ אל־חלליהם‬in v. 22.73 Additionally, and not noted in BHS, G apparently attests ἐν τῇ ῥοπῇ for M’s ‫בחרב‬, which is hardly a satisfactory translation. This last problem presumably presents an error in the transmission of the Greek text: killing Balaam with “a balance” or “scales” makes no sense.74 Margolis suggested that it should read τροπῇ instead of ῥοπῇ,75 which has a certain logic to it, but then causes the problem that the element ‫ בחרב‬would no longer be reflected in G. Another possibility would be a careless error or lost element of text in the Greek tradition. Perhaps ἐν τῇ ῥοπῇ exists as the remaining rump of ἐν (τῇ) ῥομφαίᾳ (ΡΟΠΗ from ΡΟΜΦ) as found in Num 31:8. At some point the text could have become damaged so that ῥομφαίᾳ was truncated and no longer recognizable.76 Perhaps 73 The prepositional phrase παρὰ Σηων presumably represents an attempt to clarify a Vorlage identical to M; cf. Cornelius G. den Hertog, “Studien zu griechischen Übersetzung des Buches Josua” (Giessen: Universität Giessen, 1996), 169 and Rösel and Schlund, “Arithmoi,” 640. 74 The term ῥοπή occurs rarely in G as a translation of a known Hebrew text. In Isa 40:15 and Prov 16:11 it serves to translate ‫מאזנים‬. Since no common error could have changed ‫בחרב‬ into ‫ מאזנים‬or vice versa, the best solution would be to search for a problem solely within the Greek tradition. 75 Cf. Max L. Margolis, “Τετροπωμένους Joshua 11:6,” JBL 33 (1914): 286–89 and Rösel and Schlund, “Arithmoi,” 636 and 640. 76 This is contra Greenspoon, Textual Studies, 308, who regarded the inclusion of this word behind the asterisk in the Hexapla as evidence of its stemming from Theodotion. I rather interpret it as evidence that it was OG.

4.4 Joshua 13:21–22

231

even some version of σὺν τοῖς τραυματίαις αὐτῶν went missing in this process as well. The following observations favor this theory over Margolis’: the preposition ἐν better matches ‫ ב‬than ‫אל‬, ‫על‬, or ‫( את‬the three possible readings that could be reconstructed before the term ‫ חלליהם‬in the Vorlage); the affixed object suffix (‫ )הם‬remains absent in Margolis’s reconstruction; and τροπή is never used as a translation of ‫ חלל‬in the Bible. The more likely explanation is that the Greek text was damaged during transmission, causing the end of the verse to go missing. Scribes made the best of the situation they were able. Most importantly, in the last distinction between M and G noted above, no apparent reason commends a Vorlage distinct from (proto-)M. That is not the case for the other missing elements, which can be explained as clarifying glosses in M, but cannot sensibly be attributed to intentional or unintentional deletion in G. At some later stage in transmission, a scribe added ‫ אשׁר מלך בחשׁבון‬in v. 21 and ‫ בני ישׂראל‬in v. 22 to better harmonize the text with its macro-context and remind the reader that Sihon ruled in Heshbon, as recounted in, e.g., Numbers 21 and – more importantly – Josh 13:27. Since there is hardly any reason someone would – intentionally or otherwise – remove this information from G, the best explanation is that it was added to (proto-)M at some point after the translation of G. The Septuagint’s shorter readings, except in the last case, should be favored over M’s longer readings. The G Vorlage must have read something like ‫וכל ערי המישׁר וכל־ממלכות סיחון‬ ‫מלך האמרי אשׁר הכה משׁה אתו ואת־נשׂיאי מדין את־אוי ואת־רקם ואת־צור ואת־חור‬ (?‫ואת־רבע נסיכי סיחון ישׁבי הארץ ואת־בלעם בן־בעור הקוסם הרגו בחרב )אל־חלליהם‬. This older version must be compared to the version in Numbers so that distinctions between these two texts can be appreciated. On the surface, Balaam’s appearance in Josh 13:21–22 matches that found in Numbers 31:8. Nonetheless, closer inspection reveals significant differences between the two. The contextual distinction presents the most glaring differences, but the terminological distinction cannot be overlooked either. Whereas Numbers 31 includes this list in a somewhat appropriate context – within a narrative about the Israelites destroying the Midianites – the list in Joshua 13 subsumes the list of Midianite rulers within Sihon’s domain. 77 At the same time, the list in Joshua 13 suggests that the rulers were not kings, but only princes (‫ )נשׂיאים‬and subordinates (‫ )נסיכים‬of Sihon. Taking a look at the greater context suggests that the list of princes with the notice about Balaam is secondary to its context as well. At no other point in the distribution of Transjordan among the tribes can such a list of rulers be found. True, the lists mention 77

For this reason, Richard D. Nelson, Joshua. A Commentary, OTL (Louisville: Westminster John Knox Press, 1997), 173 suggested this notice in Josh 13:21b–22 “…amalgamates Sihon with material about the five kings of Midian and Balaam taken up from Num. 31:8.” This observation suggests, in my opinion, that these verses present a later interpolation into the text of Joshua.

232

The Bible’s Balaam beyond Numbers 22–24

Sihon (13:27) and Og (13:30–31) a few times, but never with the kind of explication and enumeration of subordinates found in Josh 13:21–22. This makes the list more suspicious in its larger as well as in its immediate context. In its immediate surroundings the list of princes and the inclusion of Balaam interrupts the narrative context about Reuben’s inheritance. Joshua 13:23 presumably followed v. 21* (concluding with the phrase ‫אשׁר הכה משׁה אתו‬. At a later stage of composition, the list of rulers was added and, perhaps at an even later date, the datum about Balaam’s demise was incorporated. Nevertheless, since the list in Num 31:8 appears to have been added at one time, probability commends regarding the addition here in Josh 13:21b–22 as having been inserted at one time. Finally, the historical claims made in Joshua 13:21–22 distinguish it from its pendant in Numbers in two matters: first, Numbers does not include Sihon in the destruction of the Midianites in chapter 31 (or even in the incident at Peor in chapter 25); second, Numbers never considers Balaam “the diviner” (‫ )הקוסם‬and in fact Numbers makes no explicit claim about the specific socioreligious role that Balaam fulfilled.78 Ultimately, then, Joshua 13 and Numbers 31 transmit distinct data about Balaam and his demise. Therefore they should not too readily be included into a single literary layer. The insertion of these lists of Midianite rulers in Numbers 31 and Josh 13 may have occurred distinctly from each other. Theoretically the inclusion of Balaam could have followed at a later time, but simplicity and the placement of the repetitive ‫ חללים‬commend regarding Balaam’s attachment at the same time as the list (see the discussion to Num 31:8 above). In each case, the list of Midianite rulers fulfills a distinct purpose and Balaam fails to really match either of them, at least without the notice in Num 31:16. This observation also favors regarding the addition of Balaam simultaneously to the Midianite rulers and in Numbers before it was added in Joshua. At the same time, the notice that Balaam served as a ‫ קוסם‬in Josh 13:22 suggests a negative tendency in this case even beyond that found in Numbers 31.79 For example, the legal code in Deut 18:10–14 mandates that such practitioners be expunged from Israel. Obviously this notice should serve to increase the negative view of Balaam in the Bible. To this end, one should note that this datum appears exclusively in Joshua 13: without identifying Balaam as ‫הקוסם‬, the text proffers no other indication as to why Balaam should have died with the Midianite princes in Joshua 13. The portrait of Balaam painted in Joshua 13 remains distinct from that found in Numbers, whether in chapters 22–24 or in chapter 31. The distinction between Numbers 22–24 and Joshua 13 is more 78

Cf. Chapter Six about Balaam’s changing image in tradition history. Cf. Donner, “Balaam pseudopropheta,” 113–14 and Jacqueline Moatti-Fine, Jésus (Josué), La Bible d’Alexandrie, vol. 6 (Paris: Cerf, 1996), 170: “…la qualification de mántis suffit à déconsidérer le personnage qui subit le même sort que le rois de Madian”. For Schmidt, “Bileamüberlieferung,” 261, this term represented the nadir of Balaam’s negative evaluation. 79

4.5 Joshua 24:9–10

233

substantial: other than the common name Balaam ben Beor, the two texts have nothing in common. The same could be said for Joshua 13 and Deuteronomy 23: only the character’s name matches. Yet, even the distinction between Numbers 31 and Joshua 13, while nuanced, is hardly irrelevant. Numbers 31 and Joshua 13 both place Balaam’s demise among the same people, though they are titled differently. On the other hand, only Joshua 13 includes Balaam’s profession – a profession the Bible holds in disregard.80 The addition or, at the very least, inclusion of this profession serves to negatively evaluate Balaam in Joshua 13 even more than the negative evaluation implied by his having been killed with the Midianite rulers. Taken together, this could imply that this reference to Balaam in Joshua 13 may well be the youngest in the HB.

4.5 Joshua 24:9–10 4.5 Joshua 24:9–10

The relationship between Josh 24:9–10 and Deut 23:4–6 appears similar to the relationship between Josh 13:21–22 and Num 31:8: some dependency is clear, but the details remain somewhat murky. In this case, the connections are even more diffuse than those between Numbers 31 and Joshua 13. A detailed comparison and examination – naturally including relevant text-critical issues – is both necessary and desirable in order to consider the relationship of these traditions both to each other and to the other material about Balaam in the Bible. The discussion will open with text-critical matters, as they permit the most secure reconstruction, based on data we possess and not merely on theories we develop. After the text-critical issues have been covered, the literary analysis will follow.

80

Cf., e.g., Deut 18:10, 14; 1 Sam 15:23; 2 Kgs 17:17; Isa 44:25; and Jer 14:14.

234

The Bible’s Balaam beyond Numbers 22–24

Josh 24:9–10 M: ‫ויקם בלק בן־צפור מלך מואב וילחם בישׂראל‬ ‫וישׁלח ויקרא לבלעם בן־בעור לקלל אתכם ולא‬ ‫אביתי לשׁמע לבלעם ויברך ברוך אתכם ואצל‬ ‫אתכם מידו‬

Josh 24:9–10 G: καὶ ἀνέστη Βαλακ ὁ τοῦ Σεπφωρ βασιλεὺς Μωαβ καὶ παρετάξατο τῷ Ισραηλ καὶ ἀποστείλας ἐκάλεσεν τὸν Βαλααμ ἀράσασθαι ὑμῖν καὶ οὐκ ἠθέλησεν κύριος ὁ θεός σου ἀπολέσαι σε καὶ εὐλογίαν εὐλόγησεν ὑμᾶς καὶ ἐξείλατο ὑμᾶς ἐκ χειρῶν αὐτῶν καὶ παρέδωκεν αὐτούς

Table 18: Joshua 24:9–10 in M and G

The following distinctions between G and M in Josh 24:9–10 are readily apparent: 1) Balaam’s patronymic is missing from v. 9 in G; 2) M recounts the first verb in v. 10 in the third instead of the first person, a common variance throughout this chapter; 3) the verb “listen”, reflecting Hebrew ‫לשׁמע‬, is absent in G; 4) the name Balaam is missing in v. 10 G; 5) G includes the verb “to destroy” (ἀπόλλυμι) with a direct object, M apparently lacking both; 6) M uses the first person of the verb for saving ‫נצל‬, whereas G reflects this verb in the third person; 7) G reflects a plural enclitic particle appended to “hand” not attested in M, which reads in the singular; and 8) G continues beyond M’s reading. Each of these matters will be dealt with in turn. The Septuagint’s shorter reading in v. 9 may reflect the older reading. Hardly any reason can be found that this patronymic should have been removed, whether purposely or otherwise. Rather, M appears to attest a clarifying, harmonizing gloss vis-à-vis G. Turning to the end of v. 10, the longer reading in G may reflect an older version of the text. The verb παραδίδωμι included here reflects the Hebrew ‫נתן‬ in every other case in Joshua,81 making it likely that it can be reconstructed here. The best possibility for reconstructing the Vorlage would be ‫( ויתן אותם‬cf. Josh 24:11) or ‫( ויתנם‬cf. Josh 11:8). That being said, the reading in G also appears corrupt or at least incomplete. Who gave “them” to whom? The circumstances are by no means transparent. This opaqueness could either reflect an error in the translation or an oversight already in its Vorlage. Based on the surrounding context, the most likely replacement for or solution to the missing element in G would be εἰς τὰς χεῖρας ὑμῶν (24:8 and 11), a reflection of ‫בידכם‬. This element must have gone missing over the course of transmission, leaving the truncated reading remaining in G. At some later point in transmission, even this truncated reading must have been removed, resulting

81

Some eighteen times. The only possible exceptions to this could be: 1) in Josh 10:35, which at first appears as a translation of ‫לכד‬, but is in fact a translation of ‫נתן‬, whereby M attests a longer reading than G and thus G has no element reflecting ‫ ;לכד‬and 2) in Josh 24:33b, which has no parallel in M, however, would reflect the typical usage of ‫ נתן‬otherwise attested in Joshua.

4.5 Joshua 24:9–10

235

in the text we find in M. This removal – probably intentional – must have occurred before the third century CE, as Origen set καὶ παρέδωκεν αὐτούς behind the obelus. With the longer reading restored, the error in transmission responsible for or the intention behind M’s shorter reading can be readily recognized. The plural form ‫מידם‬, which presumably stood behind G, either became ‫ מידו‬as the result of a scribal error (deletion of ‫ מ‬and dittography of ‫ ו‬based on the following ‫ ו‬in 24:11) or someone intentionally changed the text to concentrate on Balaam. In M, God rescued the people from Balaam, whereas in G, God rescued the people from “them”, consisting of at least Balaam and Balak and probably Balak’s implied army, as well. At the same time, the reconstructed reading ‫ מידם‬increases the likelihood that ‫ בידכם‬at the end of the verse was omitted over the course transmission due to haplography. Therefore, it is possible – though admittedly a conjuncture – to reconstruct an older version that concluded the text now found in Josh 24:10 with ‫ויצל אתכם מידם ויתן אותם‬ ‫בידכם‬: “and he saved you from their hand and gave them into your hand”. Having considered both v. 9 and the end of v. 10, we can now focus on the other issues in v. 10, listed above as numbers 2–6. Two of the verbs in v. 10 reflect different subjects in M and G: whereas M presents Joshua 24 generally as a speech from YHWH, G recounts its as Joshua’s lecture. The obvious resolution to this distinction is that M changed Joshua’s speech into YHWH’s. The alternative appears unlikely; who would take statement from God and place them in a mortal’s mouth when nothing in the speech contradicts it coming from the deity? The Masoretic text perhaps also allows the original third-person character of the text shimmer through on a few occasions, most conspicuously in the use of ‫ אל־יהוה‬in v. 7, but probably also in v. 10. In M, v. 10 presents Balaam as the agent blessing Israel, whereas G describes YHWH fulfilling this role.82 The Masoretic text casts Balaam as the one doing the blessing by removing the subject and making the speech in the first-person in general, but leaving the verb ‫ ויברך‬in the third person. This transformation may have been accidental, but I would argue that it may have been intentional.83 By making YHWH deliver the speech in general, but leaving Balaam as the agent blessing Israel, M demonstrates YHWH’s absolute power over this character. The Septuagint leaves him impotent and sidelines him syntactically, whereas M pushes him into the action and objectifies him literally: YHWH’s desire in M has overwhelmed him. Simultaneously, this conforms M with the image of Balaam blessing in Num 23:20 in the older version.84 The following verb, ‫נצל‬, again has been put into the first person so that YHWH, and not Balaam, saved the people. At that YHWH saves them in M “from his (!) hand”. This salvation again demonstrates an increased focus on Balaam 82

N.b. the precisely reversed relationship between M and G in Num 23:20. Cf. Greenspoon, Textual Studies, 155–56. 84 Cf. David Frankel, “The Deuteronomic Protrayal of Balaam,” 35. 83

236

The Bible’s Balaam beyond Numbers 22–24

in M vis-à-vis G. In general then, it remains likely that M changed the speech from the third to the first person and increased YHWH’s power over Balaam at the same time. The Septuagint’s third-person readings attest the older version. This reconstruction further mandates the reconstruction of the subject in G. Thus, the first phrase in the Josh 24:10 G-Vorlage must have read something like ‫ולא אבה יהוה אלהיך‬.85 The following lexemes present the hardest problems to solve text-critically. Leaving behind the discussion about the first and third person verbs and G’s inclusion of the subject, we can focus on the missing name and verb for hearing in G. These matters may be related. Two solutions are plausible. The first possibility: the reading behind G did not reflect the lexeme ‫בלעם‬ or anything similar to M at all. Since ‫ בלע‬was only once translated as here with ἀπόλλυμι in G (Job 2:3), it might be possible to identify an alternative reading in the presumed Vorlage. The Septuagint sometimes relies on the ἀπόλλυμι to reflect the root ‫ שׁמד‬in either the Hiphil or the Niphal, twelve times in total.86 Thus, ‫ שׁמד‬presents a plausible alternative for the reading of the G-Vorlage. Both syntactically and in terms of length, the Hiphil infinitive construct with an appended ‫ ל‬and a singular second-person suffix seems the most likely choice: ‫להשׁמידך‬. This reconstruction would demand that, in the later transmission of the text, the final ‫ כ‬was lost due to its similarity with ‫ד‬, and perhaps the combination ‫ יד‬was confused with ‫ע‬. This in turn would have caused the ‫ ה‬to be deleted, most likely intentionally. That would mean, in turn, that Josh 24:10 was not initially interested in whether YHWH listened to Balaam. Rather, it simply attested YHWH’s having no desire to destroy the people. However, this postulation presents problems of its own. First, one immediately recognizes the manifold errors required in the text’s transmission for this reconstruction to be correct. That certainly questions the viability of this reconstruction, particularly whether it could present the simple explanation. Second, the form ‫ להשׁמידך‬never appears in the Bible, though one does find ‫ להשׁמיד‬with a direct object marked with ‫ את‬on a few occasions.87 That would suggest that the Vorlage might have read ‫להשׁמיד אתך‬, but that hardly fits the available space. Third, and perhaps weightier, ‫ שׁמד‬appears in Josh 24:8, but G translates it there as ἐξωλεθρεύσατε. How likely is it that the translator would have varied this widely in such a narrow context? Further, the G of Joshua never otherwise translates ‫ שׁמד‬with ἀπόλλυμι. And finally, the Masorah

85

Cf. Greenspoon, Textual Studies, 299. Takamitsu Muraoka, A Greek-Hebrew / Aramaic Two-Way Index to the Septuagint (Louvain; Paris; Walpole, MA: Peeters, 2010), 15 identified four times in which ἀπόλλυμι reflects √‫ שׁמד‬in the Niphal and seven times in the Hiphil. While the number four is correct for the Niphal (Deut 28:24, 45; Isa 48:19; and Jer 48:42), there are eight attestations which reflect the Hiphil (Deut 2:12, 21; 33:27; Est 4:8; 7:4; Isa 13:9; 14:23; and 26:14). 87 Cf. Deut 9:8, 19, 25; 2 Sam 14:16; Est 3:6; and Zech 12:9. 86

4.5 Joshua 24:9–10

237

parva identifies the phrase ‫ לשׁמע לבלעם‬as hapax legomenon, probably implicitly recognizing the distinction from Deuteronomy’s similar phrase ‫לשׁמע אל־‬ ‫בלעם‬. This similarity, at the same time, obliquely suggests that Josh 24:10 may have been emended to harmonize better with Deut 23:6.88 That would imply that the G Vorlage may have been distinct from what we now find in M, but not as distinct as this first proposed reconstruction. The second, better alternative: it is possible that some misunderstanding caused G to misinterpret the name ‫ בלעם‬as coming from the verb ‫בלע‬. Though rarely, G does attest the translation of ‫ בלע‬with ἀπόλλυμι; cf. Job 2:3 M and G. This observation at least permits such an interpretation, implying the form must have been something like ‫לבלעך‬.89 The terminal ‫ מ‬in ‫ לבלעם‬as in M would present a problem however, since it is not readily confused with ‫כ‬, the reading that would presumably have stood behind G’s σε. Another solution is available, however. The possibility cannot be entirely excluded that instead of the singular, it read the plural ‫כם‬. The Vetus Latina (Codex Lugdenensis) attests precisely this reading (vos) and lacks a translation of κύριος. These readings could imply that κύριος presents a later harmonizing interpolation with Deut 23:6 and that the Old Greek lacked it (though this is neither certain nor necessary) and read the object in the plural. But how did the object then become singular? The answer is obvious: it was converted to reflect the singular to match the ending of ‫אלהיך‬, whether intentionally or accidentally. The subsequent words ending in ‫ך‬- may have abetted this transition; n.b. the following ‫ ויברך ברוך‬as in M and implied by G and even the Vetus Latina (La). This reconstruction would thus explain the ‫ מ‬at the end of the root ‫בלע‬. An older version of the text presumably read ‫( לבלעכם‬Qal or Piʿel infinitive construct with a second masculine singular object suffix and an appended ‫)ל‬.90 It probably served as a pun in its context, a pun not available to Greek, though attested there.91 This version was corrupted 88 Others have recognized the relationship between Deuteronomy 23 and Joshua 24, particularly in the M version; cf., e.g. Rofé, Book of Balaam, 46–48 and David Frankel, “The Deuteronomic Protrayal of Balaam,” 33–34. However, they tend to disregard the text-historical aspect of the similarities. 89 Cf. Nelson, Joshua, 264, who notes, “OG (‘to destroy you’) translates Balaam’s name as an infinitive from the root blʿ.” 90 This is contra Cornelius den Hertog, “Jesus / Josue/ Das Buch Josua,” in Septuaginta Deutsch: Erläuterungen und Kommentare Band I. Genesis bis Makkabäer, ed. Martin Karrer and Wolfgang Kraus (Stuttgart: Deutsche Bibelgesellschaft, 2011), 654 and Monique Alexandre, Jésus (Josué), La Bible d’Alexandrie, vol. 6 (Paris: Cerf, 1996), 233. Cf. the note in Max L. Margolis, Joshua 19:39–24:33, vol. 5 of The Book of Joshua in Greek. According to the Critically Restored Text with an Apparatus Containing the Variants of the Principal Recensions and of the Individual Witnesses., preface by Emanuel Tov (Philadelphia: Annenberg Research Institute, 1992), 460. 91 For this reason, the position of LXX.E should be rejected; cf. Rösel and Schlund, “Arithmoi,” 654. It may not merely have been the case that wordplay stood behind ‫ בלע‬and ‫( בלעם‬as LXX.E suggests); it is highly likely. How plausible is it that a Greek translator would invent

238

The Bible’s Balaam beyond Numbers 22–24

to the reading we now find in G. The ‫ לשׁמע‬was added to harmonize with Deut 23:5–6, which explains its absence in Josh 24:10 G.92 This addition and the other errors and corrections led to the text we now find in M. With the evidence and arguments considered thus far, I propose and older stage of the text of Josh 24:9–10, even older than that which stood behind G, and read: ‫ויקם בלק בן צפור מלך מואב וילחם בישׂראל וישׁלח ויקרא לבלעם לקלל‬ ‫אתכם ולא אבה יהוה )אלהיך( לבלעכם ויברך ברוך אתכם ויצל אתכם מידם ויתנם‬ ‫בידכם‬. The ‫ אלהיך‬probably presents a later addition, apparent from its inconsistent use of the singular and its absence in La. It entered the text as a gloss from the repetitive usage of the term ‫ אלהיך‬after ‫ יהוה‬in Deuteronomistic literature.93 Over the course of transmission the verse’s ending lost some elements, ‫ מ‬as attested by G and ‫ כ‬as attested by M – in that order. After the translation into Greek, someone in the (proto-)M tradition changed the speech in Joshua into the first person. This must have occurred after the cross-contamination of Joshua 24 with Deuteronomy 23 (i.e., the insertion of ‫)לשׁמע‬. The transition from the third person to the first led to the removal of the subject from the opening clause of Josh 24:10. By this time, the text had been converted from ‫ לבלעכם‬to ‫לבלעם‬. After the translation into Greek someone also added Balaam’s patronymic in v. 9. Thus, from a longer text more or less attested – or at least implied by – the Vorlage of G we can trace the development of the text until it became more like M.94 Having clarified the text-historical issues as much as possible, a comparison between the oldest version of Joshua 24 and the oldest versions of the other Balaam texts commends itself. Even such a short text as Josh 24:9–10 distinguishes itself from the Numbers tale about Balaam in several of factors. For one thing, Joshua 24 provides no further information about Balaam than his first name, even missing his patronymic in the oldest version. That could indicate that this text originated in a narratological context subsequent to some other Balaam text. That would mean that Josh 24:9–10 stood from its inception in some narrative context after Numbers 22–24; 31; or Deuteronomy 23. The preceding discussion on Deuteronomy 23 precludes its viability, which in turn means that Josh 24:9–10 must have known one of the Numbers texts from its outset. The content of Josh 24:9–10, however, demonstrates that it must have been Numbers 22–24. this Hebrew pun that makes no sense in Greek? The Vorlage of G must have been different than M, and M was presumably later corrupted. 92 Cf. already Dillmann, Numeri – Deuteronomium und Josua, 595, who regarded the whole phrase ‫ לא אביתי לשׁמע לבלעם‬as an insertion from Deut 23:5 into Joshua. Contra Nelson, Joshua, 264. 93 Though admittedly only four other times in Joshua in M: Josh 1:9, 17; 9:9, and 24. 94 This reconstruction contradicts David Frankel, “The Deuteronomic Protrayal of Balaam,” 35–37, who regarded the metaphor of “wanting to destroy” as a late Deuteronomistic reevaluation of the situation. The evidence clearly speaks against his understanding.

4.5 Joshua 24:9–10

239

While an initial glance suggests that the basic structure of the events in Josh 24:9–10 match Numbers 22–24 (the introduction of Balak, his summoning Balaam, and the desire for Balaam to curse), many details remain quite distinct from or even contradict others. Joshua 24:9, for a start, recounts that Balak went out to battle against Israel before summoning Balaam. This statement contradicts Num 22:11, which states that Balak has not yet gone out to battle against Israel before summoning Balaam.95 Perhaps this conception in Joshua presents an interpretive development in its version vis-à-vis Numbers 22–24. That is, the author of Josh 24:9–10 extrapolated an implicit narrative scenario from Numbers 22–24 and made it explicit. Nevertheless, the precise repetition of ‫ שׁלח‬and ‫ קרא‬in Num 22:5 and Josh 24:9b commends recognizing Joshua 24 as an adaptation of the version in Numbers 22–24*. As in Deuteronomy 23, the verb for cursing in Josh 24:9–10 does not match any of the verbs for cursing in Numbers 22–24 (it does however match that used in Deuteronomy 23). Joshua 24:10 implies that Balaam must have tried to curse Israel, a claim not made anywhere in Numbers 22–24. Instead of Balaam cursing, YHWH blessed them, which stands in tension with the older version of Num 23:20, which states that Balaam blessed them. After YHWH’s blessing, or in combination with it, YHWH rescued Israel from the hand of Balaam and Balak (and the Moabites?). Numbers 22–24 lacks the root ‫נצל‬. Finally, the oldest version – apparently even older than G – implies (should our reconstruction be correct) that YHWH delivered Balak and Balaam into Israel’s hand. Numbers 22–24 makes no such claim (though Numbers 31 does state that Balaam – but not Balak – was killed by the Israelites). The phrase “to give into the hand” (‫ נתן‬+ ‫ )יד‬does not appear a single time in Numbers 22–24. The distinctions between the versions in Numbers 22–24 and Joshua 24 are thus manifold.96 The tendency shared between Josh 24:9–10 and Numbers 31 (and Josh 13:21–22, for that matter) is the negativity toward Balaam (and also Moab in Joshua 24).97 Whereas Josh 24:10 in the version reconstructed above could imply that Israel killed Balaam, Numbers 31 and Joshua 13 make his death at their hands explicit. Otherwise, the only possible similarity is the name Balaam. However, whereas Joshua 13 and Numbers 31 mention his patronymic, Joshua 95 Indeed the idea that Moab and Israel engaged in conflict during the eisodus is unique to this text in Joshua and contradicts the traditions in Numbers and Judges; cf. Hans Wilhelm Herzberg, Die Bücher Josua, Richter, Ruth übersetzt und erklärt, ATD, vol. 9 (Göttingen: Vandenhoeck & Ruprecht, 1965), 135. 96 This assessment, of course, contradicts, e.g., Soggin, Joshua, 234, who noted that “…the tradition concerning Balaam is wholly in agreement, except for its brevity, with that of Num. 22–24.” 97 Levin, Jahwist, 393 described Josh 24:9–10 thus: “Ein anekdotisches Interesse an der Gestalt des Bileam verbindet sich mit antimoabitischer Propaganda.” Schmidt, “Bileamüberlieferung,” 260 saw a connection between Josh 24:9–10 and Num 31:16, though not an explicit one: “Damit bildet Jos 24,9f eine Zwischenstufe zu der ausschließlich negativen Wertung Bileams.”

240

The Bible’s Balaam beyond Numbers 22–24

24 in the oldest version does not. Again, the traditions here seem disparate and generally fail to transmit common data. Deut 23:4–6 (Reconstructed): ‫לא־יבא עמוני ומואבי בקהל יהוה גם דור עשׂירי‬ ‫לא־יבא להם בקהל יהוה עד־עולם על־דבר‬ ‫אשׁר לא־קדמו אתכם בלחם ובמים בדרך‬ ‫בצאתכם ממצרים ואשׁר שׂכר עליך את־בלעם‬ ‫בן־בעור מארם נהרים לקללך ולא־אבה יהוה‬ ‫אלהיך לשׁמע אל־בלעם ויהפך יהוה אלהיך את־‬ ‫הקללה לברכה כי אהבך יהוה אלהיך‬

Josh 24:9–10 (Reconstructed): ‫ויקם בלק בן צפור מלך מואב וילחם בישׂראל‬ ‫וישׁלח ויקרא לבלעם לקלל אתכם ולא אבה‬ ‫יהוה )אלהיך( לבלעכם ויברך ברוך אתכם ויצל‬ ‫אתכם מידם ויתנם בידכם‬

Table 19: Deuteronomy 23:4–6 // Joshua 24:9–10

When compared to Deuteronomy 23:4–6, the similarities between the traditions become more apparent, though this may have been the result of literary development (see below in the discussion about Nehemiah 13). The reason to make such a suggestion becomes relatively conspicuous: in an older version of Josh 24:10, the first half of the verse matched Deut 23:6 down to the letter in several words. This striking similarity strongly indicates editorial copying. With the exception of these few words, however, little similarity exists between situations and characterizations in Deuteronomy 23 and Joshua 24. Deuteronomy 23 fails to mention Balak, but does mention the Ammonites, who in turn play absolutely no role in Joshua 24:9–10. Whereas Deuteronomy 23 reports that YHWH “overturned” (‫ )הפך‬Balaam’s curse and transformed it into a blessing, Joshua 24 asserts no such thing. Rather, Joshua 24 reports that YHWH did not want to destroy Israel and thus blessed and rescued them. That is, Deuteronomy 23 appears to advance beyond an idea first attested in Joshua. The rescuing motif, however, does not play a role in Deut 23:4–6. At the same time, the motif of love found in Deuteronomy 23 plays no role in Josh 24:9–10, again demonstrating that Deuteronomy must have succeeded and superceded Joshua 24 diachronically. The similarities that exist between Deut 23:4–6 and Josh 24:9–10, the name Balaam, the curse (‫)קלל‬, blessing (‫)ברך‬, and the phrase (‫ ולא אבה יהוה )אלהיך‬as attested in the reconstructed version of Joshua 24 commend regarding Deuteronomy as developing from Joshua in terms of its image of Balaam (see the discussion to Deuteronomy 23 above). Taken together, with the reflections on Numbers, it appears that the version of Balaam in Numbers 22–24 attests a development from Numbers 22–24 to Joshua 24. It was later developed again within Deuteronomy 23. These texts all evince a link, with increasingly negative characterizations of Balaam.98 The negative attitude toward Balaam in Joshua 24 shares more in common with Numbers 31 than it does with Numbers 22–24 in its oldest version. That being said, essentially 98

Contra Milgrom, Numbers, 470, who regarded Josh 24:9–10 G as favorable toward Balaam.

4.6 Judges 11:25

241

nothing beyond the name and the attitude towards Balaam connects Numbers 31 with Joshua 24.

4.6 Judges 11:25 4.6 Judges 11:25

In this case, and in this case alone, the Bible proffers a tradition about King Balak of Moab without mentioning Balaam. This verse, reported as part of Jephthah’s message to the Ammonites, exemplarily includes Balak in rhetorical questions for how to act vis-à-vis Israel. The scenario in Judges 11 seems to shuffle several groups of people together, but that may aid in appreciating how this text relates to the others discussed here. Fortunately, unlike most of the examples discussed in this survey, this text does not present substantial text-critical issues. Both the Masoretic and Greek textual traditions of this verse remained largely stable.99 Within its narrative context, this verse presents something of a non sequitur. While describing Israel’s journey through Amorite territory to the Ammonites, suddenly, Jephthah’s message includes a reference to Kemosh – a Moabite deity100 – before referencing Balak’s engagement with Israel during their exodus from Egypt (cf. Judg 11:16). The exact wording of the question runs: “and now, are you really better than Balak ben Zippor, king of Moab? Did he ever strive with Israel? Engage them in battle?” The context and pragmatic of the questions all suggest a single, simple answer: no. The relationships between this text and those discussed above becomes readily apparent. First and foremost, one cannot ignore Balaam’s absence. The motifs of blessing and cursing play absolutely no role in Judges 11, and, thus, Balaam would be superfluous and suspicious here. Judges 11 recounts the history of Israel’s journey and not some Heilsgeschichte reflecting on Israel’s blessedness nor a hagiography of Balaam. At the same time, the reference to Balak occurs essentially in a homiletic recounting of Israel’s wilderness journey. Conspicuous in this vein is the immediately preceding discussion about Sihon the Amorite, which in turn followed a tale about the king of Edom. While this recounting does spuriously mention a message to the king of Moab not reported in Numbers 20–21 (cf. Judg 11:17), it should not be overlooked that the general organization of material follows that of Numbers 20–24*: Edom, Sihon the

99 Cf. the brief discussion of the issues in George F. Moore, A Critical and Exegetical Commentary on Judges, reprint, 1895, ICC (Edinburgh: T & T Clark, 1989), 297–98. 100 Cf., e.g., the Moabite Stone and Num 21:29; 1 Kgs 11:7, 33; 2 Kgs 23:13; Jer 48:7, 13, and 46M.

242

The Bible’s Balaam beyond Numbers 22–24

Amorite, and Balak of Moab. Again conspicuous is the absence of Og of Bashan.101 Perhaps this could indicate that the Judges text here was familiar with a tradition that moved the narrative directly from Israel’s encounter with Sihon to their encounter with Balak. That would match the reconstruction of R1 presented here. Also like Numbers 22–24, but unlike Joshua 24, Judg 11:25 suggests that Balak had no military interaction with Israel.102 This element also matches R1 in Numbers 22–24. The lack of Moabite military intervention distinguishes this text most significantly from Josh 24:9–10. There, Joshua claims that Balak belligerently engaged Israel. The contradiction is conspicuous. No direct common traits exist between Deuteronomy 23 and Judges 11 at first glance, but both texts do perhaps attest a common tendency. Both Judges 11 and Deuteronomy 23 imply some connection between the Ammonites and the Moabites. Judges makes the claim more geographically, whereas Deuteronomy makes it legally. Could this combination of these peoples be an indication of a common view of these people groups? The possibility should not be prematurely excluded. Judges 11 seems familiar with a tradition about Balak more like that in Numbers 22–24, apparently already in the context of Numbers 20–21, than like that in Joshua 24.103 Deuteronomy 23 may evince a similar tendency to lump the Ammonites and Moabites together, as is more or less done in Judges 11. Judges considers them neighbors, but the connection in Deuteromony remains opaque, certainly not explicated in any fashion. Perhaps the combination of Ammonites and Moabites in this context, also with an oblique reference to the Balak story inspired the insertion of the figure Balaam into Deuteronomy at some point. At the same time, Judges 11 presumes that Israel and Moab did not engage each other militarily during the exodus from Egypt. Deuteronomy 23 might, reporting that it was Moab’s and Ammon’s failing to ingratiate themselves to Israel that precluded their admission into the assembly; the addition of Balaam into this context, however, commends regarding his engagement against Israel as heightening the conflict. That would in turn commend Deuteronomy as indeed implying that Balaam was summoned in some belligerent capacity. While Judges makes no reference to Ammon’s and Moab’s failure to provide bread and water to Israel in the desert, at least this motif does not contradict their not engaging each other in battle. The major factor distinguishing Judges 11 from the texts handled thus far is the complete absence of Balaam. However, since 101 For these reasons, George F. Moore, Judges, 295–96 suggested that the author of Judges is following E for this report. 102 Many have proffered this observation; cf., e.g., Walter Groß, Richter, HThKAT (Freiburg im Breisgau; Basel; Vienna: Herder, 2009), 593. 103 For this reason, I find the post-Deuteronomistic or even post-post-Deuteronomistic dating of Groß, Richter, 619–20 problematic. The Balak tradition of Judges does not know, or at least does not express familiarity with the Balak of Joshua 24 or the Moabites and Ammonites of Deuteronomy 23.

4.7 Micah 6:5

243

the text in Judges 11 has nothing to do with blessing or cursing, his absence does not present an insurmountable problem. Judges 11 focuses on Israel coming peacefully into the land and avoiding conflict with the peoples already there. Thus Balak, and not Balaam, serves as the interesting character for this discussion and Jephthah’s example. All things considered, Judges 11:25 and its reference to Balak remains difficult for our diachronic reconstruction here, but appears to fit into a niche between the composition of Numbers 22–24* and Joshua 24. It demonstrates familiarity with Numbers 22–24* in its secondary literary context (behind the tale of Sihon the Amorite, i.e., R1), but lacks, explicitly and poignantly, any reference to something like the battle implied between Israel and Moab as found in Joshua 24 (and then, later, presumably implied in Deuteronomy 23). It evinces essentially no conspicuous contact with Numbers 31.

4.7 Micah 6:5 4.7 Micah 6:5

Micah 6:5 presents the only text in the Latter Prophets to mention Balaam (and Balak). It stands within a context describing the exodus from Egypt. This tradition stands out in a number of manners as unique from all others dealt with here. First, and foremost, the reference to Balaam (and Balak) occurs within divine speech delivered as part of a prophetic oracle. This oracle (Mic 6:3–5) describes a portion of the people’s journey from Egypt to Gilgal, that is, from the exodus to the eisodos. The oracle refers to Egypt, the slavery from which they were redeemed, Moses, Aaron, and Miriam, Balak’s devising (√‫)יעץ‬, and Balaam’s answering him. Though this text’s form distinguishes it from the others discussed here, it does demonstrate some similarities with some other Balaam traditions, and its distinctions from the others could aid in describing its background. Unlike many of the texts discussed thus far, Mic 6:3–5 contains few textcritical problems. The most significant differences can be found in G, but they hardly demonstrate a Vorlage distinct from M. For example, G reads ἢ τί ἐλύπησά σε ἢ τί παρηνώχλησά σοι where M reads ‫הלאתיך‬. The resolution to this apparent problem is simple: G translated the verb of its Vorlage twice.104 Elsewhere in G’s Micah, one encounters this phenomenon as well (cf. Mic 4:10 and 5:3). Still three other differences between M and G must be explained, all of them in Mic 6:5. The last two are the easiest: instead of transliterating the name

104 Cf. Helmut Utzschneider, “Michaias / Micha,” in Septuaginta Deutsch: Erläuterungen und Kommentare Band II. Psalmen bis Daniel, ed. Martin Karrer and Wolfgang Kraus (Stuttgart: Deutsche Bibelgesellschaft, 2011), 2375.

244

The Bible’s Balaam beyond Numbers 22–24

‫שׁטים‬, G translated it with τῶν σχοίνων,105 and G translated M’s plural ‫צדקות‬ with a singular, which it most often did for this noun in the plural (cf. also Judg 5:11; 1 Sam 12:7; Isa 33:15; 45:24).106 Thus, one element remains for consideration: according to G – but not to M – Balak devises “against you” (κατὰ σοῦ), which most likely should be understood as an explanatory gloss inserted into the translator’s Vorlage or interpolated by the translator himself or herself. None of the differences between M and G mandate changing the M text in Mic 6:3–5. In these variants, the Twelve Prophets Scroll from Wadi Murabbaʿat (MurXII) unequivocally and unambiguously supports M in every case.107 In terms of content this text probably presents the least information about Balaam. He is the son of Beor and answered Balak, the king of Moab, who had devised some plan. The text mentions no further data about these characters. Presumably, the interaction between these two figures should have taken place during the exodus, as implied by Mic 6:4, but before the people entered the land, as suggested by v. 5b. The information surrounding the mentioning of Balaam and Balak contains especially interesting information, particularly from a narrative standpoint. That Moses, Aaron, and Miriam are mentioned as being “sent before” (‫ )ואשׁלח מפניך‬the people while they left Egypt and that the people moved “from Shittim to Gilgal” after Balak and Balaam’s interaction can hardly be overlooked. These data suggest that the author responsible for Mic 6:3–5 knew a Balaam / Balak tradition within an exodus story that included Moses, Aaron, and Miriam and recounted stages in the journey from Shittim to Gilgal. First, let us consider the individuals named. The list “Moses, Aaron, and Miriam” occurs only here in this form (cf., however Num 12:4) and could reference any number of texts that include one or more of these characters. Perhaps the verb ‫פדה‬, which is somewhat rarely used as a reference for the exodus could illuminate the situation. A quick survey of the usage of this root in describing the exodus event indicates that it tends to appear in texts of Deuteronomic or Deuteronomistic character. This intertextual affinity could indicate a similar provenience for the term in Micah, where it only occurs in the text at hand.108 However, it remains impossible to determine a specific intertext to 105 Cf. Utzschneider, “Michaias,” 2376 and Anthony Gelston, The Twelve Minor Prophets, BHQ, vol. 13 (Stuttgart: Deutsche Bibelgesellschaft, 2010), 105*. N.b. Shittim in Num 25:1; Josh 2:1; 3:1, the noun ‫ שׁטה‬and the location ‫ אבל השׁטים‬in Num 33:49. 106 Cf. Utzschneider, “Michaias,” 2376. For the rejection of the conjecture in BHS, cf. Gelston, The Twelve Minor Prophets, 105*. 107 Though it must be admitted that any pendent to G’s κατὰ σοῦ would have stood in a lacuna. However, reasons of space preclude its reconstruction; cf. the reconstruction in Beate Ego, et al., “Minor Prophets,” in Biblia Qumranica, ed. Beate Ego, et al. (Leiden; Boston: Brill, 2005), 106–7. 108 For other uses of ‫ פדה‬within or about the exodus, cf. Deut 7:8; 9:26; 13:6; 15:15; 21:8; 24:18; 2 Sam 7:23 (= 1 Chr 17:21); and – possibly, though by no means certain – Neh 1:10. No

4.7 Micah 6:5

245

which this author refers, other than some broad exodus tradition featuring Moses, Aaron, and Miriam. The references to Shittim and Gilgal can perhaps be more readily identified. These locations are rarely attested in the biblical narrative of the exodus and conquest, though Gilgal does play a prominent role in the book of Joshua. Gilgal presents the first place in Transjordan that Israel arrives, and Israel camps there for much of the book of Joshua.109 Shittim appears only in Num 25:1; Josh 2:1; and 3:1, though Num 31:49 also mentions an Abel-Shittim.110 Conspicuously, Israel moves from one of these places to the other while crossing the Jordan (cf. Josh 3:1 + 4:19). Presumably the Micah text makes some oblique reference to this event, as these places are never otherwise mentioned together in any known text, or to some otherwise unknown tradition. The difficulty remains in determining in what way this journey permitted the people ‫“ דעת צדקות יהוה‬to know the righteous deeds of YHWH”; the term ‫ צדקה‬does not even appear in Joshua 3–4. But the question remains somewhat open as to how Micah relates to the other biblical Balaam traditions. The circumstances described by Mic 6:3–5 presume a context not unlike that of Numbers 22–24 and reports about the same main characters: Balak (here without a patronymic, however), king of Moab, and Balaam ben Beor. Balak intended to do something to the people, whereby Balaam responds to this intention. Like Numbers 22–24, Micah 6 presumes that Balaam contradicted Balak’s expectations, though the Micah text implies more than states this.111 The term ‫ יעץ‬is common to both Micah 6 and Numbers 22–24,112 but to no other Balaam or Balak traditions. However, in Num 24:14b Balaam advises Balak, a reversal of the situation in Mic 6:5, in which Balak advises himself. Nonetheless, Micah’s strongest connection to another biblical Balaam tradition is to Numbers 22–24, primarily since little other material about Balaam or Balak is transmitted by the Micah text. Nothing implies that Israel killed either of these other biblical texts use the term ‫ פדה‬in the manner of these descriptions about the exodus; cf., e.g., Exod 13:13, 15; 21:8; 34:20; Lev 19:20; 27:27, 29; Num 3:49; 18:15–17; and 1 Sam 14:45. 109 Cf. Josh 4:19–20; 5:9–10; 9:6; 10:6–7, 9, 15, 43; 12:23; 14:6; and 15:7. Before Joshua, the Bible mentions Gilgal only once: in Deut 11:30. 110 The mention in Joel 4:18 is irrelevant for this survey. 111 Contrary to what some would suggest, these really present the limit of the similarities between the versions. Contra, e.g., Rainer Kessler, Micha, HThKAT (Freiburg im Breisgau; Basel; Vienna: Herder, 2000), 265–66, who readily identified the complete text of Numbers 22–24 and even Artur Weiser, Das Buch der zwölf kleinen Propheten I: Die Propheten Hosea, Joel, Amos, Obadja Jona, Micha übersetzt und erklärt, ATD, vol. 24 (Göttingen: Vandenhoeck & Ruprecht, 1967), 280, who – without explicitly stating such – read the metaphor of YHWH transforming Balaam’s curse into blessing from Deut 23:6 in the background. That goes much too far. 112 Contra Hans Walter Wolff, Dodekapropheton 4. Micha, BKAT, vol. XIV/4 (NeukirchenVluyn: Neukirchener Verlag, 1982), 149, who incorrectly stated that ‫ יעץ‬does not appear in Numbers 22–24.

246

The Bible’s Balaam beyond Numbers 22–24

characters or even battled against them. Contrary to the text of Deuteronomy, Joshua, and Nehemiah, nothing in Micah suggests that Balaam served as Balak’s ally. Rather, the Micah text, like Numbers 22–24, suggests that Balaam failed to provide Balak with his desired outcome. Micah 6 presents a largely favorable image of Balaam.113 The circumstances here, both in terms of literary and presupposed geographical context, match the version of Numbers postulated as part of the first redaction of the book; cf. especially Num 25:1, which features “Israel” at Shittim.114 The application of the term ‫ יעץ‬in Num 24:14b (first redactional layer) and Mic 6:5 commends a connection between these two texts at this diachronic stage. In other words, it appears that the Micah tradition about Balaam most matches a stage in the redaction history of Numbers identified with the first redaction, the inclusion of the Balaam story in a Deuteronomistic composition.115

4.8 Nehemiah 13:1–3 4.8 Nehemiah 13:1–3

The opening of the final chapter of Nehemiah – similarly to Deut 23:4–6 – references the story of Balaam as the basis for excluding Moabites and Ammonites from the “assembly of God” (‫)קהל האלהים‬. An obvious connection exists between the two texts, and Nehemiah seems to explicitly reference some version of Deuteronomy. This implication can be gleaned from Neh 13:1, which notes that the following material [‫נמצא כתוב בו ]בספר }תורת{ משׂה‬. However, a number of differences exist between these two texts, mandating a comparison and explication of the distinctions.

113 Cf., e.g., Greene, Balaam and his interpreters, 75, who identified a favorable tendency in Micah’s description of Balaam due to Balaam’s interfering with Balak’s plans. 114 Cf. Helmut Utzschneider, Micha, ZBK, vol. 24.1 (Zürich: Theologischer Verlag Zürich, 2005), 134, who recognized even the Deuteronomistically oriented salvation-historical theme of these verses, comparing them in this light even to Joshua 24. Wolff, Micha, 143 and 149–50 similarly recognized this sentiment. Kessler, Micha, 259–60 saw this text as post-Deuteronomistic chronologically, but still sharing substance with Deuteronomic / Deuteronomistic literature. On the other hand, Weiser, Hosea – Micha, 281 recognized the similarity to the prophecy of Hosea. 115 For the identification of Micah 6 as part of a Deuteronomistic corpus, cf. James Nogalski, Literary Precursors to the Book of the Twelve, BZAW, vol. 217 (Berlin; New York: de Gruyter, 1993), 141–44; Aaron Schart, Die Entstehung des Zwölfprophetenbuchs. Neubearbeitung von Amos im Rahmen schriftenübergreifender Redaktionsprozesse, BZAW, vol. 260 (Berlin; New York: de Gruyter, 1998), 192–93 and 227; and Jakob Wöhrle, Die frühen Sammlungen des Zwölfprophetenbuches. Entstehung und Komposition, BZAW, vol. 360 (Berlin; New York: de Gruyter, 2006), 252–55.

4.8 Nehemiah 13:1–3 Deut 23:4–6 G: οὐκ εἰσελεύσεται Αμμανίτης καὶ Μωαβίτης εἰς ἐκκλησίαν κυρίου καὶ ἕως δεκάτης γενεᾶς οὐκ εἰσελεύσεται εἰς ἐκκλησίαν κυρίου καὶ ἕως εἰς τὸν αἰῶνα παρὰ τὸ μὴ συναντῆσαι αὐτοὺς ὑμῖν μετὰ ἄρτων καὶ ὕδατος ἐν τῇ ὁδῷ ἐκπορευομένων ὑμῶν ἐξ Αἰγύπτου καὶ ὅτι ἐμισθώσαντο ἐπὶ σὲ τὸν Βαλααμ υἱὸν Βεωρ ἐκ τῆς Μεσοποταμίας καταράσασθαί σε καὶ οὐκ ἠθέλησεν κύριος ὁ θεός σου εἰσακοῦσαι τοῦ Βαλααμ καὶ μετέστρεψεν κύριος ὁ θεός σου τὰς κατάρας εἰς εὐλογίαν ὅτι ἠγάπησέν σε κύριος ὁ θεός σου

Deut 23: 4–6 (reconstructed): ‫לא־יבא עמוני ומואבי‬ ‫בקהל יהוה גם דור‬ ‫עשׂירי לא־יבא להם‬ ‫בקהל יהוה עד־עולם‬ ‫על־דבר אשׁר לא־‬ ‫קדמו אתכם בלחם‬ ‫ובמים בדרך בצאתכם‬ ‫ממצרים ואשׁר שׂכר‬ ‫עליך את־בלעם בן־‬ ‫בעור מארם נהרים‬ ‫לקללך ולא־אבה יהוה‬ ‫אלהיך לשׁמע אל־‬ ‫בלעם ויהפך יהוה‬ ‫אלהיך את־הקללה‬ ‫לברכה כי אהבך יהוה‬ ‫אלהיך‬

Neh 13:1bβ–2 M: ‫לא־יבוא עמני ומאבי‬ ‫בקהל האלהים עד־‬ ‫עולם כי לא קדמו‬ ‫את־בני ישׂראל בלחם‬ ‫ובמים וישׂכר עליו‬ ‫את־בלעם לקללו‬ ‫ויהפך אלהינו הקללה‬ ‫לברכה‬

247 Neh 13:1bβ–2 G: μὴ εἰσέλθωσιν Αμμανῖται καὶ Μωαβῖται ἐν ἐκκλησίᾳ θεοῦ ἕως αἰῶνος ὅτι οὐ συνήντησαν τοῖς υἱοῖς Ισραηλ ἐν ἄρτῳ καὶ ἐν ὕδατι καὶ ἐμισθώσαντο ἐπ’ αὐτὸν τὸν Βαλααμ καταράσασθαι καὶ ἔστρεψεν ὁ θεὸς ἡμῶν τὴν κατάραν εἰς εὐλογίαν

Table 20: Deuteronomy 23:4–6 // Nehemiah 13:1–3

Minor orthographic differences aside, Nehemiah generally proffers a shorter text than Deuteronomy here; Deuteronomy attests several elements not found in Nehemiah, but Nehemiah also attests some elements not found in Deuteronomy.116 Both conditions demand explanation. For example, whereas Deut 23:5 reads ‫קהל יהוה‬, Neh 13:1 attests ‫קהל האלהים‬. ‫ האלהים‬may be the preferred term for the deity in Nehemiah, which would explain this difference.117 More substantial differences are the inclusion of the phrases ‫גם דור עשׂירי לא־יבא להם‬ 116

Cf. Galling, Chronik, Esra, Nehemia, 250 notes these differences, but implies that they all trace their origins back to the author of Nehemiah. While this may have been the case, one should not so readily preclude recensional engagement with these versions. 117 N.b. in M ‫ האלהים‬appears 23 times, but ‫ יהוה‬only seventeen times in Nehemiah. Admittedly, the small sample size could preclude any strong conclusion on this basis.

248

The Bible’s Balaam beyond Numbers 22–24

‫בקהל יהוה‬, ‫בדרך בצאתכם ממצרים‬, ‫ולא אבה יהוה אלהיך לשׁמע אל בלעם‬, and ‫כי‬ ‫ אהבך יהוה אלהיך‬in Deuteronomy. Finally, Nehemiah (and perhaps Deuteronomy also) demonstrates some contextual changes, for instance the replacement of Deuteronomy’s second-person suffixes with third-person suffixes in Nehemiah, the reading of ‫ כי‬in Nehemiah instead of Deuteronomy’s ‫על דבר אשׁר‬, and Nehemiah’s simple ‫ וישׂכר‬for Deuteronomy’s ‫ואשׁר שׂכר‬. Fortunately, the textual tradition of these verses in Nehemiah is stable enough that this pericope necessitates no thorough text-critical engagement.118 Nonetheless, the differences to Deuteronomy must be explained. A general exegetical maxim notes that lectio brevior potior est. This principle would a priori imply Nehemiah’s priority over Deuteronomy in almost every instance. But of course exceptions to this rule do exist. Perhaps some nuances should be appreciated here. The missing of the ten generations in Nehemiah is conspicuous. Possibly Nehemiah’s editors deleted this from the Deuteronomy Vorlage in order to potentially preclude any negative reflection on David. This postulation presumes that some tradition of Ruth or at least its genealogy for David now found in Ruth 4:18–22 may have been known at the time, but that is hardly certain; cf. to this end also 1 Sam 22:3–4, which presupposes some connection between David’s family and Moab. Possibly the ten generations were inserted in Deut 23:4 to more starkly contrast the Moabites and Ammonites from the Edomites and Egyptians in Deut 23:8–9. N.b. the later addition of these generations in the preceding verse (Deut 23:3) in M when contrasted with G. Were that the case, the longer reading in Deut 23:4 should be regarded as secondary to more terse reading of Neh 13:1. However, as demonstrated above, that solution seems unlikely based on the literary history of Deuteronomy 23. Most likely Nehemiah 13 abbreviated the text of Deuteronomy to remove an inconsistency: ten generations versus eternity, as still found in Deut 23:4. Thus, Deuteronomy 23 presumably preserves the older version vis-à-vis Nehemiah 13 regarding this phrase. The phrase ‫ על דבר אשׁר‬remains sparsely attested in the Bible, and it is missing in Neh 13:2; other than its appearance in Deut 23:5, it occurs twice in Deut 22:24 and once more in 2 Sam 13:22. This rarity could indicate its originality, but perhaps the shorter reading of Nehemiah should be favored in this instance as well. No obvious solution presents itself. The phrase ‫בדרך בצאתכם ממצרים‬, absent in Nehemiah 13, could represent a deletion of this phrase, since it nowhere else appears in Nehemiah. It does occur in two other verses in Deuteronomy (24:9 and 25:17), suggesting that it may have been more acceptable in Deuteronomistic theology than in the theology of Ezra–Nehemiah, where the exodus does not play such a prominent role. At the same time, removing it creates a better accordance with Neh 13:1. That the Moabites and Ammonites 118

Cf. the small number of variants in David Marcus, Ezra and Nehemiah, BHQ, vol. 20 (Stuttgart: Deutsche Bibelgesellschaft, 2006).

4.8 Nehemiah 13:1–3

249

did not come with food and water has been divorced from a historical context of the exodus. Rather, more generally, apparently, they did not bring food and water to the Israelites. By removing the historical context of the exodus, this shorter reading in Neh 13:2 better conforms to the reading of “forever” in Neh 13:1. Balaam’s missing origins in Nehemiah 13:2 probably attest an older, shorter version.119 Nothing suggests that the authors or tradents of Nehemiah should have had a problem identifying Balaam at least by his patronymic for theological or ideological reasons. Why should they have deleted it? Nonetheless, Nehemiah 13 fails to present it and may be the older reading in this case. In this way, it matches Josh 24:9 in the older reading still attested in G. The same should be said for the common text between Josh 24:10 and Deut 23:6 but absent in Neh 13:2–3: Nehemiah 13 makes no reference to YHWH’s refusal to listen to Balaam (as in M in both Deuteronomy 23 and Joshua 24) or to destroy Israel (as in G in Joshua 24). The motif of listening or refusing to listen would have fit well into Nehemiah 13; cf. ‫ שׁמע‬in, e.g., Neh 12:42–43; 13:3 (!) and 27. One wonders what could have caused the scribes not to copy this phrase. Perhaps they only sought to abbreviate the story, although it remains possible that it had not been inserted into Deuteronomy 23 yet, just as it was not in the older version of Joshua still attested in G. Thus, in many ways, but particularly in its similarities to the older Balaam tradition in Joshua, it appears that Nehemiah 13 may attest an older form of material than that currently found in Deuteronomy 23 or even in the older version of that text reconstructed above.120 Nonetheless, in other cases, it appears to have abbreviated the text of Deuteronomy 23 it used as a Vorlage.121 In terms of content, Nehemiah 13 provides no further information about Balaam than that found in Deuteronomy 23. For that matter, Nehemiah 13 attests decidedly less data about this character. His patronymic is absent and there is nothing about his origins, social position, or profession. The common elements with the other Balaam traditions are limited to the name Balaam, some interaction with Moabites and Ammonites (as in Deuteronomy 23 exclusively), and the motifs of blessing (‫ )ברך‬and cursing (‫)קלל‬, here – as in Deuteronomy 23 – with the understanding that the audience’s God turned the curse into a blessing, but without the motif of the deity refusing to listen to Balaam. Though not explicitly negative toward Balaam, the context of this information encourages his negative characterization, referencing this incident to preclude foreigners from admission into the community. 119 Cf. the text-critical discussion of this verse above, and n.b. the addition of ‫ מפתור‬in Deut 23:5 M, which could be indicative of further expansionist editing in Deuteronomy’s referring to Balaam. 120 I.e., the absence of Balaam’s patronymic and the absence of the motif of listening. 121 I.e., the lack of the ten generations and failing to mention the historical context of the exodus.

250

The Bible’s Balaam beyond Numbers 22–24

4.9 Conclusions: Balaam in the Hebrew Bible 4.9 Conclusions: Balaam in the Hebrew Bible

After this survey, we notice many aspects about the figure Balaam and his interactions with Balak and Israel in the Hebrew Bible in the various literary iterations. Numbers 31 and Joshua 13 report the Israelite’s killing of Balaam with his Midianite allies. Deuteronomy 23 relies on Balaam as a negative example of how other peoples interacted with Israel. Joshua 24 describes God changing Balaam’s curse into a blessing and apparently Israel’s victory over Balak in an earlier version. Judges 11 implies that Balak never battled against Israel at their encounter in Moab. Micah 6 describes Balaam as someone who contradicted Balak in some capacity. Finally, Nehemiah 13 drew explicitly on Deuteronomy 23 and its image of Balaam to deny Moabites and Ammonites admission into the congregation. Numbers 22–24 lacks many of these elements, but does share some. Since some aspects of Numbers 22–24 present later interpolations, the question becomes, whether these later aspects demonstrate any relationship to these other biblical traditions. Is there a common development of the figure Balaam in the biblical literature? Apparently, the figure Balaam developed over time among the other biblical traditions just as in Numbers 22–24. To this end, we can develop a relative chronology of biblical texts about Balaam that, to some degree, reflects the same development witnessed in Numbers 22–24. None of the other biblical texts make the impression that they were familiar exclusively with the oldest core of the Balaam story identified in Chapter Three. Nothing particularly precludes that, but we should first consider what other relationships we can identify. Both the text of Judges 11 and its retelling of Balak’s engagement with Israel and the prophetic speech in Micah 6 appear to allude to, and thus demonstrate familiarity with, something like the Balaam narrative recounted in the first redaction of that story. The situation of the event between Shittim and Gilgal in Micah commends this, as does the abbreviated recounting of the events of Numbers 21 and 22–24 in Judges. Micah implies that Balak was Israel’s enemy and Balaam, then, its ally. Judges denies Balak’s military engagement with Israel. Both of these texts thus reflect a version still apparent in the first redaction. Joshua 24 presents a turn in this scheme, noting that Balak did engage Israel militarily and that Israel subsequently defeated him, should the proposed reconstruction above be accurate. That could imply that Joshua 24 knew of a distinct Balaam tradition from that of Numbers 22–24, irrespective of what version. Yet, while the interaction of Joshua 24 with Numbers 22–24 remains limited, it cannot be entirely precluded. The older version, lacking Balaam’s patronymic, for example, implies that he must be known from another context. That could suggest that Joshua 24 indeed was familiar with Numbers 22–24, but interpreted the text somewhat liberally. It thus created an actual military encounter from Balak’s desire to fight in Num 22:6 and 11. That is, Joshua 24

4.9 Conclusions: Balaam in the Hebrew Bible

251

reinterpreted Balaam’s tale more than the other texts did. Nonetheless, it remains impossible to determine a specific layer of Numbers 22–24 with which it may have interacted and upon which it may be commenting. It could just as easily be the original story as the first redaction. No specific evidence exists to ascribe it knowledge of one of the later redactions, but this cannot be entirely precluded either. Taken in the context of the first eight verses of Joshua 24, one recognizes the affiliation with elements identified with the first redaction in Numbers 22–24 (e.g., ‫ישׂראל‬, ‫שׁבט‬, Amorites before Balaam). Yet, its contradicting Judges 11 makes it unlikely that these two texts stem from a common hand or that they initially existed as portions of a common composition. The safest resolution would be to commend its affiliation with a Balaam narrative that included the first redactional layer, but which transformed the figure Balaam and his interaction with Balak to some degree. Certainly he no longer appears entirely positive in Joshua 24. That would in turn commend regarding the reference to Balak in Judges 11 as distinct from the Deuteronomistic composition that contained the edited version of Numbers 22–24 and Joshua 24. Alternatively, Joshua 24 could presume a passing familiarity with Numbers 22–24 after the second redactional that incorporated the Midianites and Numbers 31. That would explain the distinction from Judges 11, the more negative attitude toward Balaam, and the idea that Israel killed Balaam (and Balak). Much of the content of Joshua 24 makes this difficult, however. It still remains a plausible alternative – perhaps even the preferable alternative – that the Balaam tradition in Joshua 24 was inserted there after the second redaction of Numbers 22–24. At the same time, Numbers 31:16 demonstrates a similar, negative attitude toward Balaam, both in its original, abbreviated form, as well as in its redacted form. The older portion, Num 31:16a, reflects the circumstances in Numbers 22–25 at the stage of the second redaction. It combines Balaam with the Midianites and both with the situation at Peor. Its negative attitude towards Balaam matches that of Joshua 24, in tenor if not in content. While the status of the relationship of Joshua 24 to that of Numbers 22–24 remains somewhat insecure, we can pretty readily ascribe Num 31:16a to the second redaction identified in Numbers 22–24. These distinctions place it pretty firmly in some priestly – in the broadest sense of the term – school of thought. On the other hand, Num 31:16b conforms to the later redactional elements identified in Numbers 22–24 by including a reference to the ‫עדה‬, a later term for the exodus group in Numbers. Not only does this verse in Numbers 31 appear to stem from those later editorial elements, the same must be said for v. 8. This verse presents a later interpolation into a story affiliated with the second redaction. Thus, we can ascribe both Num 31:16b and 8 to the same hand. That is, these additions root firmly in post-priestly ideology (R3+). From a – presumably – somewhat later hand came Josh 13:21–22. Knauf described the scenario thus:

252

The Bible’s Balaam beyond Numbers 22–24

“Um den ganzen, sukzessiv redaktionell gewachsenen Komplex Num. 22–31 zu klammern, müssen die Midianiter nunmehr an der Schuld der Moabiter beteiligt werden, einen ausländischen Wahrsager zur Verfluchung des Gottesvolkes angeheuert zu haben; so gelangten sie redaktionell in Num. 22,4.7. Jos. 13,21 geht dann noch einen Schritt weiter und macht die Midianiterkönige aus Num. 31, weil sich ihre Geschichte an die Geschichte Bileams anschloß, die sich an die Geschichte von Sihon (und Og) anschloß (schon in ihrer Grundschicht ein nachpriesterschriftlicher Zusatz zum Pentateuch), zu Vasallen Sihons. Die scheinbar ‘beduinische’ Allgegenwart der Midianiter, ihr vermeintliches ‘Protektorat’ über das ganze Gebiet der späteren Provinzia Arabia ist kein Sachverhalt der Geschichte Palästinas und Nordarabiens, sondern das Produkt der Redaktionsgeschichte des Alten Testaments.”122

Josh 13:21–22 reports essentially the same data as Num 31:8. Likely editors copied these data from their context in Numbers as an insertion into Joshua 13. Nonetheless, it slightly expanded the information about Balaam, including – for the first time – a reference to his profession: ‫הקוסם‬.123 Not only does this term present more information about Balaam, it permits the ascription of some information to the editors responsible for this addition: they had an even more negative attitude toward Balaam than those responsible for the later additions into Numbers 22–24 and 31. With the appending of the term ‫הקוסם‬, Balaam officially became persona non grata in the biblical tradition. Finally, Deuteronomy 23 appears to have familiarity with Joshua 24 and perhaps Judges 11.124 That would put it on a similar footing to the latest redactional material in Numbers 22–24, though only a limited distinction could be made between Deuteronomy 23 and Numbers 31.125 Its familiarity with Joshua 24 seems to put it on a diachronic level consistent with the later redactions in Numbers (R3+). However, it remains impossible to say much more about it than that. Nehemiah borrowed from this tradition, apparently from an older 122 Ernst Axel Knauf, Midian: Untersuchungen zur Geschichte Palästinas und Nordarabiens am Ende des 2. Jahrtausends v.Chr., ADPV (Wiesbaden: Harrassowitz, 1988), 167–68. 123 Schmidt, “Bileamüberlieferung,” 258–59 commended regarding the addition of this title as a sign that Josh 13:21–22 attempts to paint Balaam more negatively than Deuteronomy 23, which had at least still permitted him to speak a legitimate blessing. 124 Schmidt, “Bileamüberlieferung,” 258–59 regarded the relationship between Deuteronomy 23 and Joshua 24 as the opposite of that here. One reason for this stems from his ignoring the text history of these verses in which they were adapted to one another over time. 125 For example, cf. already Donner, “Balaam pseudopropheta,” 121–22: “Die zweite, späteste Stufe der Interpretation – Num 31,8.16 und davon abhängig Jos 13,22 – setzt die in Dtn 23,5b.6 begonnene Umwertung Bileams zum pseudopropheta voraus und zieht aus ihr die letzte Konsequenz. War Bileam ein falscher Prophet, dann war selbstverständlich, daß er hatte umgebracht werden müssen. Denn falsche Propheten sind nach Dtn 13,2 umzubringen, wenn sie Israel zum Abfall von Jahwe verführen wollen. Nun folgte unmittelbar auf Num 22–24 die rätselhafte Geschichte vom Abfall der Israeliten zum Götzendienst des Baal Peor auf dem Wege der Buhlschaft mit moabitschen Frauen (Num 25,1–5), wobei im später hinzugetretenen zweiten Teil des Kapitels (Num25,|6 ff.) auch eine Midianiterin genannt war. Sollte Bileam dabei nicht seine Hände im Spiel gehabt haben, zumal er nach Num 23,28 auf dem Gipfel des Peor gestanden hatte?”

253

4.10 Balaam at Qumran and in the New Testament

version than that found in M, and, therefore, must stem from some school of thought in the wake of the expansion in Deuteronomy 23. These appear to be the latest reflections on Balaam in the Hebrew Bible. S1 R1 R2 R3+

– Judges 11:25 Numbers 31:16a Numbers 31:16b

– Micah 6:5 Joshua 24:9–10 Deuteronomy 23:4–6

–

Nehemiah 13:1–3

Table 21: The Diachronic Relationship of Numbers 22–24 with Other Biblical Balaam Texts

With that, we can roughly circumscribe the Balaam traditions and their development within the Bible.126 Having covered the material in the Hebrew Bible, we can look at the further development, where applicable, of the Balaam tradition at Qumran and in the New Testament.

4.10 Balaam at Qumran and in the New Testament 4.10 Balaam at Qumran and in the New Testament

In addition to the biblical scrolls from Qumran, a few other texts reflect on aspects of Balaam or his oracles. These will be briefly considered below before we turn to the New Testament Balaam traditions. Three New Testament texts reference Balaam: 2 Pet 2:15–16; Jude 11; and Rev 2:14. One appears to allude to it: Matthew 2. These references and allusions merit brief reflection, especially considering what elements of which Balaam tradition they accepted, adapted, and/or propagated. It is a question of reception history.127 The question must be addressed: what tendency or tendencies from the biblical texts or the Hebrew canon do the Qumran and New Testament texts favor or develop, and to what degree does this favoritism reflect tendencies that had begun already within the Hebrew Bible? This discussion will first briefly address Balaam at Qumran. Then, as with the texts of the Hebrew Bible, for those of the 126

Cf. Donner, “Balaam pseudopropheta,” 114, who recognized two negative Balaam images in the Bible, one of which was closer to Numbers 22–24 (Deuteronomy 23; Joshua 24; Nehemiah 13) and one that had essentially nothing to do with it (Numbers 31; Joshua 13). Donner’s reconstruction must be somewhat emended to afford the inclusion of the literarycritical and redaction-historical observations that demonstrated that Numbers 31 did share something in common with Numbers 22–24, but only through editorial additions. The tendency of my survey generally matches that of Ed Noort, “Balaam the Villain: The History of Reception of the Balaam Narrative in the Pentateuch and the Former Prophets,” in The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity and Islam, ed. George H. van Kooten and Jacques van Ruiten, TBN, vol. 11 (Leiden; Boston: Brill, 2008), 3–23. 127 In this study I am limiting myself to biblical literature. The sheer volume of material in the reception history of Balaam would preclude its satisfactory inclusion in a biblically oriented, primarily text and literary-historical work such as this.

254

The Bible’s Balaam beyond Numbers 22–24

Christian scriptures we will follow the traditional canonical order, without considering the relative chronology of the texts within the New Testament. Before turning to the explicit references to Balaam, we will begin our New Testament discussion with one text that appears to allude to this figure, or – more accurately – to one of his oracles. With that, we consider Matthew 2. Before that, we will provide an overview of Balaam at Qumran.

4.10.1 Balaam at Qumran Two biblical scrolls recording portions of chapters 22–24 of the book of Numbers remained at Qumran: 4Q23 = 4QLev – Numa128 and 4Q27 = 4QNumb (PreSamaritan). 129 Perhaps some elements of a third were recovered from Cave 2.130 These manuscripts are relevant to our discussions at hand due to their importance for the textual history of the Hebrew Bible. To this end, they were incorporated into the discussion about textual criticism in Chapter Two, and I will not address them further here other than with some general observations. Numbers as such apparently did not play a very important role at Qumran, as can be seen through its poor representation among the community; “…while more copies of Numbers are preserved at Qumran than copies of the so-called historical books and some of the prophetic books, Numbers is by far the least well attested, and thus seemingly the least significant, of any of the books of the Torah for the Qumran community.”131 More importantly, the attested variants of Numbers 22–24 at Qumran do not avow a more negative characterization of Balaam than that of M.132 Thus, the lesser status for Numbers as attested at Qumran nevertheless did not apparently hamper the importance they ascribed to Balaam, or – more accurately – to his fourth oracle. The community at Qumran cited relatively few Torah texts with the exception of the Temple Scroll. However, no fewer than three texts from Qumran reference Balaam’s fourth oracle: The Damascus Document (CD and 4Q266

128 “The text, especially those portions preserving Numbers, basically reflects that of the MT and can thus be described as proto-Masoretic.” Dana M. Pike, “The Book of Numbers at Qumran: Texts and Context,” in Current Research and Technological Developments on the Dead Sea Scrolls, ed. Donald W. Parry and Stephen D. Ricks (Leiden; New York; Köln: Brill, 1996), 172. 129 Cf. Pike, “Numbers at Qumran,” 172–73. 130 Cf. Eibert Tigchelaar, “A Qumran Cave 2 Fragment Preserving Part of Numbers 23:5– 7[8] (2Q29 1),” in The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, ed. George H. van Kooten and Jacques van Ruiten, TBN, vol. 11 (Leiden: Brill, 2008), 83–86. 131 Pike, “Numbers at Qumran,” 169. 132 Cf. Florentino García Martínez, “Balaam in the Dead Sea Scrolls,” in The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity and Islam, ed. George H. van Kooten and Jacques van Ruiten, TBN, vol. 11 (Leiden; Boston: Brill, 2008), 72–75.

4.10 Balaam at Qumran and in the New Testament

255

and 4Q269), 4QTestimonia (4Q175), and the War Scroll (1QM).133 Probably a fourth document alludes to it (1QSb 5,27). 134 Each of these texts describes some messianic figure(s) and applies Balaam’s prophecy to him or them. This messianic tendency demonstrates that Balaam’s oracle possessed a uniquely high status with the community, in spite of the paucity of copies of Numbers. None of these words betrays a negative attitude toward Balaam. Probably the most significant citation of this oracle, and the one I use here exemplarily, appears in the Damascus Document (CD VII 12–21) and its comments about Num 24:17. “In CD 7,19, the scepter [of Num 24,17] is interpreted as the Prince of the Congregation, who is elsewhere identified with the Branch of David. The verse is cited without interpretation in the War Scroll (lQM 11,5–7) and in the Testimonia, and is likely to have messianic connotations in both contexts. Another possible allusion is found in the blessing of the prince of the congregation in lQSb 5,27, which says ‘for God has raised you as a scepter’.”135

The relevant exegesis in the Damascus Document can be translated thus: “The star is the Interpreter of the Law who shall come to Damascus; as it is written, A star shall come forth out of Jacob and a sceptre shall rise out of Israel (Num. xxiv, 17). The sceptre is the Prince of the whole congregation, and when he comes he shall smite all the children of Seth (Num. xxiv, 17).” 136 Beyond merely quoting the text of Numbers 24, the community attests an exegetical undertaking regarding how it should be understood. “The star is equated with the ‘Interpreter of the Law,’ ‫דורשׁ תורה‬, who is best understood as the great priestly figure of the future, referred to elsewhere as ‘Messiah,’ ‫משׁיח‬, and ‘Chief Priest,’ ‫כוהן הראשׁ‬. The equation of the scepter with the ‘prince of the congregation,’ ‫נשׂיא‬ ‫)כל( העדה‬, correlates with other passages that refer to this ‘prince’ as the great messianic military leader. This passage in Numbers 24:17 was thus seen by those in the community at Qumran as referring to the two eschatological messiahs, the priestly (‘star’) and the davidic (‘scepter’). While it is true that a messianic interpretation of this passage is not unique to Qumran, the way it is woven with commentary into the Damascus Document indicates that the Qumran community had made it their own.”137

133

For the Damascus Document, cf. Joseph M. Baumgarten, ed., Qumran Cave 4.XIII: The Damascus Document (4Q266–273), DJD, vol. 18 (Oxford: Clarendon Press, 1996) and Magen Broshi, ed., The Damascus Document Reconsidered (Jerusalem: The Israel Exploration Society, 1992). For the Testimonia, cf. John Marco Allegro, ed., Qumran Cave 4.I: (4Q158–186), DJD, vol. 5 (Oxford: Clarendon Press, 1968), 57–60 and Plate 21. For the War Scroll, cf. Eliezer Lipa Sukenik, ed., The Dead Sea Scrolls of the Hebrew University (Jerusalem: Magnes Press, 1955); Schultz, “The Kittim of Assyria”; and the literature cited there. For an explanation and evaluation of the materials, cf. García Martínez, “Balaam in the Dead Sea Scrolls,” 75–82 and, particularly for the Damascus Document, García Martínez, “Two Messianic Figures”. 134 Cf. Pike, “Numbers at Qumran,” 192–93. 135 John J. Collins, “Messianism and Exegetical Tradition,” 145. 136 Pike, “Numbers at Qumran,” 183. 137 Pike, “Numbers at Qumran,” 183.

256

The Bible’s Balaam beyond Numbers 22–24

When reflecting on Balaam’s fourth oracle and its awaiting a star and a scepter, the community at Qumran apparently did not interpret this oracle as having been fulfilled. That is, they still expected someone to come in fulfillment of this oracle, demonstrating that they did not believe that David (as was probably initially intended at its composition) was the one ultimately anticipated. One other matter that stands out in the appropriation of Balaam’s purported statement is the missing ascription. Nowhere does the text of the Damascus Document address this oracle’s source. The same could be said of the War Scroll.138 Perhaps this quiescence implies reticence. They may have been uncertain about ascribing such a substantial text to a character whose image had become so burdened with negative ascriptions. Evidence for this appears in the mention of Balaam in a list of false prophets, also found at Qumran (4Q339 = 4QList of False Prophets).139 Taken together with the import afforded to his fourth oracle, though often cited without ascription, this demonstrates perhaps an ambivalent or negative attitude about the prophet, while a more positive attitude toward the prophecy. Nevertheless, the community at Qumran apparently ascribed Balaam’s fourth oracle import, and they understood the text with a view toward messianism. They liked the message, but hated the messenger. Since they cite the proto-Masoretic version of Numbers 24:17, it remains unlikely that modern exegetes should ascribe a particularly messianic tendency to the G version of Numbers. Other than mentioning him in a list of false prophets and these citations of the fourth oracle, however, Balaam and the traditions surrounding him apparently played no substantial role at Qumran.

4.10.2 Matthew 2 In its opening chapters, the Gospel of Matthew presupposes the visiting figures’, commonly referred to as “wise men” (μάγοι), and audience’s familiarity with some astral prophecy. The question and statement of the μάγοι in Matt 2:2 read ποῦ ἐστιν ὁ τεχθεὶς βασιλεὺς τῶν Ἰουδαίων; εἴδομεν γὰρ αὐτοῦ τὸν ἀστέρα ἐν τῇ ἀνατολῇ καὶ ἤλθομεν προσκυνῆσαι αὐτῷ: “Where is the one born the king of the Judeans/Jews? For we saw his star at (its) rising / in the east (ἐν τῇ ἀνατολῇ) and came to worship him”. The motif of the star leading to some royal figure features in the next several verses as well. Assuming it alludes to some biblical text, the only text to which these verses in Matthew could allude would be Balaam’s fourth oracle.140 The text of Matt 2:2–12, however, only – at best – 138

4QTestimonia does ascribe the oracle to him, however. Cf. Magen Broshi, ed., Qumran Cave 4.XIV: Parabiblical Texts, Part 2, DJD, vol. 19 (Oxford: Clarendon Press, 1995), 77–79 and Plate 11, as well as Shemesh, “Note on ‘4Q339’”. 140 Cf. Rudolf Schnackenburg, Matthäusevangelium, Band 1, NEchtB, vol. 1.1 (Würzburg: Echter, 1985), 23, who presumed that Num 24:17 stands in the background of Matthew 2. The cautious approach of Tobias Nicklas, “Balaam and the Star of the Magi,” in The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, ed. Georges H. van Kooten 139

4.10 Balaam at Qumran and in the New Testament

257

hints at a connection to the figure Balaam, but still apparently presents an important allusion to his oracles in the New Testament.141 But can we make this ascription with certainty? The Hebrew term “star” (‫ )כוכב‬appears in the singular only twice in the Hebrew Bible, in Num 24:17 and Amos 5:26.142 The other occurrences of this lexeme are all plural. Since the Amos text provides a prophetic injunction against poor behavior, and not an anticipation of some coming figure, it can be rejected as a plausible referent in Matthew 2. Add to that the combination of a star and some person, and only the text of Num 24:17 remains a plausible candidate, particularly in the form transmitted by G and presumably extant in Hebrew in its Vorlage.143 The similarity between the ἐν τῇ ἀνατολῇ in Matt 2:2 and 9 and the ἀνατελεῖ in Num 24:17 should not be overlooked.144 France suggested that Balaam’s coming from the east (Num 23:7) parallels that of the Magoi coming from the east.145 Fiedler opined that Matthew’s audience may have heard Balaam as a parallel for the Magoi and Balak as a pendent for Herod.146 With these brief observations, the certainty of Num 24:17 serving as the background for the narrative in Matthew 2 seems sufficiently secure, though it is perhaps worth reiterating that some insecurity must remain, as the author did not go to particular lengths to encourage this identification.147

and Jacque van Ruiten, TBN, vol. 11 (Leiden: Brill, 2008), 233–46 demonstrates the issues with reading Num 24:17 too quickly as the background to Matthew 2. 141 Some commentators emphasize the importance of astral imagery in the Hellenistic period for transitions between rulers; cf., e.g., Donald A. Hagner, Matthew 1–13, WBC, vol. 33A (Dallas: Word Books, 1999), 25–26 and the literature cited there. Hagner fails to even mention Balaam in his discussion of this passage; cf., further, Hagner, Matthew 1–13, 27–28. 142 The Septuagint’s usages match this, with ἄστρον appearing in the singular only in these verses. Sirach 50:6 also attests the singular ἀστὴρ, like Matthew 2, but more likely presents Onias as a different fulfiller of Balaam’s prophecy. That is, Sirach and Matthew both reference the same background tradition found in Balaam’s fourth oracle, but see the fulfillment in two distinct individuals. On the other hand, Sirach qualifies this star as the “morning” (ἑωθινὸς) star, an aspect foreign to Numbers 24. The tenfold repetition of ὡς and the idyllic imagery in Sir 50:6–10 remind the reader of the fourfold ὡσει / ‫ כי‬in Num 24:7. This repetition and imagery, combined with the astral metaphor, evoke the oracles of Balaam in a way similar to, but still distinct from that of Matthew 2. 143 Cf. the discussion to Num 24:17 in Chapter Two, on the textual criticism of Numbers 22–24. 144 Cf. William D. Davies and Dale C. Allison, A Critical and Exegetical Commentary on the Gospel According to Saint Matthew, Volume One, ICC (Edinburgh: Clark, 1988), 236 and Peter Fiedler, Das Matthäusevangelium, Theologischer Kommentar zum Neuen Testemament, vol. 1 (Stuttgart: Kohlhammer, 2006), 57. 145 Cf. Richard T. France, The Gospel of Matthew, NICNT (Grand Rapids, MI: Eerdmans, 2007), 62. 146 Cf. Fiedler, Matthäusevangelium, 56–57. 147 Cf. Nicklas, “Balaam and the Star,” 237–38.

258

The Bible’s Balaam beyond Numbers 22–24

But what information or attitudes about Balaam could this text transmit? The answer is, of course, limited and indirect. On the one hand, the author of Matthew (henceforth, just “Matthew”) must have granted Balaam’s prophecy sufficient import to allude to it in the narrative of the birth of Jesus. Simultaneously, this allusion presumes the audience’s familiarity, apparent through the characters’ familiarity, with the same prophecy. Taken together, these data suggest that Balaam’s prophecy – at least the prophecy transmitted in Num 24:17 – must have had an important role in Matthew’s community.148 At the same time, Matthew did not transmit the prophecy literally, though several other biblical citations occur in the first two chapters of the book. One citation even transmits the name of the prophet to whom the quote was attributed (Matt 2:17–18). Not so with the allusion to Balaam: neither his name nor the exact contents of the prophecy are transmitted. On one occasion Matthew 2 records a citation for which no source material is known (Matt 2:23). That at least raises the possibility that Matthew here cites some source no longer available to us, but the common connection of an ascending star as a sign of an anticipated royal accession seems too marked to be merely coincidental. These observations have important ramifications for the understanding of Balaam. First, it appears that, with the inclusion of this allusion in the narrative about the birth of Jesus, Matthew implies to his audience that this prophecy had thus far remained unfulfilled. At least, Matthew or his audience must not have regarded Num 24:17 as referring exclusively to some historical event of the past. That tends to match the circumstances at Qumran. Second, around the transition between eras, Balaam’s oracle must have been important to at least this group. That is, the Balaam tradition in Numbers must have remained relevant to them. This observation matches Green’s consideration about messianism: “…early Christian writers gave the figure of the messiah a diachronic dimension. They situated the messiah’s origin not in the present but in Israelite antiquity and thus established the Hebrew scriptures as a sequence of auguries. Reading scripture became, and to a large extent has remained, an exercise in deciphering and tracing a linear progression of portents.”149

Third, the lack of an ascription for the oracle is suspicious. Perhaps this indicates that attitudes about Balaam among Matthew and/or his audience were not the same as they were towards someone like Jeremiah.150 This again matches 148

As it apparently did at Qumran at a somewhat earlier period. William Scott Green, “Introduction: Messiah in Judaism: Rethinking the Question,” in Judaisms and Their Messiahs at the Turn of the Christian Era, ed. Jacob Neusner, William Scott Green, and Ernest S. Frerichs (Cambridge: University Press, 1987), 5. 150 Similarly, William Foxwell Albright and Christopher S. Mann, Matthew. Introduction, Translation, and Notes, AncB, vol. 26 (New York: Doubleday, 1971), 14–15 noted that this prophecy lacks the typically Matthean ascription “that it might be fulfilled”. That could also 149

4.10 Balaam at Qumran and in the New Testament

259

perhaps the tendency at Qumran. Further, Matthew witnesses to a version of Balaam’s oracle more like that of the Septuagint than that of the Masoretic text. That could speak to a plurality of versions of Numbers still extant around the transition of the eras. However, none of this data reflects explicitly on the character Balaam, only on his import for Matthew and his audience. Finally, Matthew only alludes to Numbers 22–24, or more specifically Num 24:17, and to no other text about Balaam or the Balaam tradition known from the Hebrew Bible.

4.10.3 2 Peter 2:15–16 This, the longest passage in the NT dealing with Balaam, misrepresents his patronymic, focuses on his greed, and notes his interaction with his donkey.151 The text can be translated, “Leaving the straight path, they were led astray, following the path of Balaam of Bosor, who loved the wages of unrighteousness. But he was rebuked for his own lawlessness: a mute donkey, speaking with a human’s voice, forbade the prophet’s madness.” Very little overlap exists between this interpretation of Balaam and any found within the Hebrew scriptures. Yet some elements should not be overlooked. The most conspicuous distinction is obviously Balaam’s patronymic, which in the oldest reconstructed NT text is τοῦ Βοσόρ. This actually distinguishes this Balaam from every other mention of Balaam in the bible. Scribes of other NT manuscript traditions noticed this problem and emended the text to more closely resemble the proto-Masoretic data about Balaam (B et al. read Βεωρ and S et al. read a combination of both variants, namely Βεωρσορ). Presumably this distinction indicates that the author of 2 Pet 2:15–16 was not copying from a text, but referencing it from memory. Otherwise it could present an attempt to further impugn Balaam’s character, making Balaam a “son of the flesh” (from Hebrew ‫בשׁר‬, as also attested in the Rabbinic tradition.152 Still other factors suggest that the author probably was not merely copying from a text: the term for donkey in 2 Pet 2:16 (ὑποζύγιον) does not match that used in Numbers 22* (ὄνος) and 2 Pet 2:16 identifies Balaam as a prophet (προφήτης), a term never applied to him otherwise in the Bible. These matters all suggest that the imply some reticence about the prophecy’s origin. Alternatively, Davies and Allison, Matthew 1–7, 235 regarded the missing ascription as the result of this prophecy, in contrast to others in this chapter, lacking a geographical location. 151 Cf. the more expansive survey of these verses and their context in Tord Fornberg, “Balaam and 2 Peter 2:15: ‘They Have Followed in the Steps of Balaam’ (Jude 11),” in The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, ed. George H. van Kooten and Jacques van Ruiten, TBN, vol. 11 (Leiden: Brill, 2008), 265–74. 152 Cf., e.g., Richard Bauckham, Jude, 2 Peter, WBC, vol. 50 (Waco: Word Books, 1983), 267–68; Charles Bigg, A Critical and Exegetical Commentary on the Epistles of St. Peter and St. Jude, ICC (Edinburgh: Clark, 1946), 283–84; and the primary literature cited there.

260

The Bible’s Balaam beyond Numbers 22–24

author of 2 Peter was not merely copying this data from a Vorlage, but more likely reciting from memory. That would explain why some motifs are present (a talking beast of burden, a man with preternatural ability) without the nomenclature matching other known texts about Balaam. One motif present, even prominent, in 2 Pet 2:15–16 without any pendant in either the Hebrew Bible or its Greek translation is Balaam’s fiduciary interest; in contrast to the tradition of Numbers 22–24–the only biblical tradition in which compensation of any nature could be found in a text about Balaam – 2 Peter implies that Balaam accepted or at least anticipated some form of payment. Numbers 22:18 and 24:13 both imply that Balaam could not have changed his response for any amount of money, and Numbers 22–24 never informs the reader that Balaam received any financial recompense for his attempts. The first attested instance of money in the Balaam story in Numbers 22–24 is the Vulgate, which in Num 22:5 indicates that Balak’s messengers had divinationis pretium in manibus.153 Second Peter contains this line of understanding, but no evidence suggests that this understanding came directly from the Bible. Thus, while 2 Peter mentions a few motifs from biblical Balaam traditions – all of them found in Numbers 22–24 – it adds one as well. The paucity of lexical overlap implies recitation and reference from memory rather than from a Vorlage. The focus on finance in 2 Peter (and Jude, see below) did not develop in a vacuum, however. While this understanding was not immanent in the biblical texts, it apparently was also not merely the creation of New Testament authors. Rabbinic exegesis developed similar ideas about Balaam. For example, in how they understood Num 22:18. “To the modern reader, this [Num 22:18] appears to be a refusal to sell the word of God, but the Rabbis understand it as yet another sign of Balaam’s greed.”154 Second Peter and Jude (see below) match this understanding.155 The pinnacle of such interpretation may be found in the Vulgate of Numbers, which goes so far as to insert this payment, not found in Numbers 22–24 up to that point, into the text at Num 22:7 for the curious plural ‫קסמים‬. That was a move that not even the Rabbis made. “The simplest explanation [of ‫]קסמים‬, appearing in Numbers Rabbah xx. 8, is that they were the objects or instruments of divination which Balaam was to use.”156 How the New Testament authors understood the ‫ קסמים‬of Num 22:7 must remain unclear, but in general, they appear to have accepted the greedy image of Balaam, known also from Rabbinic literature, though not ascribed to or affiliated with the term ‫ קסמים‬there. 153

The Septuagint’s term τὰ μαντεῖα never refers to payment for services rendered. Vermes, Scripture, 133. 155 Cf. Lewis R. Donelson, I & II Peter and Jude, NTL (Louisville, KY: Westminster John Knox Press, 2010), 254 regarding the precise relationship between these texts. 156 Vermes, Scripture, 130. 154

4.10 Balaam at Qumran and in the New Testament

261

Bearing the context of the citation in mind, perhaps Numbers 31, combined with Numbers 25, should be understood in the background. Second Peter 2 uses the term ἀσέλγεια (“licentiousness”) three different times (2:2, 7, and 18) and references other episodes understood as sexual misconduct in Hebrew Bible (Genesis 6 in 2 Pet 2:4–5; Genesis 19 in 2 Pet 2:6–8).157 Viewed in this light, Balaam could be understood in the role ascribed to him in Num 31:16,158 in which he supposedly convinced the Midianite women to lead the Israelites astray (Numbers 25). Though no particular comment is made to this end in 2 Peter 2, the larger context of the chapter suggests that this combination of texts stands in the background. Therefore, the author more directly cites the tale in Num 22:22–35 and broadly alludes to the story of Numbers 22–24, while more indirectly alluding to Numbers 25 and 31. This distinguished him from, e.g., the Gospel of Matthew, which only alluded to Numbers 24. Second Peter 2 references the broader biblical Balaam tradition, but in many ways also understood the Balaam figure in ways shared with Rabbinic literature.

4.10.4 Jude 11 Like 2 Pet 2:15–16, Jude 11 references Balaam, reflecting on his greed. Like 2 Peter 2, Balaam appears with other figures known from the Hebrew Bible, albeit with different characters than those mentioned in 2 Peter 2.159 Jude 11 reads, “Woe to them! For by the path of Cain they were brought, and by the deception of Balaam’s wages they abandoned themselves, and in Korah’s dispute they perished.” Though these figures – naturally, other than Balaam – remained absent in 2 Peter 2, Jude does include an oblique allusion to Genesis 6 (in v. 6) and a more direct reference to Genesis 19 (in v. 7), just like 2 Pet 2:5 and 7. At the same time, Jude demonstrates an affinity for literature that became extrabiblical, i.e., was not incorporated into the canon of most Christian groups. The references to Enoch (v. 7 and 14–15) and the Testament of Moses (v. 9) demonstrate this familiarity and affinity. As noted above, Jude focuses on Balaam’s desire for compensation for services rendered, just as was the case in 2 Peter 2, and, as noted above, this element remains unknown in texts about Balaam found in the Hebrew Bible. Nonetheless, this interpretation is consistent with Rabbinic interpretations about Balaam. This has been addressed above and demonstrates that Jude adds nothing new to the figure Balaam that was not found also in 2 Peter. Quite the contrary, 2 Peter’s focus on sexual misconduct goes beyond anything found in Jude. However, like 2 Peter, Jude fits within a line of Rabbinic interpretation 157

Interestingly, the author fails to comment on Lot’s relationship with his daughters, noting only that Lot was righteous and tormented by this righteousness during his time in Sodom. 158 Cf. Bo Reicke, The Epistles of James, Peter, and Jude, AncB, vol. 37 (New York: Doubleday, 1964), 169–70. 159 Namely Noah (2 Pet 2:5) and Lot (2:7).

262

The Bible’s Balaam beyond Numbers 22–24

that understood Balaam negatively and greedily,160 in this case listing him together with Cain and Korah. Nonetheless, the allusion here remains apparently focused on Numbers 22–24; nothing here suggests a relationship or regard for any other biblical Balaam tradition.

4.10.5 Revelation 2:14 Revelation’s reference to Balaam is somewhat more difficult than the others in the New Testament. Here, the text regards Balaam as Balak’s teacher, who instructed Israel to eat food offered to idols and engage in fornication. Clearly Num 31:16, combined in some inexplicable fashion with Numbers 22–24 and 25 stands behind the text of Revelation 2.161 Each of these elements would be necessary for the reconstruction of every element in Rev 2:14: the interaction with Balak probably stems from Numbers 22–24;162 the idolatry and fornication stem from Num 25:1–5; and Numbers 25 must have been understood in light of Numbers 31 in order to affiliate either of these characters with it, since neither of them appear in Numbers 25.163 Taken together, these elements combined to make Balaam the prototypically bad teacher.164 Unlike the other biblical traditions about Balak and Balaam, however, Rev 2:14 implicates Balak 160

Cf. Bigg, Epistles of St. Peter and St. Jude, 332; Reicke, James, Peter, and Jude, 205–8; Bauckham, Jude, 2 Peter, 81–83; Hubert Frankemölle, 1. Petrusbrief, 2 Petrusbrief, Judasbrief, NEchtB, vol. 18–20 (Würzburg: Echter-Verlag, 1987), 137; and Donelson, I & II Peter and Jude, 186. 161 Cf., e.g., Robert Henry Charles, A Critical and Exegetical Commentary on the Revelation of St. John, Volume 1, ICC (Edinburgh: Clark, 1920), 63–64; Hubert Ritt, Offenbarung des Johannes, NEchtB, vol. 21 (Würzburg: Echter-Verlag, 1986), 28; and Jan Willem van Henten, “Balaam in Revelation 2:14,” in The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, ed. George H. van Kooten and Jacques van Ruiten, TBN, vol. 11 (Leiden: Brill, 2008), 247. 162 Though Deuteronomy 23; Joshua 24; and perhaps even Micah 6 could have served, since only the names of the characters and the fact that they interacted matter. 163 van Henten, “Balaam in Rev 2:14” identified these as specific concerns from the author of Rev 2:14, with fornication implying taboo sexual relationships or idolatry, and the offerings contravening Jewish dietary restrictions. However, “fornication” in this context could also allude abuse of power and exploitation. 164 Cf. Robert H. Mounce, The Book of Revelation, NICNT (Grand Rapids, MI: Eerdmans, 1998), 80–81. The apparent heresy addressed here is that of the Nikolaitans; cf. Josephine Massynberde Ford, Revelation. Introduction, Translation, and Commentary, AncB, vol. 38 (Garden City, NY: Doubleday, 1975), 400. It is possible that “Balaam’s teaching” developed as a technical term for eating meat sacrificed to idols and fornication; cf. Hermann Lichtenberger, Die Apokalypse, Theologischer Kommentar zum Neuen Testament, vol. 23 (Stuttgart: Kohlhammer, 2014), 97. At the same time, the reference to Balaam here could legitimately reflect the name of the teacher of the opposition, but it certainly appears to at least have a specific teacher in mind; cf. Brian K. Blount, Revelation. A Commentary, NTL (Louisville, KY: Westminster John Knox Press, 2009), 58–59.

4.11 Conclusions

263

personally for the events recounted in Numbers 25, even though that text does not mention him explicitly. Thus, Rev 2:14 attests something new in the Balak tradition: only in this instance is he made culpable for the Israelite apostasy, presumably at Peor, although this name remains conspicuously absent from that chapter. Still, Balaam stands behind him in Rev 2:14 and was ultimately responsible in the eyes of the author, in that he provided Balak with a solution to draw the Israelites into disfavor with their God. Like Jude and 2 Peter, this verse in Rev 2:14 has a slight parallel with a Jewish interpretation of the text.165 In Josephus’ Ant. 4.129 Balaam explicitly instructs Balak (and the Midianites) to send women to lead Israel astray. In the subsequent material, however, Balak alone does not send the women. Rather the Midianites – without Balak – send the women to lead the Israelites astray. The common element, at least, is Balaam’s instructing Balak to send the women. That can be found in both Rev 2:14 and Ant. 4.129. So, while biblically speaking, this datum is unique to Revelation, parabiblical Jewish literature, in this case Josephus, attests a parallel. This makes the Revelation text like the other New Testament references to Balaam. Other biblical literature evinces no identical claims, but other extrabiblical materials demonstrate similar understandings and apparently a common background and interpretational tradition. Revelation, like 2 Peter, covers a spectrum of biblical Balaam traditions, which distinguishes it, with 2 Peter, from other New Testament Balaam texts.

4.11 Conclusions 4.11 Conclusions

Several conclusions avail themselves from the preceding study. The distinctly negative attitude toward Balaam in virtually all of the biblical texts outside of Numbers 22–24 and Micah 6 stands out markedly. None of the other versions represents him in any positive light, contrasting them to the oldest core of the version of the Balaam tradition found in Numbers 22–24. Micah, if not somewhat positive towards him, could be understood as neutral towards him. Perhaps this impugning of Balaam’s character in the other biblical traditions – here naturally referring only the traditions within the Hebrew corpus – influenced or paralleled the redaction history of the text now found in Numbers 22– 24. This possibility will be explored further below. While the other biblical traditions remain generally distinct from Numbers 22–24, one notes at the same time that these traditions bear only limited resemblance to each other. Of course there are exceptions: Nehemiah’s excerpting from Deuteronomy 23 and the list of deceased Midianites with Balaam in Numbers 31 and Joshua 13. The exceptions, however, indicate scribal copying more 165

Cf. David Edward Aune, Revelation 1–5, WBC, vol. 52A (Waco: Word Books, 1997), 187–88.

264

The Bible’s Balaam beyond Numbers 22–24

than anything else. The historical claims of the different traditions do not match, even sometimes within what we currently regard as a single book, as in Joshua. The various traditions can be placed into families with similar claims and perhaps even chronologically organized on this basis. The text that least contradicts Numbers 22–24* in its oldest form is Mic 6:4– 5. This Micah text presumes one primary element that matches Numbers 22– 24*, but none of the other Balaam traditions. Micah 6 recounts, namely, that Balaam and Balak disagreed, or at least that Balaam apparently did not go along with what Balak demanded. Micah places this conversation during a period when the people of YHWH (cf. Mic 6:3: “my people”) had left Egypt and were still at Shittim in Transjordan, before they had arrived in Cisjordan at Gilgal. This itinerary, which matches that found in the first redaction identified in the Numbers, could imply the Micah text’s familiarity with that version of the Balaam story and the eisodus account. One other text that has a similar element to Numbers 22–24* is Judges 11:25. It presumes – like Numbers 22– 24* – that Balak did not engage Israel militarily after they had conquered the land of the Amorites. The contextualization among the Amorites suggests something like the first redaction of the Balaam tradition in Numbers 22–24*, in which Israel engaged Sihon (but not Og) militarily before Balak undertook his attempt to curse them, but never attacked them. However, unlike Numbers 22–24, Balaam plays no role whatsoever in Judges 11. The case differs with Joshua 24, which reports that a bellicose Balak battled the exodus generation before beckoning Balaam, which stands in direct contradiction to both the Judges text and Numbers 22–24*. At the same time, the Joshua texts introduced new elements: YHWH’s willingness – or more accurately YHWH’s lacking the willingness – to destroy Israel, but rather saving them from Balaam. Probably an older version of Josh 24:9–10 even concluded with the idea that YHWH delivered Balak and Balaam into Israel’s hands. The military overtones are hard to miss and distinct from all earlier iterations in the Bible. Joshua 24 thus evinces for the first time a more negative attitude toward Balaam. Otherwise the circumstance in the chapter as a whole commend regarding it as familiar with R1 in Numbers 22–24. At least, no concrete element of R2 appears in Josh 24:9–10. Perhaps Joshua 24 presents the earliest impugning of Balaam’s character, opening the door for others. Numbers 31 presents precisely such a defamation of his character. Because Balaam encouraged the Midianite women to lead the Israelites astray at Peor (v. 16), he merited death and was summarily executed with the Midianite rulers (v. 8), upon whom the Israelites also avenged themselves. These references reflect attitudes like, and suggest knowledge of the second (R2) and third redactional layers (R3+), respectively, identified in Numbers. Balaam’s demise in Numbers 31 presents one of the worst attitudes toward him in the Hebrew Bible. No reference is made to his interaction with Moab or to his potentially blessing Israel – whether it had been his intent or not. Nothing mentions his

4.11 Conclusions

265

predicting the downfall of Moab and other enemies of Israel at the hands of some Israelite monarch in the future. With the end of his life recounted in Numbers 31, Balaam had no chance for redemption. This text irreparably recasts this character in a negative mold and permanently colors any subsequent understanding about him. The copying of the data from Num 31:8 into Josh 13:21–22 goes even further, identifying him unambiguously as ‫הקוסם‬. That presents an even deeper nadir than Numbers 31 for his characterization.166 Similarly, Deuteronomy 23 describes distinct elements within the Balaam tradition. For the first time the reader encounters the idea that YHWH turned his curse, which Deuteronomy 23 presumes he uttered, into a blessing. This transformation cannot be found in any tradition other than Deuteronomy (and Nehemiah, which relied on some version of it). Likewise, Deuteronomy 23 makes no explicit reference to Balak, combines the Moabites with the Ammonites, and makes Balaam a Mesopotamian. Comparing Deuteronomy 23 with Nehemiah 13 (and even Joshua 24) suggests that Deuteronomy 23 may have been edited on a few occasions. Perhaps later hands added Balaam’s geographical background, patronymic, and the reference to YHWH’s refusal to listen to Balaam. The first two could have been introduced from Numbers 22–24* (though Balaam’s Mesopotamian background would have had to been created from disparate elements in Num 22:5 and 23:7), while the last element could have been introduced from an earlier version of Joshua 24, one that G of this chapter still reflects to a greater degree. Yet, while these texts (particularly Deuteronomy 23 and Nehemiah 13, but also Joshua 24) demonstrate something of an antiMoabite character, they are not particularly negative toward Balaam. The Moabites (and the Ammonites in Deuteronomy and Nehemiah) hired Balaam and he merely undertook their demand. These texts still permit him to bless, even if his blessing was created from an attempted curse. At most Balaam was guilty by association and Israel’s God ignored or averted his attempt. Such a presentation of Balaam, however, does not mitigate or particularly contradict the more negative opinions of this figure. Guilty by association still appears enough to make him guilty for later reception. The negative attitude about Balaam most broadly influenced later Jewish and early Christian understandings of his character. These can be found in the writings of the Rabbis and in Josephus, even influencing the Medieval Masoretic accent tradition: “The Masoretic accentuation [of Num 24:14] is disjunctive, indicating that the information itself was omitted. Rabbinic tradition, in the light of 31:16, supplied Balaam’s information: 166

In general, this tendency to move from a positive characterization toward negativity about Balaam matches the conclusions of Levine, Numbers 21–36, 453–54. However, I think that Num 31:8 (and Josh 13:22), which promulgate Balaam’s merited slaughter present the pinnacle of negative attitudes towards Balaam, as opposed to only one rung in an increasingly negative latter of depreciation.

266

The Bible’s Balaam beyond Numbers 22–24

‘Prepare inns and employ seductive women to sell food and drink at lowered prices. And these people [Israel] will come eat, drink, and have sexual intercourse with them; and they will reject their God. And they will be delivered into your hand shortly, and many of them will fall.’ [Targ. Jon.; cf. Sanh. 106a] Other midrashim develop this theme into a total abandonment of Jewish life. [Philo, 1 Mos. 294–301; Josephus, Ant. 4.126–130; TJ Sanh. 28cd].”167

They dominate his image in the New Testament as well. In these writings, Balaam’s blessing aspect has been completely lost. They fully reject him as an ally of the people of YHWH, and ascribe to him greed and licentiousness. These assertions completely fail to conform to the portrait Numbers 22–24 painted of Balaam. “In the biblical narrative, Balaam foretells the destruction of Moab by Israel, but Jewish tradition presents him as advising the king how to outwit the Israelites by inducing them to sin against God.”168 Rather, the combination of other traditions from texts beyond Numbers 22–24 led to this defamation of his character and thus reversed the earliest biblical understanding of Balaam. “Die solcherart [negative] belastete Nachgeschichte der Bileamgestalt im Judentum und im Christentum aber geht nicht in erster Linie von Num 22–24 aus, sondern alttestamentlichen Notizen, die sich außerhalb dieses Komplexes finden, und zwar hauptsächlich von Dtn 23,5 f. und Num 31,8.16.”169

In other words: “Tradition records in detail how Balak followed Balaam’s counsel, and thereby establishes a direct link between chapters xxiv and xxv. Israel’s debauchery was due to Balaam’s advice to the king, and caused the death of twenty-four thousand people. So although Balaam was prevented by God from cursing Israel, he nevertheless obtained what he sought, the destruction of many of its sons.”170

The general acceptance of this more negative image within the later tradition, with essentially no attempt to salvage his positive characterization, could suggest the chronological development of an increasingly negative attitude toward Balaam. From Numbers 22–24* and Micah 6*,171 through texts like Joshua 24, we ultimately arrive at Numbers 31 and Deuteronomy 23. “Later tradition acknowledges almost nothing of Balaam the obedient servant of the Lord, who could not be bribed by all the wealth of Moab. He is, instead, the archetypal enemy of Israel, a Pharaoh or Haman, whose power would threaten to annihilate Israel were it not for the intervention of Israel’s God. Yet both traditions, the saint and the sinner, have their roots in Scripture, indeed, in these very chapters of the ‘Book of Balaam.’”172

167

Milgrom, Numbers, 206. Vermes, Scripture, 162. 169 Donner, “Balaam pseudopropheta,” 113. 170 Vermes, Scripture, 164. 171 Balaam’s absences in Judg 11:25 makes its inclusion in this spectrum superfluous. 172 Milgrom, Numbers, 471. 168

4.11 Conclusions

267

Similar or identical scribes may have simultaneously edited this increasingly negative characterization back in to subsequent versions of Numbers 22–24. Possible indications of this would be the similar nomenclature for the group that should have been cursed. For example, Josh 24:9 refers to Balak’s going out to battle “Israel” (= R1), whereas Deut 23:4 identifies them simply as “you” or even – implicitly – as ‫ =( קהל יהוה‬R3+). Nehemiah 13 on the other hand, while referencing the ‫קהל האלהים‬, calls the exodus group ‫בני ישׂראל‬. This phraseology matches Numbers 31, which also identifies them as ‫ =( בני ישׂראל‬R2, particularly in 31:16a), while at the same time identifying the group at Peor as ‫ =( עדת יהוה‬R3+, 16b). One recognizes the same tendencies in the nomenclatures of the redactional layers of Numbers 22–24. The first redaction includes the term Israel, which matches the Balaam tradition identified in Joshua 24 (and Judges 11:25). The second (and possibly later) redaction(s) use ‫בני ישׂראל‬, matching Numbers 31:16a and Nehemiah 13. The oldest core of the story in Numbers 22–24 refers to the exodus group merely as “the people”, which is most similar to Micah 6:3–5, a text in which YHWH addresses them as “my people”. The tendencies of these texts also favor a connection deeper than the simple nomenclature. The Micah text maintains a disposition similar – or at least not contradictory – to the oldest core of the Balaam narrative in Numbers 22–24*, whereas Numbers 31 contains a substantially more negative view of Balaam, precisely as found in the later additions to and editions of Numbers 22–24, as in, e.g., Num 22:22–35. Thus, one has a spectrum of texts and their inherent tendencies toward Balaam that can be chronologically oriented, from a more favorable disposition to a substantially negative reevaluation. But, as a control, the opposite situation should be considered: could it not be the case that a more negative view of Balaam became more positive over time?173 What precludes such a reconstruction? The primary argument against this interpretation can be found – in my opinion – in the early reception history. The Rabbis, Josephus, and early Christians tended to look unfavorably, even hostilely upon Balaam. Apparently the negative views of Balaam became the dominant views of Balaam. Perhaps this could even explain how a text like Num 24:17 became so important in messianic interpretations, but at the same time lost the ascription to its “author”. One notes that of the Qumran texts that use this passage, only one of them ascribes it to Balaam.174 The same might be said of Matthew 2 in the New Testament: should Num 24:17 actually stand in the background of the prophecy of the star (and Num 24:17 presents the most likely known candidate), it remains conspicuous that the author made no such citation or ascription. The absence of references to Balaam could be indicative 173 David Frankel, “The Deuteronomic Protrayal of Balaam”, e.g., ultimately adopted this, in my opinion quite untenable, position. 174 Cf. CD (= Damascus Document) VII, 18–21 and 1QM (= War Scroll) XI, 6–7, which do not mention Balaam, but 4QTestimonia (= 4Q175), which does.

268

The Bible’s Balaam beyond Numbers 22–24

of an interest in the content, without wanting to affirm its supposed “author”. Similar evidence of hesitation about the prophet Balaam, while yet approving of elements of the prophecy itself, appears in the writings of the Church Fathers.175 Another factor suggests that the positive impression of Balaam was the oldest, with the negativity developing over the course of time. Quite simply, only one version clearly reflects positively on Balaam, and editors have conspicuously expanded it on a few occasions to include more negative characterizations of this figure. Were the opposite true, that Balaam was initially hated and then only later liked, we should have expected to find some diachronic evidence for such a postulation. Such is simply not the case. Later scribes and tradents seemed to have only emended the negative written traditions in order to make them more consistent, i.e., more consistently negative. But this is true even for the Balaam traditions beyond Numbers 22–24. For example, Deut 23:4–6 appears to have been expanded by a number of phrases, particularly when contrasted with Josh 24:10 and even compared to Neh 13:1–2. Joshua 24 underwent editing as well, changing it from Joshua’s speech to YHWH’s; as part of this process, the references to Balaam were edited slightly, but never in such a fashion that would grant him a more positive reception. Bearing the reception-historical and even text-historical evidence in mind, it remains most unlikely that a negative impression of Balaam became more positive over time. Rather, the alternative must have been the case; through the course of biblical redaction and transmission, Balaam became decreasingly well-received. Therefore we can conclude with Donner: “Denn wie literarisch aus dem negativen Bileam, den die Israeliten schließlich erschlugen, der positive jahwetreue Seher geworden sein sollte, ist schlechterdings nicht auszudenken. Der umgekehrte Weg jedoch ist möglich: so nämlich, daß man spätestens in nachexilischer Zeit die klassische Bileampreikope Num 22–24 unter negativem Vorzeichen las und deutete.”176

In this way, the development of the traditions about Balaam in the Hebrew Bible generally seem to match the processes in the development of Numbers 22–24, particularly regarding the characterization of Balaam. Yet, the images of Balaam strewn throughout the Hebrew Bible also generally affirm the redaction-historical image of Numbers identified in Chapter Three. Micah appears familiar with either the oldest version or – more likely – that of the first redaction identified in Chapter Three. The attribution of Micah 175 For a brief summary of some Patristic evidence and how they dealt with issue surrounding Balaam’s character, cf. Judith R. Baskin, “Origen on Balaam: The Dilemma of the Unworthy Prophet,” VC 37, no. 1 (March 1983): 23–25. For some minor insight on the Syriac tradition regarding Balaam, cf. Jacques-M. Vosté, “Les Oracles de Balaam d’après Mar Išoʿdad de Merw (c. 850),” Bib 29, no. 3 (1948): 169–94. 176 Donner, “Balaam pseudopropheta,” 120.

4.11 Conclusions

269

to some Deuteronomistic composition, like a book of Four Prophets, would favor the first redactional stage of Numbers as that with which it was familiar. Nothing speaks against Judges 11 knowing either the oldest version or that of the first redaction either. Balaam’s absence raises questions as to whether he was always part of the narrative of Balak’s conflict with Israel. Yet, no such version could have been reconstructed in Numbers 22–24. What speaks in favor of some knowledge of the oldest version or that of the first redaction without the impact of later additions is the anticipated answer that Balak did not engage Israel militarily, something that does begin to find expression with the version in Joshua. The ending of Joshua may also present some portion of Deuteronomistic editing as well, though this text does not demonstrate the same more or less positive attitude toward Balaam found in the redacted text of Numbers at this stage.177 That would imply that Joshua 24 could evince some post-Deuteronomistic editing in this chapter that began the transition towards the negative image of Balaam that came to dominate the biblical text. The version of Balaam in Deuteronomy 23 also shares some elements with the version in Joshua 24, but also seems to attest a further thematic development. The addition of this Balaam material in a context about the rules for joining the ‫ קהל‬in Deuteronomy 23 could evince some relation to the third or a later redactional layer. Nehemiah demonstrates familiarity with this version, but this could have occurred at any point after the development of Deuteronomy 23. Each of these versions – Joshua 24 and Deuteronomy 23 – further developed the negative attitude towards Balaam. Finally, Balaam’s appearance in Numbers 31 and Joshua 13 must belong to the latest redactional stages, certainly after the incorporation of priestly material with the Deuteronomistic composition. With these layers of supplemental material about Balaam, the negative understanding of Balaam reached its full potential and greatest fruition. At the latest, by this point in the development of the Pentateuch, there was no going back to some more idealized, positive version of this seer. With that, we can recognize that the same layers and ideas about Balaam are strewn throughout the Bible as within Numbers 22–24 alone. Each of these traditions can be dated to roughly their parallel attitudes found in Numbers 22–24, with the oldest core as the product of the northern monarchy and the latest layers demonstrating affinity with Holiness and Theocratic elements from post-priestly circles. Similar groups presumably stood behind

177 Cf. David Frankel, “The Deuteronomic Protrayal of Balaam,” 38, who regarded the Deuteronomistic material about Balaam as the transition to more negative portrayals of this figure, and Vuilleleumier, “Bileam Zwischen Bibel und Deir ʿAllā”, who regarded the donkey story as the beginning of Balaam’s downward slide.

270

The Bible’s Balaam beyond Numbers 22–24

these processes. Having concluded the survey of biblical literature about Balaam in its literary-historical development, we can turn to Balaam outside of the Bible as attested in one epigraphic source from the ancient Orient.

Chapter 5

An Inscription from Deir ʿAlla. Balaam in Transjordan 5.1 Introduction 5.1 Introduction

The study of the ancient figure Balaam gained new impetus on March 17th, 1967, when a Dutch team of excavators discovered several fragments of inscriptions in Tell Deir ʿAlla.1 This site has often been identified with biblical Sukkoth,2 and appears to have hosted a village from the Late Bronze Age into the Iron Age. 3 These epigraphic fragments had been written with ink on a smoothed plaster surface attached to underlying bricks or perhaps to some other figure, such as a stele, attached to the wall. No elements of the inscription were found still attached to the wall or figure upon which they had been written. At some point after the inscription of this writing on the wall, as it were, an earthquake destroyed much of the building in which it was housed, separating the plaster from the wall and scattering it about on the floor. Yet, in spite 1 Cf. the anecdote about the discovery in Gerrit van der Kooij and Moawiyah M. Ibrahim, eds., Picking up the Threads… A Continuing Review of Excavations at Deir Alla, Jordan (Leiden: University of Leiden Archaeological Centre, 1989), 63–64. The editio princeps appeared as Hoftijzer and van der Kooij, ATDA. These, however, were not the first written materials discovered in Deir ʿAlla; cf. Hendricus Jacobus Franken, “The Stratigraphic Context of the Clay Tablets Found at Deir ʿAlla,” PEQ 96 (1964): 73–78. 2 The identification of Tell Deir ʿAlla with Sukkoth goes back to Selah Merrill, “Identification of Succoth and Penuel,” BSac 34, no. 136 (October 1877): 742–54. He particularly based this on the Talmudic name Terʿalah, which he saw as essentially a cognate spelling of Deir ʿAlla. The first criticism of this identification appeared in J.A. Paine, “Succoth and Penuel not Yet Identified,” BSac 35, no. 139 (July 1878): 492–97, who demonstrated that the transition from Talmudic Hebrew name to the Arabic name is not without problems. J. Garrow Duncan, “Notes on the Sites of Succoth and Penuel as Bearing Upon the Routes of Gideon and Jacob,” PEQ 59, no. 2 (April 1927): 89–96; J. Garrow Duncan, “Notes on the Sites of Succoth and Penuel as Bearing Upon the Routes of Gideon and Jacob (Concluded),” PEQ 59, no. 4 (October 1927): 188–91; and Hendricus Jacobus Franken, Excavations at Tell Deir ʿAllā, Volume 1: A Stratigraphical and Analytical Study of the Early Iron Age Pottery, DMOA, vol. 16 (Leiden: Brill, 1969), 4–8; and H. Neil Richardson, “Skt (Amos 9:11): ‘Booth’ or ‘Succoth’?” JBL 92, no. 3 (September 1973): 381 rejected this identification. Nonetheless, the identification of Deir ʿAlla with Sukkoth remains common in the literature. 3 Cf. Zeidan Kafafi and Gerrit van der Kooij, “Tell Dēr ‘Allā During the Transition from Late Bronze Age to Iron Age,” ZDPV 129, no. 2 (2013): 121–31.

272

An Inscription from Deir ʿAlla. Balaam in Transjordan

of the poor condition in which they found them, the discoverers were able to cast the plaster into new forms and publish a first edition of the texts and the circumstances of their discovery in 1976.4 Hence, they have become known as the Deir ʿAlla Plaster Texts (DAPT).5 In this first critical edition, the editors combined many of the fragments into larger groups, which they called “Combinations”, numbering them individually into about fifteen groups. Two of these combinations, the largest, were identified as the cores of two longer texts to which many of the smaller pieces must have belonged. These two main texts were identified as Combination I and Combination II, with Combination I having been inscribed above Combination II.6 While the initial editors considered these two combinations likely components of a longer text or elements of a series about a common figure,7 this view has since lost favor and fallen out of fashion.8 Grounded in observations about the form and content of the texts, this rejection of their common horizon and relationship to a common figure is probably accurate: Combination I consists of a tale about a seer explaining his vision to his people, whereas Combination II transmits a wisdom text.9 Enough about the distinctions in content, in terms of form, Combination II is clearly much longer than Combination I, but is also physically wider than Combination I. While Combination I covers an area of roughly 32 x 32 cm, Combination II is almost 40 cm wide and some 70 cm long, at least according to contemporary reconstructions.10 This difference – 4 Cf. the description in Jacob Hoftijzer, “The Prophet Balaam in a 6th Century Aramaic Inscription,” BA 39, no. 1 (March 1976): 11–17, as well as the first publication in Hoftijzer and van der Kooij, ATDA. 5 Since this study concentrates on combination A, I use the abbreviation DAPT also to refer to that inscription alone. 6 Cf. the reconstruction in Kooij and Ibrahim, Picking up the Threads, 64. 7 Cf., e.g., Hoftijzer and van der Kooij, ATDA, 280 and still in Kooij and Ibrahim, Picking up the Threads, 64–65. 8 Cf. particularly Erhard Blum, “Die Kombination I der Wandinschrift vom Tell Deir ‘Alla: Vorschläge zur Rekonstruktion mit historisch-kritischen Anmerkungen,” in Berührungspunkte: Studien zur Sozial- und Religionsgeschichte Israels und seiner Umwelt. Festschrift für Rainer Albertz zu seinem 65. Geburtstag, ed. Ingo Kottsieper, Rüdiger Schmitt, and Jakob Wöhrle, AOAT, vol. 350 (Münster: Ugarit-Verlag, 2008), 573–601; Erhard Blum, “‘Verstehst du dich nicht auf die Schreibkunst..?’ Ein weisheitlicher Dialog über Vergänglichkeit und Verantwortung: Kombination II der Wandinschrift vom Tell Deir ‘Alla,” in Was ist der Mensch, dass du seiner gedenkst (Psalm 8,5). Aspekte einer theologischen Anthropologie. Festschrift für Bernd Janowski zum 65. Geburtstag, ed. Michaela Bauks, Kathrin Liess, and Peter Riede (Neukirchen-Vluyn: Neukirchener Verlag, 2008), 33–53; and Erhard Blum, “Die aramäischen Wandinschriften von Tell Deir ʿAlla,” in Texte aus der Umwelt des Alten Testaments. Neue Folge Band 8. Weisheitstexte, Mythen und Epen, ed. Bernd Janowski and Daniel Schwemer (Gütersloh: Gütersloer Verlagshaus, 2015), 459–74. 9 Cf. the introductory material to each combination in Blum, “Kombination I” and Blum, “Kombination II”. 10 Cf. Blum, “Kombination II,” 34.

5.1 Introduction

273

particularly in breadth – could easily be understood as a literally superficial indication that the texts remain distinct compositions. It’s hardly likely that one contiguous composition took on two distinct physical manifestations.11 Shortly after the initial publication, scholars began adding smaller combinations or even single fragments from combinations into the two main texts, particularly to Combination I. They also pulled the component fragments of Combination I closer together. For example, already in 1977 – one year after the initial publication – Caquot and Lemaire proposed moving Fragments Ic and Id up two lines, bringing them closer to Fragments Ia and Ib. They also joined Fragments VIIId and XIIc, placing this joined piece between Fragments Ia and Id.12 This insertion and readjustment of the pieces has found broad consensus.13 Weippert and Weippert moved Ic closer to Id, reducing the lacuna in line 6 to only about two letters.14 Lemaire added (from right to left) Fragments IIIf, IIIh, IIIe, and IIId to the beginning of line 1 and Fragment VIIa at the beginning of line 11. 15 These transformations have also since found broad acceptance. Lipiński proposed inserting Fragments Vc and Vd below Fragment Ic,16 which Blum – the foremost scholar still currently publishing on the inscriptions – has also accepted.17 Since Lipiński’s additions, Blum has added several pieces beginning at line 9 and continuing to the end. These include Vb, Ve, Vg–h, Vj, and Vk–n (all connected to Vd), Va, Vf, and Vi (to the right of Vc), and Vq at the beginning of lines 16–17, as well as a variety of fragments from Combinations IV, VI, VIII, XIII.18 One can accept most of these insertions as accurate, 11 Contra Giovanni Garbini, “L’iscrizione di Balaam Bar-Beor,” Hen 1 (1979): 171–72, who argued that Combination I flowed contiguously into Combination II. Cf. Müller’s critique in Hans-Peter Müller, “Die aramäische Inschrift von Deir ʿAlla und die älteren Bileamsprüche,” ZAW 94 (1982): 231. This also contradicts the reconstruction in Meinert Dijkstra, “Is Balaam Also Among the Prophets?” JBL 114 (1995): 43–64, who read both combinations as an essentially contiguous whole. 12 Cf. André Caquot and André Lemaire, “Les Textes Araméens de Deir ʿAlla,” Syria 54 (1977): 193–94. 13 Cf., e.g., already Rofé, Book of Balaam, 62; P. Kyle McCarter, “The Balaam Texts from Deir ʿAllā: The First Combination,” BASOR 239 (1980): 51 and Jo Ann Hackett, The Balaam Text from Deir ‘Allā, HSM (Chico, Calif.: Scholars Press, 1980), 21. More recent studies have followed suit; cf. Edward Lipiński, Studies in Aramaic Inscriptions and Onomastics II, OLA (Leuven: Peeters, 1994), 113–15, as well as Blum, “Kombination I” and Blum, “TUAT.NF.”. 14 Cf. Manfred Weippert and Helga Weippert, “Die ‘Bileam’-Inschrift von Tell Dēr ‘Allā,” ZDPV 98 (1982): 82. Republished as Manfred Weippert and Helga Weippert, “Die ‘Bileam’Inschrift von Tell Dēr ‘Allā,” in Jahwe und die anderen Götter, Manfred Weippert, FAT, vol. 18 (Tübingen: Mohr Siebeck, 1997), 136. 15 Cf. Andre Lemaire, “Les inscriptions de Deir ʿAlla et al littérature aramée antique,” CRAI 129, no. 2 (1985): 279. 16 Cf. Lipiński, Studies II, 115. 17 Cf. Blum, “Kombination I,” 576 and Blum, “TUAT.NF,” 466. 18 For the individual locations, cf. Blum, “TUAT.NF,” 466.

274

An Inscription from Deir ʿAlla. Balaam in Transjordan

with the exception of IVa, which cannot stand where Blum has located it.19 Due to the extensive compiling of these combinations and fragments, Blum has since proposed changing the nomenclature, referring to Combination A and Combination B.20 This helpful convention is followed here, except where explicit reference is made to older published versions. But how old are these literary works? The first short note about the archaeological season at Deir ʿAlla in 1967 suggested that they stemmed from the Persian Period.21 Naveh shortly thereafter suggested a date in the first half of the eighth century BCE.22 An initial paleographical analysis in an unpublished dissertation commended a dating around 800 BCE.23 However, a stratigraphic analysis of the pottery found in the layer of destruction (identified as M in the editio princeps, but later as Phase IX) commended a somewhat later date, probably about 750 BCE.24 A third analysis, still in the initial publication, but this time radio-carbon dating (C14), suggested that the paleographical analysis was closer to the actual date: a 66% chance of 800 BCE ± 70 years.25 A later study on the radio-carbon dating affirmed the dating around 800 BCE;26 this date has been accepted in the newest literature and editions of the text and will be favored here.27 While it cannot be known precisely when the aforementioned destructive earthquake occurred, Lemaire has suggested that it could have been the same earthquake mentioned in Amos 1:1. That would mean that it should be dated to around 760 BCE.28 While it would be exciting to correlate these texts in this 19

Cf. already Lipiński, Studies II, 115, who noted that Fragment IVa must have belonged to a now lost element of Combination II, and the following discussion. 20 Cf. Blum, “TUAT.NF,” 461 fn. 10. 21 Cf. Franken, “Texts”. 22 Cf. Joseph Naveh, “The Date of the Deir ʿAllā Inscription in Aramaic Script,” IEJ 17 (1967): 256–58. 23 Cf. the citation in the editio princeps, Hoftijzer and van der Kooij, ATDA, 12. 24 Cf. Hoftijzer and van der Kooij, ATDA, 12. Cf. also Hendricus Jacobus Franken, “Deir ʿAlla Re-Visited,” in The Balaam Text from Deir ʿAlla Re-Evaluated: Proceedings of the International Symposium Held at Leiden, 21–24 August 1989, ed. Jacob Hoftijzer, Gerrit van der Kooij (Leiden; New York; Copenhagen; Cologne: Brill, 1991), 9: “…It would be very difficult to date this pottery [used for dating the stratum of the find] later than the first half of the eighth century B.C. It could be earlier but not later.” 25 Cf. Hoftijzer and van der Kooij, ATDA, 16. 26 Cf. Moawiyah M. Ibrahim and Gerrit van der Kooij, “The Archaeology of Deir ʿAlla Phase IX,” in The Balaam Text from Deir ʿAlla Re-Evaluated: Proceedings of the International Symposium Held at Leiden, 21–24 August 1989, ed. Jacob Hoftijzer, Gerrit van der Kooij (Leiden; New York; Copenhagen; Cologne: Brill, 1991), 16–29 and Monique E. Vilders, “The Stratigraphy and the Pottery of Phase M at Deir ʿAlla and the Date of the Plaster Text,” Levant 24 (1992): 187–200. 27 Cf., e.g., Lipiński, Studies II, 104; Blum, “Kombination I”; and Blum, “TUAT.NF.”. 28 Cf. Lemaire, “Les inscriptions de Deir ʿAlla et al littérature aramée antique,” 272. Robert Wenning and Erich Zenger, “Heiligtum ohne Stadt – Stadt ohne Heiligtum? Anmerkungen zum

5.1 Introduction

275

way, doing such cannot escape the realm of speculation. After all, the region experienced substantial tectonic activity throughout its history.29 Van Seters disavowed Lemaire’s much older dating of the inscription, or rather the text from which it was copied,30 instead he related the inscription to the specific time and circumstances in Ammon.31 At any rate, it must have occurred sometime in the 8th century BCE, though it remains impossible to say more than that.32 The influence of the Aramaic language suggests that this area stood under their dominion during the time of composition. That would in turn suggest a dating not too late in the first half of that century.33 Fascinating, perhaps, but why should this interest anyone involved in biblical studies? In the fifth line of the portion the editors identified as Combination I stands the name ‫“( בלעם‬Balaam”). Just before, the second line of Combination I reads the phrase ‫“( ברבער‬son of Beor”, written together, in Aramaic) following an ‫“( מ‬m”) after a lacuna. Combining this evidence with the phrase ‫בר‬.‫“( ובלעם‬and Balaam son of”) from line 2 of Fragment d of Combination VIII, one can readily reconstruct the Aramaic form of the name of the figure Balaam son of Beor (‫ )בלעם בר בער‬known from the Hebrew Bible ( ‫בלעם בן‬ ‫)בעור‬. Further study of the textual fragments found in Deir ʿAlla demonstrated that this Balaam had – at the very least – some superficial similarities with the figure identified in the Bible, particularly of that figure as presented in Numbers 22–24. Nonetheless, these similarities should not be overstated.34 archäologischen Befund des Tell Dēr ʿAllā,” ZAH 4 (1991): 185 dated the destruction rather uncritically to the year 762 BCE based on Amos 1:1 and Zech 14:5. 29 Cf. David Horst Kallner-Amiran, “A Revised Earthquake-Catalogue of Palestine,” IEJ 1 (1950–51): 223–46 and David Horst Kallner-Amiran, “A Revised Earthquake-Catalogue of Palestine,” IEJ 2 (1952): 48–62. 30 Cf. Van Seters, Life of Moses, 411–12. 31 Cf. Van Seters, Life of Moses, 412–13. 32 The lack of conspicuous Assyrian influence as well as the stratigraphy of the pottery from Phase VI commend the conclusion of Phase IX within the eighth century BCE; cf. Ibrahim and van der Kooij, “The Archaeology of Deir ʿAlla Phase IX,” 27. Others leave the range more open, such as, e.g., Garbini, “L’iscrizione di Balaam Bar-Beor”, who dated it between the eighth and sixth centuries BCE. 33 Cf. Erhard Blum, “Die altaramäischen Wandinschriften vom Tell Deir ʿAlla und ihr institutioneller Kontext,” in Metatexte. Erzählungen von schrifttragenden Artefakten in der alttestamentlichen und mittelalterlichen Literatur, ed. Friedrich-Emanuel Focken and Michael R. Ott (Berlin: de Gruyter, 2016), 24. 34 Cf., e.g., Martin Rösel, “Wie einer vom Propheten zum Veführer wurde. Tradition und Rezeption der Bileamgestalt,” Bib 80 (1999): 507, who was – in my opinion – somewhat overanxious to identify these commonalities: “Es kann m.E. kein Zweifel daran bestehen, daß der biblische Bericht über Bileam auf diesen offenbar bekannten Gottesseher anspielt. Dafür sprechen neben anderem die auffällige Verwendung ‫ חזה‬im 3. und 4. Bileamorakel wie in der Deir ʿAlla-Inschrift (Zeile 1), die geographische Nähe des Inschriftfundes und die Lokalisierung des in Num 22–24 berichteten Geschehens im Jordangraben und die ebenfalls bemerkenswerte Übereinstimmung bei der Nennung der šadajin-Götter (Z. 8) und von ‫( שדי‬Num 24,4.16).”

276

An Inscription from Deir ʿAlla. Balaam in Transjordan

5.2 The Inscription’s Language 5.2 The Inscription’s Language

One would imagine that determining the language of an inscription like this would be easy, but this issue is still debated to some degree.35 Several features find consensus: it contains many elements that suggest an Aramaic background for the text. This covers both the forms and vocabulary apparent in the text.36 However, the inscription simultaneously demonstrates features and terminology more consistent with southern Canaanite or Hebrew.37 The most conspicuous examples of this “Canaanite” influence are ‫ראה‬, which is foreign to Aramaic, and the narrative or ‫ו‬-imperfect form, which was commonly regarded as a feature of southern Canaanite dialects like Hebrew and Moabite.38 However, since the discovery of the Tel Dan Inscription, it has become increasingly likely that the narrative form was known in Old Aramaic before dying out in later periods. Nonetheless, some of the Canaanite features cannot be so quickly disregarded. Others have also argued for Arabic influence; cf. particularly Lipiński’s argumentation and evaluation. 39 Still others have focused on the mixing of grammatical elements and the geographical location and have argued for calling it Gileadite: “What language or dialect are we dealing with then when we discuss the DAPT? Although Deir ʿAlla is not that distant from Biblical Ammon proper, there is now hesitation to use the term ‘Ammonite’ for it. Ammonite, as known from the limited corpus of material available, displays a different set of linguistic features. The term ‘Southern Canaanite’ is best preserved for the Canaanitisms in the Amarna correspondence and related texts from the second millennium. Although the comparisons that have been made by this writer and by others are perforce with Biblical Hebrew, the language of DAPT is clearly not Hebrew, as known from

35 Cf. the overview in Holger Gzella, “Deir ʿAllā,” in Encyclopedia of Hebrew Language and Linguistics, vol. 1, ed. Geoffrey Khan (Leiden; Boston: Brill, 2013), 691–93. 36 Cf. the catalogue of Aramaic-like features in P. Kyle McCarter, “The Dialect of the Deir ʿAlla Texts,” in The Balaam Text from Deir ʿAlla Re-Evaluated: Proceedings of the International Symposium Held at Leiden, 21–24 August 1989, ed. Jacob Hoftijzer, Gerrit van der Kooij (Leiden; New York; Copenhagen; Cologne: Brill, 1991), 87–89. 37 Cf. the listed features in McCarter, “The Dialect of the Deir ʿAlla Texts,” 89–90. Victor Sasson, “The Book of the Oracular Visions of Balaam from Deir ʿAlla,” UF 17 (1986): 284– 85 suggested even Ammonite. 38 This seems more appropriate than regarding them as archaic or vestigial elements of Aramaic that remains present due to the location’s regional isolation as Dennis Pardee, “The Linguistic Classification of the Deir ʿAlla Text Written on Plaster,” in The Balaam Text from Deir ʿAlla Re-Evaluated: Proceedings of the International Symposium Held at Leiden, 21–24 August 1989, ed. Jacob Hoftijzer, Gerrit van der Kooij (Leiden; New York; Copenhagen; Cologne: Brill, 1991), 105 suggested. 39 Cf. Lipiński, Studies II, 117–39.

5.2 The Inscription’s Language

277

the major literary works preserved in the Hebrew Bible. Taking geographic factors into consideration it may best be called Gileadite. It is a local dialect, close to the Canaanite of its time, sharing with it essential morphological, syntactic, phonological and lexical features.”40

At the same time, Blum in particular has argued for the exclusively Aramaic character of the text and identified it as indistinguishable from the Aramaic evinced by the inscription from Tel Dan.41 While there may be some similarities in syntax, I would note that significant differences cannot be overlooked. For example, ln. 3 of the inscription from Tel Dan attests the form ‫ יהך‬for the third person masculine singular imperfect of perhaps ‫הוך‬: “he went (repeatedly)”. However, in the DAPT, this verb in the imperative plural appears unambiguously from ‫ הלך‬in v. 21, a form unknown in and generally inconsistent with Aramaic. Should the root have been ‫ הלך‬in Aramaic, it would always be the ‫ ל‬that elided, precisely the opposite of what we find in Tell Deir ʿAlla. More importantly than the verbal forms are the letter forms. Of course, one has to take this with a grain of salt, since the medium inscribed was basalt in the case of the Tel Dan Inscription and plaster at Tell Deir ʿAlla. One was engraved and the other was applied with a stylus. That being said, the shapes of the letters ‫א‬, ‫ה‬, ‫ו‬, ‫י‬, ‫מ‬, ‫ק‬, and ‫ ת‬remain particularly distinct between the two inscriptions. It remains doubtful that one scribal school produced the scribes behind these writings. So, while there may be some relationship between the language and letters of the Tel Dan Inscription and Combination A of the plasters inscriptions from Tell Deir ʿAlla, the case should not be overstated. They share almost no common vocabulary (the prepositions ‫ מן‬and ‫ על‬and the narrative form ‫ויעל‬, perhaps the object suffix ‫)הם‬, making a comparison tentative at best. That being said, there is hardly any satisfactory answer to the question of the inscription’s language other than Aramaic with some southern influences on it.42 Terms like ‫ אתה‬for “come, walk” and ‫ פעל‬for “make, do” and the 40 Jonas C. Greenfield, “Philological Observations on the Deir ʿAlla Inscription,” in The Balaam Text from Deir ʿAlla Re-Evaluated: Proceedings of the International Symposium Held at Leiden, 21–24 August 1989, ed. Jacob Hoftijzer, Gerrit van der Kooij (Leiden; New York; Copenhagen; Cologne: Brill, 1991), 118. 41 Cf. Blum, “TUAT.NF,” 460; cf. also Stephen A. Kaufman, “Aramaic,” in The Semitic Languages, ed. Robert Hetzron (New York: Routledge, 1997), 114–15. 42 Cf. Baruch Halpern, “Dialect Distribution in Canaan and the Deir Alla Inscriptions,” in ‘Working with No Data’. Semitic and Egyptian Studies Presented to Thomas O. Lambdin, ed. David M. Golomb (Winona Lake, IN: Eisenbrauns, 1987), 119–39, who described the language of the inscription as part of a linguistic continuum, presenting elements of both Hebrew and Aramaic, without being a representative of either purely. Cf. Hans-Peter Müller, “Die Sprache der Texte von Tell Deir ʿAllā im Kontext der nordwestsemitischen Sprachen mit einigen Erwägungen zum Zusammenhang der schwachen Verbklassen,” ZAH 4 (1991): 2: “Am ehesten repräsentiert die in einer Randlage bezeugte Sprache der DAT [Deir ʿAlla Text] also den relativ späten Rest eines nordwestsemitischen Zustands vor der endgültigen Trennung des Aramäischen vom Kanaanäischen. Relativ ist das späte Datum eines solchen Rests insofern, als die

278

An Inscription from Deir ʿAlla. Balaam in Transjordan

relative particles ‫ זא‬and ‫( זי‬should this reflect an accurate reconstruction), as well as morphological phenomena like the plural ending ‫ן‬- (although also present in Moabite)43 and the suffixed preposition ‫ אלוה‬tend to favor Aramaic as the principle language of the inscription. However, it may not have been a pure form, perhaps instead a local dialect as indicated by the appearance of roots like ‫ הלך‬for “come, walk” and ‫ ראה‬for “see”. “These [linguistic] details show that while the Deir ʿAlla dialect resists classification as Aramaic or Canaanite in categorical terms, it fits well into its geographical context. It is remote from Phoenician. It is strongly linked to Hebrew and Moabite. It is also strongly linked to Aramaic. These associations are easy to understand in terms of dialect geography. Hebrew and Moabite were spoken and written in nearby or contiguous communities. The Aramaic region lay farther away to the north, but we know that Damascus exercised a substantial political and cultural influence on northern and central Jordan until the fall of Damascus in 732 B.C. The adoption of the Aramaic alphabet for writing Ammonite is powerful testimony to this influence.”44

In general we can regard it as a ninth or eighth century Aramaic inscription with elements of Canaanite influence. This could permit identifying it with the Aramean expansion into Transjordan under Hazael of Damascus, should the notice in 2 Kgs 10:32–33 be trustworthy.45 That seems a likely circumstance behind this inscription, since it appears that Damascus exercised some influence over the region during this period, probably during the reign of Hazael in the second half of the ninth century BCE.46 Both the language and the implied historical circumstances of this find in Transjordan could suggest an Aramaic origin for the figure Balaam.47 After all, an Aramaic inscription featuring this Datierung der Inschrift bislang umstritten bleibt, aber auch deshalb, weil die Frage kaum zu beantworten ist, wann die Trennung von Kanaanäisch und Aramäisch in Zentren wie den phönizischen Städten einerseits und den Herkunftsorten der bekannten früharamäischen Inschriften andererseits erfolgt ist.” 43 Cf. also the other examples of this phenomenon in Ernst Axel Knauf, “War ‘BiblischHebräisch’ eine Sprache? Empirische Gesichtspunkte zur linguistischen Annäherung an die Sprache der althebräischen Literatur,” ZAH 3 (1990): 16. 44 McCarter, “The Dialect of the Deir ʿAlla Texts,” 97. 45 This is precisely the dating that Blum strives to ascribe it; cf. Blum, “TUAT.NF,” 459– 60. 46 Cf. Gershon Galil, “The Boundaries of Aram-Damascus in the 9th–8th Centuries BCE,” in Studies in Historical Geography and Biblical Historiography Presented to Zecharia Kallai, ed. Gershon Galil and Moshe Weinfeld (Leiden; Boston, Cologne: Brill, 2000), 35–39 and Younger, Political History of the Arameans, 622–24. 47 Jo Ann Hackett, “Response to Baruch Levine and André Lemaire,” in The Balaam Text from Deir ʿAlla Re-Evaluated: Proceedings of the International Symposium Held at Leiden, 21–24 August 1989, ed. Jacob Hoftijzer, Gerrit van der Kooij (Leiden; New York; Copenhagen; Cologne: Brill, 1991), 83 has argued against appreciating an Aramean characterization of the figure Balaam on the grounds that the Aramaic in the text is inconsistent with Aramaic from around Damascus. However, regarding the DAPT as attesting Aramaic influence on the local language could resolve this issue. It wasn’t Arameans per se writing about Balaam at Deir

5.3 The Text, Its Reconstruction, and Its Translation

279

figure from a time in which Aram appears to have exercised hegemony over the region (or shortly thereafter) could easily commend such an interpretation. The later sections of this chapter, involving the inscription’s Sitz im Leben, will consider these issues.

5.3 The Text, Its Reconstruction, and Its Translation 5.3 The Text, Its Reconstruction, and Its Translation

We can reconstruct the opening of the inscription with some certainty.48 Line one consists of what can be regarded as the work’s title followed by a verbless clause before continuing on into two verbal sentences. The second of these verbal sentences ends in the inscription’s second line. The title identifies the work as “The Admonitions of the Book of Balaam” (v. 1). The subsequent participial sentence explains who Balaam was: “A man who saw the gods was he” (v. 2).49 Ln. 1

2

Text ‫[ב]ר בע[ר‬.]‫בל[עם‬.]‫[ס̊פר‬.]‫יסר̊י‬

V. 1

‫[הא‬.]‫אלהן‬.‫ח]ז[ה‬.‫אש‬.

2

‫בלילה‬.‫אלהן‬.‫אלוה‬.‫ויאתו‬ 2 ̊ ̊ [‫[ ̊ב]מ‬.‫ ̊א[ ̊ת]ו‬.] ̊. ‫ח ז ה‬ ̊ .‫אל‬.‫כשמא‬ ̊ .‫בר בער‬.‫ל]בלע[ ̊ם‬.‫̊ויאמרו‬ .‫כה‬ [.]̊‫יפעלה‬ .‫אח ראה‬.‫ז̊א‬

3 4

‫לרת̊ה‬.‫אש‬

8

‫[עת‬.‫טב]הם‬

9

5 6 7

Translation THE ADMONITIONS OF THE BOOK OF [BAL]AAM SO[N OF BEO]R. A MAN WHO S[AW] THE GOD[S] was he. And they came to him, the gods, at night. They came in a vision 2 as per El’s statement. And they spoke to Balaam thus: HE / SOMEONE / THEY WILL DO IT! THAT WHICH A BROTHER SEES / OF THE BROTHER’S SEEING! A MAN DID NOT PITY IT! (= NO ONE HAD PITY / COMPASSION) PREPARE THEM NOW! / PLEASE THEM NOW!

Ln. 1

2

ʿAlla. It was Transjordanians writing a Balaam legend from Aram that thus included elements of the Aramaic and autochthonous languages. 48 For the purposes of this publication, the inscription’s red passages are italicized in the Aramaic transliteration, small capitals in the English translation. 49 Such clauses occur with some regularity in Aramaic; cf., e.g., line 2 of the Zakkur Inscription (KAI 202), which contains no fewer than two, and Stanislav Segert, Altaramäische Grammatik mit Bibliographie, Chrestomathie und Glossar, 4th ed. (Leipzig: Verlag Enzyklopädie, 1990), 422. That the term ‫ חזה‬should be regarded as a participle seems most likely from the age of the text; otherwise one would expect ‫ חז‬for the perfect; cf. Segert, Altaramäische Grammatik, 297–301. This identification commends regarding the ‫ הא‬in the line as part of this clause. This interpretation remains not without its difficulties, since the color of the text changes at this point such that ‫ אלהן‬is red, but ‫ הא‬is black. Nonetheless, the syntax favors this understanding which alone provides a satisfactory explanation for the function of the ‫ הא‬in this line.

280 3

An Inscription from Deir ʿAlla. Balaam in Transjordan ̊ .‫מן‬.‫בלעם‬.‫ויקם‬ .‫מחר]ת‬ .[‫וה‬...‫]א‬.‫י ̊]ו[מן‬.‫י ̊[ל̊ל ̊ת‬.‫וילל‬.‫קם‬

10 11

̊ .‫ויצ[ ̊ם‬.‫ולי ̊כ]ל‬

12

̊ .‫ ה‬4 ‫̊ ו ̊בכ‬ .‫יבכה‬ ̊ .‫עמה‬.‫̊ויעל‬ .‫אלוה‬ ̊ ‫[ ̊ל‬.‫]ויאמרו‬ .‫[ ̊בער‬.]‫בר‬.‫בלעם‬

13 14 15

5

̊ [.]‫לם‬ ‫תצם‬ .‫[תבכה‬.]‫ו ̊[ל ̊ם‬.] .‫להם‬.‫ מר‬5 ‫ויא‬ .‫ש]ב ̊[ו‬ ̊ .‫מה‬.‫אחוכם‬ [.]‫ יפעל[ו‬.‫שד]ין‬

16 17 18 19 20

6

.‫לכו‬ ̊ .‫ראו‬ ̊ .‫אל]ה ̊[ן‬.‫פעלת‬ ̊ .‫אתיחדו‬ .‫̊א̊ל]ה ̊[ן‬ ̊ ‫מו‬ ̊ .‫שדין‬.‫ונצבו‬ .‫עד‬ ̊ [.]‫לש]מ[ ̊ש‬.‫ואמרו‬ ̊ .‫שמין‬.‫סכרי‬.‫תפרי‬ .‫ ̊ גה‬7 ‫נ‬.‫ואל‬.‫שם חשך‬.‫בעבכי‬

21 22 23 24 25 26 27

.‫[ ̊ס ̊מרכי‬.‫ ̊וא]ל‬.‫עטם‬ ̊ .‫תהבי‬ .‫חשך‬.‫בע[ ̊ב‬.]‫חתן‬

28 29

.‫עלם‬.‫עד‬.‫ואל תהגי‬ .‫נשר‬.̊ ‫ פת‬8 ‫חר‬.‫ססעגר‬.‫כי‬ ̊ .[‫וק]ל‬ .‫רח ̊מן‬ .‫ ̊נחץ‬.‫[ ̊בבי‬.‫ ̊ח]רפת‬.‫יענה‬

30 31 32 33

̊ ̊ .‫אנפח‬ .‫אפרחי‬.‫וצרה‬

34

9

.‫ ]י ̊[ון‬9 .‫נשרת‬.‫דרר‬ ̊ .‫מע]ז[ין‬.‫יחז‬.‫וצפר‬ ̊ .‫[ה‬...]‫̊ ו‬ .‫[מטה‬...]‫יבת‬ .‫ארנבן‬.‫חטר‬.‫ייבל‬.‫רחלן‬.‫באשר‬

35 36 37 38

10

.‫ ]י[ ̊ח ̊ד‬10 .‫אכלו‬ ̊ [‫חפש]ן‬ .‫חב ̊ ̊בן זית‬ ‫אכל‬.‫[ ̊בת‬.‫את]א[ו]ו‬ .‫חמר‬.‫שתיו‬ 11 ‫ש‬.‫[גרי‬.]‫מוסר‬.‫שמעו‬.‫וקבען‬ ̊ ̊] .‫על‬ ̊ ̊ [.]‫קר‬ ̊ .‫̊ ומן‬ .‫הלכו‬.‫חבצן‬ [‫כתר]ן‬

39 40 41 42 43

.‫יקחך‬.‫[לחכמן‬.‫בט]לן‬

45

4

7

8

11

44

So Balaam got up on the morro[w. He arose and lamented ex]pressly for d[a]ys […?] And he became fee[ble and he fas]ted And he wept desperately. Then his people came to him, [and they spoke] to Balaam son of Beor, “Why are you fasting? [And] why are you weeping?” And he said to them, Si[t]! I will tell you what the Shadda[yin] are [doing]! Come! See the go[d]s’ deeds! The go[d]s assembled themselves. The Shaddayin stood as an assembly. They spoke to Sha[ma]sh, Break the bolts of heaven! In your midst set darkness and not luminescence, obscurity and n[ot] your shimmering! Provide terrors in the [mid]st of darkness, that you should not glow forever! For the swift reviled the eagle And the screech [owl] the vulture The scops owl re[viled] the children of the hawk (?) And the wallcreeper the chicks of the heron. The swallow tears at the [d]ove And the sparrow grasps ? And […] ? […] staff At the place of ewes the rod leads hares. They eat [to]gether. Bat[s] are lovers of olives They long for a measure of food. They drank wine. And hyenas heard the exhortation of jackal whe[lps.] And from the city the crown[ed] went as the exiled. Idle[rs] laugh at the wise

3

4

5

6

7

8

9

10

11

5.3 The Text, Its Reconstruction, and Its Translation

12

13

14

15

16

17 18

̊ .‫רקחט‬.‫ועניה‬ [.]‫מר‬ .‫וכהנה‬ [.]‫תפגר‬.‫[ ̊ר ̊ש ̊ ̊ען‬.]‫]בכ[ ̊ף‬ ̊ [.‫מר[ ̊כ ̊ב ̊]ן‬.‫חלחל]ה‬.‫תר ̊פת‬

46 47

.‫קרן‬.‫אזר‬.‫̊לנשא‬ .‫חשב‬.‫חשב‬ ̊ .‫ ]שב‬13 ‫ח‬.‫חשב‬ .‫אתנתק‬.‫ק[חך‬ [...]‫̊מר‬ [Lacuna] ‫רחק‬.‫מן‬.[‫חרש]ן‬.‫ושמעו‬ ̊ .‫[לם עבק]כ[ם‬...] ‫ד‬

49 50 51 52 53 54 55 56

‫חית‬.‫נ[געה‬...]‫א‬ ̊ .‫קקב‬ ̊ .‫[ד‬...] […]‫חז‬ .‫קקן‬.‫חזו‬.‫וכל‬ ‫ ̊ק]ו‬15 ‫ל‬.‫ועשתר‬.‫שגר‬ [...]‫י‬.‫[זי‬.]‫[י‬... .‫[ ̊אכמכם‬...] .‫ו̊לנמר‬.̊ ‫ל[ ̊צבי‬.]‫תר‬.‫[יחת‬...] .‫[ססה‬.‫ ]י‬16 ‫בנ‬.‫הקרקת‬.‫חניץ‬ .‫ל[ב‬.]‫[אכתי‬...] ̊ .‫מלת‬ [...]‫פר ̊ה‬ .‫ועין‬.‫אברן‬.‫[משן‬...] […].‫כשד‬

57 58 59 60 61 62 63 64 65 66 67 68

12

48

And the poor-woman mixes myrrh But the priestess faints by the hand of the wicked. A shudd[ering] woman releases the [ch]ariot[s] The belted-one does not lift horns. Consider! Consider! Consider! Con[sider!] The one who laughs destroys himself! Bitterness (?) [Lacuna] And the dea[f] hear from afar. D […] to them / why (?) he hastens you (pl.) ʾ […] his/her/its plague animals […] D a plague / herbage (?) of ḥz (?) And everyone saw oppression. Offspring and young suffer (?) ? which ? […] blinded (?) you (?) […] and to the panther. A piglet flees the sons of the mare. […] Angry of heart (?) The word of his cow / silliness From a tooth, a limb, and an eye. Like a field (?)

281

12

13

14

15

16

17 18

Table 22: The Deir ʿAlla Plaster Text, Combination A

The rest of the second line continues the narrative, reporting that the gods spoke to Balaam and what they said (vv. 3–5). Their speech consists of a few elements that are difficult to interpret. The imperfect form of the verb appears to have an object suffix (v. 6), as is probably the case. However, the numerus of the verb is unclear: context demands a plural, but the form is singular. If singular, it must refer to something that El or ‫( שׁדין‬if singular) is going to do, if plural to something the gods or the ‫( שׁדין‬if plural) are going to do. The form provides no explicit resolution to this issue. The following phrase ‫זא אח ראה‬ (v. 7) is syntactically enigmatic, but explicable. ‫ זא‬presents the object, the subject is ‫אח‬, and the verb is ‫ראה‬. Problematic is the verb, both in form and in content: for a perfect, one would expect ‫ רא‬in Old Aramaic, suggesting that this might have the aspect of a participle; on the other hand, ‫ ראה‬is otherwise

282

An Inscription from Deir ʿAlla. Balaam in Transjordan

unknown in Aramaic, meaning it must be a loan-word, most likely from Hebrew or a similar Canaanite dialect.50 The geographical site of the finding does permit this interpretation, and it remains the best option. The phrase ‫אש לרתה‬ (v. 8) consists of the subject ‫אש‬, the negation ‫ל‬, and the verb ‫ רתי‬with a feminine singular object suffix ‫ה‬-.51 This translation must then read, “a man did not pity it”, or – more idiomatically – “no one pitied it”. The final phrase of line 2 (v. 9) consists perhaps of the verb ‫ טבב‬as attested in Syriac ( ), meaning “to report” or similar.52 Alternatively, it could be related to the Syriac root ‫ ܒ‬/ (Heb. ‫ טוב‬/ ‫)טיב‬, meaning “to prepare”, or from the Aramaic ‫טיב‬, meaning “to please”. Reading ‫ טיב‬in some form might increase the difficulty of filling the lacuna before the end of the line, but we will see below that this is not an issue. In terms of form, the verb appears to be a masculine imperative singular, which in its attested Imperial Aramaic form at least did not contain a mater lectionis.53 An object suffix presumably must have followed this verb, but it would be difficult based on the context to determine what it should have been. The most likely candidates would be third person singular feminine (‫ה‬-), second person singular masculine (‫כ‬-), or third person plural (‫הם‬-) based on the space available and the context. The third option, ‫הם‬, only becomes available if no preposition proceeded ‫עת‬, as Blum has reconstructed.54 Rather than reconstruct ‫כ‬.‫ ה‬in the lacuna as Blum has done, I prefer to reconstruct .‫ הם‬or perhaps even .‫הם‬., as the direct object, naming whom Balaam should prepare or please.55 The possibility of reading a ‫ מ‬in this position is as likely as a ‫כ‬, as Hackett and McCarter already noted based on the apparent downstroke beneath the ‫ ע‬in line 1.56 Thus the line would end with, “prepare them now!”. This Balaam does in the subsequent narrative sequence by warning his people of the impending danger. Alternatively, it could mean “please

50 Cf. J. Hoftijzer and K. Jongeling, Dictionary of the North-West Semitic Inscriptions, HdO. Erste Abteilung, Der Nahe und Mittlere Osten, vol. pt. 1–2 (Leiden; New York: Brill, 1995), 1041–42. 51 The verb ‫ רתי‬most likely carries the meaning “to be compassionate” as in Aḥiqar and in Samaritan Aramaic; cf. Hoftijzer and Jongeling, DNWSI, 1088, the literature cited there, and Blum, “Kombination I,” 584–85. A meaning reflecting the verb ‫ רתת‬is contextually impossible. 52 Thus Blum, “Kombination I,” 585. 53 Cf. Segert, Altaramäische Grammatik, 292. 54 Cf. Blum, “Kombination I,” 585. 55 It is possible that the “heavy” object suffix ‫ הם‬might have been separated from the verb and the following adverb by dividers. Aramaic inscriptions sometimes attest this; cf., e.g., line 10 of the inscription from Tel Dan. 56 Cf. McCarter, “Balaam Texts,” 51–52 and Hackett, Balaam Text, 36. Lipiński, Studies II, 120 restored a ‫ פ‬at this point. But since this was based on McCarter’s preceding conjectures, which seem unlikely based on the reconstruction since achieved by Blum, in which Fragments IIIa and XVa–b fit into the lacuna, a ‫ מ‬seems the better choice.

5.3 The Text, Its Reconstruction, and Its Translation

283

them now!”, meaning Balaam should appease the gods before their terrible undertaking comes to pass. The inscription’s third line returns to the narrative following the directly dictated divine speech, noting that Balaam arose in the morning and lamented to a great extent for days (vv. 10–11).57 The antepenultimate lexeme in line 3 (v. 12) has sometimes been translated as “he was not able” from ‫( ליכל‬the negative particle ‫ ל‬+ ‫)יכל‬,58 which is quite possible from the space available, but syntactically difficult. One would anticipate a subsequent infinitive ending in ‫מ‬, which is difficult to reconstruct based on the context and available evidence. Alternatively, I would suggest the possibility of reconstructing ‫ ל( לאכל‬+ ‫)אכל‬, “he did not eat”, which has the advantage of matching the context. While recovering an ‫ א‬from the damaged surface remains difficult, it is by no means impossible in my opinion.59 Therefore, I regard this as a viable option. If keeping the common reconstruction (‫)ליכל‬, I would proffer the translation “he became feeble” for the phrase “he was not able”. This interpretation makes sense of the context, phrasing in a more adequate English idiom. Admittedly there is a paucity of attestations for precisely this understanding. The conclusion of the third line and the beginning of the fourth (v. 13) rephrase Balaam’s mourning, repeating his lamenting, exchanging that verb with the verb for crying ‫בכי‬. Here it means “weeping” or “lamenting”. The story continues with “his people” (‫ )עמה‬coming (‫ )עלל‬to him,60 at which point they address him.61 They inquire about his fasting and weeping, to which he responds at the end of ln. 4 and in ln. 5 (vv. 19–22) with two commands and two introductions, set up as parallel structures (see below in the commentary). The lexeme ‫ אחוכם‬appears to be a paʿel imperfect first-person singular of the root

57

Here I follow Blum’s reconstruction as plausible based on the available evidence; cf. Blum, “Kombination I,” 576–77 and Blum, “TUAT.NF,” 466–67. 58 Cf. McCarter, “Balaam Texts,” 51–52; Hackett, Balaam Text, 37; and Blum, “Kombination I,” 577. 59 Cf. also Segert, Altaramäische Grammatik, 272–73, which notes that ‫ א‬after a prefix may elide as early as in Imperial Aramaic. This could permit an iterative or durative understanding of the imperfect. 60 Hackett, Balaam Text, 37 suggested reading this verb as the Hebrew √‫עלה‬, but another root seems more likely, either √‫ עלי‬or √‫עלל‬. Since √‫ עלי‬generally remains outside of regions this far south, being attested in Phoenician and Punic, the reading √‫ עלל‬is favored here; cf. Lipiński, Studies II, 122 and Blum, “Kombination I,” 577. The editors of the editio princeps reconstructed ‫ אלקה‬instead of ‫ ;אלוה‬Lemaire and Caquot followed them in this; cf. Hoftijzer and van der Kooij, ATDA, 190 and Caquot and Lemaire, “Les Textes Araméens,” 194. Cf., however, Hackett, Balaam Text, 37(as above) for a corrective. 61 With Lipiński, Studies II, 122 I insert and read Fragment VIIId here as ‫בר‬.‫לבלעם‬. Blum apparently reads something else here, interpreting the passage as a vocative; cf. Blum, “Kombination I,” 577 and Blum, “TUAT.NF,” 466–67. It remains unclear what letter Blum reads before ‫בלעם‬.

284

An Inscription from Deir ʿAlla. Balaam in Transjordan

‫חוי‬.62 In this form, and the Haphʿel, the root takes the meaning “to explain, demonstrate, make known”.63 He will report on what the Shaddayin are going to do or are doing. Following these introductory remarks, an account of his vision begins towards the end of the inscription’s fifth line (v. 23) and continues until its conclusion. In spite of all of the attempts to sketch precisely what this vision entails, much of it must remain a mystery due to the severely damaged condition of many lines. The first, most frustrating lacuna in this case appears in the gap in the name of the deity to whom the Shaddayin speak in lin. 6 (v. 25). The deity was certainly feminine, as obviously indicated by the verb forms ‫( תפרי‬v. 27) and ‫תהבי‬ (v. 29) as well as the repeated object suffix ‫כי‬- in ln. 7 (v. 27). In the initial publication, Hoftijzer suggested reconstructing the name as ‫ שׁגר‬as the female pendent to the masculine ‫( אשׁתר‬cf. the end of ln. 14, v. 60),64 but this reconstruction has been generally abandoned. Caquot and Lemaire commended reading the name as ‫שׁמשׁ‬, at first glance a difficult reading due to the genus. However, they note that the solar deity was feminine in Ugarit.65 McCarter suggested the deification of the concept of the world of the dead, reading ‫שׁאל‬, which is feminine in Hebrew.66 However the context makes little sense for such a reading and only the gender of the deity could favor this interpretation. Hackett generally followed Caquot and Lemaire, yet noted that the Hurrian storm goddess Šala, who is often pictured with Adad, should not be dismissed out of hand.67 However, as Lipiński expressed (and Blum reconstructed), the stroke at the very right edge of Fragment Id, to the right of the word ‫תפרי‬, solidly affirms the reading ‫שׁמשׁ‬.68 The references to light and darkness in the context of the inscription (vv. 27–30) commend the solar interpretation as well. The following verb ‫( תפרי‬ln. 6, v. 26) comes from ‫ פרר‬as Caquot and Lemaire, as well as Rofé, suggested and McCarter, Hackett, and Weippert and Weippert followed. 69 This reading seems preferable to the reading ‫תפקי‬, as

62

Cf., e.g., Lipiński, Studies II, 123 for this interpretation. Cf. Hoftijzer and Jongeling, DNWSI, 353–54 for attestations and usages. 64 Cf. Hoftijzer and van der Kooij, ATDA, 272–74; Hans-Peter Müller, “Einige alttestamentliche Probleme zur aramäischen Inschrift von Dēr ‘Allā,” ZDPV 94 (1978): 64; Rofé, Book of Balaam, 61; Moshe Weinfeld, “The Balaam Oracle in the Deir ‘Alla Inscription,” Shnaton 5– 6 (1981–82): 146; etc. 65 Cf. Caquot and Lemaire, “Les Textes Araméens,” 196–97. 66 Cf. McCarter, “Balaam Texts,” 53–54. 67 Cf. Hackett, Balaam Text, 42. 68 Cf. Lipiński, Studies II, 127; Blum, “Kombination I,” 583 fn. 36; and Blum, “TUAT.NF,” 467 fn. 37. 69 Cf. Caquot and Lemaire, “Les Textes Araméens,” 196–97; Rofé, Book of Balaam, 61; McCarter, “Balaam Texts,” 53; Hackett, Balaam Text, 42–43; and Weippert and Weippert, “‘Bileam’-Inschrift,” 91–93 = Weippert and Weippert, “‘Bileam’-Inschrift,” 147–48. 63

5.3 The Text, Its Reconstruction, and Its Translation

285

suggested by Weippert and followed by Lipiński and Blum.70 In terms of form, it is a short imperfect, with the force of a jussive. Thus, it should be translated like an imperative: “break!”. The direct object is ‫סכרי שׁמין‬, “the bolts of heaven”. The following phrase (v. 27), continuing in line 7, picks up the tenor of the jussive and advances it with an imperative of ‫שׂים‬, albeit without an apparent feminine ending, ‫י‬-.71 The lexeme ‫ עב‬in line 7 (vv. 27 + 29) refers not to clouds, as Blum translated,72 but to one’s midst, as Lipiński has demonstrated.73 Šamaš should set darkness instead of brilliance in her midst and give or provide (‫ )יהב‬terrors in the midst of this darkness. The term ‫ עטם‬in v. 28 probably refers to obscurity or similar, as Sasson proposed.74 The term ‫ סמר‬applies to the sons radiance, a parallel to ‫נגה‬.75 While no certainty or consensus exists concerning the word ‫ואלתהגי‬, Lipiński provides the best explanation, particularly regarding the context in which it occurs. Most likely, it consists of three elements and is missing a divider: the first element is the conjunction ‫ו‬, upon which the verbal negation ‫ אל‬follows, before the jussive verb ‫תהגי‬. The root behind this verb could be related to the Arabic ‫( ﻭﻫﺞ‬whj), which means “to glow”.76 At this point, the text becomes quite difficult to interpret. For the textual reconstruction, I essentially follow Blum for the remainder. Only in minor instances does my translation differ from his. Line 14 (v. 56) begins with a ‫ ד‬as Blum initially reconstructed,77 and not with a ‫ תו‬as he apparently later supposed.78 Though it might be possible to reconstruct a ‫ ת‬before the first remaining letter, this remains pretty speculative. Between these publications, he also 70 Cf. Manfred Weippert, “The Balaam Text from Deir ‘Allā and the Study of the Old Testament,” in The Balaam Text from Deir ‘Allā Re-Evaluated: Proceedings of the International Symposium Held at Leiden, 21–24 August, 1989, ed. Jacob Hoftijzer and Gerrit van der Kooij (Leiden: Brill, 1991), 157 fn. 21 = Manfred Weippert, “Der ‘Bileam’-Text von Tell Dēr ‘Allā und das Alte Testament,” in Jahwe und die anderen Götter, Manfred Weippert (Tübingen: Mohr Siebeck, 1997), 168 fn. 19; cf. Lipiński, Studies II, 128; Blum, “Kombination I,” 577; and Blum, “TUAT.NF,” 467 fn. 39. 71 Contra Hackett, Balaam Text, 43–44, who interpreted it as coming from the same verb √‫שׂים‬, but identified the form as an infinitive. She denied the possibility that ‫ שם‬could be an imperative based on the forms ‫ תפרי‬and ‫תהבי‬. But neither of these forms is an imperative; rather, they are jussives. To this end, cf. Hackett, Balaam Text, 46. 72 Cf. Blum, “Kombination I,” 577 and Blum, “TUAT.NF,” 467. 73 Cf. Lipiński, Studies II, 128. 74 Cf. Victor Sasson, “Two Unrecognized Terms in the Plaster Texts from Deir ʿAlla,” PEQ 117, no. 2 (July 1985): 102. 75 Cf. Puech, “Balaʿam and Deir ʿAlla,” 35–36. 76 Cf. Lipiński, Studies II, 130. The alternative readings fail to satisfy the context at hand in terms of solar imagery. For example, Blum, “TUAT.NF,” 467, which interprets the root to mean “grumble”, but in what way does the sun grumble? 77 Cf. Blum, “Kombination I,” 578. 78 Cf. the somewhat unclear image in Blum, “TUAT.NF,” 466, which appears to begin line 14 with ‫תו‬.

286

An Inscription from Deir ʿAlla. Balaam in Transjordan

apparently began to regard the downstroke on the right edge of ‫ ד‬as longer, permitting him to reevaluate it as a ‫ו‬. The root upon which he bases his translation in TUAT.NF remains obscure. He apparently regards the form as ‫תוהו‬, coming from the biblically attested ‫ תוה‬in peʿal. While this remains possible, the reconstruction of a ‫ ו‬loses out against the line beginning with ‫( ד‬or even ‫ ;)תד‬no remnants favor reading a ‫ ו‬at the beginning of line 14. Further in the line, at the boundary between Vf and Vc as combined by Blum (still v. 56), I would reconstruct a ‫ כ‬instead of Blum’s ‫ע‬.‫ן‬. The shape of the ‫ נ‬that he reconstructs varies substantially in its angle from the others in the text; cf., e.g., the ‫ נ‬also directly above his reconstructed ‫ נ‬in line 13. The remains of the letter at the right edge of Vc better match a ‫ כ‬than an ‫ ;ע‬any such ‫ ע‬would have been little more than a dot on the surface of the plaster. After the ‫ מ‬in v. 56 there is a word divider and an ‫א‬. At that point, Blum inserts Fragment IVa, which seems erroneous. N.b. the red stroke on the upper left edge, which looks like the, now vestigial, tail of a ‫ ת‬rather than merely an accidentally superfluous mark. The original editors certainly thought that the first line of this fragment had originally been written in red ink.79 Add to that the strikingly variant slope of the last line of this fragment when compared to the material surrounding it in Blum’s reconstruction (ln. 15) and one must likely accept that this piece does not belong here. Should Blum’s placement of the fragments be correct for the rest of ln. 14,80 I would read the line before returning to Fragment Id as [...]‫חית‬.‫̊ געה‬ ̊ [.]‫[חז‬.]‫[קקב‬.]‫ד‬. The original editors transcribed the first sign of IVj as ‫ר‬,81 but even based on their own sketches, this ‫ ר‬would have no rounded head. Blum followed this reading.82 The letter appears to have consisted of two lines joined at a point, like the head of an arrow pointing generally upward. This would favor reading it as a ‫ג‬. Reading it this way provides no readily identifiable, satisfactory reconstruction of the word, however. Nonetheless, several iterations are possible. No root ‫ געה‬is known. That datum commends regarding the ‫ה‬- as a possessive or an object suffix (masculine or feminine). A few known forms ending in ‫גע‬- are known:83 ‫ בגע‬and ‫ נגע‬are known from Qumran Aramaic, ‫אגע‬, ‫יגע‬, ‫מגע‬, ‫נגע‬, ‫פגע‬, ‫מפגע‬, ‫רגע‬, and ‫ שׁגע‬are known from Hebrew. The overlap is apparent: many of the forms reflect ‫( נגע‬Aramaic and Hebrew),

79

Cf. Hoftijzer and van der Kooij, ATDA, 255. From right to left: IVj, IVf, IVg, XVc. 81 Cf. Hoftijzer and van der Kooij, ATDA, 175. 82 Cf. Blum, “TUAT.NF,” 468. 83 Cf. Ruth Sander and Kerstin Mayerhofer, Retrograde Hebrew and Aramaic Dictionary, Journal of Ancient Judaism Supplements, vol. 1 (Göttingen: Vandenhoeck & Ruprecht, 2010), 196 for Aramaic possibilities (albeit from Qumran) and Sander and Mayerhofer, Retrograde Hebrew and Aramaic Dictionary, 109 for Hebrew. 80

5.3 The Text, Its Reconstruction, and Its Translation

287

namely ‫אגע‬, ‫בגע‬, ‫יגע‬, ‫מגע‬, and ‫נגע‬. 84 That could commend the inscription’s mentioning some kind of “affliction”, “contact”, “touching”, or “plague”. 85 The alternatives, ‫פגע‬, ‫מפגע‬, ‫רגע‬, and ‫שׁגע‬, would imply some reference to “his/its destiny”, “moving/flowing”, or “maddened speech”. Of the available options, “his/her/its plague” or similar seems preferable based on context; cf. the following phrase. With that, we can tentatively reconstruct a ‫ נ‬before ‫געה‬, giving a reading like “his/her/its plague” followed by “animals” (‫)חית‬. It would be tempting to reconstruct something like ‫ אבד‬in the lacuna following the ‫א‬ (recreated by removing fragment IVa from Blum’s reconstruction), but there would be space for four to five letters and a word divider after that ‫א‬. That seems to leave too much room for such a reconstruction, as it is unlikely that a form like ‫ אביד‬either stood there or would fill the gap. Contextually, in a text about a world turned upside down, one would also most likely want to consider a root that would have the opposite meaning; i.e., the plague should not destroy the animals (assuming for the moment that they are the direct object of some missing verb), but should rather aid them. If you’ll permit me such speculation, possibly the particle ‫“( אזי‬then”, “thereafter”) stood there and was followed by a verb in perfect, whatever that might have been. While quite speculative, it would at least fit the context and form of the text. Several of the following terms remain at least as problematic: for ‫קקב‬, Hoftijzer suggested “herbage” or some kind of tree (based on the Arabic ‫ﻗﻀﺐ‬ [qḍb]),86 whereas Blum has suggested “plague”.87 The ending of ‫ חז‬is missing in v. 58, precluding any identification of this term. Perhaps it had something to do with seeing or a vision; cf. the next phrase. After this uncertainty, however, the rest of line 14 and the beginning of 15 can be translated with some certainty. At the end of ln. 14, the phrase ‫ועשתר‬.‫( שגר‬v. 60) probably does not refer to deities, as the original editors suggested and many have followed. 88 More likely, the text returns to animal imagery at this point, as Caquot and Lemaire, as well as McCarter and Hackett noted.89 Blum has also accepted this understanding with increasing certainty.90 The final letter of ln. 14 is clearly a ‫ ל‬and, assuming the placement of Fragment Va at the beginning of lines 14–15 is correct, ln. 15 begins with either ‫ ק‬or ‫ט‬. Blum, who first added this piece at this 84 N.b., however, that the attestation of ‫ בגע‬has been disputed; cf. Edward M. Cook, Dictionary of Qumran Aramaic (Winona Lake, IN: Eisenbrauns, 2015), 152. 85 Cf. Gesenius, et al., Gesenius, 780–81; Hoftijzer and Jongeling, DNWSI, 1.715; and Heinz-Josef Fabry, ed., Ulrich Dahmen, ṭhr–str, vol. 2 of Theologisches Wörterbuch zu den Qumrantexten (Stuttgart: Kohlhammer, 2013), 869–74. 86 Cf. Hoftijzer and van der Kooij, ATDA, 255–56. 87 Cf. Blum, “TUAT.NF,” 469 fn. 49 and the literature cited there. 88 Cf. Hoftijzer and van der Kooij, ATDA, 218–19 and Lipiński, Studies II, 137–38. 89 Cf. Caquot and Lemaire, “Les Textes Araméens,” 201–2; McCarter, “Balaam Texts,” 56; and Hackett, Balaam Text, 54. 90 Cf. Blum, “Kombination I,” 579 versus Blum, “TUAT.NF,” 469.

288

An Inscription from Deir ʿAlla. Balaam in Transjordan

point, read a ‫ ט‬and added a ‫נ‬, apparently to make a plural participle from the biblical Aramaic ‫לוט‬, “to (be) curse(d)”. Perhaps one could also consider the possibility of reading a ‫ק‬, which might permit the reconstruction of ‫לקו‬, the third person plural from ‫לקי‬, postulated to mean “suffer”; cf. the Akkadian laqāʾu.91 While this remains conjecture, it at least would match the context. From this point the reconstruction becomes increasingly uncertain. Therefore it will not play much of a role for the rest of this discussion.

5.4 Commentary and Evaluation 5.4 Commentary and Evaluation

The opening phrases of Combination A, the only combination that interests us here, state the name of the work and identify its main character (vv. 1–2). The title, in rubrum, is the “The Admonitions of the Book of Balaam Son of Beor”. The mention of the title commends recognizing this as a known, extant work. Perhaps the appellation “admonitions” demonstrates that this version attests an except from the longer ‫ספר בלעם בר בער‬.92 The next phrase consists of a verbless sentence identifying Balaam: “A man who saw the gods was he”. The following narrative focuses on this figure and what he sees. This second phrase in line 1 both references the preternatural ability of the main character and simultaneously opens the narrative. Balaam is an abnormal figure in exceptional circumstances. However, it seems to contain a scribal error: the particle ‫ הא‬is the wrong color. This story’s exposition, the narrative circumstance follows immediately after Balaam’s introduction (v. 3). The gods visit him at night. In this phrase we note a second scribal error, in which the phrase ‫ אלוה‬stands supralinearly, having apparently been overlooked or forgotten initially. If the scribe writing on the wall copied a Vorlage, this mistake could have stood there. The only evidence to favor that suggestion is the precise layout of the inscription, which could not have afforded the addition of this word and maintained the division of the first two lines into perfect black and red halves. However, it seems more likely that it was added at this point because it stood in the Vorlage but had been forgotten or overlooked. This godly visitation apparently occurred as part of a divine vision, commissioned by El, as per the fourth phrase (v. 4) in lines 1–2. The gods then speak to Balaam (v. 5), proffering their message in a few short phrases (vv. 6–9). Like the work’s title, the gods’ message is inscribed in red ink, contrasting it markedly from the surrounding black text. This contrast presumably should emphasize the message of the deities and distinguish it from the rest. Alterna-

91 92

Cf. Hoftijzer and Jongeling, DNWSI, 584 for the literature. Cf. Dijkstra, “Balaam,” 52.

5.4 Commentary and Evaluation

289

tively, it could indicate the incorporation of another document into this context.93 However, this is somewhat speculative, and it would be going too far to suppose too much from these circumstances without having more data about scribal conventions of this era in this medium. Still, I find it likely. The gods’ message remains somewhat obscure. Scholars often regard the first verb as a plural (v. 6), but that interpretation is hardly orthographically secure. One would expect it to have the appropriate plural ending, particularly as other plural conjugations appear conspicuously in the inscription, specifically the imperatives in ln. 5 (vv. 19 and 21–22) as well as the perfects in lines 1–2 (vv. 3 and 5) and lines 5–6 (vv. 23–25). More likely the text reflects a singular masculine with an object suffix. Some uncertainty remains with this interpretation both in form and content. On the one hand, no other case exists in this inscription in which a clearly recognizable plural has an object suffix, so we do not know what that would even look like. On the other hand, it remains unclear who the subject “he” implied by the singular verb should be. The phrase (v. 7) seems to mean “He will do it”, where both the subject and object are unnamed. The following phrase also remains somewhat obfuscated. The letters are all pretty clear, but the meaning cannot be determined with certainty. The first word seems to be the feminine relative particle ‫זא‬, demanding that this relative clause refer to the feminine direct object of the preceding phrase (v. 6). The most likely explanation for the subsequent phrase in v. 7 is the subject “brother” and the verb, probably a participle, “see”.94 Combined then, these phrases read “he will do it, that which a brother is seeing!” The subsequent phrase (v. 8) states the reason for the impending action: “no one had compassion”. Finally, the divine message concludes with the command to prepare the people for what is coming (v. 9). The situation seems to imply that the negative aspects of the subsequent vision, the world turned upside down, will come about due to the audience’s (or some other group’s) lack of the compassion. At this stage, line 3 returns to the narrative, twice stating that Balaam arose before noting that he lamented and undertook other signs of penance, like fasting and weeping (vv. 10–13). His actions lead his people to inquire as to the reason for these gestures (vv. 14–17).95 To this inquiry, Balaam responds with 93 Cf. Millard, “Epigraphic Notes,” 24–25 and Alan R. Millard, “In Praise of Ancient Scribes,” BA 45, no. 3 (Summer 1982): 149, who regarded the red color as indicating a heading in the Egyptian style. 94 Curiously and conspicuously missing a word divider. 95 Nothing immanent to the text implies that this is some ritual format, contra Greenfield, “Philological Observations,” 119: “In the DAPT there is no inkling of sacrifices being made. Instead there are possible indications of the use of incubation rituals such as fasting and weeping (I,4). I understand the passage not as a means of expressing grief, or a post-factum expiatory act, but the means used to induce a vision.” The most conspicuous problem with Greenfield’s theory is the apparent narratological inconsistency in that Balaam appears to already have had his vision before weeping, thus precluding weeping and fasting as the means of having a vision.

290

An Inscription from Deir ʿAlla. Balaam in Transjordan

the content of what the gods are doing, the narrative introduction of which appears in v. 18. Balaam’s report to his people begins in ln. 5 with two parallel structures: each consists of an imperative followed by an introduction (vv. 19– 20 and 21–22). In the second case, the introduction also begins with a command. The parallel phrases commend regarding the Shaddayin and the gods mentioned in the line as the same group. This suggests that the Shaddayin are not a sub-group of the pantheon.96 The meaning of the term Shaddayin remains obfuscated, but suggestions include relations to fields, mountains, breasts, and pouring, should it be based on a Semitic root. Fertility presents the common theme to most of these – i.e., all but mountains. Perhaps this is the background for the term: the gods may have presented some aspects of fertility to their adherents. Alternatively, one could consider an Egyptian background, which might favor understanding the term as coming from the Egyptian for “savior”.97 Nothing in this context commends regarding them outright as “powerful, evil deities”, as Sasson opined.98 Nevertheless, the parallel structure of vv. 20 and 22, as well as 24–25 commends interpreting ‫ שׁדין‬as an eponym for all of the deities in the story. Again, the circumstances remain somewhat veiled. It appears that the Shaddayin are speaking to the solar deity Shamash, who is feminine in this text (v. 25). They command her to transform the state of heaven (vv. 26–30). She should break the bolts of heaven, perhaps suggesting some kind of deluge. Instead of light, she should provide darkness, even terrorizing creation with this darkness.99 After these brief introductory imperatives, the text transforms into a recounting of the perverted world, the world turned upside down. This interpretation goes back to McCarter and has found general acceptance.100 Several phrases demonstrate this understanding clearly. The image of the world turned upside down begins with several reversed aviary relationships (vv. 31–36). The swift appears mightier than the eagle; owls demonstrate greater strength than raptors. After this, the text lists some 96

Cf. Hackett, “Response,” 76; Weippert, “Balaam Text,” 156; Lipiński, Studies II, 123; and Choon Leong Seow, “West Semitic Sources,” in Prophets and Prophecy in the Ancient Near East, by Martti Nissinen, contributor Choon Leong Seow and Robert K. Ritner, Writings from the Ancient World (Atlanta: SBL, 2003), 208. Another interesting idea, proposed by Jo Ann Hackett, “Some Observations on the Balaam Tradition at Deir ʿAllā,” BA 49, no. 4 (December 1986): 219, is that ‫ אלהן‬is the general term for the deities, while ‫ שדין‬specifically references their function in the divine council. 97 Cf. the discussion to ‫ שׁדין‬/ ‫ שׁדי‬in Chapter Six. 98 Cf. Sasson, “The Book of the Oracular Visions of Balaam from Deir ʿAlla,” 295. 99 For ‫ עטם‬in v. 28, Seow, “West Semitic Sources,” 212 suggested its relation to Akkadian eṭemmu, and translated it as “eeriness”. In the same verse, he recognized ‫ סמר‬as related to, “the Arabic samaru (used of the radiance of the moon), but also Akkadian samaru, a term used of ornaments.” 100 Cf. McCarter, “Balaam Texts,” 58–59, as well as the versions from, e.g., Hackett, Lipiński, and Blum.

5.4 Commentary and Evaluation

291

unclear relationships between different animals (vv. 37–43), before turning briefly to transformed interpersonal relationships (vv. 44–49). Way recognized the reversal of relationships in the animal as reflecting the story of Balaam’s donkey in Numbers 22: “One convention that is common to both the DAPT [Deir ʿAlla Plaster Text] and Numbers 22–24 is ‘role reversal.’ In Combination I of the DAPT, many scholar point out the motif of the ‘world upside down’ with specific reference to animal activity: it is the weaker animal that threatens the stronger. For example, ‘the swallow reproaches the eagle’ (lines 7–8).”101

However, this description remains problematic, in that Balaam presumably should not be regarded as an animal in Numbers 22–24. The role reversal thus occurs between members of distinct taxonomic or even ontological groups, unlike those in the DAPT. Way particularly found this interpretation attractive, since he regarded the term ‫ חמר‬in ln. 10 as referring to a donkey. “Another feature that is common to both the DAPT and Numbers 22–24 is the presence of the donkey. […] The term ḥmr does clearly occur in Combination I, line 10.”102 “The interpretation ‘ass(es)’ makes the best sense in light of the following animal word (wqbʿn, ‘and hyenas’) and in light of the context of lines 7–10, which describe the unnatural activity of various animals.”103 “The interpretation of ḥmr as ‘ass(es)’ is especially inviting because the biblical account gives such prominence to the donkey (although the word used in Numbers 22 is ʿātôn, rather than ḥămôr).”104

The parallel with ‫חמר‬, however, is food (‫ )אכל‬and not hyenas (‫)קבען‬. The hyenas in v. 43 parallel the bats (‫ )חפשן‬of ln. 40 in a chiasm. Therefore, these assertions can be rejected. In the context of human relations, the text notes that those crowned left the city as exiles, and that lazy idlers deride the wise. While a poor woman mixes expensive myrrh, a priestess falls at the hands of the wicked.105 Something like an admonishing interlude directed to the audience interrupts this image of the world inverted in vv. 50–52. How one should understand the fourfold ‫ חשב‬remains obscure: perhaps they are all imperatives; perhaps their form varies be-

101

Way, Donkeys, 61. Way, Donkeys, 62. 103 Way, Donkeys, 63. 104 Way, Donkeys, 63. 105 The context of the world turned upside down precludes Greenfield’s identification of the ‫ עניה‬in ln. 11, v. 46 as an “answerer”; cf. Greenfield, “Philological Observations,” 119: “The ʿnyt is not a ‘poor woman’ but an ‘answerer’, the equivalent of the apiltu, known from an earlier period in the Mari texts, and of the ragintu ‘speaker’ known from neo-Assyrian texts. The rqḥt mr describes the specific function of preparing myrrh, probably for libanomancy, and the khnh ‘priestess’ presents no problem. It may be assumed that in our text these three terms are used of one and the same person who preformed rites for Bileam bar Beor. Needless to say this is all sheer speculation.” 102

292

An Inscription from Deir ʿAlla. Balaam in Transjordan

tween perfects and infinitives or something else. Perhaps there is some wordplay behind this admonition lost to modern readers.106 Sasson suggested almost a dramatic understanding, in which the speaker addressed different sections of the audience.107 The culmination of this interruption appears somewhat clearer: the one who laughs ultimately destroys himself. The world on its head is no laughing matter. Beginning again in the middle of ln. 13 (vv. 53–54), the text perhaps continues with the description of the topsy turvy world,108 but the broken context makes this anything but certain. At least the image of the deaf hearing from afar in v. 55 would commend this understanding. But until v. 59, the text is more or less impossible to decipher. Phrases 59–60 note that people will see suffering and their offspring will suffer as well. Again, the context and lexemes remain unclear. Thereupon follow a number of animals again (vv. 63–64), suggesting that the description of the world turned upside down continues. Verse 64 provides perhaps the best evidence of this. Nonetheless, the last several phrases prohibit any concrete reconstruction and anything beyond mere guesswork about what actually followed. However, one thing seems certain: the text of Combination A presumably ended after line 18, with perhaps some elements at the beginning of a nineteenth line. Based on Blum’s reconstruction, one can identify a dividing mark after the eighteenth line of text.109 While there is no text immediately above that portion of the divider that has survived, it is possible that a nineteenth line began and covered about half of the line. Should this have been the case, no evidence remains of this. Since the text clearly ends at this point in the eighteenth or theoretical nineteenth line, however, one matter remains clear: the story apparently had no denouement as inscribed on the wall; in this form, the conflict never resolved. As far as the context permits, it appears that the plaster inscription in Combination A concluded with elements of the vision of the world turned upside down. No room remains for a satisfactory narrative ending, perhaps even for the vision’s end.110 It remains highly unlikely that Balaam or his people could have reacted to this report in any way in order to prevent its occurring or that the narrative continued, describing their downfall and destruction. This implies that the remnant found at Tell Deir ʿAlla must have been part of a longer work that is only excerpted here. Perhaps this form of the plaster inscription represents the beginning of a longer literary Balaam tradition 106

Thus Blum, “Kombination I,” 591 fn. 86. Cf. Sasson, “The Book of the Oracular Visions of Balaam from Deir ʿAlla,” 303. 108 Contra Sasson, “The Book of the Oracular Visions of Balaam from Deir ʿAlla,” 304–6, who suggested that Balaam then cursed the responsible deities before a passage predicted the restoration of the normal order of things. 109 Cf. the image of the reconstruction in Blum, “TUAT.NF,” 466. 110 Cf. Blum, “Institutioneller Kontext,” 28: “…nichts deutet aber auf eine Wiederaufnahme der erzählten Szenerie von V. 1–9 hin.” 107

5.5 The Inscription’s Form

293

known in Transjordan in the ninth or eighth centuries BCE. Alternatively, this could indicate that the contents of the vision were interpolated from another context, a context that preserved them more fully.

5.5 The Inscription’s Form

5.5 The Inscription’s Form

Unlike biblical texts before their collection and scribal editing, we can say something with certainty about the form of the realiter of the plaster inscription. This version of the text apparently did not contain the end of the story from which it was drawn. Lines clearly demarcate three of the four sides of the text. Red lines define the top and the bottom, as well as the left edge. The right border has no such line, but was maintained through a justified edge.111 A red zigzagging line between two parallel red lines defined the bottom boundary, distinguishing it from the other edges. The text defined by these boundaries is roughly square, about 32 cm in both height and breadth.112 Both the title and the citation of divine speech were recorded in red ink, while the remaining text was written in black ink. Someone, perhaps the same scribe who wrote the rest, emended it at least once; other than the size of the letters, almost nothing distinguishes this writing from the rest. This insertion could evince scribal reliance on a Vorlage,113 with Millard going even so far as to recognize that this portion of text was copied from a column of a scroll.114 It therefore seems likely that the scribe was relying on a specific Vorlage from which he copied and perhaps towards which he corrected, though it would certainly be impossible to ascertain if the supralinear addition in the first line appeared in the Vorlage or presents a correction toward that Vorlage. 111

Cf. the image in Blum, “Institutioneller Kontext,” 25. Measuring with the scale provided in the editio princeps shows that even in that larger reconstruction, in which Fragments Ic and Id were lower than they are regarded now and were spaced more widely from each other, the measurement comes in at about 33.5 cm x 33.5 cm. 113 Cf. Kooij and Ibrahim, Picking up the Threads, 65. Cf. also Andre Lemaire, “Les Inscriptions sur Plâtre de Deir ʿAlla Det Leur Signification Historique et Culturelle,” in The Balaam Text from Deir ʿAlla Re-Evaluated: Proceedings of the International Symposium Held at Leiden, 21–24 August 1989, ed. Jacob Hoftijzer, Gerrit van der Kooij (Leiden; New York; Copenhagen; Cologne: Brill, 1991), 41–45, who argued for a collection of Balaam oracles in a manuscript used as a Vorlage for the Deir ʿAlla inscriptions. This collection was 100–200 years older than the inscription and came from the Damascus region. 114 Cf. Millard, “In Praise of Ancient Scribes,” 149. Wenning and Zenger, “Heiligtum,” 192 postulated: “Dagegen [gegen kurze Texte wie Kuntillet ʿAǧrud] sind die Texte von TDA umfangreiche, literarische Texte im eigentlichen Sinn, nur vergleichbar den Aufzeichnungen auf Schriftrollen. So legt es sich nahe, daß die Texte von TDA als direkte Übertragung bzw. Exzerpt eines auf einer ‘Schriftrolle’ (vgl. den Anfang von Kombiation I) vorgegebenen Textes (Papyrus?, Tafel?) verstanden werden.” Cf. also Lemaire, “Les Inscriptions sur Plâtre de Deir ʿAlla Det Leur Signification Historique et Culturelle,” 49–55. 112

294

An Inscription from Deir ʿAlla. Balaam in Transjordan

Syntactically, few differences distinguish the various portions of the text. Several narrative forms (‫ו‬-imperfect) appear in the opening frame of the story (vv. 3, 5, 10, 11, 14, 18, and perhaps in 12 and 15, if the reconstruction happens to be accurate). These end after v. 18, giving way to perfect and ‫ו‬-perfect forms with a jussive (v. 30), as well. However, the narrative opening also attests perfect forms. Otherwise, the placement of verbs in the sentences varies in both the narrative frame and the report of the gods’ deeds. The author of the narrative portion rephrases several elements in parallel ways, repeating information with different tenses. For example, phrases 3–4 include essentially identical statements in parallel, the first with a narrative form and the second with a perfect. Verses 10–11 demonstrate precisely the same feature. Leading into the vision, one notes the similar structure between 19–20 and 21–22. Verses 23– 24 stand in parallel, as do the commands in 27–30. In general, this duplicate structure defines the text. Most of its report flows from couplets, yet several phrases break this scheme: vv. 5, 18, 25, 26, at least some portion of vv. 39–43, and perhaps v. 52, though its broken context precludes that certainty. Following v. 52 it becomes impossible to say anything about the form beyond mere conjecture. The feature common to the independent stichs in the narrative portion is their introducing speaking. Within the vision report, the first phrase stands independently (v. 26). Thereupon follow at least seven couplets (vv. 27/28; 29/30; 31/32; 33/34; 35/36; 37/38), before the scheme becomes more opaque for vv. 39–43. Verse 39 fails to fulfill the couplet scheme. Thereupon vv. 40–43 follow a pattern distinct from the rest, ABB’A’, in which A and A’ are about animals and B and B’ are about sustenance. Verse 44 returns to the couplet scheme, featuring at least four pairs of couplets before the admonition in v. 52. After that, there can be no certainty about the structure, but vv. 59–60 and 63–64 are parallels, which could imply that the other phrases were too. That would imply at least six couplets, perhaps eight or more before the end of the inscription. That all being said, no overarching structure for the whole composition becomes readily apparent. That could be the result of the missing conclusion, but that remains in the realm of speculation. Summary of the Inscription’s Structure:115 Vv. 1–2: Title / Superscription Vv. 3–5: Narrative Exposition Vv. 6–9: Divine Message Vv. 10–22: Narrative: Balaam and His People Vv. 23–49: Balaam Reports His Vision Vv. 23–30: The Divine Assembly Vv. 31–36: The World Perverted, Part I: Ornithology 115

Contrast this with the structural proposal of Sasson, “The Book of the Oracular Visions of Balaam from Deir ʿAlla,” 292–306.

5.6 The Inscription’s Sitz im Leben

295

Vv. 37–43: The World Perverted, Part II: Zoology Vv. 44–49: The World Perverted, Part III: Anthropology Vv. 50–52: Balaam’s Admonition to His Audience Vv. 53–68: Balaam Continues His Report about the Perverted World (?)

In sum, the composition can best be described as a narrative vision report with a quasi-poetic structure. It remains possible that diachronic processes stood behind the composition of this narrative. For example, the narrative could have been written to embed an older, extant vision report; the gods’ message falls somewhat out of the context and does not entirely mesh with the rest of the material. Such possible diachronic processes must be developed further below. Grammatically, linguistically, and syntactically, Combination A is not really distinct from Combination B. Even in terms of the form, it is noteworthy that Combination B also seems to rely on couplets to deliver its message.116 However, the sheer size of the text, its physical format, distinguishes it from Combination A. Whereas Combination A was roughly a 32 cm x 32 cm square, Combination B had “eine minimale Maße [von] 39 x 70 cm...”.117 Particularly the divergence in the breadth demonstrates its distinction from Combination A. Combined with its distinct content, therefore, one need not regard Combination B as part of some Balaam tradition.118

5.6 The Inscription’s Sitz im Leben 5.6 The Inscription’s Sitz im Leben

Having discussed the form and context of the inscription, we can turn to its origin and background. The text did not come to exist in a vacuum. Therefore, we will turn to the culture and circumstances behind its inception. Who made this inscription and to what end? What background does this inscription evince? Did it serve a particular purpose, and, if so, what? The following discussion addresses these questions.

116

Cf. the reconstruction in Blum, “Kombination II,” 36–40. Blum, “Kombination II,” 34. 118 For these reasons, I will not address the postulations of Rofé regarding Combination B, its Midianite (according to him) language, or any presumed relationship to P or cultic prostitution; cf. Rofé, Book of Balaam, 59–70. 117

296

An Inscription from Deir ʿAlla. Balaam in Transjordan

5.6.1 The Scribe Some have described the person who inscribed this wall as, e.g., “a highly experienced scribe” or similar.119 But does the inscription really merit this identification? Some letter forms are similar to those of Aramean scribes from Damascus and Hamath. Yet, some forms more typical of the Transjordan.120 The first case could imply that the scribe had travelled to learn to write. The second circumstance favors a local scribe. Perhaps taken together, both observations commend a local development over time of an imported scribal tradition.121 At any rate, it seems unlikely that a master scribe from the court of Damascus, for example, copied this inscription onto the wall. For that matter, I feel somewhat uncomfortable exhorting the greatness of this scribe without considering some of the problems in the inscription. Presumably the person who wrote this on the wall must have been of above average ability, particularly when contrasted to the society as a whole. Simply the fact that he was literate probably puts him ahead of most of his contemporaries. Even more, he took the time to write, or at least copy, a literary text and not just a sales receipt or similar. That is, he must have had some education. At the same time, I count up to seven errors in the inscription: 1) the ‫ הא‬of the second clause not matching the color of the rest of that phrase; 2) the addition of the supralinear ‫ אלוה‬in the third phrase; 3) the missing word divider between ‫אח‬ and ‫ ראה‬in v. 7; 4) perhaps a missing ‫ א‬in ‫ ויכל‬in v. 12; 5) the missing feminine ending for ‫ שׂם‬in v. 27; 6) the missing word divider between ‫ שׂם‬and ‫ חשׁך‬in that same verse; and 7) the missing word divider between ‫ ואל‬and ‫ תהגי‬in v. 30. So while the scribe responsible may have been a person of significant talent, particularly when contrasted with his average contemporaries, said competence should not be exaggerated to be regarded as perfection.122 Perhaps that implies a student of high caliber rather than a master. The alternative understanding would be to regard the scribe as a master who copied even the errors from a Vorlage. While that remains possible, it would still mandate a reason: to what end should someone have copied these errors? The only logical answer would be that the text had some kind of canonical 119 Cf., e.g., Kooij and Ibrahim, Picking up the Threads, 65. They did note the error in the text of the first line, identifying it as homoioarkton (without using that term) and ascribing it to the scribe’s needing to become accustomed to writing on a vertical surface. 120 Cf. Kooij and Ibrahim, Picking up the Threads, 66. 121 This would support the observations of McCarter, “The Dialect of the Deir ʿAlla Texts,” 98, who remained hesitant to ascribe the inscription’s Aramaic characteristics to an Aramean community in the Transjordan. That would fail to explain the local dialectic elements apparent in the inscription’s language. 122 Contra Millard, “Epigraphic Notes,” 25, who commented that the use of word dividers at Deir ʿAlla, “joins the evidence of the script to indicate that the writer was well-trained and experienced in his art”. He does not even mention the errors, however.

5.6 The Inscription’s Sitz im Leben

297

status, even in its imperfect form. However, I find that improbable. Nothing about the find’s circumstances commends it as a particularly holy text or as some unassailable literary prototype. The more likely explanation, also connected to the circumstances of the find would be that a student, probably an advanced or skilled one, copied the text onto the wall. Perhaps the correction with the supralinear ‫ אלוה‬attests a master’s correction to a student’s oversight. This suggestion could in turn reveal something about the context of the find, to which we will return below. The language of the inscription does suggest that the scribe was familiar with Aramaic, but was also familiar with elements from the local Canaanite dialect or language. The Aramaic character of the language could reflect the political situation behind the copying of this text. Two scenarios commend themselves to this end: 1) Arameans dominated the area at the time of the inscription or shortly before; or 2) Arameans had been sent there as exiles. The second scenario only seems likely if one dates the inscription about a century later than currently thought.123 That commends the first scenario as the more likely. We will return to this matter further below.

5.6.2 The Location The function of the room that housed the inscription deserves some attention. Within the room itself, a raised area along the wall appears to have served as a bench of sorts.124 That would imply that people were sitting there. Thus, older reconstructions tended to favor the identification of the structure in which the inscription was found as a temple.125 Much of this understanding rested on the identification of the DAPT as a religious text. Therefore, it must have stood in a religious context. While the text could be considered religious in the broadest sense of the term,126 it does not appear to have any kind of liturgical or cultic function. Rather, it is narrative. The only cultic function one might consider it having, would be as a record of some cult-prophetic exhortation. This must be further considered in an examination of the text’s function. Nevertheless, one

123 Cf., e.g., Al Wolters, “The Balaamites of Deir ʿAlla as Aramean Deportees,” HUCA 59 (1988): 101–13, who argued precisely this point: the Assyrians exiled Arameans from Damascus at Deir ʿAlla around 700 BCE. However, accepting the inscription’s date as around 800 BCE entirely precludes this scenario. 124 Cf. the reconstruction of a room with something like a bench following the walls in Kooij and Ibrahim, Picking up the Threads, 64. 125 For the following discussion, cf. the observations in Blum, “Institutioneller Kontext,” 35. 126 For example, it mentions deities.

298

An Inscription from Deir ʿAlla. Balaam in Transjordan

must not regard a cultic background as the only understanding of the context of the text’s discovery.127 Another option makes better sense of the room in which the text was found. As mentioned before, along the wall, the excavators identified something like a bench. While that might be indicative of a temple or religious setting,128 one could just as easily recognize it as a classroom.129 That is, the benches served as seating for students learning to write. Add to that the plastered wall on which the writings were made and which still had more room for other writings, and it begins to looks like a classroom. That in turn would provide some information about the text’s function, both in the context of the find and more generally. Blum goes even so far to identify this as the room for the masters’ class.130 The Aramaic influence on the inscription’s language commends identifying it with a governing institution. That is, Hazael or his successor must have established the school to teach locals about Damascene state functions and their culture. Deir ʿAlla hardly makes the impression of needing a private school, but placing such an official school at such a convenient (in terms of transportation) location and so deep in occupied territory could only benefit their empire. 131 While it probably must remain impossible to determine too much about the location in which the text was discovered, the best possibilities remain some cultic location or – my preferred choice – something like an academic or educational setting.132

5.6.3 The Function The question still remains unanswered: what purpose did this text serve? Two aspects to this inquiry appear necessary. 1) To what end did someone compose this work? 2) Why did someone inscribe it at this location? Different answers to these questions appear likely, depending on how one describes the location. When understood as a cultic site, the purpose of the text could either be something like a hagiography about Balaam or a narrative manner to embed his divinely inspired vision and implied social critique. Either of these sounds like a reasonable explanation. The cultic interpretation of the site, however, creates a 127 Cf. the critique in Wenning and Zenger, “Heiligtum,” 184, that “…im Architekturbefund der Räume keinerlei besonderen Nutzungselemente gegeben sind, die eine Deutung als Teil eines Kultkomplexes anzeigen.” 128 People at home in modern European or North American churches could easily recognize something like a pew. 129 Cf. Lemaire, “Les Inscriptions sur Plâtre de Deir ʿAlla Det Leur Signification Historique et Culturelle,” 54 and Blum, “Institutioneller Kontext,” 36–40. 130 Cf. Blum, “Institutioneller Kontext,” 37–38. 131 Cf. Blum, “Institutioneller Kontext,” 41. 132 Wenning and Zenger, “Heiligtum,” 192 went even further, imagining the room as place for prophets and their students to gather and pass on their stories and oracles.

5.6 The Inscription’s Sitz im Leben

299

problem when considering why the text was inscribed where it was inscribed. Nothing about the text in itself has anything to do with typical cultic activities. One could hardly consider it something for normal visitors to a cultic site to engage with.133 It seems unlikely, both based on content and form, to be something read before a group as part of a ritual. The lack of a recognizable cultic function, therefore, commends interpreting the text’s location as a schoolroom setting. But would that interpretation change the function of the text? When understood as a classroom, two possible interpretations present themselves. Either it was something like an exam or it was something for students to copy. Possibly it was both. These suggestions at least explain why the text was found where it was. Someone copied it either as a Vorlage for others to copy as an exercise or someone wrote it down as part of an exercise or examination. Either of these possibilities would make sense in a classroom setting. Neither of them explains the original function of the text. Presumably no one composed it with the intention of using it for an examination or merely to serve as a scribal Vorlage. Therefore, one must inquire as to its original intent. With that, we can turn our attention back to the text’s supposed functions described in a presumed cultic setting. Either the story presents some kind of hagiography or it served to keep and transmit some prophetic uttering of the seer Balaam. Once again, it could be both. The missing conclusion precludes any certainty. Should the text have primarily functioned as the means to transmit Balaam’s oracular material, we could consider whether the text always existed as a unit.134 It remains possible, that if the text served as a literary structure for the prophetic material, that the oracles existed independently of the elements of narrative framing. The combination of the prose and poetry commends perhaps understanding the text in its composite state as a kind of moral teaching (‫ יסרי‬in ln. 1, v. 1) from the gods through Balaam to his audience and ultimately to the public of this inscription. The message is clear: moral inadequacy is dangerous.135 More so than the biblical Balaam oracles, the pairs presented in DAPT do not rely on their narrative or historical context. No presuppositions jump out in these statements and one would be hard-pressed to a concrete historical situation into which they might speak. These observations could suggest that the oracles originally existed independently of the narrative. Nonetheless, this cannot be stated as fact, but only as a possible thesis. Should it have been part of 133 Cf. Greenfield, “Philological Observations,” 119–20: “Who read this text? A priest, or a scribe, in all likelihood, rather than schoolboys, as has been suggested. This is surely not the way that reading was taught in the ancient world. Literacy was still very | limited in the mideighth century and inscriptions, when visible, were intended to make an impression by their presence rather than by their contents, since these were not readily available to the average attendant at a shrine.” 134 Cf. Blum, “Institutioneller Kontext,” 28–29. 135 Cf. Blum, “Institutioneller Kontext,” 29.

300

An Inscription from Deir ʿAlla. Balaam in Transjordan

some story about Balaam, something that I have anachronistically identified as a “hagiography”, one wonders why there is so much focus on his message and so little on his actions. The text says almost nothing about him. That suggests that he may have been known in more literature. The fragmentary nature of this text makes any conclusions more difficult. At the same time, it seems relevant that Balaam delivered this message. The title clearly ascribes it to him, and he personally engages with the gods. Therefore, we could regard this as an excerpt of a more substantial corpus of Balaam literature. Still, it remains impossible to discover whether this corpus consisted of more oracular or narrative material about Balaam. With these considerations and observations, we can now turn to the primary interest of this chapter: in what way(s) are the biblical and extrabiblical materials about Balaam related, if at all? To this we now turn.

5.7 The Relationship to the Bible’s Balaam(s) 5.7 The Relationship to the Bible’s Balaam(s)

Of course the primary reason that this text has received as much attention as it has is the reference to the figure Balaam son of Beor. While many have attempted to demonstrate quite a strong link between these figures, I would urge caution about taking the similarities too far. There is, for example, hardly enough evidence of a connection to identify the account of Numbers 22–24 as an Elohistic attempt to incorporate something like the Deir ʿAlla tradition into such an eponymous, postulated seventh-century pentateuchal source.136 As far as I see, at most five factors commend regarding the oldest text of Numbers 22–24* and the Deir ʿAlla inscription as outgrowths of a common “literary” tradition in the broadest sense. One is the name including its patronymic. Another is the fact that the main character has some kind of preternatural ability, or at least receives some divine nocturnal visitation. Each of these elements can be found in the oldest level of Numbers 22–24*. Less surely, the generally Aramaic character of the inscription’s language and the identification of Balaam with an Aramaic patronymic (‫ )בר‬could commend another common feature: Balaam’s originally Aramaic background, an origin in the region around Damascus. 137 While the DAPT permits that interpretation, the Bible makes it explicit (Num 23:7). Further, the divine term (‫ שׁדי)ן‬appears in both the oldest biblical Balaam narrative and in the DAPT. That could imply that in the oldest layer, the author(s) responsible for Numbers 22–24 still thought to connect Balaam to some ‫ שׁדין‬tradition, even if we no longer understand with certainty what that meant. The distinction in the form of the names ‫( שׁדין‬i.e., without the concluding ‫ נ‬in the Bible) presents a minor difficulty. Finally, in terms of style, both DAPT and the Bible present Balaam’s orations in the form 136 137

Cf. Robert K. Gnuse, “Redefining the Elohist,” JBL 119, no. 2 (Summer 2000): 204–6. Cf. Puech, “Balaʿam and Deir ʿAlla”.

5.7 The Relationship to the Bible’s Balaam(s)

301

of parallel couplets. Perhaps coincidentally, the oldest version present three sets of oracles, each consisting of seven pairs of parallel phrases, the same number of parallels attested in vv. 27–38 of DAPT. Other than these few things, the common elements remain thin and probably root in a more common idiom.138 Let me be explicit here, I do not think that it would be possible to affirm or really even suggest a common textual tradition behind them. Rather, we must consider a common tradition-historical background for the distinct literary figurations.139 Considering the redacted text of Numbers 22–24, one notes the commonality of the divine names in Numbers 24* and the inscription from Tell Deir ʿAlla. Both El and Shadday(in) appear in these texts. Unlike the term ‫( שדין‬as mentioned above), the form ‫ אל‬is consistent in both the Bible and the DAPT. How should one evaluate these common divine elements? Perhaps there is some common traditional element behind the redaction and the inscription, but this seems less certain and certainly less specific than in the aforementioned commonalities. The Bible uses the terms ‫ אל‬and ‫ שׁדי‬in a number of cases, particularly when describing the God of Israel in either older periods or when attempting antiquating language as in the book of Job.140 Since the terms ‫ אל‬and ‫ שׁדי‬appeared already in the oldest version of Balaam’s biblical oracles, editors could just appropriate and copy it into new circumstances. At the same time, one notes that ‫עליון‬, attested with both ‫ אל‬and ‫ שׁדי‬in Num 24:16, remains absent from the DAPT, as does ‫יהוה‬. Further, the solar deity (even understood just as the sun) is absent from the biblical text of Numbers. That could evince a more general common background for the deities mentioned in both texts; that is, ‫ אל‬and ‫ שׁדי‬both occur in both Numbers and DAPT as the result of coincidence, a happy accident. To maintain the presumably antiquated style, later editors increased the usage of ‫שׁדי‬. It remains most fundamentally impossible that later editors were familiar with this version of this Balaam text, since it was buried under the ruins of the building after the earthquake that destroyed it. Otherwise, the texts evince a general paucity of common language, particularly of less generic terms. For example, the term ‫ לילה‬has often been cited as a distinguishing feature. Yet, how else should one have referred to the natural, astronomical phenomenon of night? At most, one could argue for a common appreciation of the possibility of divine nocturnal visions. The common vocabulary consists essentially of common Hebrew or Aramaic words that do not 138

Cf. Hans-Peter Müller, “Alttestamentliche Probleme,” 57–58; Weinfeld, “Balaam Oracle,” 141–42; Puech, “Balaʿam and Deir ʿAlla,” 42–43; and Blum, “Institutioneller Kontext,” 30. 139 This evaluation conforms with that of David Carr, “Method in Determining the Dependence of Biblical on Non-Biblical Texts,” in Subtle Citation, Allusion, and Translation in the Hebrew Bible, ed. Ziony Zevit (Sheffield; Bristol, CT: Equinox, 2017), 48–49. 140 Cf. the discussion to these terms in the next chapter.

302

An Inscription from Deir ʿAlla. Balaam in Transjordan

permit the identification of specifics within a common traditional background. The form of embedding oracles in a narrative setting is not unique to DAPT and Numbers, but appears in many biblical texts as well. Stating that the form commends a common literary background of all such texts would be preposterous. Thematically, both versions of Balaam are unique. In the text from Deir ʿAlla he pronounces doom for his people and describes a world turned upside, a world in which the visible natural order has been reversed. Balaam in Numbers does quite the opposite. Rather than explain his people’s coming doom, he tells a foreign king about the blessedness of another foreign people. At least in the oldest version, this was the extent of his proclamation. In later redacted versions, he also explains to the king who summoned him how the blessed people will one day destroy the king’s own people, thus elucidating their doom. The blessedness of the one becomes the accursedness of the other. Nonetheless, Balaam remains outside both of these peoples in Numbers 22–24. Only in later texts like Numbers 31 does the reader encounter Balaam as a victim of this curse as well. Only at this later stage does he become a member of the doomed people. Additionally, the godly visions in Numbers 22 result from an inquiry, whereas Tell Deir ʿAlla mentions no such thing. Balaam receives his divine visitation without prompting from him or anyone else in the DAPT. Yet in spite of this distinction, the common background for each Balaam figure – that of the oldest version of Numbers 22–24* and that of Tell Deir ʿAlla – remains clear: he must have been associated with negative divinely inspired messages. The DAPT makes this explicit, but Numbers implies it as well. Were he not affiliated with such negative messages, there would have been no reason for Balak to summon him in Numbers. The character Balak and the implied reader should both expect Balaam to be able to deliver such a message. He certainly did not summon him to bless his enemy, as he reiterates in Numbers 23–24. Common to both Balaam figures are the manner of his divine revelations; in neither Numbers 22–24* (in its older version) nor in the Deir ʿAlla text does Balaam receive his vision because of something he does: he receives divine nocturnal visitation apparently because of who he is. In the DAPT he is not the ‫ קוסם‬of Josh 13:22, nor is he accused of ‫ לקראת נחשׁים‬as in the interpolation in Num 24:1. Balaam does nothing to influence the message in the oldest version; the texts simply make him the messenger. So, while it would be inappropriate to suggest that Combination A from Tell Deir ʿAlla and Numbers 22–24* have a strong literary or even tradition-historical background, the commonalities cannot be overlooked. In some fashion, the same figure must stand backstage of both dramas, to borrow a metaphor from theater. In this, we can consider the comments of Van Seters: “There is, therefore, no reason to believe that the biblical tradition of Balaam is directly dependent upon DAT [ = Deir ʿAlla Text]. One can account for everything in Numbers 22– 24 simply on the basis of a continuous popular Transjordanian tradition about a seer, Balaam,

5.7 The Relationship to the Bible’s Balaam(s)

303

son of Beor, who was associated with the region of Deir ʿAlla. The gods, El and the Shaddayin, may have been those of the region and hence their use in the biblical tradition with Balaam.”141

Both versions of Balaam presume that he had some preternatural connection to the divine. Both attest that he did not motivate this connection, but only received it passively – at least in the oldest version of the Numbers text. At the same time, the oldest version of Numbers apparently places his origin in Aram far to the north of Tell Deir ʿAlla. Though the inscription’s generally Aramaic language could imply such, the Deir ʿAlla text does not explicate this affiliation. Neither does it explicitly preclude it. Balaam might have been a figure of interest in a local tradition of local background. However, based generally on the language of the text, one should not preclude Combination A presupposing his Aramaic background. The contrast comes from Balaam speaking to and about his own people in Combination A, but to and about two other people in Numbers. In sum, we can identify the following connections between the Balaam tradition of Numbers 22–24* and the DAPT, Combination A: 1) the figure Balaam, who is the son of Beor; 2) this figure receives divine visitation at night; 3) he apparently does nothing to provoke this visitation, receiving it only passively; and 4) he delivers the message to an audience. Less certain, 5) he was affiliated with (‫ ;שׁדי)ן‬6) he may have been an Aramean; 7) both in the Bible and in the context of the find, Balaam appeared in Transjordan; and 8) his divinely inspired messages were delivered in pairs of parallels. However, the following distinctions must not be overlooked: 1) Numbers’ Balaam is certainly from Aram, whereas the origin of Balaam in Deir ʿAlla is at best ambiguous; 2) the later iteration of Balaam in Deuteronomy 23 is from Mesopotamia; 3) he delivers the divine message to a foreigner in Numbers 22–24* but to his own in Combination A; 4) his audience consists of only one person in Numbers 22–24, but a collective at Tell Deir ʿAlla; 5) Moab plays no role at all in DAPT, but the central role in Numbers 22–24 and in many other Balaam traditions; 6) Numbers 22–24* describes multiple divine visitations, whereas Combination A reports only one; 7) Balaam regrets the message he should give in Combination A, but shows no qualms about this in Numbers 22–24; and, finally, 8) Balaam delivers a message of hope to an object of scorn in Numbers 22–24, whereas he delivers a message of hopelessness to his own people in Combination A. The versions of these stories about Balaam demonstrate a common tradition-historical background, but one that should not be overstated. 142 The commonalities of both texts revolve exclusively around the figure Balaam, who 141

Van Seters, Life of Moses, 411. For example, Levin, Jahwist, 386 regarded the opening of the DAPT and Num 24:4b, the opening of the oldest oracle in his pre-Yahwistic version of Numbers, as essentially identical. That’s stretching the similarities too far. 142

304

An Inscription from Deir ʿAlla. Balaam in Transjordan

he was, and what his capabilities were. The other circumstances and implications of the texts remain otherwise distinct, precluding overstatement of similarities between the versions.

5.8 Conclusions: What Does the DAPT Reveal about the Bible’s Balaam? 5.8 Conclusions: What Does the DAPT Reveal about the Bible’s Balaam?

Four matters in the DAPT reveal substantial information and points of contact about the epigraphic and biblical versions of Balaam. First, the inscription’s Aramaic language generally affirms Aramaic domination of the region during the ninth or eighth century. The archaeological and paleographical dating of this inscription around the year 800 BCE commend ascribing this domination to Hazael or his successor. In terms of relating this to the oldest biblical version of Balaam in Numbers 22–24*, both appear to presuppose cultural contact between Aram-Damascus and the Transjordan around the period of the composition of each text. The time of original composition appears similar, with both texts, DAPT and the oldest core of Numbers 22–24 stemming from around 800 BCE ± 50 years. Second, the Aramaic language of the inscription could indicate an Aramaic background for the tradition of the Balaam figure, rather than a Transjordanian (or Mesopotamian) heritage for the tradition.143 This character as attested at Deir ʿAlla may represent an alien tradition brought along with the conquering forces as a cultural good.144 This also matches the image of Balaam presented in the oldest version of Numbers 22–24*, in which he comes from Aram (Num 23:7). Despite these Aramean connections, however, one cannot overlook the Canaanite characteristics of the inscription’s language. That implies, third, that Transjordan was not always or even necessarily very long part of the Damascene empire, if I might use that term loosely. The Canaanite characteristics described in the survey above could indicate that, by the time of this particular composition, Aramean hegemony of the region had come to an end or that it had not completely replaced the Canaanite dialect spoken and apparently written there. Perhaps Aramaic only had the opportunity to leave a thin layer on top of the local linguistic tradition. At any rate, the Canaanite characteristics could indicate a local variety of a common flavor of Balaam material. Finally, and most importantly, both the oldest biblical version and the DAPT version of Balaam demonstrate knowledge of this figure as some kind of a 143 Contra, e.g., Mathias Delcor, “Le texte de Deir ʿAllā et les oracles bibliques de Balaʿam,” in Congress Volume, Vienna 1980, VTSup, vol. 32 (Leiden: Brill, 1981), 52–73, who sought an Ammonite background for the figure Balaam. 144 Cf. Wenning and Zenger, “Heiligtum,” 192: “Die Texte von TDA [Tell Deir ʿAlla] erlauben auch nicht, irgendeine Verbindung zwischen dem Wirken des historischen Bileam und dem Ort TDA zu rekonstruieren.”

5.8 Conclusions: What Does the DAPT Reveal about the Bible’s Balaam?

305

mantic or – more specifically – a visionary or seer, apparently a well-known and competent one. The dating of the DAPT to the ninth or the eighth century BCE commends regarding the oldest biblical narrative, the one most similar in content and tone to the DAPT, as coming from that same period.145 To this end, then, it appears that the oldest narrative layer behind Numbers 22–24* demonstrates an Israelite appropriation of a perhaps initially Aramean figure for a literary conflict against the Moabites. Taken together, this all suggests that cultural contacts between Aram, Israel, and the Transjordan occurred during the ninth or eighth centuries BCE and that probably some corpus of literature about Balaam must have developed or been passed on during that period, perhaps even among these cultures. How improbable would it be, if we had access to the only two pieces of literature about this same figure from almost 3000 years ago? Nonetheless, the question must remain open at this point whether this oldest Aramean Balaam was a historical personage or merely a literary figure in some dramatic works of fiction.146 To aid in answering this question, among others, we can now turn our attention to the tradition-historical background of the various iterations of Balaam.

145

Though it cannot be precluded that even earlier versions or collections of material about Balaam existed. In accepting this dating and our ability to date this oldest biblical text indirectly on archaeological grounds, I am admittedly more optimistic than, e.g., Ziony Zevit, “Scratched Silver and Painted Walls: Can We Date Biblical Texts Archaeologically?” HS 48 (2007): 23– 37. 146 Not only does Mathias Delcor, “Texte de Deir ʿAllā,” 73 date this figure to the middle of the eighth century BCE, but also regarded him as a historical figure. While that is possible, it must remain uncertain and unverifiable. Cf. Ashley, Numbers, 439: “The Deir ‘Allā texts do not demonstrate or even confirm that this Balaam was a historical character as the Bible portrays him.”

Chapter 6

Balaam as a Character His Origin and Background The different biblical and extrabiblical materials about Balaam remain largely reticent about essentially all facets his background, apparently favoring innuendo or even disparate information to concrete data. This remains particularly true for Balaam’s special ability, which only once ever receives a specific name in the biblical literature, and it is pejorative. Yet, there is some information that can be gleaned from the witnesses. Various texts provide oblique data – sometimes contradictory – about his origin, profession, and religious context. Here, these data will be briefly considered, compared, and contrasted in order to fully appreciate the disparate background and station of the character Balaam. My considerations will begin with his geographical or ethnic background, before turning to his profession and religious identity in the traditions. More importantly, this chapter will address the tradition-historical backgrounds of the various authors and editors of the Balaam material, as demonstrated through their attitudes about him.

6.1 What’s in a Name? 6.1 What’s in a Name?

To date there has been no satisfactory explanation of the name Balaam.1 Two primary possibilities exist: the name either comes from the combination of ‫בל‬ + ‫ עם‬or from ‫ בלע‬+ ‫מ‬. In the first case either element could be theophoric, meaning either “Bel is my paternal uncle / protector” or “(My) lord is Amm”. While both are possible, neither is certain or particularly clear. The greatest difficulty determining the correct understanding comes from the paucity of attestations of the west Semitic ‫ בעל‬being abbreviated to ‫בל‬.2 Considering the 1 To the following cf. Gray, Numbers, 324–25 and Erasmus Gaß, “Bileam.” http://www.bibelwissenschaft.de/stichwort/15381/. 2 Cf. Scott C. Layton, Archaic Features of Canaanite Personal Names in the Hebrew Bible, HSM (Atlanta: Scholars Press, 1990), 171: “No other Canaanite PNN in the Hebrew Bible have the DN Baʿal spelled without the ʿayin, and neither the MT vocalization nor the LXX supports this interpretation.” Further, cf. the attestations and discussion in Ran Zadok, On West Semites in Babylonia During the Chaldean and Achaemenian Periods (Jerusalem: H. J. & Z. Wanaarta & Tel Aviv University, 1978), 69–71.

6.1 What’s in a Name?

307

other alternative, the moniker coming from the verbal root ‫בלע‬, one first would have to explain the meaning of the root. Two possibilities present themselves again here: either it means something like “swallow, destroy” (as in the Hebrew and Talmudic Aramaic) or it is related to the Arabic blġ (‫ )ﺑﻠﻎ‬and means something like “well-spoken”; cf. Job 37:20. 3 The evidence remains ambiguous. Further, the concluding particle ‫ מ‬would still have to be explained.4 The last possible meaning available resorts to folk-etymology and identifies Balaam as the “swallower / destroyer of the people”, ‫ בלע‬+ ‫עם‬.5 This seems to be the most unlikely possibility for the original meaning, but it still may be relevant for the later appreciation of the character in the negative evaluations that developed around him. Balaam’s name presents the most important aspect and it is the only consistent factor in the various traditions about him. It remains the only reason that texts as diverse as Numbers 22–24; Deuteronomy 32; Joshua 31; 24; Micah 6; and the DAPT are regarded as related to one another. No other elements in these texts demonstrate such affinity that a relationship could satisfactorily be postulated. The name Balaam in the various biblical texts and its extrabiblical occurrence dramatically increase the plausibility of a broader ancient Balaam tradition, whether in some oral or literary form, about a special person named Balaam and his substantial capability (see further below). However, our inability to ascertain beyond mere speculation the meaning of Balaam’s name and his patronymic limits the tradition-historical yield about how the authors evaluated him. That is, the name tells us nothing of value about the kind of groups that might have initially composed stories about Balaam or where Balaam, should he have been a historical figure, might have originated. It remains impossible to determine based on the etymology his name, if one should reckon with a Canaanite, Aramean, or some other heritage, for example. The most common descriptor for Balaam is his patronymic, “son of Beor”. It can be found in most biblical pericopes about Balaam and even in the inscription from Tell Deir ʿAlla, which identifies Balaam only in this manner,

3 These Hebrew and Arabic definitions are also noted in Gesenius, et al., Gesenius, 153–54. For the Talmudic Aramaic, cf. Marcus Jastrow, Dictionary of the Targumim, the Talmud Babli and Yerushalmi, and the Midrashic Literature, compiled by Marcus Jastrow (New York: Title Publishing, 1943), 174–75. Epigraphically, only Nabatean attests ‫ בלע‬epigraphically; cf. Hoftijzer and Jongeling, DNWSI, 67. 4 Due more or less to the lack of better alternatives, Layton, Archaic Features, 170–72 regarded this ‫ מ‬as an enclitic particle. 5 Cf., e.g., Keil, Leviticus, Numeri und Deuteronomium, 312. Indeed, this etiology has led some to identify Balaam with Luqmân, known from the Qurʾan; cf. Eduard Meyer, Die Israeliten und ihre Nachbarstämme, reprint, 1906 (Darmstadt: Wissenschaftliche Buchgesellschaft, 1967), 378–79. Against this position, cf. already William F. Albright, “Islam and the Religions of the Ancient Orient,” JAOS 60, no. 3 (September 1940): 288 and the literature cited there.

308

Balaam as a Character. His Origin and Background

providing him no other background information.6 In spite of the almost ubiquitous attestation of Balaam’s patronymic (it is curiously absent in the quote of Deuteronomy 32 found in Nehemiah 31 and in Josh 24:9 G), little can be said of its meaning. Exegetes encounter the name Beor also in Gen 32:63 and 1 Chr 34:1, each of which connects this name with the Edomites. 7 Ascribing Balaam’s background to the Edomites seems most unlikely based on all of the other available evidence. However, the notices in Gen 32:63 and 1 Chr 34:1 presumably lie behind some scholarly attempts to emend the text of Num 7:32 to read ‫ אדם‬instead of ‫ ;ארם‬cf. the conjectural footnote to this effect in BHS. Even as the name Balaam remains ambiguous, the name of his father presents no fewer difficulties. The root ‫ בער‬in Hebrew can mean “burning”, “cleaned out or cleared out”, or “to be stupid or uncultivated”.8 None of these seem adequate explanations for the name, and yet Balaam’s father’s name is almost as well attested as Balaam’s in the witnesses, in most biblical texts and the Deir ʿAlla Plaster Text (DAPT). Nonetheless, Balaam’s name and his patronymic remain at best ambiguous and at worst deride him, at least in folketymological interpretation of the names (“People-swallower Stupidson”). The identification of Balaam as Beor’s son unfortunately does not aid in defining any traits of his background. Yet this remains among the most consistent data about Balaam attested in the primary literature. Other than perhaps the possible folk etymology, the name of Balaam’s father does not seem to serve any particular purpose other than identifying this Balaam as the important one. The virtual ubiquity of the patronymic suggests its import in spite of the paucity of real data it presents.

6.2 Balaam’s Heritage 6.2 Balaam’s Heritage

In the various witnesses one finds a number of fleeting references to Balaam’s ethnic or geopolitical background. These will be covered here, beginning with Numbers, followed by the other biblical materials, and then bringing in any information from the Deir ʿAlla Inscription. The usage of these terms will be considered in its current and any reconstructed earlier contexts.

6 Cf. Num 22:5; 24:3, 15; 31:8; Deut 23:5; Josh 13:22; 24:9; Mic 6:5. The inscription attests this name in, e.g., ln. 2 and it can be readily reconstructed in ln. 1. 7 Cf. the discussion of these verses in Chapter Four. 8 Cf. Gesenius, et al., Gesenius, 164–65. The Talmudic Aramaic understanding generally only reflects the meaning “to burn”; cf. Jastrow, Dictionary of the Targumim, 182–83. No epigraphic usages supply a reliable meaning of the root; cf. Hoftijzer and Jongeling, DNWSI, 185.

6.2 Balaam’s Heritage

309

6.2.1 ‫פתורה אשׁר על־הנהר‬ According to M and G, Balaam comes from Pethor on the river. Even before the deciphering of cuneiform, some already sought this city in Mesopotamia,9 probably based on the notice in Deuteronomy 23 M. Since the identification of the city Pitrû in Akkadian inscriptions, scholars quickly presumed these two locations to be identical, with the river in question being identified as the Euphrates.10 In general, this identification has been accepted in scholarship.11 Yet, it remains problematic.12 The city Pitrû was not on the Euphrates, but on the Sāǧūr, its tributary. Such an identification in Numbers presumes that the audience would be familiar both with this city and with the river it was on to the point that the river would not even need to be named. The renaming of Pitrû to Ana-aššur-utter-aṣbat by – at the latest – the reign of Shalmaneser III (Monolith Inscription col. II ln. 85–86) further complicates its identification with biblical Pethor. Others have gone another route. Because of the location of the discovery of the Deir ʿAlla Inscription, Levin follows Smr, S, and V and read ‫עמון‬ regarding Balaam’s origin in Num 22:5 (see below). “Die Unsicherheiten über seine Herkunft, wie die Textüberlieferung von Num 22,5 sie wiedergibt, lassen sich danach entscheiden: Bileams Wohnort Petor lag ‘im Lande der Ammoniter’, nicht am Euphrat, wie Spätere annahmen.”13 This is probably exactly the reverse of what happened. The latest text traditions (S and V) affirm Balaam as an Ammonite (probably under the influence of Deuteronomy 23), whereas G does not. Most likely the biblical reading attests a place otherwise unknown or even fictional: the suggestion that ‫ פתורה‬as in M and G should be understood as “visionary land” (cf. ‫ הפותר‬as presumed in V and T) does commend itself.14 Such an interpretation could even suggest a fictional locale; that later biblical authors could easily subsume it into Mesopotamia probably stems from some readiness to identify such matters with Mesopotamian religion. While it thus seems highly improbable that the earliest authors and redactors of Numbers 9

Cf., e.g., Keil, Leviticus, Numeri und Deuteronomium, 316. Cf., e.g., Gray, Numbers, 325–26. Josephus also identifies this river as the Euphrates; cf. A.J. 4.104. Because of the location of the discovery of the inscription from Deir ʿAllā, Barré, “Portrait,” 259 identified this river as the Wadi Jabbok. 11 Cf. Mathias Delcor, “Balaʿam Patorah, intrete des songes au pays d’Ammon d’apres Nombres 22:5. Les temoignages épigraphiques paralleles,” Semitica 32 (1982): 89–91; Manfred Görg, “Die ‘Heimat’ Bileams,” BN 1 (1976): 24–28; and the commentaries. HansPeter Müller, “Alttestamentliche Probleme,” 61 rejected Pitrû as Balaam’s homeland, but failed to reject the identification of Pitrû and ‫פתור‬. This presumably spurious identification has even made its way into the lexica: cf. Gesenius, et al., Gesenius, 1090. 12 Cf. the issues presented and literature cited in Layton, Archaic Features, 35–37. 13 Levin, Jahwist, 384. 14 Cf. the discussion about this verse in the chapter on textual criticism. 10

310

Balaam as a Character. His Origin and Background

22–24 regarded Balaam as coming from Pitrû, later scribes apparently did think he came from some city called Pethor that was located by a river. The mention of the river could potentially be understood as a method of differentiating this Pethor from another, perhaps even more familiar, Pethor. Unfortunately it is impossible to reconstruct what they may have been thinking of and any attempt to do so presents little more than speculative guessing, since no other location called Pethor is known. Perhaps it was a city in Transjordan. Perhaps Pethor is fictional. I favor this latter resolution, regarding it as a kind of folk etymology: Balaam is a visionary from Seertown.15 The current contextual usage presents the most pressing issue however. Disregarding for the moment the term ‫ פתורה‬and concentrating only on the reference to the river, it seems apparent that the river should serve as some boundary. At the very least, the mention of Balaam being summoned from elsewhere presents the reader with just that datum: he is neither from Moab nor – presumably, though this is less certain – Cisjordan. Even in the oldest version of the text, it should be clear that Balaam travelled from another region near a river; the reticence about the place’s name suggests that it should remain unknown, at least to this point in the story. Should it have been intended to refer to the Euphrates, one must question why the scribes should not have included the name. Other usages in the Hebrew Bible demonstrate that they were certainly capable of this, suggesting that they intentionally left it ambiguous and unresolved for narratological reasons. This ambiguity, in combination with a potentially fictional town, suggests intentional obfuscation, while at the same time making Balaam’s background seem legitimate and historical by providing his home town with a name, albeit a metaphorical one. By naming a particular town in an unnamed place by an unnamed river, they lent credibility to the account, making it less fanciful, while at the same time permitting or implying the association with interpreting dreams by keeping ‫ פתר‬and leaving Balaam’s geopolitical and ethnic background initially opaque. During the reception history of this verse, the tradition changed from its original sense in that it increased in specificity. The original datum merely implied that Balaam came from elsewhere, a city of interpreters in an unnamed land, which may have even manifested in Balaam’s legendary status, the very reason that Balaak summons him in the tale in Numbers 22–24. Later scribes – as in V and S – make this connection explicit and identify Balaam as “the interpreter”. Other later tradents, such as those behind Deuteronomy 23 M turned him into a Mesopotamian, an ascription followed in G (see below) and which many modern scholars have affirmed. 15 At the same time, one cannot entirely preclude a coincidence. There are presumably gastroenterologists in Darmstadt (literally, “intestine-city”), but no one would therefore deny the existence of Darmstadt based on a story about a particularly gifted gastroenterologist who happened to come from there.

6.2 Balaam’s Heritage

311

6.2.2 (‫ארץ בני־עמו)ן‬ In ML and G, Balak sends delegates to reach Balaam in “the land of the sons of his people”, a rather curious and ambiguous phrase. One could presume that Balaam was in the land of his kin even without this information. Some have suggested that the ‫ עמו‬in this phrase refers to Balak and not Balaam.16 That would mean that Balak, not actually a Moabite but still the king of Moab, was sending back to his homeland to retrieve Balaam. No evidence supports this interpretation.17 The other attempt to rectify and clarify this text comes from modern exegetes who – with recourse to the Idrimi inscription – identify an area in the Levant as “Amau”.18 While this is admittedly possible, the basis for this identification is rather slim.19 Others have suggested that the term ‫ עמו‬refers to ʿAmau known from Egyptian usage.20 Both of these understandings of ʿAmau remain problematic.21 The phrase ‫ ערץ בני־עמו‬in this context reiterates that Balaam comes from elsewhere and was not a Moabite. Leaving his background general allows the audience to interpret more freely whence he came or even consider him “the foreigner” in general. He represents not a particular ethnicity or political group, not a particular people, but rather all other non-Moabite and non-Israelite peoples. The phrase ‫ ארץ בני־עמו‬identifies Balaam as a foreigner – even the foreigner – in the story, indeed as someone foreign to both other peoples in Numbers 22–24. For this reason particularly, later scribes appended this ascription. The division of peoples into three groups occurs elsewhere in the Balaam oracles, in 24:9b. The first redaction added this piece and therefore probably also the phrase ‫ ארץ בני־עמו‬in 22:5. Balaam thus fulfills the capacity of the “one who blesses” for R1, as the opposite of Balak’s “one who curses”. 16

Cf., e.g., Sutcliffe, “A Note on Numbers XXII,” 441–42; Eric Burrows, The Oracles of Jacob and Balaam, Bellarmine Series, vol. 3 (London: Burns, Oates & Washbourne, 1938), 78; and Otto Eißfeldt, “Die Komposition der Bileam-Erzählung. Eine Nachprüfung von Rudolphs Beitrag zur Hexateuchkritik,” ZAW 57 (1939): 226. This understanding has its roots in Rabbinic interpretation, as seen in Midrash Rabbah and Rashi. 17 Cf. the discussion in Gross, Bileam, 113. 18 Cf., e.g., William F. Albright, “Some Important Recent Discoveries: Alphabetic Origins and the Idrimi Statue,” BASOR 118 (April 1950): 15–16 and Y. Liver, “The Image of Balaam in Biblical Tradition [Hebrew],” EI 3 (April 1954): 97–100. 19 Cf. the discussion in Seebass, Numeri III, 15. 20 For Egypt, cf. Abraham Shalom Yahuda, “The Name of Balaam’s Homeland,” JBL 64 (1945): 547–51. 21 Yahuda’s suggestion, that this references the land ʿĀmu found in Egyptian records from the third and second centuries BCE remains chronologically problematic. Were this the case, Num 23:5 would be the only known attestation of this identification in the whole of Semitic literature, a rather improbable scenario. Cf. also Gross, Bileam, 107–8. For a refutation of Albright, cf. Scott C. Layton, “Whence Comes Balaam? Num 22,5 Revisited,” Bib 73, no. 1 (1992): 44–45.

312

Balaam as a Character. His Origin and Background

The curious phraseology of ML may have been the reason – or presumably was at least one of the reasons – for the text-critical variants in many Masoretic manuscripts, Smr, S, and V that turned Balaam into an Ammonite.22 The ascription to Balaam of an Ammonite heritage appears to even have made its way into the Arabic literary tradition, though not that of the Qur’an.23 The witnesses distinct from ML and G sought to clarify his origin through the addition of ‫ ן‬and transformed him into an Ammonite.24 Another factor behind this variant reading may have been the connection of Moabites and Ammonites in the biblical literature; cf., e.g., Gen 19:30–38; Judges 11; and especially Deut 23:4–6. The tradition of Balaam as an Ammonite reflects a genuine attempt by later scribes to make a difficult text more cogent in itself and more consistent with the greater biblical corpus.25 He therefore became an Ammonite living in Mesopotamia, as he is in the Masoretic version of Deut 23:4–6. Through the course of reception history, the tradition was adapted for the purposes of consistency, perhaps with some theological or ideological impetus. Balaam was neither a resident of Amau nor an Ammonite in the oldest biblical versions. Later tradents and modern scholars appended these attributes to his character. Initially scribes added this characterization of his being in the land of the sons of his people in order to emphasize that Balaam was neither part of the exodus group nor an autochthonous Moabite.

6.2.3 ‫ארם‬ Balaam opens his first oracle in Num 23:7 with that statement that he is from Aram, without any further specification. Generally, in the Bible “Aram” without any modifiers – such as Padan (‫ )פדן‬or Naharaim (‫ – )נהרים‬refers to the region around Damascus in the Syrian steppe.26 Almost exclusively, it refers to 22 Such is also the position of, e.g., Layton, “Whence Comes Balaam?” 46–50. Layton based this ascription on 1) narrative logic (travelling from Aram would take too long on a donkey) and 2) the findings from Deir ʿAlla. To number 1: first he must demonstrate the unity of these data in Numbers 22–24, which he nowhere did. To number 2: I’m hesitant to simply interpret this biblical text based on an archaeologically discovered epigraphic find without further consideration. Both may attest some Balaam tradition in the region, but reading them together should be demonstrated as necessary and not simply presumed as such. 23 See the helpful discussion of the Arabic sources about Balaam in al-Ghul, “Question of the Homeland”. Al-Ghul demonstrates that the Arabic sources tend to affirm Balaam’s Ammonite, Transjordanian background, though some sources claim he came from Yemen. However, contrary to what al-Ghul states, the Arabic sources do not affirm the “original” biblical understanding of Balaam as an Ammonite. More likely, this geographical understanding of Balaam’s origin reflects Arabic familiarity with textual traditions that mention that Balaam was an Ammonite, i.e., the Samaritan Pentateuch, the Peshitta, or even the Vulgate. 24 Contra, e.g., Lust, “Balaam”, who regarded his Ammonite heritage as the older version. 25 Contra Barré, “Portrait,” 255. 26 Cf. the discussion in Younger, Political History of the Arameans, 64–104.

6.2 Balaam’s Heritage

313

this polity during the monarchic period.27 These data suggest 1) that Balaam came from the kingdom of Dasmascus and 2) this ascription comes from the monarchic period. Later editors did not always accept or transmit this ascription, editing it in a number of ways. At the same time, the descriptor “Aram” was vague enough to permit his identification with a variety of peoples. Many biblical places are identified with the epithet “Aram”: Padan-Aram, Aram-Damascus, Aram-Maacah, Aram-Naharaim (= Mesopotamia), and Aram-Zobah. Thus, texts like the later Balaam tradition in Deuteronomy 23 noted that he came from ‫ארם־נהרים‬. The Septuagint expounds the reading in Num 23:7, professing his having come ἐκ μεσοποταμίας to meet Balak. This presumably occurred under the influence of Deut 23:5. Numbers 23:7 identifies Balaam as an Aramean in some form without clarifying the specifics. But, as in the previous examples, the reception of this text demonstrates the continuing interpretation of the character Balaam beyond the original tradition, here particularly as seen in G. Perhaps G sought to smooth out the Numbers text with Deuteronomy; perhaps G viewed Mesopotamians as either a more widely known people or as a more appropriate origin for someone versed in augury or some related field. These interpretations nonetheless represent later interpretations of the term Aram. The most common usage of “Aram” without any qualifiers does commend understanding Balaam as coming from the region around Damascus. That presents the standard use of this term in the Hebrew Bible and is probably the oldest ethnic or geographical ascription of this character’s heritage in Numbers and in the Bible as a whole. This matches, perhaps, the background of this figure in the inscription from Tell Deir ʿAlla. The Aramaic language of this text could support an Aramean tradition-historical background for the figure Balaam. Admittedly, the term Aram does not appear in the inscription, but the connection should not be cast aside prematurely. The most likely oldest presentation of Balaam, both in the Bible and beyond, appears to have some connection to Aram, i.e., Damascus, and the Aramaic language. This ascription, whether biblical or not, appears to go back to at least the monarchic period in Israel. Numbers 22–24 presumes a historical background in which it was conceivable that a Moabite would contact an Aramean for assistance in overcoming Israel. Few such historical circumstances stand out, but the most probable would be some time in the ninth or eighth century BCE, when – as we know from other biblical and extrabiblical sources – Israel’s relationship with these two peoples was hardly free of tension.28 Balaam comes from a kingdom that 27

The exceptions are Gen 10:22–23; 22:21; Judg 3:10; 10:6; 1 Chr 1:17; 2:23; and 7:32. Cf. the Mesha Inscription, the Tel Dan Inscription, 1 Kgs 20 MT; 22; 2 Kgs 3; 6–7; 10:32– 33; 12:17–18; 13, and the historical reconstruction of the period in, e.g., Robker, Jehu, 285– 302. 28

314

Balaam as a Character. His Origin and Background

had difficult relations with Israel during the monarchic period, and visits a kingdom with their own problems with Israel during this same era. Taken together, these data commend ascribing Balaam an Aramean heritage in the oldest biblical tradition, but perhaps also in the DAPT. Later interpreters developed this tradition, transposing his background from the region around Damascus to Mesopotamia.

6.2.4 ‫מהררי־קדם‬ In Num 23:7bα, Balaam iterates that he came to visit Balak “from the mountains of the east” or “from the mountains of Qedem”. This reference to the east or Qedem seems to stand in tension with other data about Balaam in Numbers. Keil, without really citing further evidence, identified ‫ קדם‬with Mesopotamia.29 Mowinckel understood the mountains of the east as a reference to Edom and placed Balaam’s background entirely in that context. 30 That appears to have been both premature and incorrect. None of the other descriptors in Numbers could be consistent with “east” unless Pethor was somewhere east of Moab. The city Pitrû – should it be identified with Pethor – is neither to the east of Moab nor among Syrian mountains.31 The plains of Mesopotamia – to the east of ‫ – ארם‬hardly present a possible candidate for consistency with the phrase “mountains of the east”; to reach mountains near Mesopotamia, one would have to go all the way to the Zagros Mountains of Iran. Such information makes Num 23:7bα G and Deuteronomy’s identifications of Balaam as a Mesopotamian problematic. Three viable possibilities remain: either Balaam should be understood as coming from Qedem or Ammon or the phrase ‫ מהררי־קדם‬is not about Balaam.32 The first option is possible. “From the Egyptian story of Sinuhe, Qedem seems to designate a specific territory in the Syrian desert (east of the Phoenician coast). This accords with its designation here as being synonymous with Aram. [...] However, the second term ‘hills’ is hardly appropriate since the specific territory of northern Aram / Qedem is not conspicuously mountainous.”33 This understanding would add further support to Numbers’ original identification of Balaam as an Aramean. When read chiastically, this appears to be the favorable understanding: the inner terms (“Balak” and “king of Moab”) refer to the same 29

Cf. Keil, Leviticus, Numeri und Deuteronomium, 327. Cf. Mowinckel, “Ursprung,” 236–37. 31 This is contra Gray, Numbers, 346, who – based on his identification of Pethor as Pitrû – had to have Balaam reference the mountains of Syria here in spite of Pitrû’s not being in the mountains. 32 Qedem could reference the region between the Antilebanon Mountains and the Euphrates, which would match much of the other geographical data delivered regarding Balaam, at least in Numbers 22–24; cf. Yahuda, “Name,” 549–51. 33 Milgrom, Numbers, 196. 30

6.2 Balaam’s Heritage

315

individual, just as the outer terms (“Aram” and “mountains of Qedem / the east”) refer to the same locale. The second option, the Ammonite resolution, presents many of the same problems as the aforementioned: while it is true that Ammon has mountains (the modern Jordanian capital Amman has an average altitude of about 1000 m above sea-level), it is hardly consistent with the identification of Balaam as having come from Aram, unless of course this region had been incorporated into some Aramean “empire”. The ascription of Ammonite heritage to Balaam represents a later scribal development (see above). Should the Transjordan have been under Damascene hegemony during the period of the composition of the oldest layer of Numbers 22–24*, one could understand this again as a reference to Balaam’s Aramean heritage. That would suggest that he was an Aramean in Transjordan when Balak summoned him. But this understanding remains particularly tenuous. To the third option: Since these two identifiers – Aram and Qedem / east – occur within the same sentence and no conspicuous literary-critical basis for separating them exists,34 one could re-examine the syntax and make a decision based on that. The two clauses that make up Num 23:7bα could be regarded as one conjoined phrased rather than as two distinct phrases; i.e., “From Aram Balak, the King of Moab from the mountains of the east, called me,” as opposed to “From Aram Balak called me, the King of Moab from the mountains of the east”. Three matters must be mentioned in this regard. 1) The Masoretes may have understood the text in this way. Support for such an interpretation can be seen in their use of conjunctive (as opposed to disjunctive) accents here. 2) This phrase further distinguishes Balaam’s background from Balak’s; one is Aramean and the other is Moabite. 3) Such an understanding of this phrase implies that Balaam is taking on the perspective that Moab is east of him, i.e., Balaam speaks from an Israelite perspective. Precisely Balaam’s acceptance of the Israelite perspective presents an interesting aspect. Although Numbers’ narrator and the character Balaam himself have reiterated Balaam’s generic foreignness, here the tenor changes: Balaam takes on Israel’s perspective of Moab as someone in Cisjordan, as an insider in Israel. At this point it becomes clear how the story and oracles will develop; indeed, it’s here in the first line of Balaam’s first oracle. From here on out, it will be impossible for Balaam to turn against Israel in the older iterations of the story. The first redactor seized this poignant moment and changed its understanding through the addition of the new pinnacle of Balaam’s (redacted) oracles in 24:9b: “whoever blesses you is blessed, and whoever curses you is cursed”. In 24:9 the oracles come full circle and Balaam belongs to the group of the blessed in that he blesses Israel (23:20 G); in 23:7 he has already taken on this perspective in his first oracle’s opening. Such foreign acceptance of an 34

See the discussion in Chapters Three.

316

Balaam as a Character. His Origin and Background

Israelite perspective reminds the reader of another biblical Aramean: Naaman (2 Kings 5). A brief summary of the results of this examination from Numbers 22–24 is in order. Numbers 22–24 presents hardly any unambiguous claims about Balaam’s geopolitical or ethnic background. The primary objective of the first identifiers are to distinguish him from the Israelites and Moabites. While most commentators accept that he should be identified as coming from the city Pitrû, this remains implausible. Probably his city is a pun on the ‫פתר‬. His affiliation with Ammon remains just as tenuous based on the manuscript evidence and must be regarded more appropriately as a creation of the earliest stages of reception history. Only the ascription to Aram remains, and this seems the most likely identification of his background in the text of Numbers 22–24, generic though it is. The phrase “from the mountains of the east / Qedem” in Num 23:7 is difficult and cannot be resolved with certainty. Initially it probably pointed toward an Aramean origin, but it later served to distinguish Balaam from Balak and identify his perspective as identical with Israel’s. In this way, the edited narrator reverses expectations about Balaam and starkly contrasts his foreignness with that of Balak. This presents one of the primary foci of the primary redaction of the Balaam narrative of Numbers 22–24. Thus, the older versions (S1 and R1) apparently regarded Balaam as an Aramean.

6.2.5 ‫מדין‬ Numbers 31 includes Balaam among a list of Midianite rules, although it fails to mention whether he was a Midianite or not.35 These verses, Num 31:8 and 16, are later redactional elements that serve to combine Numbers 22–24 and 25 with Numbers 31 into a more cogent narrative whole.36 It may be one of the latest elements in the book of Numbers and suggests that later tradents viewed Balaam as being connected to the Midianites, even if they did not necessarily say he was a Midianite. This inclusion fails to clarify his background in that it contends with all other data about him. Joshua 13:22 also records this tradition (and it may be younger than Num 31), but qualifies Balaam more by identifying him as ‫( הקוסם‬see below). Ultimately Balaam’s affiliation with the Midianites provides little background information about his character. It does, however, demonstrate that the editors who expanded the book of Numbers (R2 and R3+) found fit to include him among Midianites, even cooperating with them to the detriment of the Israelites.

35 36

For general information about the Midianites and Midian, cf. Knauf, Midian, 1–6. Cf. Chapters Three and Four.

6.2 Balaam’s Heritage

317

6.2.6 ‫מפתור ארם נהרים‬ Deuteronomy 23:5 M suggests that Balaam came “from Pethor, Aram Naharaim” (i.e., Pethor in Mesopotamia). The matters regarding Pethor have already been considered above. Here, I can briefly reiterate that this identification is again text-critically insecure (G and even V do not proffer any evidence of Pethor having been in their Vorlagen, whereas Smr reads the identical form found in Num 22:5 – clear evidence of expounding and scribal copying).37 The identification with Pitrû remains even more problematic in that it is not strictly within the bounds of Mesopotamia.38 Balaam’s Mesopotamian heritage only agrees moderately with the Hebrew text in Numbers, which has Balaam coming from Aram and not Aram Naharaim.39 It seems plausible that the author or an editor in Deuteronomy either 1) expounded on the Numbers tradition of Balaam coming from Aram, 2) copied from something like the Vorlage of the Greek version of Num 22–24, or 3) was familiar with a different origin for this (in Deuteronomy) antagonist. The matter cannot be decided with any certainty, but I tend to favor the idea that the party responsible for this note in Deuteronomy combined elements from Num 22:5 (‫ *הפותר‬and ‫ )נהר‬with an element from Num 23:7 M (‫)ארם‬. Thus, the author responsible for this notice in Deut 23:5 was probably familiar with the text of Numbers 22–24*, though willing to adapt it for later historical circumstances.40 This interpretation developed perhaps due to a greater appreciation of Mesopotamia as locus for such mantic activities. Even in modernity,

37 Cf. the discussion to Deuteronomy 23 in Chapter Four for a more in-depth text-historical analysis. 38 Cf. however the notice and literature in Gray, Numbers, 326: “…Aram-naharaim is not merely the country between the Euphrates and the Tigris; it is the Naharin (River-country) of the Egyptian inscriptions, called Nahrima or Narima in the Tel el-Amarna tablets – a district which appears to have extended from the valley of the Orontes eastwards across the Ephrates.” The issue regarding the particular identification of biblical ‫ ארם נהרים‬is somewhat thorny; cf. already Jacob Joel Finkelstein, “Mesopotamia,” JNES 21 (1962): 84–86, who did not think that the form refers to the two rivers. At the same time, the attestations of ‫ארם נהרים‬, irrespective of whether the form is specifically dual or not, tend to regard the region we currently – and from the Hellenistic Period onward – refer to as “Mesopotamia”, more specifically “Upper Mesopotamia”. Part of the difficulty in identifying the geographical antecedent of the phrase ‫ ארם נהרים‬stems from readings of the biblical materials that do not undertake text-critical evaluations and therefore identify, based solely on Deut 23:5 M and the (mis)identification of Pethor with Pitrû, are forced to expand the boundaries from the eastern bank of the Euphrates to include the western bank and even areas further afield as well. Cf. most recently Younger, Political History of the Arameans, 96. 39 The exception to this is of course Num 23:7 G. 40 Cf. Martin Rose, 5. Mose 12–25. Einführung und Gesetz, ZBK (Zürich: Theologischer Verlag Zürich, 1994), 326.

318

Balaam as a Character. His Origin and Background

some have sought to affiliate Balaam with the religious tradition of Mesopotamia.41 In the context of Deuteronomy this notice has the same function as the notices in Numbers: it establishes that Balaam does not belong to either of the conflicted peoples.

6.2.7 Conclusions about Balaam’s Ethnicity and Geopolitical Background Little uniformity and clarity define the biblical background for Balaam. The oldest text in Numbers 22–24* identified him as coming from “interpreter’s city upon the river” and from “Aram”. Other elements, like Qedem, remain unclear. To these older ascriptions, the first editor added his coming from “the land of the sons of his people”. Later scribes emended these ascriptions within the Numbers tradition, making him Mesopotamian (G) or Ammonite (Smr, S, and V). His Mesopotamian identity also appears in Deuteronomy 23. To this disparate image, one should also add his inclusion with the Midianites in later textual traditions, particularly in Numbers 31 and Joshua 13. While neither of these texts clearly identify him as a Midianite, the association with them and their common context could possibly imply some Midianite heritage for the character Balaam. Such an interpretation would transpose him from the northern Levant into the south, even toward Arabia. Thus the biblical description of Balaam’s background remains quite disparate and vague.42 Even less can be said about his origin in the DAPT, in which only the language of the composition could play a role in identifying this character’s background. In the Bible, one finds a smattering of inconsistent notes about where Balaam is from, many of which appear to be the result of interpolations or corruptions in the proto-Masoretic and various other textual traditions. Numbers 22–24* apparently regarded him initially as an Aramean, as the character himself professes in Num 23:7. The other nomenclature ascribed to him in Numbers 22–24 resulted from errors and later interpretation: he is not from Pitrû and he is not an Ammonite. The text of Deuteronomy locates his origin in Mesopotamia, which seems to present a conglomeration of data found in Numbers 22–24. Still other texts mention nothing about his background, most notably Neh 13:2 and Mic 6:5; the latter mentions only his patronymic while the former mentions only his name. Ultimately, the character Balaam’s origin remains a mystery, making it difficult to offer any certain reconstruction of the background of any such historical figure, should he have ever existed, or his literary heritage. At most – relying on the notice in Num 23:7 and the site and language 41

Cf., e.g., Daiches, “Balaam” and René Largement, “Les oracles de Bileʿam et la mantique suméro-akkadienne,” in Mémorial Du Cinquantenaire. 1914–1964, École de langues orientales anciennes de l’Institut catholique de Paris (Paris: Bound & Gay, 1964), 37–50. 42 Cf. the curious attempt to make sense of all of these factors by compiling them in Selah Merrill, “Balaam’s Birthplace,” The Hebrew Student 1, no. 4 (July 1882): 75.

6.3 Balaam’s “Profession”

319

of the DAPT – one could consider Aram or possibly (northern) Transjordan. These remain the most likely options for the origin of this historical or literary figure. Balaam’s foreignness seems to be of import to Numbers 22–24, and the general ambiguity about where he is from allows him to be understood as a kind of chimera for foreigners in general. Particularly for the editors, beginning with R1 in Numbers, this appears to be the case. What remains most significant narratologically is that he is neither Israelite or Moabite in Numbers 22–24. While particularly true of the text in Numbers 22–24, Deuteronomy 23 also denotes his foreignness vis-à-vis the other peoples. In the end, the best conclusion about the oldest Balaam tradition is that he was from the region of Damascus or Transjordan under Damascene hegemony. These considerations do not aid in determining whether this Balaam was a historical figure or only a literary product, but only describe at least his cultural characterization in the oldest traditions. Many other elements in his character merit further examination and may aid in determining whether he was historical or literary. Thus the question: Can one glean any other meaning from the generally ambiguous data about Balaam’s background in the Bible (and from DAPT, for that matter)? To further investigate this matter, we must consider Balaam’s role in the various witnesses.

6.3 Balaam’s “Profession” 6.3 Balaam’s “Profession”

Like Balaam’s ethnic and political background, one finds a variety of disparate data reflecting on his – for lack of a better term – profession.43 Again, the data vary to some degree between the different textual traditions. Was he a prophet? Was he a magician? Or something else?44 Only a few nouns are used to reflect upon his job for Balak; one must primarily reconstruct what his job is based on the verbs about his attempts to influence Israel for his benefactor. These will be addressed below, beginning with Numbers 22–24, then considering the 43 Contrast this survey with the rather superficial coverage in Walter C. Kaiser, Jr., “Balaam Son of Beor in Light of Deir ʿAllā and Scripture. Saint or Soothsayer?” in “Go to the Land I Will Show You.” FS Dwight W. Young, ed. Joseph E. Coleson and Victor Matthews (Winona Lake, IN: Eisenbrauns, 1996), 95–106. Cf., however, the helpful summary of ancient opinions about Balaam in Gilmore H. Guyot, “Balaam,” CBQ 3, no. 3 (July 1941): 235–37 and the primary and secondary literature cited there. 44 Cf. Otto Eissfeldt, “Wahrsagung im Alten Testament,” in La divination en Mésopotamie ancienne et dans les régions voisines. XIVe Rencontre Assyriologique Internationale (Strasbourg, 2 - 6 juillet 1965) (Paris: Presses Université de France, 1966), 141–46 (= Otto Eissfeldt, “Wahrsagung im Alten Testament,” in Kleine Schriften. Vierter Band, ed. Rudolf Sellheim and Fritz Maass [Tübingen: Mohr Siebeck, 1968], 271–75) in which Eissfeldt presumes that the development of prophecy in Israel superseded such mantic practices known from other cultures.

320

Balaam as a Character. His Origin and Background

other biblical texts about Balaam, and ultimately considering any relevant data that can be found in the DAPT. These matters will also each be addressed in their own context.

6.3.1 ‫( פותר‬Num 22:5) The Peshitta and Vulgate identify Balaam in this manner, as an interpreter, whereas the lexeme in the other witnesses refers to a geographic locale and not a profession (see above). Some have argued for this reason that this was the original understanding in Numbers 22–24.45 All other occurrences of the Hebrew root ‫ פתר‬can be found in Genesis 40–41 and Sir 50:27.46 One can also consider the Hebrew and cognate Aramaic root ‫ פשׁר‬found in Qoh 8:1, Sir 34:18, the 31 attestations in Daniel, and more than 100 attestations at Qumran: “to solve, interpret; to annul a charm” etc.47 “Der kontextuelle Zusammenhang ist in allen [biblischen] Belegen überraschend einförmig: es geht um eine Deutung eines von einem König vorgetragenen Problems, fast überall um die Deutung eines Traumes des Königs.”48 Balaam’s interaction with Balak parallels Daniel’s interaction with Nebuchadnezzar and Joseph’s interaction with Pharaoh and his servants, and since this verb has prophetic and apocalyptic undertones in the Daniel and Joseph story, one could identify the same undertones in the Balaam story. Understood this way, the Numbers versions in S and V suggest that Balaam functioned primarily as an interpreter (of dreams?) who could potentially annul Israel’s blessing, something that Balak apparently desired of him. However, unlike Joseph, Numbers 22–24 never unequivocally mentions that Balaam was a dreamer himself. Rather, God came to him in the night, with it specifically being a dream.49

45

Cf., e.g., Layton, Archaic Features, 38–42 and the literature cited there. Gen 40:5, 8, 12, 16, 18, 22; 41:8, 11–13, 15. Michael S. Moore, The Balaam Traditions: Their Character and Development, SBLDS (Atlanta: Scholars Press, 1990), 81, in fact, regarded the dream report in the DAPT as closer to that of Joseph (in Genesis 37) than to that of Balaam in Numbers 22. Problematic with this interpretation is that neither the DAPT nor Numbers 22 explicates that God or the gods came to him in a dream. 47 Cf. the definitions in Gesenius, et al., Gesenius, 1008 and 1094 and Hoftijzer and Jongeling, DNWSI, 946–47. 48 Heinz-Josef Fabry and Ulrich Dahmen, “‫פשׁר‬,” in Theologisches Wörterbuch zum Alten Testament, vol. 6 ‫קום–עזז‬, ed. Heinz-Josef Fabry and Helmer Ringgren (Stuttgart; Berlin; Köln: Kohlhammer, 1989), 812. 49 Frank Schnutenhaus, “Das Kommen und Erscheinen Gottes Im Alten Testament,” ZAW 76 (1964): 14–19 regarded ‫ בוא‬in the context of visitation as the most common, imprecise kind of epiphany in the HB. He related it to incubation rituals known from Mesopotamia and Egypt, which had the simple goal of relaying a divine message. However, Schnutenhaus regarded it primarily as evoking dream language, something missing in Numbers 22–24, though he listed Balaam as someone who received these dreams. 46

6.3 Balaam’s “Profession”

321

The text’s ascribing this quality to Balaam at the time of his summoning indicates that Balaam must have practiced this profession even prior to his contact with the Moabites. That is, he is known as such a professional. Like Joseph and Daniel, he represents a foreigner being summoned by a monarch to interpret and thus influence a potentially disastrous situation for the monarchs’ subjects (the drought and famine in Egypt; the downfall of Babylon; the devastation of Moab). On the other hand, Balaam remains distinct from Joseph and Daniel in that Balak orders him to come from another territory, whereas Joseph and Daniel were already living as foreigners in the domain of the monarch who enlisted their aid. At the same time, no text ever explicates that Balaam was a dreamer or an interpreter of dreams. That again distinguishes him from Joseph and Daniel. By describing Balaam in this manner, S and V make the wordplay of his ascribed domicile more explicit; he becomes the interpreter, and not merely a citizen of “interpreter-ville”. A similar conception of Balaam stands behind the DAPT, in which the gods appear to Balaam at night, even though the DAPT does not attest ‫ פתר‬or its Aramaic cognate ‫פשׁר‬.

6.3.2 ‫( ארר‬Num 22:6, 12; 23:7; and 24:9) The basic meaning of this Hebrew root is “to curse” or “to be cursed”, but such a definition hardly clarifies what it means.50 This root is epigraphically attested in Hebrew (cf. KAI 191), but the inscription KAI 191 does not illuminate what the curse entails, merely that it results from opening the grave.51 In cognate languages, Akkadian attests this root with a variety of meanings (arāru A).52 These will be dealt with further below. The antonym of ‫ ארר‬is ‫ברך‬.53 50 For the following terminology, cf. Herbert Chanan Brichto, The Problem of “Curse” in the Hebrew Bible, JBL Monograph Series, vol. 13 (Philadelphia: Society of Biblical Literature and Exegesis, 1963), 77–117; Rüdiger Schmitt, Magie im Alten Testament, AOAT, vol. 313 (Münster: Ugarit-Verlag, 2004), 123; and the discussion about blessing and curses within the confines of family religion in Rainer Albertz and Rüdiger Schmitt, Family and Household Religion in Ancient Israel and the Levant (Winona Lake, IN: Eisenbrauns, 2012), 410–13. “Blessings and curses were not magical acts that merely manifested the inherent power of words but were ritual acts of speech that mediated or anticipated God’s favor or disfavor and were performed by a person endowed with authority, even when that authority was merely situational.” (Albertz and Schmitt, Family, 410). 51 Cf. Herbert Donner and Wolfgang Röllig, Kanaanäische und aramäische Inschriften (Wiesbaden: Harrasowitz, 2002), 44. To epigraphic sources mentioning cursing, cf. Albertz and Schmitt, Family, 491: “The most common context for curses in epigraphic sources is tomb inscriptions to prevent the family’s cave from being disturbed or robbed.” 52 For the various attestations and meanings, cf. Oriental Institute of the University of Chicago, The Assyrian Dictionary Volume 1: A, Part 2, CAD (Chicago: Oriental Institute, 1968), 234–38 and below. 53 For the meanings and usages of ‫ברך‬, cf. Aimo Murtonen, “The Use and Meaning of the Words Lebårek and Beråkåh in the Old Testament,” VT 9, no. 2 (April 1959): 158–77. Murtonen

322

Balaam as a Character. His Origin and Background

Several things can be cursed, the most common of which are individuals. The others include animals (Gen 3:17; Deut 28:18), objects (Gen 3:17 [the land]; 5:29 [the land]; Deut 28:17 [basket and kneading bowl]), emotions or abstracts (Gen 49:7 [anger]; Jer 20:14 [the day of one’s birth]; Mal 2:2 [blessings]), and finally a group of people (Gen 9:25 [Canaan, although narratologically here it refers to a person], the people/Israel (Num 22:6, 12; 23:7), and the Gibeonites (Josh 9:23). From this list, one notes that cursing something other than an individual is more the exception than the rule. Of what did curses consist? A rather broad spectrum of usages better circumscribe its semantic field than a simple definition. Though the Hebrew Bible attests this root some 63 times, no unambiguous meaning can be gleaned from the contexts. Most occurrences just state something like “cursed is…” without providing an explanation of what the curse entails.54 Only in a few cases, does one discover the specific content of a curse and what it implies for those afflicted. The specific content of cursedness (‫ )ארר‬in the Hebrew Bible includes: 1) the serpent wandering on its belly, eating dust, enmity with humans (YHWH to the serpent in Gen 3:14); 2) the land becoming difficult to work (YHWH about the land in Gen 3:17; Noah’s naming in Gen 5:29); 3) difficulty in working the land and finding a home (YHWH to Cain in Gen 4:11); 4) being a slave (Noah curses Canaan in Gen 9:25; Joshua cursing Gibeonites in Josh 9:23); 5) being divided and scattered as a tribe (Jacob curses the anger of Simeon and Levi in Gen 49:7); 6) pain, vaginal discharge, hysterectomy55 (the bitter water that causes the curses; in the judgment of an unfaithful wife in Num 5:18–19, 22, 24, 27); 7) the death of two children (Joshua curses the person who rebuilds Jericho in Josh 6:26); 8) darkness (those who curse the sea in Job 3:8, but the context remains murky); 9) drying out and inhabiting arid regions (those who trust in people instead of in YHWH in Jer 17:5); 10) being overthrown (Jeremiah curses the day of his birth and the man who reported his birth to his father in Jer 20:14–15); 11) the rejection of offerings, faces covered in the dung of offerings, cast out of God’s presence (YHWH curses the false priests in Mal 2:2); and 12) being cursed implies having poor agricultural yields (cursed are those who rob YHWH of his due in Mal 3:9).

regarded the usages in Numbers 22–24 as “general blessings”, but identified the most general understanding of this root as “fertilization”. 54 Cf. Gen 12:3; 27:29; Exod 22:27; Num 24:9; Deut 27:15–26; 28:16–19; Judg 5:23; 21:18; 1 Sam 14:24, 28; 26:19; 2 Kgs 9:34; Ps 119:21; Jer 11:3; 48:10; and Mal 1:14. 55 For this ritual, one could consider a connection to the Akkadian arāru C: “to rot, to discharge a putrid liquid, to defecate”; cf. Oriental Institute of the University of Chicago, CAD 1.2, 237–38.

6.3 Balaam’s “Profession”

323

Apparently it generally remains reserved for some kind of exclusion from a group or from well-being.56 Cursedness often implies poor agricultural circumstances or yields.57 The most common usages in which the contents of the curse can be reconstructed suggest subservience, pain, difficulty, enmity, or poor living standards (Gen 3:14, 17; 4:11; 9:25; 49:7; Numbers 5; Josh 9:23; Jer 17:5; 20:14–15; Mal 2:2; and 3:9), with death being the exception (Josh 6:26; 1 Sam 14:24, 28?). A distinct case appears in Josh 9:23 regarding the Gibeonites: their being cursed will lead to their enslavement. Behind Gen 9:25 lies the same idea about Canaan, and even though the text narratologically addresses him as an individual, it is clear that he serves as a synecdoche for the whole group.58 All of these examples about peoples develop the same theme: the devaluation of the people vis-à-vis another group or contrary to current conditions. Thus, biblically speaking, cursing in the sense of ‫ ארר‬suggests that someone will have a difficult position or be enslaved to someone else, without the details being further clarified. Finally, in the fewest cases, one could regard the content of the curse as the opposite of specific blessings named. This is precisely the case in Numbers 22– 24 and only in Numbers 22–24. The authors and editors never explicated the content of the curse specifically, but one can regard them as the opposite of named blessings. That is, cursedness in Numbers 22–24 would entail living among the nations, not counting itself alone, and being a nation with a finite number members, should one contrast Num 23:9b–10a with Num 23:7. In Num 22:6 (and also therefore 22:12), the implication is that cursing the people would enable their military defeat. Taken together, then, we can appreciate the understanding of ‫ ארר‬in Numbers 22–24. Balak wishes the people to be cursed so that he might be able to defeat them in battle (Num 22:6). He hopes to weaken them, since they are mightier than him, and then drive them from the land. Balak seeks to demean their conditions in respect to his and in respect to their current condition. Only indirectly would the cursing have anything to do with death. Balak seeks Balaam in order to manifest the devaluation of the people. Further evidence for this understanding of ‫ ארר‬can be found in the Akkadian cognate arāru, which can mean “to curse” and “to treat with disrespect”.59 This disrespect must have been understood to have concrete adverse effects on the people or person cursed, without always specifying what that adverse effect was. At the same time, other meanings from Akkadian suggest an important relationship to the 56

Cf. the discussion in Josef Scharbert, “‫ארר‬,” in Theologisches Wörterbuch zum Alten Testament, vol. 1 ‫גלה–אב‬, ed. Gerhard Johannes Botterweck and Helmer Ringgren (Stuttgart: Kohlhammer, 1973), 437–51. 57 Cf. Gen 3:17; 4:11; 5:29; Jer 17:5; and Mal 3:9. 58 Cf. Albertz and Schmitt, Family, 412: “The Noah curse on his son Canaan in Gen 9:25– 27 serves more as an etiological synecdoche for Israel’s superiority over Canaan.” 59 Cf. Oriental Institute of the University of Chicago, CAD 1.2, 234–36.

324

Balaam as a Character. His Origin and Background

Hebrew root: on some occasions Akkadian arāru can take on the meaning “to beset with fear” or “become agitated”. These meanings would overlap with the sense of treating the people with disrespect, which would then take on the manifestation of fear. 60 In Numbers 22–24 the demanded curse would be detrimental to Israel’s ability to engage in combat. The first redactor (R1) inserted this term for cursing into the oldest version of the Balaam story, as demonstrated in Chapter Three. It was foreign to the oldest layer, but has intruded into elements that we can identify in that layer.

6.3.3 ‫( קסם‬Num 22:7; 23:23; and Josh 13:22) Attested in a variety of cognates in a number of Semitic languages (e.g., ancient Southern Arabic),61 the root ‫ קסם‬possesses the basic meaning of “separate, decide” and can thus be understood in the context of oracles as “augur, seek an oracle” or “cast lots”.62 That is, it represents a general term for deductive or instrumental mantic practice.63 With the probable exception of Prov 16:10, this root is never used in a positive manner in the Bible, only pejoratively.64 The Septuagint aids in appreciating this understanding in that it consistently translates this Hebrew root with μαντεία and μαντεύω, etc. The usage of this root in the various portions of the Bible leads to the impression that ‫ קסם‬presents “ein Thema dtn/dtr und schriftprophetischer Literatur”.65 The Deuteronomic / Deuteronomistic literature in general loads this term with negative baggage; cf., e.g., Deut 18:10, 14, 1 Sam 15:23; and 2 Kgs 17:17, as well as the occurrences in Jeremiah and Micah. Sometimes this behavior is ascribed to nonIsraelites, as is the case in Josh 13:22; cf. 1 Sam 6:2; Ezek 21:26–27. First Samuel 6 suggests that Philistine ‫ קסמים‬and priests undertook some activity in order to divine how to end their suffering, but the text does not describe how they arrived at their answer, only that they did so. (And it was apparently appropriate and efficacious, though 1 Samuel 6 does not explicitly state as much; it nonetheless presumes it worked.) The story of the medium at Endor (1 Sam-

60

Cf. arāru B in Oriental Institute of the University of Chicago, CAD 1.2, 236–37. Cf. Walter W. Müller, “Altsüdarabische Beiträge zum hebräischen Lexikon,” ZAW 75, no. 3 (1963): 314. 62 Cf. the etymological discussion in Lothar Ruppert, “‫קסם‬,” in Theologisches Wörterbuch zum Alten Testament, vol. 7 ‫שׁכן–קוץ‬, ed. Heinz-Josef Fabry and Helmer Ringgren (Stuttgart; Berlin; Köln: Kohlhammer, 1990), 79. 63 Cf. T. Witton Davies, Magic, Divination, and Demonology Among the Hebrews and Their Neighbours Including an Examination of Biblical References and of the Biblical Terms, reprint, 1898 (New York: Ktav Publishing House, Inc., 1969), 44–47 and 78–79 and Rüdiger Schmitt, Mantik im Alten Testament, AOAT, vol. 411 (Münster: Ugarit-Verlag, 2014), 46–47. 64 Cf. Ruppert, “‫קסם‬,” 83–84. 65 Ruppert, “‫קסם‬,” 80. 61

6.3 Balaam’s “Profession”

325

uel 28) demonstrates the efficacy of such practices and that they were undertaken within Israel. Endor is elsewhere described as within the boundaries of Manasseh (cf. Josh 17:11). That which Samuel’s ghost reports even comes true. Still, this Samuel text fails to acknowledge precisely what the woman did to raise Samuel’s ghost. It only recounts that his ghost came and told the future without noting how they inspired his ghost to return from the realm of the dead. At least in this case, but distinct from the others, the term ‫ קסם‬appears to represent a kind of necromancy. Ezekiel 21 provides an essentially unique example in the Bible of what this practice might have entailed. At least it puts it within the context of other mantic practices. Ezekiel 21:26–28 describes the king of Babylon making a decision about whom to attack. Four manners of his determining how to act are listed in the text: “to decide by oracle” (‫)לקסם־קסם‬, to shake arrows, questioning the teraphim, and examining the liver. Each of these methods are known from Akkadian sources. In this context, the king’s choice can only have two outcomes, suggesting that the ‫ קסם‬could only offer two possible responses, like flipping a coin. At the same time, this decision may have been guided by which hand contained the oracle: Ezek 21:27 notes that Jerusalem’s oracle is “in his right hand” (‫)בימינו‬, suggesting that Rabbah’s oracle may have been in his left hand, though the text never explicitly states as much. At least in the usage of Ezekiel, however, it seems likely that relying on ‫ קסם‬could have resulted in one of two answers, with other possible methods of divine inquiry listed subsequently. This understanding of ‫ קסם‬distinguishes Ezekiel from the other observations proffered thus far. Regarding Balaam, only Josh 13:22 identifies him as ‫הקוסם‬, though the root ‫ קסם‬can be found in Num 22:7 and Num 23:23. In the first instance in Numbers it is unclear to what exactly this should refer. The standard interpretation for Num 22:7 (“payment for prophecy”) is most improbable, since no evidence for this understanding can be found until Jerome translated the Hebrew for the Vulgate (divinationis pretium).66 Others have tried to demonstrate the opposite, but never with evidence; cf., e.g., Wagner’s interpretation of Num 22:15–17: “je größer das Honorargebot und je prominenter die Bittsteller, um so gewisser die Hoffnung auf Erfüllung des Wunsches.”67 Rather, in its current context, it

66 Cf. the brief notice to this effect in Robker, “Balaam Narrative,” 339–40. To aid in this understanding, Ashley, Numbers, 447 regarded the ‫ קסמים‬in 22:7 as “fees for divination” based on the comparison to ‫ בשׂורה‬meaning both “wages for good news” (2 Sam 18:20, 25) and “good news” itself (2 Sam 4:10; 18:22). Michael Heltzer, “‘True’ and ‘False’ Prophecies [Hebrew],” EI 23 (1993): 74–77 regarded the advanced payment for prophetical undertakings as the sign of a prophecy’s falsehood. 67 Siegfried Wagner, “Offenbarungsphänomenologische Elemente in den Bileam-Geschichten von Numeri 22–24,” in Ausgewählte Aufsätze zum Alten Testament, ed. Dietmar Mathias (Berlin: de Gruyter, 1996), 88.

326

Balaam as a Character. His Origin and Background

appears that the messengers brought the oracle with them that they wanted Balaam to explicate.68 They were trying to persuade him by bringing their message with them. This notice, which is presumably a late addition into its context, should reflect negatively on Balaam. The recognition of ‫ קסם‬as a kind of implement in Ezekiel 21 elucidates the usage in Num 22:7, which reports that the elders have this in their hands when they set out to call Balaam (S2). Perhaps originally, S2 reflected some other magical undertaking, as proposed in the literary-critical reconstruction in Chapter Three. If that reconstruction is correct, the usage here did not even initially apply to Balaam. Ezekiel 21 could also inform the usage in Num 23:23 (R1), though this is less certain. Here, ‫ קסם‬parallels ‫ נחשׁ‬and may indicate some destructive practice undertaken against Israel (see below to ‫)נחשׁ‬.69 In Num 23:23, Balaam himself states in a double entendre that there is no ‫ קסם‬in Israel and that no ‫ קסם‬is efficacious against Israel. He nowhere admits to be someone trying to proffer a ‫ קסם‬against Israel. No evidence contradicts his statement. Balaam’s proclamation in this case demonstrates – quite effectively – his positive relationship to Israel. In the usage of ‫ קסם‬in Numbers 22–24 we again identify a case in which the root was used in an older version (Num 23:23 belongs to the first redaction) before another hand interpolated another meaning (with the addition of Num 22:7a). Originally, it attested Balaam’s positive view of Israel, before it was placed in the hands of Balak’s messengers to suggest that they were coming to visit Balaam with his message already fit to order, to deliver as they saw fit. The later Balaam tradition attested in Josh 13:9–10 identifies Balaam pejoratively with this term, representing the culmination of the interpretative undertaking against this figure as attested in the HB.

6.3.4 ‫( קבב‬Num 22:11, 17; 23:8, 11, 13, 25, 27; and 24:10) This root again means “to curse” or “to be cursed”.70 Biblical texts attest the ‫ קבב‬significantly less often than √‫ארר‬, with the majority of occurrences in Numbers 22–24: Num 22:11, 17; 23:8, 11, 13, 25, 27; 24:10; Job 3:8; 5:3; Prov 11:26; and 24:24. Most of the usages again defy easy semantic clarity (Num 22:17; 23:11, 13, 25, 27; 24:10; Job 3:8; Prov 11:26; and 24:24), but some information can be gleaned from the context of a few verses. For example, Num 22:11 expresses the same sense as 22:6, namely that the one cursed should be able to be more easily defeated. The context of Job 5:3 indicates that the accursed have children in danger with no one to save them and have difficulty 68

Cf. Hurowitz, “The Expression ûqsāmîm beyādām” and Victor (Avigdor) Hurowitz, “What Did Balak’s Emissaries Bring to Balaam? [Hebrew],” EI 23 (1993): 83–86. 69 Cf. the also discussion in Rouillard, Balaam, 301–9. 70 Cf. Brichto, Problem, 200–202; Rouillard, Balaam, 81–86; and Rüdiger Schmitt, Magie, 124.

6.3 Balaam’s “Profession”

327

with their agriculture and water resources (cf. Job 5:4–5). These ideas overlap to some degree with those of ‫ארר‬, but it would be inappropriate to suggest a total agreement. The ‫ קבב‬fails to cover the important aspect of subservience and servitude covered by the ‫ארר‬. The antonym of ‫קבב‬, like that of ‫ארר‬, is ‫ברך‬.71 That this root can be regarded as generally synonymous with ‫ ארר‬in Numbers 22–24 cannot be doubted. The singular proof necessary to this effect can be found in their ability to be interchanged. Comparison of Balak’s message to Balaam and Balaam’s narrating the situation to God, as well as God’s response initially demonstrates this: Balak asks Balaam to come curse (‫ )ארר‬Israel in Num 22:6; Balaam tells God that Balak wants him to curse (‫ )קבב‬Israel (Num 22:11); in Num 22:12 God refutes Balaam’s inquiry and says that Israel is blessed and cannot be cursed (‫)ארר‬. Additionally, according to the first missive in Num 22:6, Balaam should come to curse (‫ )ארר‬Israel, and in the second missive (Num 22:17) he should come to curse (‫ )קבב‬Israel. Finally, in Balaam’s first oracle, the terms are used as parallel elements of a chiastic structure. Thus it is apparent that ‫ ארר‬and ‫ קבב‬generally present perfect synonyms in Numbers 22–24, though the literary-critical evaluation of these chapters demonstrated that these terms stem from different authors. They both cover the aspect of weakening the people and making them easier to conquer. At the same time, ‫ ארר‬may include the aspect of enslavement or servitude, which may have been foreign to ‫קבב‬. Based on the small sample of attestations, however, it is impossible to confirm this understanding. Unlike ‫ארר‬, ‫ קבב‬occurred in the oldest version of Numbers 22–24, which attests the majority of biblical appearances of this root. It presumably appeared in a few other cases in those chapters, which later scribes then overwrote with ‫( ארר‬Num 22:6a and 23:7). The DAPT supposedly contained the ‫ קבב‬according to the original publication in Combination II, ln. 17. However, Hackett revised the reconstruction and interpretation of the DAPT, ultimately removing this root from its text based on her observations.72 Blum has followed her in his most recent reconstruction.73 Otherwise, ‫ קבב‬is attested in Phoenician and Punic with a similar meaning. In Ugaritic, the root means “to bend”. These northern cognates could aid in identifying the provenance of the original narrative of Numbers 22–24; they might imply a northern origin for the tale. A similar root, qabû, is known from Akkadian with the meaning “to speak, recite, make a proclamation,

71

Cf. the discussions in Hoftijzer and Jongeling, DNWSI, 977–78 and Helmer Ringgren, “‫קבב‬,” in Theologisches Wörterbuch zum Alten Testament, vol. 6 ‫קום–עזז‬, ed. Heinz-Josef Fabry and Helmer Ringgren (Stuttgart; Berlin; Köln: Kohlhammer, 1989), 1138–39. 72 Cf. Hackett, Balaam Text, 73–74. 73 Cf. Blum, “Kombination II”.

328

Balaam as a Character. His Origin and Background

etc.”.74 This can even take on the tone of delivering omens or reciting prayers.75 Such an understanding would round out the image of Balaam in the DAPT and HB, should a connection to Akkadian be clearly discernible.

6.3.5 ‫( עלה‬Num 23:2 and 4) While strictly speaking, Numbers 23–24 does not really present Balaam as the one making offerings (only in the interpolation in 23:4), he does instruct Balak to do so on several occasions in the later iterations. In this he takes on an almost priestly capacity in these chapters, at least in terms of his advisory role.76 He tells Balak what must be done in preparation for Balaam’s encounters with God, including building altars. For this reason, Moore identified Balaam here as an “exorcist” engaged in his role as a “purification-priest”.77 Perhaps, but supporting evidence for this claim remains conspicuously narrow. Wagner regarded the offerings in this passage as evidence that divine revelations must have had their place in cultic contexts, at least in a way such that the audience would recognize in this story what was happening.78 Again, the evidence is sparse, and perhaps this recounting carries a negative connotation about such activities no longer apparent to us. The text mentions nothing about what kind of altars were built, leading Gray to opine that they were “…doubtless of material on the spot, such as earth or undressed stones (cp. Ex. 2024f.)…”79 While this postulation possesses a certain logic, it certainly goes beyond the bounds of the text as it stands in Numbers 22–24. While the details remain somewhat shrouded (How big were the altars? From what material were they made? Was the same material reused for each successive set of altars? Does the color of the animals matter? Could they have imperfections? For which deity or deities were the offerings provided?), Numbers suggests that Balaam must have some experience with offering, or at least with instructing monarchs how to do it.80 Fundamental in this regard is that the original story apparently lacked any such capacity for Balaam (see Chapter Three). In the oldest narrative Balak presumably sacrificed at his own behest.

74 Cf. Oriental Institute of the University of Chicago, The Assyrian Dictionary Volume 13: Q, CAD (Chicago: Oriental Institute, 1982), 22–42. 75 Cf. Oriental Institute of the University of Chicago, CAD 13, 29–30. 76 Cf. the similar observation in Greene, Balaam and his interpreters, 59. 77 Cf. Michael S. Moore, “Another Look at Balaam,” RB 97, no. 3 (July 1990): 374–75. 78 Cf. Wagner, “Offenbarungsphänomenologische Elemente,” 92. 79 Gray, Numbers, 342. 80 For this reason, Rost, “Fragen um Bileam,” 379–80 rejected Mesopotamia as a legitimate background for Balaam, since Mesopotamians did not offer burnt sacrifices. He argued this position, contra Daiches, “Balaam”. Ashley, Numbers, 436–37 went in a distinct direction from Rost and Daiches, seeing the offerings as evidence of Balaam working as an ašipu.

6.3 Balaam’s “Profession”

329

The postulated later hand would have made Balaam culpable for these sacrifices, while at the same time making him responsible for their increase in number. The redacted narrative could imply some kind of priestly background – at least in his advisory capacity – for the character Balaam, though this remains inconclusive. Should one accept Daiches’ argumentation, the editors R3+ would have converted Balaam into a Mesopotamian bārū, a status he did not initially obtain.81 At the same time, the kind of sacrifice that Balaam demands Balak matches nothing otherwise known in the Bible. “These are the most expensive (hence, efficacious) animals in the sacrificial system...”82 Several other biblical texts mention bulls and rams being offered together, yet obviously none of them recount several altars. Most of them do not even present identical number of bulls and rams, but which Balak and Balaam purportedly sacrifice in Numbers 23–24. Regarding the sacrificial regulations, the most similar texts can be found in Num 29:1–11 (the festival of trumpets and Yom Kippur) and 29:35– 38 (the last day of Sukkoth). These observations could imply that Balak proffers kind of an anti Yom Kippur or Sukkoth at Balaam’s behest in the edited version. However, it must be noted that the offerings for the Israelite festivals all require offerings not found at Balak’s sacrifices. In narrative texts, the most similarities can be found with the offering of thanks that the Levites prepare in 1 Chr 15:26 (though it mentions no altar there, and it is a ‫ זבח‬and not an ‫עלה‬ as in Numbers 22–24); David’s offerings for the building of the temple in 1 Chr 29:21 (thousands of animals plus libations); and Hezekiah’s offering at the restitution of the temple in 2 Chr 29:21 (also includes lambs and male goats but is initially referred to as a ‫)חטאת‬. In the framing narrative of Job the friends are commanded to bring seven rams and bulls to Job as a burnt offering so that Job might pray for them, that they should not be punished for their folly. This sounds like Balaam and Job are not all that different from one another in the edited version of Numbers 22–24.83 Finally, the book of Ezekiel presents the same number of bulls and rams as burnt offerings in 43:23–25; 45:23–24; and 46:6–7 (here with six lambs and grain offerings). What becomes readily apparent is that there is no clear parallel to Balaam’s recommended offering. The most similar case can be found in Job 42:8 in which the friends should submit an offering and be prayed for so that God might be appeased. Like Balaam, Job is not called a priest, but he does submit offerings in the framing narrative; cf. Job 1:5. Should appeasing God also be regarded as the sense of Numbers 23–24? Should Balak’s offerings make God 81

Cf. Daiches, “Balaam”. Milgrom, Numbers, 194. 83 N.b. the plus in the Epilogue of Job in G identifies Job as the son of Balak ben Beor, king of Edom. Perhaps the responsible scribes recognized this commonality. Cf. Kepper and Witte, “Job,” 2048–50 and the literature cited there. 82

330

Balaam as a Character. His Origin and Background

more receptive to Balaam’s inquiry?84 Job’s prayer results in a positive outcome for the friends in Job 42:9. Such an understanding could be the background for Balaam’s suggestions in the redacted text, but one could hardly expect the same of Balaam’s encounters with God in Numbers. While Balak’s singular attempts to influence YHWH’s decision in the oldest tale in Numbers 22–24 remained inefficacious, even Balaam’s sevenfold increase in the redacted version of the offerings could not change the outcome.85 This increase in the number of sacrifices and the ascription to his character both suggest volition on his behalf. In the later version of the text, which places the responsibility for Balak’s sacrifices on the character Balaam, the reader is encouraged to see Balaam as both willing and attempting to change the outcome of YHWH’s blessing. These aspects of Balaam – his willingness and his attempting to try to change YHWH’s mind – were both missing in the oldest narrative level. They were secondarily added to negatively impact his image and make the story more consistent with the data found in Deuteronomy 23, namely that Balaam sought to curse Israel, but that this curse was changed into a blessing. The addition of Balaam’s suggesting that Balak make sacrifices added this ingredient to the mixture of Numbers 22–24 for the first time and perhaps even invites his identification as a Mesopotamian bārū, again reinforcing the understanding of Deuteronomy 23.

6.3.6 ‫( זעם‬Num 23:7–8) This term usually means “to feel anger” or “to incite anger” in the Bible.86 The cognates outside of the Bible – Arabic, Ancient Southern Arabic, and Syriac – generally, though not exclusively, favor the meaning “anger, to speak in anger”.87 Similarly, in Akkadian zenû bears the meanings “to be angry” or “to anger”,88 though as far as I can tell, the biblical and Hebrew lexica lack this observation. There are limited attestations with the meaning “to curse”: Num 84

E.g., Dennis T. Olson, Numbers, IBC (Louisville, KY: John Knox Press, 1996), 145 suggested that the offerings attempt to sway God’s demeanor, causing God to permit the curse. Cf. also Achenbach, Vollendung, 407–9. 85 “Die 7-Zahl [der Opfer] soll die Vollständigkeit des Rituals gewährleisten. Schließlich geht es um die Abwehr eines der größtmöglichen denkbaren Unheile. Da die Erzählung aber im Sinne des israelitischen Monotheismus stilisiert ist, wird hier nicht die Anrufung einer 7Zahl von Göttern symbolisiert, sondern die Übertragung auf den einen Gott Israels.” (Achenbach, Vollendung, 408) “In der Bileam-Erzählung kommt zu dem Aspekt der Befragung der einer Abwehr des Unheils von einem drohenden, zahlenmäßig mächtigen Feind. Hierdurch ist die 7-zahl motiviert.” (Achenbach, Vollendung, 409) 86 Cf. Brichto, Problem, 202–3; Rouillard, Balaam, 216–20; and Rüdiger Schmitt, Magie, 124. 87 Cf. Gesenius, et al., Gesenius, 307–8. 88 Cf. Oriental Institute of the University of Chicago, The Assyrian Dictionary Volume 21: Z, CAD (Chicago: Oriental Institute, 1960), 84–86.

6.3 Balaam’s “Profession”

331

23:7–8; Isa 30:27; Mic 6:10; Prov 22:14; 24:24; and SirC 3:16. The usage in Dan 11:30 could be of significance in that Numbers 22–24 seems to have been reworked toward this Daniel text at some point (see the comments on Num 24:24 in Chapter Two). Again, this root presents more or less a synonym with ‫ ארר‬and ‫ קבב‬in Numbers; its antonym is ‫ברך‬.89 The term ‫ זעם‬can only be found three times in Num 23, all of them in vv. 7– 8. Determining its semantic range presents a challenge insofar as the terms with which it is affiliated cannot be entirely explained. Nor does context really aid in determining what it might mean. The only text that might be able to help is Isa 30:27; the subsequent context suggests that the curse in this verse has something to do with Assyria’s destruction, which many will celebrate (cf. Isa 30:31–33). In Num 23:7 ‫ זעם‬parallels ‫ארר‬, though this probably presents a later contamination from an original ‫קבב‬. In Num 23:8 it parallels ‫קבב‬. Thus, its meaning must certainly overlap with these ‫ קבב‬in the oldest version and ‫ ארר‬in later editions in terms of semantic range concerning the aspect of “cursing”. However, one might consider some aspect beyond the devaluation known from ‫ קבב‬and the subservience determined by ‫ זעם ;ארר‬seems to have a somewhat more provocative sense, i.e., perhaps cursing more akin to the English sense of using foul language to express anger or elicit an angry response.90 Balaam refuses to provoke Israel to anger or provoke an angry response from them by befouling them with his language. Curiously enough, the only occurrences of this root occur in Balaam’s first oracle and already in the oldest core; this suggests that he came up with this idea of provocation and insult on his own according to this story. That is, Balak did not summon him to curse Israel with the term ‫זעם‬. The ‫זעם‬, like ‫ קבב‬is at home in the oldest layer of the Balaam story in Numbers 22–24. No occurrences can be identified in the redactional material of these chapters.

6.3.7 ‫( נחשׁ‬Num 23:23 and 24:1) The ‫ נחשׁ‬occasionally refers to some kind of mantic practice in the Hebrew Bible. The root is, however, attested in other Semitic languages, especially Aramaic and Arabic. Epigraphic usage in the Semitic languages generally reflects the meaning of “bronze” or similar,91 but the mantic meaning has been identified as well.92 It could also present some connection to “sorcery”, i.e., trying to

89 Cf. the discussion about the ambiguity of the usages in Hebrew in Bertil Wiklander, “‫זעם‬,” in Theologisches Wörterbuch zum Alten Testament, vol. 2 ‫חמץ–גלולים‬, ed. Gerhard Johannes Botterweck and Helmer Ringgren (Stuttgart: Kohlhammer, 1977), 621–26. 90 Cf. also the German schimpfen / beschimpfen as opposed to fluchen / verfluchen. 91 Cf. Hoftijzer and Jongeling, DNWSI, 726. 92 cf., i.e., Hatra 67:1–2 = KAI 254.

332

Balaam as a Character. His Origin and Background

influence circumstances by magical / religious means beyond merely trying to interpret signs about coming eventualities. The biblical cases could mean “augury” or “omen”, also “reading omens”.93 The narrative portions of the Bible (outside of Deuteronomistic evaluations) present no explicitly negative connotations. The biblical legislation and the Deuteronomistic redaction of Kings do denounce this practice (Lev 19:26 and Deut 18:10) or list it as a reason for Israel’s downfall (2 Kgs 17:17) and that of Judah (2 Kgs 21:6 = 2 Chr 33:6). The Bible apparently regards ‫ נחשׁ‬as a foreign practice (Gen 30:27 and 1 Kgs 20:33), but does not evaluate it negatively when undertaken by foreigners. Whereas 2 King condemns Manasseh precisely for this reason (among others), Gen 44:5 and 15 describe Israel’s son Joseph as practicing ‫נחשׁ‬, but fail to negatively evaluate him for this reason. Nevertheless, both the paucity of appearances and the general reticence of biblical authors about what it specifically consisted make understanding this root particularly difficult. Genesis 30:27 (Laban); 44:5, 15 (Joseph); Lev 19:26; Num 23:23; 24:1; Deut 18:10; 1 Kgs 20:33 (Ben-Hadad’s soldiers); 2 Kgs 17:17; 21:6; and 2 Chr 33:6 present the only biblical attestations of this root with some mantic meaning; otherwise it refers to snakes or bronze, though it is unclear how these lexemes may be related.94 Additionally, several verses in the book of Samuel mention this as an Ammonite name.95 Of these biblical texts that reference mantic undertakings, only Gen 30:27 secures a context in which it means something like augury. Laban has augured that Jacob brought him fortune. None of the other uses necessitate this meaning, suggesting that it could mean something else. The case in Num 23:23 appears particularly curious: to what end should one augur a negative outcome for Jacob? Rather than appearing as some kind of inquiry, ‫ נחשׁ‬appears to express influencing the object in this case. That could imply that it has some other meaning than interpreting fortune, particularly here. More likely, one should regard it as determining fortune or hexing.96 For that matter, most of the other biblical uses could fit this meaning. This interpretation would work for Gen 44:4, 15; Lev 19:26; Num 23:23; 24:1; Deut 18:10 (?); 1 Kgs 20:33 (?); 2 Kgs 17:17; 21:6; 2 Chr 33:6. The only biblical text that would not match this interpretation would be Gen 30:27. That could open up a new interpretive possibility for this lexeme.

93

Cf. Milgrom, Numbers, 200. Cf. the etymological discussion in Davies, Magic, Divination, and Demonology, 50–53 and 81–83 and Heinz-Josef Fabry, “‫נָ ָחשׁ‬,” in Theologisches Wörterbuch zum Alten Testament, vol. 5 ‫עזב–מרר‬, ed. Gerhard Johannes Botterweck, Helmer Ringgren, and Heinz-Josef Fabry (Stuttgart: Kohlhammer, 1986), 385–86. 95 Cf. 1 Sam 11:1–2; 12:12; 2 Sam 10:2; and 17:25, and 27. 96 Cf. Harry Torczyner, “Zu ‫ חנשתי‬Gen. 30,27,” OLZ 20, no. 1 (1917): 10–12; Rüdiger Schmitt, Magie, 110–12; and Rüdiger Schmitt, Mantik, 47. 94

6.3 Balaam’s “Profession”

333

While describing Balaam’s undertakings in Numbers 22–24, Moore described him in precisely this twofold capacity, albeit not with this particular root in mind: “Rarely does a military leader in the ancient Near East view an army apart from the supernatural forces which empower it. Balak’s instinctive response, therefore, is to ‘hire’ a wellknown magico-religious specialist (1) to ascertain the will of the | gods regarding the future of his country at this moment of military crisis, and (2) to exorcise (‘drive out’), via the appropriate apotropaic rituals, the foreign people with its foreign god(s)/daimon(s) threatening to violate the vulnerable door to his ‘house’.’”97

That is, Balak summoning Balaam probably included both the element of determining the divine will and trying to influence it.98 This dual understanding of the ‫ נחשׁ‬promotes precisely this image, particularly in the edited versions of Numbers 22–24. However, the biblical taboos presumably precluded further information being passed down about precisely what this practice entailed. Ultimately, the evidence of the Hebrew Bible provides no clue as to what this term meant. One can perhaps find some aid in identifying this term, or at least how it was understood in later antiquity, by considering the evidence of G. In Genesis 44 and Numbers 23, this noun is translated in each case by the Greek ὀιωνισμός, the mantic practice of identifying omens based on the flight patterns or calls of birds. 99 In Num 24:1, the translator maintained this connection, choosing ὀιωνός as the equivalent of the Hebrew √‫ ;נחשׁ‬here it could mean either that Balaam was looking for the birds themselves or the omen they portended.100 The Vulgate followed this understanding perfectly in Numbers 23–24, translating with augurium, which first means “avian omen” and then only in a broader sense “omen”. This understanding may have been lost in Hebrew and it is no longer noted in modern lexica.101 It could also present a later interpretation for a term that was no longer recognized. In the context of Numbers 22–24 it is virtually impossible to determine the precise meaning of ‫נחשׁ‬. In 23:23a it is parallel to ‫( קסם‬see below), and 24:1 states that Balaam stopped practicing this once he realized that blessing Israel curried favor with God. Nonetheless, no other text mentioned that Balaam had been engaging in such a practice up to that point. Rather he had been meeting God. The first use of ‫נחשׁ‬, in Numbers 23:23, comes from the first redactional layer of Numbers 22–24 and focuses on the inability to either identify such a 97

Michael S. Moore, Balaam Traditions, 97–98. Cf. Jastram, “The Text of 4QNumb,” 190, who recognized that ‫ נחשׁ‬must refer to harmful enchantments in the context of Numbers 22–24. 99 These verses, together with Sir 34:5, present the only usages of this noun in G. Indeed, Michael S. Moore, Balaam Traditions, 69–71 has gone so far as to identify Balaam as ornithomantic in the DAPT due to the mention of birds in ln. 7–9. 100 This represents a hapax legomenon in G. 101 Cf., e.g. Gesenius, et al., Gesenius, 806–7. 98

334

Balaam as a Character. His Origin and Background

practice in Israel or – more likely – on the inefficacy of such an undertaking against Israel. Both understandings may be possible and desirable.102 The appearance of ‫ נחשׁים‬of Num 24:1 presents an even later editorial insertion to reflect negatively on Balaam. If my literary-critical reconstruction is accurate, in this case, the root likewise probably described an activity that the elders of Moab and the Midianite elders undertook against the Israelites using the ‫קסמים‬ (Num 22:7a + 24:1*) in order to affect a negative outcome for the Israelites, not merely to augur information about the outcome. They sought to impact the outcome. In the oldest story about Balaam, then, the root did not appear. In the first redaction it did reflect any undertaking of his, but merely commented on a practice either supposedly unknown or inefficacious or both in Israel (Num 23:23). In the use of this root, a later interpolator successfully adapted a term referring to other agents from the earlier version of the text to fulfill his or her own needs.103 Suddenly Balaam’s shadowy character was illuminated with the light of a condemned practice, suggesting to the audience that he undertook such reprehensible steps. At the same time this refiguring cast him as an ironic figure who tried to evoke negativity toward Israel (Num 24:1) in spite of his previously identifying it as inefficacious (Num 23:23).

6.3.8 ‫( ותהי עליו רוח אלהים‬Num 24:2) The opening of the fourth oracle in Numbers 22–24 presents Balaam in a new manner. No longer does he go out to receive some word of God, rather the spirit of God seems to overwhelm him. There is no reason to regard this as God overwhelming Balaam, who had intended to curse Israel.104 “The Spirit of God coming upon Balaam marks the high point of his portrait in Numbers 22– 24. It is rare for an Israelite author to claim that the Spirit of God (or of Yahweh) comes 102 Cf. Rösel, “Wie einer,” 512–13: “In der Auslegungsgeschichte gibt es gute Gründe für beide Lösungen, doch mir will scheinen, daß man sich hier nicht zwischen Alternativen entscheiden muß, beide Verstehensmöglichkeiten sind ja richtig. Es gibt – in Übereinstimmung mit Dtn 18 – keine solchen verbotenen Dinge, daher muß ein fremder Seher geholt werden. Doch auch wenn dieser einen Fluch aussprechen sollte, was Bileam ja nicht will, selbst dann könnte er nicht gegen Israel ausrichten, denn da Gott auf Israels Seite ist, sind Zaubereien nutzlos.” 103 As an example of the changing meaning of terminology based on edited context: At the beginning of the Twentieth Century, Holzinger, Numeri, 107 noted “…der Gebrauch von ‫יַ ְהוֶ ה‬ und ‫ֹלהים‬ ִ ‫ ֱא‬ist, wie Sam. und LXX nahelegen, in der Bileamperikope zum Teil gründlich in Verwirrung geraten und leistet keinen Führerdienst.” As an appropriate example for how meanings can change over time, I would anecdotally note that the phrase “leistet keinen Führerdienst” took on quite a different tone when quoted in 1938 in Rudolph, Der “Elohist”, 103. 104 Cf. Guyot, “Balaam,” 240, who insisted without evidence immanent to Numbers 22–24 that Balaam had been trying to curse Israel the whole time, until the spirit of God overwhelmed him here and force him to bless Israel. His reading conforms Numbers 22–24 to Joshua 24 and Deuteronomy 23.

6.3 Balaam’s “Profession”

335

upon a non-Israelite, enabling him to prophesy like one of Israel’s prophets. Within the dramatic movement of the narrative, the fact that the last pro-Israelite oracle comes to Balaam under the influence of the Spirit of God constitutes the supreme authentication of these oracles as revelations from Yahweh. God has indeed blessed Israel!”105

Though not explicitly stated, this datum could identify Balaam as a legitimate prophet in biblical understanding, although more generally, this phrase more generally applies to special ability that has nothing to do with prophecy. Genesis 41:38 implies that Joseph had it. Exodus 31:3 and 35:31 mention that Bezazel was filled with this spirit. The spirit of YHWH clothes itself with Gideon in Judg 6:34. Often YHWH’s spirit interacts with the Judges (cf. Judg 3:10; 11:29; 13:25; 14:6, 19; and 15:14). At his anointing, the spirit of YHWH avails itself of David (1 Sam 16:13). David’s last words as reported in 2 Samuel 23 mention that the spirit of YHWH spoke through him in v. 2. Obadiah ascribes this spirit to Elijah, mentioning that YHWH’s spirit will protect him in 1 Kgs 18:12. One story in 2 Chr 15:1 mentions that the spirit of God enabled Azariah ben Oded to prophesy to King Asa of Judah, while another note in 2 Chr 24:20 ascribes a similar ability to Zechariah. Otherwise, the spirit of God only interacts with Saul and his subordinates (1 Sam 10:6, 10; 11:16; 16:23; 19:20, and 23), sometimes in a negative fashion (1 Sam 16:14–16; 19:9). These interactions with Saul and in Chronicles often have the characteristic of prophetic activity. To this end, one also recognizes the same phenomenon in 1 Kgs 22:24 = 2 Chr 18:23; 2 Chr 20:14; Ezek 11:5; and Mic 3:8.106 In this way, the coming of the spirit certainly marks Balaam as a special individual, though it remains ambiguous as to whether one could rightfully call him a “prophet” in the biblical understanding. At any rate, the biblical authors conspicuously failed to ever apply the ‫ נבא‬to him. Nonetheless, this interaction between Balaam and God’s spirit marks both the delivery of the third oracle and its deliverer as special in their contexts. God’s special interaction with Balaam in this manner goes back to the oldest narrative layer of Numbers 22–24 and distinguishes the version in Numbers from all others. Even the DAPT makes no similar statement about Balaam. It can hardly be doubted, then, that from the outset, the presumed author and implied audience of Balaam’s oracles – particularly this third one – regarded them as legitimate and their content as true, though they never identified him as a ‫נביא‬. Only later editors sought to emend this image of him and his words. Nonetheless, the presence of God’s spirit should not be regarded as a permanent state for Balaam from that point onwards.107 105

Barré, “Portrait,” 263. The other usages in the prophetic literature do not contain this aspect of prophetic undertaking; cf. Isa 11:2; 40:7, 13; 59:19; 63:14; Hos 13:15; and Mic 2:7. 107 Cf. Erasmus Gaß, Menschliches Handeln und Sprechen im Horizont Gottes. Aufsätze zur biblischen Theologie, FAT (Tübingen: Mohr Siebeck, 2015), 100. 106

336

Balaam as a Character. His Origin and Background

6.3.9 ‫( הגבר שׁתם העין‬Num 24:3 and 15) This phrase is unclear in M, which resulted and results in many interpretive difficulties. The root ‫ שׁתם‬remains somewhat ambiguous, largely because Num 24:3 and 15 present the only occurrences of this root in the Hebrew Bible. The general understanding of it has been “open” as found in Middle Hebrew and Jewish Aramaic.108 The Peshitta follows this understanding. Such an interpretation does not match all ancient translations however, as can be seen in the Vulgate’s reading obturatus (“closed”); cf. also M Lam 3:8 (‫שׂתם‬, an alternative form of ‫ )סתם‬for the meaning “closed”. The Septuagint arrived at a different conclusion, offering neither “open” nor “closed”; rather, it reads “true” (ἀληθινῶς). Apparently it parsed the word into the relative particle ‫ שׁ‬and the root ‫תם‬.109 Others have sought help from the Arabic, but the results have been neither convincing nor widely accepted.110 Even with these aids, however, the meaning of M remains obscure. How could this issue be resolved in M? First, 24:3 may have been pointed for consistency with 24:4, which notes that Balaam’s eyes are open (‫)וגלוי עינים‬. Second, perhaps M presents an alliterative pun with the following story in Numbers 25, which presents Israel at Shittim (‫)שׁטים‬. The Septuagint’s reading of the relative particle strengthens the case for reading ‫ שׁ‬instead of ‫שׂ‬. Bearing these data in mind, it remains essentially impossible to determine what this phrase about Balaam meant. Either his eyes were opened or they were closed. Or they were complete and saw the truth. Essentially no means of determining the meaning in the pre-Masoretic composition exists. “Since the meaning is doubtful, it is hardly possible to give a correct interpretation of the phrase.”111 And yet the only relevant matter for the pericope being discussed is that this term suggests that Balaam has some special power, particularly when contextualized with its parallel phrases in the third and fourth oracles: “one who hears the words of El” and “one who sees the vision of Shaddai”. Beyond his special insight in some way, this phrase adds almost nothing to the discussion of Balaam’s function or profession. Nor was this moniker probably present in the oldest version of the tale. None of the other texts about Balaam reference anything like this. The first editor (R1) added it in Num 24:15 and a later editor (R3+) copied it into Num 24:3.

108

Cf. Gesenius, et al., Gesenius, 1418. Cf. Gesenius, et al., Gesenius, 1441–42 for the relevant meanings and also, for a somewhat different position, Wevers, Notes on the Greek Text of Numbers, 402–3. 110 Cf. such an attempt in John Marco Allegro, “The Meaning of the Phrase šeṯūm hāʿayin in Num. XXIV 3, 15,” VT 3, no. 1 (January 1953): 78–79 and the response in Wagner, “Offenbarungsphänomenologische Elemente,” 96. 111 Guyot, “Prophecy of Balaam,” 332. 109

6.3 Balaam’s “Profession”

337

6.3.10 ‫( שׁמע אמרי־אל‬Num 24:4 M and 16) Balaam’s third and fourth oracles open with a number of phrases in which he describes his perceptions of the divine. The theological elements will be dealt with below, but the verbal elements will be briefly considered here. These opening statements focus on what Balaam senses and perceives. The first moniker identifies him as “the hearer of El’s words”. This fact should demonstrate his perception as one who receives and presumably understands (and even interprets) the words of El via audition.112 His auditory talent connects him to the god El. This matches to a degree the statement in ln. 1–2 of the DAPT, in which the gods appeared to Balaam in a vision as per El’s statement (‫אל‬.‫)כשמא‬. An editor (R1) inserted this phrase as part of the Bible’s fourth Balaam oracle and – through the course of textual transmission – later scribes erroneously or intentionally added it into the fourth oracle. In the oldest version, Balaam was a seer and not a listener. This transition from sight to hearing could evince hesitation in describing a physical manifestation of the deity, though that remains uncertain.

6.3.11 ‫( מחזה שׁדי יחזה‬Num 24:4 and 16) Balaam succeeds his first claim by noting that he “sees [or will see] Shaddai’s vision”. The noun ‫ מחזה‬is seldom attested in the Bible. In this pericope we find it only in Num 24:4 and 16; Gen 15:1; and Ezek 13:7 represent the other nonarchitectual usages (i.e., when it refers to “windows”; cf. 1 Kgs 7:4–5). Ezekiel uses it pejoratively, modifying it with the term “false” (‫)שוא‬. Coincidentally, the inscription from Tell Deir ʿAlla attests this term for a vision in ln. 1.113 Implicitly, the usage of this term in Numbers 24 suggests that Balaam must be considered a seer (‫ )חזה‬of visions, in this case from Shaddai. The aspect of sight plays a crucial role in understanding Balaam both in Numbers and in the DAPT. Terminologically, Balaam’s seeing (‫ )חזה‬must be distinguished from his normal seeing (‫)ראה‬.114 “Das Verb ‫ חזה‬wird … des Öfteren für visionäre Erfahrungen mit Gott oder für Visionen in einem breiteren Sinn gebraucht. Als ein ‫ חזה‬ist er sicherlich auch mit divinatorischen Praktiken vertraut.”115 This 112

Wagner, “Offenbarungsphänomenologische Elemente,” 89 considered the nocturnal visits as a form of audition as well, though Numbers 22 remains reticent on that point. The text commends interpreting the visitation as a more physical rather than merely auditory manifestation. 113 Following Hackett, Balaam Text, 25 and 33 (who in turn followed an unpublished presentation of Hamilton) and Blum, “Kombination I,” 577. 114 For observations about the use of this root in this pericope, cf. Hans F. Fuhs, Sehen und Schauen. Die Wurzel ḥzh im Alten Orient und im Alten Testemant. Ein Beitrag zum prophetischen Offenbarungsempfang, FB (Würzburg: Echter Verlag, 1978), 109–66. 115 Gaß, Menschliches Handeln, 105.

338

Balaam as a Character. His Origin and Background

terminological distinction also develops a differentiation in the content of what is seen. “Die gewöhnliche Sicht Bileams auf Israel kulminiert schließlich im vierten Orakel in einer tatsächlichen Vision. Während er zuvor nur ein visionäres Erlebnis hat, kommt es schließlich zu einer Vision, was mit dem Wechsel von ‫ ראה‬zu ‫ חזה‬auch sprachlich ausgedrückt wird”116 Milgrom recognized a distinct development in Balaam’s characterization specific to the idea of sight. “In the first oracle, he [Balaam] is endowed with normal physical sight. In the second, he attains the spiritual power to see Israel’s invulnerable state in the present. Now [in the fourth] his vision soars from the plane of the present to behold the distant future.”117 The story of the donkey turns this motif on its head, demonstrating that Balaam could not even see as well as his donkey. This famous seer was not as insightful as previously thought.118 The oldest layer of the Balaam oracles contained this attribute of Balaam’s, but later editors copied (R1) and fruitfully expounded it, even going so far as to make it ironic (R3+).

6.3.12 ‫( נפל וגלוי עינים‬Num 24:4 and 16) Balaam next identifies himself as one who falls, yet has uncovered eyes. This text’s sense remains somewhat opaque, which would explain why G proffers an alternative reading; the translator apparently understood this falling as indicative of sleep (ἐν ὕπνῳ); cf. Gen 2:21 (hiphil) and 15:12, which both explicitly mention “sleep”. The implication here may be that Balaam possesses a gift for receiving revelation via dreams, though this remains at most implicit. This interpretation largely stems from the reported nightly divine visits in Numbers 22, as well as the conception that the gods visited Balaam at night in the DAPT. It is difficult to ascertain what else might be implied by this statement, perhaps an ecstatic trance?119 Wagner suggested this could be a gesture of cultic or ceremonial behavior,120 but the evidence for this understanding is even slimmer. Thus, one could suggest that Balaam regarded himself as either a dreamer or an ecstatic, but there can be no certainty about which it was or even if both should be regarded as mutually exclusive.

116

Gaß, Menschliches Handeln, 106. Milgrom, Numbers, 207. 118 Wagner, “Offenbarungsphänomenologische Elemente,” 90 recognized the distinct characterization of the “theophany” of the donkey pericope when contrasted with similar circumstances in Numbers 22–24: “Theophanie ist Sehvorgang des Menschen, dessen Auge von Jahwe dafür enthüllt worden ist. Jahwe ist gleichwohl auch ohne menschliches Sehen bzw. Gesehenwerden im Erfahrungsraum des Menschen anwesend, doch sind die Augen des Menschen gleichsam bedeckt. Erst ihre ‘Entdeckung’ durch Jahwe initiiert den Wahrnehmungsvorgang, der dann ein ganzheitlicher ist. Das Auge ist pars pro toto.” 119 Cf. 1 Sam 19:24. Such is also the interpretation of Rüdiger Schmitt, Mantik, 51. 120 cf. Wagner, “Offenbarungsphänomenologische Elemente,” 97–98. 117

6.3 Balaam’s “Profession”

339

Beyond the falling however, we note that Balaam’s eyes stay uncovered. It is probably of significance that the verb is passive, i.e., that his eyes were opened.121 This second mention of his eyes emphasizes his sight more than his other senses. For that matter, this presents the third mention of his sight within each of the third and fourth oracles. This emphasis on his sight demonstrates that his capability with this sense must have been most respected by the author. Otherwise, only one of Balaam’s senses is mentioned only once in the third oracle: his hearing.122 The other senses remain elusive, and the focus remains on his eyesight. It is precisely this heightened sense of his that is mocked in the story of the donkey (the occurrence in Num 24:4 stems from R3+, just as the donkey story does), probably for the very reason that it was so emphasized in the older version of the story (R1 in 24:16): Balaam, who himself notes on three occasions in his fourth oracle the exemplary quality of his eyes and vision, fails to see the true divine messenger – a messenger that even his donkey can see. The focus on the sense of sight was present in the oldest diachronic layer about Balaam in the Bible, but later editors increasingly emphasized it and used it to their own ends.

6.3.13 ‫( וידע דעת עליון‬Num 24:16) Finally in Numbers 22–24, Balaam identifies himself as “the one who knows Elyon’s knowledge”. Curiously, this phrase only appears in Num 24:16 and not in 24:5. Achenbach suggested that “…Bileam erscheint wie Daniel als ‫ידע דעת‬ (Dan 1,4).”123 This ascription remains problematic, however, in that Daniel is not described as knowing the knowledge of ‫ עליון‬and in that this phrase applies equally to Daniel’s colleagues Hananiah, Mishael, and Azariah in the context of Daniel 1. This description of Balaam distinguishes itself from his other descriptions of himself in that it does not really reflect on one of his senses (sight or hearing). Rather, he simply claims to know what Elyon knows, without specifying how he came to this knowledge, be it through some kind of special revelation or through some other manner. In this way, I would argue that this phrase distinguishes Balaam the most; his connection to Elyon remains so strong that he can know what Elyon knows without the need to explain how he came to this knowledge, almost as if the two were of one mind, like some kind of clairvoyance.124 No seeing, hearing, sleep, or ecstatic state is necessary in this case. Like the addition of Balaam’s hearing El’s words, his knowing the 121

Cf. Wagner, “Offenbarungsphänomenologische Elemente,” 97. Cf. Wagner, “Offenbarungsphänomenologische Elemente,” 96. 123 Achenbach, Vollendung, 422. 124 Cf. the discussion in Wagner, “Offenbarungsphänomenologische Elemente,” 97, who proffered, “Erfahrener in Gotteserfahrungen, Fachmann im Umgang mit Gott” as possible translations. 122

340

Balaam as a Character. His Origin and Background

mind of Elyon was added at a later stage (R1) and not initially included in the oldest version of the story or oracles. The secondary addition of these other senses indicate a more nuanced appreciation for Balaam’s interaction with the divine. Rather than only perceive with his eyes, he interacts with the divine through his hearing and has privileged access to diving knowledge.

6.3.14 ‫ קלל‬and ‫( קללה‬Deut 23:6; Josh 24:9; and Neh 13:2) Turning our attention to Joshua 24 and Deuteronomy 23 (with Nehemiah 13), we notice that the authors or editors used terminology distinct from Numbers regarding Balaam’s activities. In these cases, the reader encounters the verb ‫ קלל‬in connection with Balaam in Deut 23:5, which clearly connects with Neh 13:2. This verbal usage is unique within Deuteronomy and Joshua, though the noun is more common in Deuteronomy (see below). This root commonly translates as “to curse” but it does not possess this meaning at the most basic level. The Hebrew root ‫ קלל‬presents the basic meaning of “to be small or worthless” in Qal, but “to count as little” or “to evaluate as worthless” in Piʿel.125 This understanding carries with it an undertone of cursing in a broader sense, such that distinguishing it entirely from the meaning of ‫ ארר‬becomes impossible.126 One important difference (though not entirely relevant for the passage at hand, is recognizable theologically: while ‫ קלל‬can be used of God, one never finds ‫ ארר‬used in this way.127 Like the other Hebrew and cognate terms for cursing addressed above, the Bible fails to explicate precisely what this term entails. Based on the meaning “to be small or worthless”, the most obvious understanding of cursing would be “to deride or regard with contempt”. This understanding can be found within the biblical texts (cf. Judg 9:27 and 2 Samuel 16). However, the specific manifestation of such derision remains unclear. When directed toward God, ‫קלל‬ results in the death penalty; cf. Leviticus 24. Cursing in the name of YHWH results in children being devoured by she-bears in 2 Kgs 2:24. Nehemiah’s cursing of those who intermarry (Neh 13:25) manifests itself in his striking (‫ )נכה‬some of the men, making them bald (‫)מרט‬, and chasing one of them away (‫)ברח‬. So, while the root may imply death in some cases, this remains neither clearly nor universally attested. As with the synonyms listed above, no unambiguous meaning behind ‫ קלל‬can be identified – particularly on the basis of the Bible alone.

125

Cf. the discussion in Brichto, Problem, 118–99 and Rüdiger Schmitt, Magie, 124. Cf. the definitions and attestations in Gesenius, et al., Gesenius, 1170–71 and Josef Scharbert, “‫קלל‬,” in Theologisches Wörterbuch zum Alten Testament, vol. 7 ‫שׁכן–קוץ‬, ed. HeinzJosef Fabry and Helmer Ringgren (Stuttgart; Berlin; Köln: Kohlhammer, 1990), 40–49. 127 Cf., e.g., Ps 37:22 and Scharbert, “‫קלל‬,” 43. 126

6.3 Balaam’s “Profession”

341

Other Semitic languages also attest this lexeme with similar meanings. It is related to the Akkadian cognate qalālu (“to lose importance, become discredited” etc.),128 and it can be found in Aramaic as well.129 The root appears in the causative stem (ya-qí-ìl-li-ni) as a Canaanite gloss in one of the Amarna letters from Megiddo (EA 245:38).130 The geographical importance of this find should not be overlooked. In these cases the root again does not bear the apparent meaning “to curse”, but rather affirms the aspect of ‫ קלל‬meaning “to despise or disregard” or “to lose importance”. Nothing suggests a particular manifestation of having been disrespected in these cases, as Deuteronomy, Joshua, and Nehemiah seem to anticipate. The impression of this verb is strengthened through the related noun ‫קללה‬, which occurs again in Deuteronomy, Joshua, and Nehemiah and only bears the meaning “a curse” – both the spoken element and the resultant state.131 In some cases, the noun ‫ קללה‬even stands parallel to the verbal adjective (= passive participle) of ‫ ;ארר‬cf. Deut 28:15–16. It should however, be noted that the noun ‫ קללה‬finds more substantial usage in Deuteronomy, especially in chapter 11, though it is not as common in Deuteronomy as ‫ארר‬. In Deut 30:19 life (‫ )החים‬parallels the blessing (‫)הברכה‬, death (‫ )המות‬parallels the curse (‫)הקללה‬. This parallel indicates that for Deuteronomy, the curse could be understood as some kind of death sentence. Deuteronomy, Joshua, and Nehemiah indicate that Balaam wanted to evaluate Israel negatively, but they do not explicate what concrete implications that would have had or whether his attempts could have been efficacious – under circumstances not directly contravened by YHWH. Only by considering some other biblical uses, particularly in the Deuteronomistic literature, it becomes more likely that Balaam’s curse should have evoked Israel’s death. This dramatic understanding advanced beyond the negative evaluation or subjugation that one can identify in the vocabulary of Numbers 22–24. Deuteronomy and its related texts increased the effect of Balaam’s curse, going beyond that which Numbers intended, transforming Moab’s interaction with Israel into a life and death struggle into which YHWH intervened in love for the sake of YHWH’s people. Thus, the term ‫ קלל‬adds another layer of understanding to the conflict between Moab and Israel, Balaam and God, in the retelling of these events in Joshua, Deuteronomy, and Nehemiah. Beyond merely describing Balaam’s activity regarding Israel with distinct vocabulary, Josh 24:9 and Deut 23:5 (and, thus, Neh 13:2) present a different 128

Cf. Oriental Institute of the University of Chicago, CAD 13, 55–58. In Aramaic Aḥiqar 141; cf. Herbert Niehr, Aramäischer Aḥiqar, JSHRZ (Gütersloh: Gütersloher Verlagshaus, 2007) for a translation and the relevant editions and secondary literature. 130 Cf. the grammatical discussion in Anson F. Rainey, Canaanite in the Amarna Tablets: A Linguistic Analysis of the Mixed Dialect Used by Scribes from Canaan Volume 2: Morphosyntactic Analysis of the Verbal System, HdO (Leiden; New York; Cologne: Brill, 1996), 191–92. 131 Cf. Scharbert, “‫קלל‬,” 44–45. 129

342

Balaam as a Character. His Origin and Background

situation than that recounted in Numbers 22–24. In these three instances, Balaam actually seeks to curse Israel, which is simply not the case in Numbers 22–24. At no point in Numbers 22–24 does Balaam express interest in despising Balak’s enemies. Quite the contrary, he insists throughout that he will only perform that which YHWH instructs him. On the other hand, Deuteronomy, Joshua, and Nehemiah suggest that YHWH and Balaam are adversaries, that Balaam is Israel’s adversary, and that YHWH must stop him from his action. Such an interpretation goes quite a bit further in regard to Balaam than anything found in Numbers 22–24. The chosen lexeme ‫ קלל‬adds a more ominous element to the recounting, implicitly including the impending threat of death.

6.3.15 Balaam’s Profession in the Inscription from Tell Deir ʿAlla Combination A of the Plaster Texts from Deir ʿAlla (DAPT) simultaneously provides less, yet more specific information about Balaam’s station. The reconstructed text only permits limited ascriptions to his character, but these become more precise than the disparate picture proffered by the Bible.132 Should the reconstruction of the first line of text be accurate, it describes Balaam as “a man, a seer of the gods” (‫אלהן‬.‫חזה‬.‫)אשׁ‬. Lines 1–2 affirm this description noting that he had a “vision according to El’s speech” (‫אל‬.‫כמשׂא‬.‫)מחזה‬. Both of these phrases identify him as a visionary.133 The terminology of ‫ חזה‬indicates that Balaam was an intuitive mantic, as opposed to a deductive mantic (‫)קסם‬.134 While the conclusion of ln. 2 is unclear, it presents one of two imperatives: either Balaam should prepare them (i.e., his people) or please them (i.e., the gods).135 Both of these understandings, the second more than the first, suggest that Balaam is not merely a passive part of the story, but indeed must undertake a mission, must fulfill a command. He should report what he has seen. The rest of the text describes his vision within another narrative frame. The people come to him, apparently also recognizing him as a figure with something to say, and inquire as to his current poor disposition (lines 3–4) before Balaam reports what he saw beginning in line 5.136 132 For the reconstruction of the first combination, cf. Blum, “Kombination I”. Like Blum, I consider the first combination to present the Balaam text, with combination II presenting an (at least apparently) unrelated wisdom text; cf. Blum, “Kombination II”. 133 While his vision apparently occurred at night, the DAPT never mentions that the gods appeared to him “in a dream”. The same must be said of Numbers 22, which mentions no dream. Contra Michael S. Moore, “Another Look at Balaam,” 369, who identified him in the DAPT as an oneiromantic, a dream-driviner. 134 Cf. Rüdiger Schmitt, Mantik, 42–43. 135 Cf., alternatively, Blum, “Kombination I,” 585–86. 136 The mention of birds in ln. 7–9 in the DAPT fails as a sufficient criterion to regard Balaam as an ornithomantic in that context; contra Michael S. Moore, “Another Look at Balaam,” 370.

6.3 Balaam’s “Profession”

343

With Balaam’s description as a seer, we can gather somewhat more clarity about him than the biblical texts offer. For one thing, this office, ‫חזה‬, is known from another Aramaic inscription, namely that of Zakkur, king of Hamath.137 We can glean some evidence from this inscription about what a “seer” was. In the context of the inscription, Zakkur seeks aid against his enemies, and to this end he attempts to contact the Lord of Heaven (‫ בעלשׁמין‬by raising his hand to him (ln. A11). The Lord of Heaven apparently answered the king (ln. A12) by the hand of “seers” (‫ )חזין‬and “diviners” (‫)עדדן‬,138 who apparently assured King Zakkur of the Lord of Heaven’s aid (ln. A13–15). The Lord of Heaven’s support apparently sufficed for Zakkur to overcome his enemies and expand his kingdom, but the context of the B side of the inscription is so fragmentary that the details remain mysterious. A few clues in the Zakkur Inscription can illuminate precisely what a seer did. In the case of the Zakkur Inscription we note, first of all, that there was more than one seer active. There is a plurality of unnamed people with this function. At the same time, the seers were apparently affiliated with another group, the diviners – though precisely what this group did remains elusive.139 The seers and diviners together were considered legitimate messengers of the Lord of Heaven, who answered by their hand (ln. A12). These people apparently were of some repute and took on an advisory capacity for the king. That is, they had access to royalty and were probably not just anyone off the street. At the same time, the context does not explicate any connection to any official cult or any official office they might have held. They could have operated outside of official institutions, but no definitive answer to this question can be offered. More importantly, they demonstrated divine support for the king and gave him an oracle affirming his position against his enemies. Read in this light, perhaps we can consider more about Balaam in the DAPT. That text identifies him as a seer, but does not comment specifically whether he might have been part of some group of seers. Perhaps the reference to a “brother” in ln. 2 could be indicative to his belonging to some such group, but this remains speculation. Balaam in the DAPT does not respond to a royal inquiry, but delivers instead a divine message to his people (‫ )עמה‬in general. The message does not reflect divine salvation from the people’s enemies, but rather a vision of the world turned on its head, a vision of chaos. In all of these ways, Balaam seems to depart from the vision of a seer presented in the Zakkur Inscription. The ambiguity about whether he served in some official function presents the single similarity – other than the title ‫ – חזה‬between the DAPT and the Zakkur Inscription. In this way, Balaam appears somewhat distinct from 137 Cf. KAI 202 and the translation, cited literature, and brief discussion in Robker, Jehu, 275–78. 138 To ‫עדדן‬, cf. Rüdiger Schmitt, Mantik, 46 and the literature cited there. 139 Cf. Hoftijzer and Jongeling, DNWSI, 827–28.

344

Balaam as a Character. His Origin and Background

the seers known from Hamath. At least, he appears distinct based on the limited evidence available. Rather than disappear in to a group, the text identifies Balaam by his name and by his patronymic. He reports not to the king, but to the people in general. The divine message he presents is one of desperation and not one of support. Thus, Balaam appears distinct from others in the role of seer as known from Hamath.

6.3.16 Conclusions about Balaam’s “Profession” The biblical texts ascribe a number of attributes to Balaam that one would expect to elucidate his character. These attributes changed to some degree over time, primarily in tenor. Some texts are less forthcoming about any abilities or functions that Balaam might have had. Numbers 31 and Micah 6 present Balaam without any particular power other than advising or answering inquests. These texts bring nothing potent to the character Balaam beyond an advisory capacity, perhaps one that could even be regarded as common. Other texts proffer the most common recurring theme: Balaam possesses some preternatural power. Joshua 13:22 notes that Balaam was an augur, and identified him as such with a negatively-charged term to this end; it is hardly accidental that the text identifies him as ‫ הקוסם‬within the immediate context of Israel’s killing him. This ascription of mantic ability presents a later reinterpretation of Balaam’s character. The oldest versions of his story, found in Numbers 22–24 and the DAPT do not ascribe him any such power. At the same time, Balaam, particularly in Numbers 22–24, Deuteronomy 23, Joshua 24, and Nehemiah 13, clearly must possess faculties beyond those of normal people. These texts anticipate that Balaam could have potentially “cursed” Israel, using a number of terms to describe his precise undertakings (‫קבב‬, ‫זעם‬, ‫ארר‬, ‫)קלל‬. Yet, in spite of this manifold nomenclature – perhaps even because of it – it remains obscure what exactly such cursing might entail. At most one can appreciate the idea that he would have spoken badly of Israel and that this linguistic endeavor would have been detrimental to their well-being. Some of these terms bear a slight inflection that can aid in interpreting precisely what the expected outcome of the curse would have been. The earliest version (S1) smacks of derision (‫ )קבב‬and provocation (‫)זעם‬. A later editor (R1) included a slight aftertaste of subjugation (‫)ארר‬. Such detriment presents Balak’s objective in commanding Balaam to come to him. After all, Balak invokes fear as his motivation in his missive to Balaam, and thus he would like to have them weakened by any means possible. While the text (or at least the character Balak) ascribes Balaam with this capability, Balaam never really seeks to harm Israel at all in Numbers 22–24. So, that story does not comment on his ability to actually act in the manner ascribed to him. In Numbers, he seeks to do Israel no harm.

6.3 Balaam’s “Profession”

345

And yet, Balaam only actively tries to denounce Israel in the text of Josh 24:9 and Deut 23:5–6 (with Neh 13:2). That is, later interpreters of this redacted version of Numbers 22–24 relied on a distinct root to describe his potential undertaking (‫)קלל‬. The version of events cited in Deuteronomy, Joshua, and Nehemiah implies (or at least suggests) that death might have resulted from Balaam’s action (‫)קלל‬. But Balaam fails in his endeavor in Deuteronomy, Joshua, and Nehemiah because of YHWH’s intervention. So, if his station – or job, if you will – was foremost to curse, he remained unwilling in Numbers and unsuccessful in the other versions; his attempt did not exist in Numbers 22–24 and was inefficacious in Deuteronomy, Joshua, and Nehemiah. Numbers 22–24 ascribes Balaam other abilities as well. While the characters in the text presume that Balaam has the ability to curse, his character describes his own other preternatural abilities. He senses things that others presumably do not – the words of El, the visions of Shaddai, the knowledge of Elyon. The visions of the Shaddai present his only divine attribute in the oracles in the oldest version of Numbers 22–24. The other deities (or divine monikers) and Balaam’s appreciation or discernment of them were added in a later phase. Additionally, the later text (from the first redaction adding Num 24:15–16) implies that he either has the ability to dream or fall into an ecstatic state (or both) “with eyes uncovered” or with “a truthful eye”. These ascriptions also suggest he has a preternatural ability in some such capacity as well. The narrator mentions (nocturnal) communion with the divine. Without it being specifically enumerated, the narrative portions of Numbers 22–23 describe Balaam’s encountering the divine without providing concrete details about how this occurred. The context suggests that sometimes it happened at night (Numbers 22), but could also happen during the day (Numbers 23 fails to indicate when the encounters reported there occurred, but the first one may have been in the morning; cf. ‫ בבקר‬in 22:41). Balaam’s third oracle does not come from an encounter with the divine, but Num 24:2 states simply that “the spirit of God was upon him” (‫)ותהי עליו רוח אלהים‬. These descriptions belonged even to the oldest narrative of Numbers 22–24. In spite of these descriptors, little can be said about what Balaam’s social station should have been. The text of Josh 13:22 makes it explicit: it pejoratively names Balaam an “augur”. On the other hand, the texts of the Vulgate and Peshitta demonstrate that Balaam was an interpreter of dreams – someone along the same lines as Joseph in Genesis 44. Other than these descriptions, the Bible offers little indication about Balaam’s social background. The texts generally present him as commanding supernatural ability in some form, while simultaneously failing to explicate precisely what that ability (or those abilities) would have entailed, what he could have done. Simultaneously, a number of traditions permit him access to the king of a foreign people, suggesting that he must have been a person of some import (Numbers 22–24; Deuteronomy 23; Joshua 24, Micah 6).

346

Balaam as a Character. His Origin and Background

Ultimately modern exegetes encounter a disparate and somewhat unclear image of Balaam. While some texts suggest that he would at least be theoretically capable of cursing Israel, he only ever attempts it in Joshua, Deuteronomy, and Nehemiah. He never succeeded at it. Particularly the text of Numbers 22–24 focuses on more positive aspects of his ability and indicates that he must have possessed abnormal, superhuman insight and senses that placed him in a special relationship to God/the gods. This developed more in a later version of the story, such that he saw their visions, heard their words, knew their thoughts, and could interpret dreams or fall into ecstasy. In the final edition of Numbers 22–24 he took on a somewhat priestly advisory capacity, suggesting to Balak that he build altars and sacrifice thereupon. Joshua 13 describes him with a term that generally must be regarded as pejorative in biblical literature. In the combination of these abilities, Balaam certainly appears unique in biblical literature.140 But Balaam remains unique not only regarding biblical literature. The DAPT introduces him as a seer and demonstrates his connection to the divine in the first lines of text. Yet, Balaam distinguishes himself from other known seers in the ancient Aramean world in that he is known by name and not merely an unnamed member of a faceless guild. He does not act in cooperation with ‫עדדן‬, whatever they may have precisely been. He fails to deliver a message of salvation for the king or some people in the DAPT. The gods interact with him and command him to present their message, which he does. That remains the common element between the texts: the seer receives a divine message (or divine messages with the biblical Balaam) and reports – apparently faithfully – that which should be recounted. This still places him in a ubiquitously recognized religious role, the ecstatic,141 even if in a somewhat distinct form from that found elsewhere.

140

Due to the plurality of roles that he fulfills, particularly regarding his deductive and inductive mantic methods, e.g., Johannes Lindblom, Prophecy in Ancient Israel (Oxford: Basil Blackwell, 1962), 90–95 identified him as a kahin. Such an identification seems premature. 141 Cf. Johannes Lindblom, “Zur Frage des kanaanäischen Ursprungs des altisraelitischen Prophetismus,” in Von Ugarit nach Qumran: FS Otto Eissfeldt, ed. Johannes Hempel and Leonhard Rost (Berlin: Töpelmann, 1958), 98: “Die Religionsgeschichte lehrt uns also, daß Ekstatikertum und ekstatisches Prophetentum nicht auf bestimmte Gebiete begrenzt und nicht an besonderen Völker oder Rassen gebunden sind; sie finden sich überall, wo Menschen leben, in mehr oder weniger ausgeprägten Formen. Die Ekstase ist eine allgemein menschliche Erscheinung. Das ist gewiß nicht so zu verstehen, daß alle Menschen ekstatisch veranlagt sind, aber so, daß Menschen mit Begabung für ekstatische Zustände überall auftreten können. Überall in der Welt hat es auch Männer und Frauen gegeben, die die Eigentümlichkeit gehabt haben, in einem ekstatischen Zustande Offenbarungen aus der unsichtbaren Welt zu erhalten und sich unter göttlichem Zwange genötigt gefühlt haben, diese Offenbarungen anderen Menschen zu verkünden. Dies ist es ja, was einen Menschen zum Propheten macht.”

6.4 Balaam’s “Religious Affiliations”

347

6.4 Balaam’s “Religious Affiliations” 6.4 Balaam’s “Religious Affiliations”

Balaam maintains contact with or has special knowledge of several deities in Numbers 22–24 and the DAPT. These various deities and divine names will be examined here to consider in what religious or theological framework we can include Balaam at what stages in the diachronic development of the text. The text of Numbers will be given priority, with a consideration of the deities in the DAPT noted at the appropriate points.

6.4.1 ‫יהוה‬ The first deity named in the Balaam narrative of Numbers 22–24 is ‫יהוה‬, the God of Israel, in Num 22:8. Balaam promises to bring YHWH’s word back to Balak’s messengers and report to them exactly what YHWH said. YHWH plays a most profound role in the story, receiving ever increasing recognition from non-Israelites. Balaam pretty readily identifies himself as an adherent to YHWH, referring to “YHWH, my God” in Num 22:18 M and 24:13 in an older version.142 The usage of the phrase “my God” in Num 22:18 and 24:13 demonstrates – implicitly – that Balaam is contrasting his deity with some other god. In this case, he distinguishes YHWH from Balak’s god(s).143 By its denouement even Balak recognizes that YHWH is at work in the story: in 24:11 Balak tells Balaam, “YHWH has refused to honor you”. Both of these data affirm two important factors about this narrative: its tradents were apparently believers in YHWH and – more significantly for our purposes here – it identifies Balaam as obeisant to Israel’s God. Another character in the story, a foreigner at that, even recognized Balaam’s subjugation to this deity without calling it into question (Num 23:17). Implicitly Balak recognized through this interaction that he acknowledges YHWH’s power. Whether this obeisance remained exclusive (i.e., should Balaam be recognized as monotheistic) is another issue to be dealt with below. At any rate, the deity YHWH thoroughly saturates the Balaam narrative, even going back to the oldest layer. In the last stage of composition, however, this aspect changed in Numbers 22–24. In the inserted story of the donkey in Num 22:21aβ + 22–35a, the narrator generally refers to YHWH’s messenger (‫)מלאך יהוה‬. There are three exceptions to this rule, all of them readily explicable. In Num 22:22 the majority of Masoretic textual witnesses (there is one exception) refer to ‫אלהים‬, whereas 142

cf. the text-critical notice to this verse in Chapter Two. Eissfeldt considered the possibility that Balaam’s reference may be of a category in which the proclaimer identifies his deity with a deity other than YHWH. Cf. Otto Eissfeldt, “‘Mein Gott’ im Alten Testament,” ZAW 61 (1945–48): 4 (= Otto Eissfeldt, “‘Mein Gott’ im Alten Testament,” in Kleine Schriften. Dritter Band, ed. Rudolf Sellheim and Fritz Maass [Tübingen: Mohr Siebeck, 1966], 36. 143

348

Balaam as a Character. His Origin and Background

one M manuscript, Smr, and some G manuscripts (Ambrosianus, Venetus, 53) all refer to ‫יהוה‬.144 While it remains impossible to decide precisely which of these readings is older, I tend to favor the reading ‫יהוה‬. This reading was probably evidence of a compositional break, with ‫ אלהים‬giving Balaam permission to go with the emissaries (22:20) and ‫ יהוה‬becoming angry at his going. Further, ‫ יהוה‬serves as a supporting actor in the donkey story, which focuses on Balaam’s interaction with the ‫מלאך יהוה‬. Both other occurrences of ‫ יהוה‬without ‫ מלאך‬demonstrate this: ‫ יהוה‬opens the donkey’s mouth (22:28) and Balaam’s eyes (22:31). Otherwise only the ‫ מלאך יהוה‬interacts with Balaam (and his donkey). What these uses of the name ‫ יהוה‬without the qualifier ‫מלאך‬ demonstrate is a distinct theology: the deity – whether called ‫ יהוה‬or ‫– אלהים‬ no longer speaks directly with Balaam, but engages the character through a messenger. This should not suggest that ‫ יהוה‬is inactive: where the basis of the action is concerned (22:22) or where the miraculous occurs (vv. 28 and 31), the narrator duly notes that ‫ יהוה‬acted. In this way the narrator distinguishes divine aspects: ‫ יהוה‬remains the instigator whereas the ‫ מלאך יהוה‬serves as the interlocutor. Such a division of labor was unnecessary in the older version of the story, but was included in a later stage to distance Balaam from the divine. It is hardly coincidental that this distinction occurs uniquely in the donkey pericope, which reflects more negatively on Balaam than any other portion of Numbers 22–24. Micah 6:5 contains no explicit connection between YHWH and Balaam, but it does contain the name ‫יהוה‬. The opening of Micah 6 clearly references the text as a speech of ‫( יהוה‬Mic 6:1–2). The transition to the first person in v. 3 indicates that YHWH is speaking in the following verses. Part of this speech demands that the audience remember Balak and Balaam “in order to recognize YHWH’s righteousness” (‫)למען דעת צדקות יהוה‬. Therefore, the text of Micah demands the association of Balaam with YHWH’s righteousness and thus with YHWH, even though no explicit comment about Balaam’s religious background can be found in this notice. Micah 6 clearly subsumes Balaam’s activity to YHWH’s righteous dealings, even without explicitly joining the two or claiming that Balaam knew or believed in YHWH. Other texts doubt Balaam’s submission to YHWH. Numbers 31:16 rejects Balaam’s submission to YHWH and indicates that he had encouraged the Midianite women to lead the community of YHWH astray. The recounting clearly demonstrates that YHWH and Balaam are enemies.145 The same must be said of Deuteronomy. Deuteronomy 23:5–6 notes that YHWH reversed Balaam’s attempted curse. In this case, not only does the text mention YHWH, it identifies YHWH as “your God” (‫)אלהיך‬, concentrating on the personal relationship of the 144 It should also be noted that a variety of other Greek manuscripts and daughter translations present the reading κύριος ὁ θεός or similar, which could evince familiarity with both variants. 145 Numbers 31:8 does not affiliate Balaam with any deity.

6.4 Balaam’s “Religious Affiliations”

349

God YHWH with each individual in the community. Joshua 24:9 also presents this interpretation and has YHWH himself explicitly state – through his servant Joshua – that he rescued Israel from Balaam’s hand.146 These data demonstrate a theological development around the character Balaam and what – if any – relationship he might have had to YHWH. The oldest narrative in Numbers 22–24 indicates that Balaam worked for YHWH and that he regarded himself as a legitimate messenger of YHWH’s missives. Barré described the theological background thus: “If the seer’s positive oracles concerning Israel are true divine revelations, how could they not come from Yahweh? Conversely, if Balaam divines by a god other than Yahweh, how could his oracles about Israel be true?”147 While not claiming this connection between the deity and the messenger as explicitly, Mic 6:5 certainly indicates that Balaam’s activity should have been understood within the bounds of YHWH’s righteous undertakings. The other texts – Numbers 31, Deuteronomy, and Joshua – revise the theological image of Balaam away from being a follower of YHWH and transform him into an opponent of both Israel and its God. While Balaam appears to subject himself to YHWH in Numbers 22–24*, other, later biblical texts witness YHWH rejecting Balaam. The chronological division of these understandings seems readily reconstructable.

6.4.2 ‫ אלהים‬/ ‫אלהן‬ The other primary epithet for God in the Balaam narrative recounted in Numbers 22–24 is the generic Hebrew term for God or gods, ‫אלהים‬. It is “God” who interacts with Balaam in Num 22:9–12 and 20. Based on the variants in the textual transmission, one sees that the tradents of the text understood the epithets ‫ יהוה‬and ‫ אלהים‬as interchangeable.148 Balaam seems to react in the same manner, in that he anticipates ‫’יהוה‬s nocturnal visitation (Num 22:8), but does not seem surprised when it is ultimately ‫ אלהים‬who comes (v. 9). For the character Balaam, like the later tradents of the text, these two monikers reference the same being. The other biblical texts about Balaam use this term primarily – though not exclusively – together with the Tetragrammaton. So, for example, Deut 23:5 mentions ‫ יהוה אלהיך‬and Josh 24:2 opens Joshua’s speech with an ascription to ‫יהוה אלהי ישׂראל‬. Such uses demonstrate a nuanced aspect of YHWH, rather than a distinct reference to ‫אלהים‬. The one exception to this is Nehemiah 13, which 146

Joshua 13:22 does not mention any deity in connection with Balaam. Barré, “Portrait,” 262. 148 Contra Gaß, Menschliches Handeln, 81: “Anscheinend ist ‫ יהוה‬in der Bileamerzählung jener Gott, der die einzelnen Ereignisse bestimmt und lenkt. Im Gegensatz dazu ist ‫ אלהים‬lediglich ein unpräziser Ausdruck für eine Gottheit. Insofern verwundert es nicht, dass sowohl Bileam wie auch der Erzähler den Namen ‫ אלהים‬in Num 22,38; 23,4.27 verwenden können.” 147

350

Balaam as a Character. His Origin and Background

does not include the name ‫יהוה‬, but does make recourse to “our God” (‫)אלהינו‬. This usage focuses attention more on the communal connection to the deity (especially when contrasted with its almost identical parent text in Deut 23:5), while at the same time iterating that Balaam was not a believer in this God and thus was not part of the community. The audience in Nehemiah and the audience of Nehemiah should probably thus be understood as hearing this explicit contrast between them and Balaam. The term ‫ אלהן‬also occurs in the DAPT, here however it refers to a plurality of deities and not to a single god. Similarly to Num 22:8–9, these ‫ אלהן‬come to Balaam at night in ln. 1 and speak to him in ln. 2 of the DAPT. Lines 5–6, though somewhat incomplete, recount that Balaam seeks to make his people aware of what the gods, also identified as the ‫( שׁדין‬see below), were planning. The gods were assembled as the ‫ שׁדין‬when they began to address the goddess ‫ שׁמשׁ‬about her plans and asking her to avoid making darkness last forever. The relevance of Balaam’s relationship to the gods cannot be overlooked: he relates to them in a manner similar to that known from Numbers 22–24, particularly in the oldest version of the tale. Balaam brings the message from the divine to the target audience. However, unlike Numbers he has not been commanded by a royal figure to come and curse another people. Rather, he deals with his own people.

6.4.3 ‫אל‬ The noun ‫ אל‬appears in several Semitic languages, either as a common noun or a divine name. However, the meaning in the HB is not always clear or precise. In a survey of the attestations of ‫ אל‬in the HB, Eissfeldt recognized three types of biblical usages of El: 1) El as a designation for YHWH (the overwhelming majority of instances); 2) YHWH appropriates the name El, a name which was not originally ascribed to YHWH, taking it as a proper name to the exclusion of other candidates to this epithet; 3) El as a deity distinct from YHWH.149 To none of these categories does he ascribe the usages in Numbers 23–24, which he does not even mention in his article. Thus, Eissfeldt left the riddle unsolved here. Perhaps the appearance of ‫ אל‬in other Semitic languages could help. “In the Phoenician, Aramaic, Punic and Neo-Punic inscriptions the noun ʾl is generally used as appellative in the sense of ‘god, godhead’ or as adjective ‘divine’. This use of the term is

149 Cf. Otto Eissfeldt, “El and Yahweh,” JSS 1 (1956): 26–29 and, for a reprint of the German original, Otto Eissfeldt, “El und Jahwe,” in Kleine Schriften. Dritter Band, ed. Rudolf Sellheim and Fritz Maass (Tübingen: Mohr Siebeck, 1966), 387–89.

6.4 Balaam’s “Religious Affiliations”

351

also known from the Ugaritic texts of Ras Shamra and from the HB. Yet, El was also used as proper name, e.g. when El is mentioned alongside other gods.”150

While Herrmann opines that the apposition of El with other gods implies that El represents a proper name, he quickly revises this opinion for the cases in Numbers 23–24: the usages in Num 23:8, 19, 22–23; 24:4, 8, 16, and 23 are appellative, as they feature the name YHWH “supplemented by a genitive employed in apposition”.151 While this is possible, it remains insecure. Perhaps some combination of the two possibilities is more likely. While at the earliest stages of the composition and transmission of the Balaam story in Numbers 22–24 it is possible that El referenced another deity, by the conclusion of the text and in its current context, El presumably references YHWH obliquely, serves as a moniker for that God. Cross tends to support this notion, noting that generally, “ʾĒl is rarely if ever used in the Bible as the proper name of a nonIsraelite, Canaanite deity in the full consciousness of a distinction between ʾĒl and Yahweh, god of Israel.”152 At any rate, the appearances of ‫ אל‬in Numbers 22–24 cannot be explained with certainty. The other biblical texts about Balaam make no reference to El; it remains unique to Numbers 22–24, and there it remains exclusive to the oracles. Two occurrences of this divine epithet parallel the Tetragrammaton, such that one could consider them identical. Examples of this usage are 23:8 in which the parallel is most obvious, but also in 23:22, in which the usage parallels that of v. 21. The other appearances of this name do not associate him directly with YHWH, but syntactically, one could commend such an interpretation for 23:19, which presents El as the subject until 23:21b, which changes the subject to YHWH, before 23:22 changes the subject back to El. In these cases – some of them from the oldest stratum (23:8 and 19) – it is relatively clear that El and YHWH should be understood more or less as parallel and probably, therefore, identical.153 Two references to El, one of them in the oldest version (24:8), make no reference to YHWH, but indeed to traditional El imagery. Numbers 23:22 and 24:8 both affirm that El is leading Israel out of Egypt, with Num 23:22 being a redactional interpolation based on Num 24:8.154 More importantly, they both 150 Wolfgang Herrmann, “El ‫אל‬,” in Dictionary of Deities and Demons in the Bible. Second Extensively Revised Edition, ed. Karel van der Toorn, Bob Becking, and Pieter W. van der Horst (Leiden; Boston; Cologne: Brill, 1999), 276–77. 151 Herrmann, “El,” 278. 152 Frank Moore Cross, Canaanite Myth and Hebrew Epic. Essays in the History of the Religion of Israel (Cambridge, MA: Harvard University Press, 1973), 44. 153 The same could also be true for ‫ שׁדי‬and ‫ ;עליון‬cf., e.g., Manfred Weippert, “Erwägungen zur Etymologie des Gottesnamens ʾEl Schaddaj,” ZDMG 111 (1961): 61. 154 Cf. Knauf, Midian, 140: “Es gibt keinen Hinweis darauf, daß der Exodus jemals mit einem anderen Gott verbunden gewesen wäre als Jahwe (das Israel des 13. Jh. v.Chr. war das wohl): Num. 24,8 ‘El, der es (sc. Israel) aus Ägypten führte’ ist kein archaischer, sondern ein

352

Balaam as a Character. His Origin and Background

describe El with horns like a bull. Such bovine imagery is common for El, even from the earliest time in the ancient Near East. That YHWH’s name should stand in such proximity to El and an explicit bull metaphor suggests that by the time of the composition of this story, some overlap between these deities had already taken place. The notice that El is bringing Israel out of Egypt appears out of place, even unique in the Bible, until one regards the parallelism with the name YHWH in 23:21–22. This parallelism reuses this motif from 24:8 (there without mentioning YHWH), which could be understood with some ambiguity. Read in the context of the whole, however, ‫ אל‬in Num 24:8 presumably too should be understood as an eponym for YHWH. The only other occurrence of the name El that remains without some kind parallel to YHWH is 23:23, in which it is asked who will inquire of Jacob/Israel what El has done. Perhaps this instance indicates recourse to an old El tradition, but that is far from certain. More likely, one should consider El as an epithet for Israel’s God YHWH in this case as well, as it belongs to the same redactional layer responsible for adding Num 23:22 and the parallelism there.155 Each of these parallels appears to relate to Num 23:8, which belonged to the oldest diachronic layer of Numbers 22–24. The later redactional layers in Numbers 22–24 adapted the moniker El as well, again only in the oracles. The first redaction added El in its adaptation of Num 24:8 in Num 23:22, there as a parallel for YHWH, present in v. 21. It included a referent to El in a rhetorical question in Num 23:23. Further, the first editor added El in the context of Balaam’s listing of his own abilities at the opening of his final set of oracles (Num 24:16). Balaam “hears the words of El”. Perhaps this final case does in fact imply a separate deity, but this is hardly certain. It cannot be overlooked that the name El occurs alongside ‫ עליון‬and ‫שׁדי‬ (see below). The first redaction in Numbers 22–24 thus introduced an association of divine speech to El in Numbers 22–24. A later redaction (Num 24:23) attributes activity to El (“when El sets this”) and could even identify El with the god affiliated with the coming of the Kittim in 24:24. This remains syntactically difficult. Should that be the case however, perhaps the later redactor sought to place El in a negative light: El is the deity behind the coming destructions. This interpretation would cast El in a more ambiguous light. El appears in Combination A of the DAPT as well, but only plays a minor role there. At the opening of the story, the inscription reports that the gods came to visit Balaam in the night and he had a vision “like El’s speech” (ln. 2). archaisierender Text, für den wohl schon El und Jahwe als der Eine Gott zusammengefallen waren…” 155 This understanding matches the findings of Volkmar Fritz, “Jahwe und El in den vorpriesterschriftlichen Geschichtswerken,” in “Wer ist wie Du, HERR, unter den Göttern?” Studien zur Theologie und Religionsgeschichte Israels für Otto Kaiser zum 70. Geburtstag, ed. Ingo Kottsiepter, et al. (Göttingen: Vandenhoeck & Ruprecht, 1994), 111–26, though he regards this aspect of ‫ – אל‬or rather, ‫ אל‬as an aspect of YHWH – as typical for the source E.

6.4 Balaam’s “Religious Affiliations”

353

The texts fails to clarify further precisely what this means, but at least it makes clear that a god known as El was part of Balaam’s pantheon and was perhaps included under the rubric ‫אלהן‬. It becomes clear as well that El could communicate and that Balaam’s vision should be understood as part of such a communication. The inscription presents no other details about the god El. Perhaps the name El should thus be understood merely as the adjective “divine” as in some other cases in Aramaic.156

6.4.4 ‫שׁדי‬ Like El, Shaddai appears a few times in Numbers 22–24, always in the oracles, and in no other biblical text about Balaam. Together with Gen 49:25 (Jacob blesses Joseph), the usages in Num 24:4 and 16 are the only attestations in the Pentateuch (or Hexateuch or Enneateuch, for that matter) that do not combine it with ‫ אל‬or include it as a theophoric element in a personal name; cf. Gen 17:1; 28:3; 35:11; 43:14; 48:3; Exod 6:3; Num 2:12; and 10:19 for cases of this combination.157 Psalm 68:15 also attests ‫שׁדי‬, which could reinforce our consideration of a relationship between the Balaam text in Numbers 22–24 and Psalm 68 as described in Chapter Two. Other usages without El can be found in every occurrence in Job; Ps 91:1; Ezek 1:24; and Ruth 1:20.158 The disparate character of these attestations preclude combining them under a common heading. It remains somewhat unclear what this name means.159 Some have suggested a relationship to the mountains from the Akkadian šadu.160 Others have commented that it could reflect a connection to the field, from the West Semitic ‫שׂדה‬.161 These suggestions could more generally reflect an understanding like “God of the wilderness” (reflecting references to fields or mountains). This “wild” aspect could support an understanding of Shaddai as a storm or warrior god.162 Another alternative would ascribe an Egyptian background to the name, 156

Cf. Hoftijzer and Jongeling, DNWSI, 53–55 for examples. Outside of the Pentateuch ‫ אל שׁדי‬occurs only in Ezek 10:5. 158 The majority of occurrences of ‫ שׁדי‬can be found in Job, some 23 times of 39 total. These will not be addressed here, since they provide no relevant evidence for the case at hand. 159 For general information, further secondary literature, and an introduction, cf. Manfred Görg, “Šaddaj – Ehrenrettung einer Etymologie,” BN 16 (1981): 13–15; Ernst Axel Knauf, “El Šaddai,” BN 16 (1981): 20–26; and Ernst Axel Knauf, “Shadday ‫שׁדי‬,” in Dictionary of Deities and Demons in the Bible. Second Extensively Revised Edition, ed. Karel van der Toorn, Bob Becking, and Pieter W. van der Horst (Leiden; Boston; Köln: Brill, 1999), 749–53. 160 Cf., e.g., William F. Albright, “The Names Shaddai and Abram,” JBL 54 (1935): 180– 93. 161 Cf., e.g., Weippert, “Erwägungen zur Etymologie des Gottesnamens ʾEl Schaddaj”. 162 Cf., e.g., Cross, Canaanite Myth, 54–60 and Naftali Herz Tur-Sinai, “El Shaddai [Hebrew],” EI 3 (1954): 39–41, who conjectured a meaning related to “thundering”. David Biale, “The God with Breasts: El Shaddai in the Bible,” HR 21, no. 3 (February 1982): 245 followed 157

354

Balaam as a Character. His Origin and Background

either affirming a direct connection to the deity Shed or more generally reflecting this deity’s protective or soteriological aspect. Such an understanding would provide the general meaning “savior”.163 Some connection to “breasts” should not be discounted outright.164 An affiliation with “breasts” could commend understanding Shaddai with an aspect of fertility, as could be understood in many of its appearances in Genesis.165 In the first half of the twentieth century, some tried to identify this divine epithet with private religious piety, but more recent scholarship has tended to reject such a position. “In the biblical references, Shadday is a rather universal/cosmic god; not a single attestation refers to the level of ‘family religion’...).”166 As was the case for El, Shaddai could be found in the oldest version of Numbers 22–24, but was also inserted into another context by a later editor.167 However, in contradistinction to the editor’s use of El, the phrase containing Shaddai was essentially copied into a new context, from Num 24:4 to Num 24:16, in which it became part of a list of various deities and Balaam’s perception of them. In this way, the sense of the two occurrences of the word ‫ שׁדי‬can hardly be addressed as distinct from each other. “For Num 24:4.16, Shadday is just another epithet for Yahweh like El or Elyon…”168 While this appears generally true, the combination of the three epithets El, Shaddai, and Elyon into one list permits at least speculation about whether the composer responsible wanted to define them as a group distinct from YHWH. To this end it remains conspicuous that the name YHWH never occurs in the final set of oracles in Num 24:15–24. this understanding, but only for the later biblical texts that mention Shaddai. Others – cf. particularly Harriet Lutzky, “Shadday as a Goddess Epithet,” VT 48, no. 1 (January 1998): 15– 36 – have gone farther, arguing even that ‫ שׁדי‬should be understood as identifying a feminine deity, most likely Asherah. Taking this thesis to the logical extreme, Harriet Lutzky, “Ambivalence Toward Balaam,” VT 49, no. 3 (July 1999): 421–25 then argued that the biblical ambivalence toward Balaam stemmed from biblical polemic against Asherah. This thesis remains exceptionally speculative. 163 Cf. the discussion in Mathias Neumann, “(El) Šadday – A Plea for an Egyptian Derivation of the God and Its Name,” WO 46 (2016): 244–63. Archaeology could provide anecdotal support for an Egyptian understanding; cf. Franken, “Deir ʿAlla Re-Visited,” 14: “It is clear that the Valley of Succoth was the scenery of inter-regional trade, which in the Late Bronze Age and the beginning of the Iron Age was largely controlled by Egypt.” 164 Cf. Ezek 23:21; Song 1:13; and 8:1. Albright, “The Names Shaddai and Abram” suggested that breasts and mountains were connected. Cross followed this understanding as well; cf. Cross, Canaanite Myth, 52–54. For a more in depth discussion, cf. Biale, “God with Breasts”. 165 Cf. Biale, “God with Breasts,” 247–48. 166 Knauf, “Shadday,” 751. 167 This would support Biale’s hypothesis, that “…Shaddai was an ancient divine name which was more or less abandoned, but not entirely forgotten, after the literary activity of the early monarchy, only to become popular once again in late texts with definite archaizing tendencies.” Cf. Biale, “God with Breasts,” 244. 168 Knauf, “Shadday,” 751.

6.4 Balaam’s “Religious Affiliations”

355

This absence could indicate intention: these names are not used of YHWH in 24:16. The usage in 24:4, however, could be regarded as a parallel to YHWH, but this remains speculative. In the text of Numbers it seems that ‫ – שׁדי‬like ‫ – אל‬might have originally referenced some other deity (or even a group of deities), and that its current context remains ambiguous about whether it should be understood as an epithet for YHWH.169 In addition to being mentioned in the third and fourth oracles of Balaam, Combination A of the DAPT mentions the ‫ שׁדין‬twice in ln. 5–6. In the initial publication of the inscription, Hoftijzer suggested that “the šdyn (and their female counterparts) are identical with the gods as such or possibly represent an important group from among the gods”.170 Müller interpreted them (based on the usage of the term ‫ אל שׁדי‬in Genesis) as a group of ancestral gods,171 but the evidence for this is hardly overwhelming, if even plausible. No basis for such an identification would be possible based on the DAPT. Moore suggested that, “they enact a demonic role in the ‘divine council’ judgment scene in DAT I, a chthonic function repeatedly exercised by the much/feared Sibitti in Anatolia and Mesopotamia, the ‘Satan’ (Job 1:7–12l cf. Num 22:22) and perhaps the rûaḥ šeqer (‘lying spirit,’ 1 Kgs 22:22) in biblical conventions of the divine council, and the šydyn in later Aramaic bowl incantations.”172 Nothing supports this claim, however. Since the inscription’s initial publication, the general consensus has followed Hoftijzer’s suggestion that this reference identifies a specific group of gods within the totality rather than all of the gods.173 However, the syntax and narrative context of the inscription do not commend understanding them as a distinct group of deities within the greater pantheon of deities (‫)אלהן‬. Rather, the parallel structure of vv. 19–22 and vv. 23–24 imply that these terms apply to the same group. In the DAPT, the ‫ שׁדין‬implored ‫ שׁמשׁ‬to unleash her anger, so that darkness would dominate the heavens forever (ln. 6– 7, vv. 25–30). Therefore, the ‫ שׁדין‬of the DAPT function similarly to ‫ שׁדי‬in 169 This is contra Achenbach, Vollendung, 391: “Die Unterstellung, nicht allein die Väter Israels hätten Jahwe als El Schaddaj verehrt, sondern auch der im Ostjordanland besonders verehrte Prophet Bileam, ist dabei das eigentliche Novum. Es setzt die nach-dtr. Ausbildung eines radikalen Monotheismus voraus, der es sich verbietet, mit der Wirkmächtigkeit eines Gottes zu rechnen, der nicht mit Jahwe identisch sein könnte, erst recht dann nicht, wenn diese Wirkmächtigkeit Israels Segnung betrifft.” Achenbach regarded the reference to El and Shaddai as referring to the same deity, though nothing in Numbers 22–24 commends this interpretation. Further, he presumed that this ascription preserved an epithet for YHWH, something that, while possible, remains unexplained in the text. Using those data to then ascribe this material to an editorial layer (i.e., “post-Dtr”) seems therefore fanciful, or at least insecure. 170 Hoftijzer and van der Kooij, ATDA, 275. 171 Cf. Hans-Peter Müller, “Alttestamentliche Probleme,” 65–67 and Hans-Peter Müller, “Aramäische Inschrift,” 223. 172 Michael S. Moore, Balaam Traditions, 87. 173 Cf., e.g., Weippert and Weippert, “‘Bileam’-Inschrift,” 88–92 (= Weippert and Weippert, “‘Bileam’-Inschrift,” 143–47).

356

Balaam as a Character. His Origin and Background

Num 24:4 and 16 on the one hand. In both cases, Balaam sees a vision of the (‫שׁדי)ן‬. But on the other hand, the content of the visions of the (‫ שׁדי)ן‬remains distinct. The vision of Shaddai remains without further explanation in Numbers 24, whereas the DAPT reports Balaam’s vision of the ‫ שׁדין‬as that of the perverted world order. When all of these data are considered together, still little can be said about what the term ‫ שׁדי‬should represent in Numbers 24 (or in the DAPT, for that matter). The singular in Numbers (and generally in the Bible) permits the conclusion that one god should be identified with the appellation, whereas the plural of DAPT suggests several deities being known under this rubric. These gods interact with the sun deity in the DAPT in order to provoke her anger. Nothing like this can be said of ‫ שׁדי‬in Numbers (or in the Bible in general). Numbers 24 provides the sole datum that Balaam either saw a vision of Shaddai or from Shaddai, though the syntax does not clarify which. Considering the parallel structure to “words of El” and “knowledge of Elyon” increases the plausibility of the understanding “vision from Shaddai”.

6.4.5 ‫עליון‬ Like ‫שדי‬, the usage of ‫ עליון‬in the Pentateuch (and Hexateuch/Enneateuch) generally appends it to ‫( אל‬cf. Gen 14:18–20 and 22). Otherwise it presents an adjective or comparative (as in Gen 40:17; Deut 26:19; 28:1; Josh 16:5; 1 Kgs 9:8; 2 Kgs 15:35 [emend to G]; 18:17). The only similar usages to that in Num 24:16 are Deut 32:8 (in the Song of Moses) and 2 Sam 22:14 (= Ps 18:14: David’s Psalm of Thanksgiving). “Curiously, the OT traditions rarely attest ʿElyôn standing alone, without modification. In the Aramaic sections of Daniel, however, references to Yahweh as ʿElyôn (ʿlyʾ/ʿlʾh) often stand independently, without modification, although the intended referent is clearly Yahweh (Note that qdyšy ʿlywnyn is also attested). A similar phenomenon is evidenced in the frequent references to ʿElyôn (hypsistos [altus in 2 Esdr]) in the apocryphal books (1 and 2 Esdr, Tob, Jdt, Add Esth, Wis, Sir, Pr Man, 2 and 3 Macc).In Sir, it is the most common divine name after kyrios. The epithet also occurs in varous [sic] pseudepigraphical works, particularly in T. 12 Patr.”174

In Numbers, Elyon may refer to an independent deity, but this is not assured. “In contrast to the mixed evidence to support the identification of ʿElyôn as autonomous, there is a wide range of evidence to suggest that ʿElyôn was a common epithet in the West Semitic region, applied at different times and in different cultures to any god thought to be supreme.”175 It is used with El and 174 Eric E. Elnes and Patrick D. Miller, “Elyon ‫עליון‬,” in Dictionary of Deities and Demons in the Bible. Second Extensively Revised Edition, ed. Karel van der Toorn, Bob Becking, and Pieter W. van der Horst (Leiden; Boston; Cologne: Brill, 1999), 298. 175 Elnes and Miller, “Elyon,” 295.

6.4 Balaam’s “Religious Affiliations”

357

Baal and even references Zeus among the Greeks. Thus, it remains unclear if this epithet refers to YHWH or some other god, which in turn precludes advancing our understanding of Balaam much. Perhaps one can identify this name as a sobriquet for the God of Israel in its current position, but it could as easily have referenced a distinct or different deity. The circumstances are even more insecure for the epithet ‫ עליון‬than was the case for either ‫ אל‬or ‫שׁדי‬. Like ‫ שׁדי‬and ‫אל‬, ‫ עליון‬appears only in the Numbers recounting of Balaam’s story. Once again, it occurs only in the oracles, this time in the fourth oracle, together with both of the aforementioned. In its context, just like the others, it is redactional and not original to the narrative (as is the whole of the final set of oracles in Num 24:14b–24). The findings about its meaning and an appropriate interpretation preclude any secure resolution. Perhaps it should be understood as another name for YHWH, but perhaps it should not. Certainly it is an editorial addition. Possibly a redactor included these names in the final oracles – without explicit reference to YHWH – in order to suggest that Balaam indeed references other deities, that he is polytheistic. While this is possible, and I indeed favor this interpretation, admittedly little supporting data for this position exists.

6.4.6 Conclusions about Balaam’s Religion The biblical text creates a discontiguous image of Balaam’s religiosity. Numbers 22–24 present him primarily as a follower of YHWH, though the oracles at least permit his being a polytheist. They mention other deities (or at least they ambiguously mention other sobriquets for the same deity) to whom he relates. Nonetheless, Balaam’s polytheistic nature remains at best implicit, as the text can suggest that these names of other deities present epithets for the God of Israel, particularly in the final form of the text. At the same time, one should not overlook the diachrony: the oldest version of the story generally presents Balaam as a faithful presenter of YHWH’s message. Secondary additions question this by introducing more divine names into other contexts, precisely into contexts lacking the name YHWH. Plausibly they undertook such endeavors to caste Balaam in more ambiguous mold. Similarly, biblical authors and tradents working beyond the book of Numbers revised Balaam’s theological background and demonstrate that – regardless of what Balaam may have thought of YHWH – YHWH rejected Balaam. The inscription from Tell Deir ʿAlla suggests that Balaam was a polytheist in their version, being that he was apparently “a man [who] saw the gods” (‫ )אשׁ חזה אלהן‬and knew about the ‫שׁדין‬. However, like the original tale in Numbers 22–24, in the DAPT, Balaam faithfully presents the divine message given to him to the appropriate audience. In general then, the sources again provide a disparate image of Balaam’s background, in this case regarding his religious proclivities.

358

Balaam as a Character. His Origin and Background

6.5 Conclusions about the Character Balaam 6.5 Conclusions about the Character Balaam

Little can be said to circumscribe the character Balaam given the available sources. They provide hardly any compatible information, which suggests that the historical background for the various groups responsible for the data were distinct, whether chronologically, geographically, or socially. The most common identifier of Balaam is his name, unclear in its provenience, followed by his patronymic, “son of Beor”. This patronymic occurs in most biblical witnesses and even the extrabiblical Tell Deir ʿAlla Inscription. Balaam’s presumed preternatural ability presents the second most common theme about him. His special power or insight can also be found in a majority of biblical texts and in the DAPT. Balaam possessed supernatural faculties in the majority of witnesses, and yet even exactly what he was capable of remains somewhat unclear. The terminology is inconsistent and it is unclear if he actually could have affected a change. His divine vision and interaction with deities form the basis for the oldest narratives about him. Both Numbers 22–24 and the DAPT present him as an able and trustworthy messenger of a God or the gods. The idea of cursing remains unique to some biblical witnesses, but this power may have been understood as implicitly based on his interaction with the divine. Either he possessed many functions at the same time or more functions were ascribed to him over time. The later seems more likely based on the redactionhistorical evidence.176 The final significant matter attested in many witnesses is his interaction with foreigners and non-Israelites. While this can again be found in most biblical texts about Balaam, it is absent in the DAPT; in the DAPT, Balaam probably addressed his own people, as indicated by ‫ עמה‬in v. 14 in ln. 4. Nothing indicates that they might have been Israelites. In terms of his heritage, the Bible places his background somewhere in what can only be identified as “greater Aram”, covering an area from Mesopotamia, an area near the modern Turkish border with Syria, or as far south and west as Damascus, and perhaps elsewhere. The oldest version apparently places his origin somewhere near Damascus, but the later literary versions of Balaam expanded this horizon to include Mesopotamia as a possible point of origin. The DAPT contributes no information about his background, but the location of its discovery could suggest a background in the Transjordan, at least for traditions about him. Balaam’s interaction with the Moabites according to the biblical witness – particularly in the oldest version of the story – would superficially 176 Michael S. Moore, Balaam Traditions, 109 appears to favor the former: “In the Bible, Balaam enacts roles as ‘seer,’ ‘oneiromantic,’ and ‘oracle-reciter’ within an overall ‘divineseer’ role-set. But he also enacts roles as ‘purification-priest,’ and (potential) ‘sorcerer’ within a (now severely restricted) ‘exorcist’ role-set. Thus the Balaam we see in the biblical texts is not a simple figure, but a complex one enacting a plurality of roles for which there is not a minimal but a considerable degree of role overlap and preemption.”

6.5 Conclusions about the Character Balaam

359

tend to support such an interpretation. The language of the inscription, and the Aramean domination of the Transjordan around the time of the inscription’s composition, could also commend that geographical background as his original. That would make it consistent with the Bible’s oldest version of this figure. Finally, Balaam’s religious characteristics are disparate in the various witnesses. He has some special relationship to a deity or deities in Numbers 22– 24 and the DAPT. However, the deity rejects him in other biblical texts. For this reason, one must postulate either disparate traditions about Balaam’s character, extant at the same time in diverse contexts, or one must consider the development of the tradition over time to reflect different chronological or social circumstances. The combination of literary-critical, redaction-historical, and tradition-historical findings suggests that the tradition about Balaam moved increasingly from his serving as a faithful servant of a deity or deities to his becoming an enemy of Israel’s deity. The development of the tradition about Balaam over time presents the easiest resolution to these disparities. While the oldest traditions presented him favorably, his interaction with foreigners seems to have biased later biblical tradition against him. Part of this bias was included in Numbers through the addition of various divine titles. Other biblical traditions reflected a version of events in which Balaam’s egregious action against Israel was stopped by their loving God. Yet, the question has remained open up to this point about whether the figure Balaam is historical or literary. The preceding survey cannot really answer that question fully. For Mowinckel, “…das ‘Geschichtliche’ an ihm ist der Name; sonst gehört er dem Märchen.” 177 However, Mowinckel regarded his background as Edomite or Northern Arab,178 which must be rejected based on the other evidence. Other scholars remained more positive about the certainty of Balaam’s historicity.179 The distinct versions of this character with some common elements could speak to his having actually lived. Certainly the literary traditions about him suggest that the authors and even later editors considered him an important figure with something to say about history. They ascribed and even inscribed weighty messages to him. They seem to have accepted the veracity – at least initially – of what he said. Later editors did not seem to doubt this, though they did relativize it and incorporate the biblical materials into other structures. Certainly Balaam was a literary figure of some import, probably from an Aramean milieu historically. Whether he might have actually

177

Mowinckel, “Ursprung,” 237. Cf. Mowinckel, “Ursprung,” 237. 179 Cf., e.g., Wagner, “Offenbarungsphänomenologische Elemente,” 84: “Die Geschichtlichkeit des Bileam unterliegt keinem Zweifel.” 178

360

Balaam as a Character. His Origin and Background

lived must unfortunately remains unanswered, but it is certainly not improbable.180 With this in mind, the basic sketch of the oldest Balaam tradition contains the name and the patronymic (Balaam bar / ben Beor) of a man with some special ability to deliver messages from a deity to an audience. He used this ability accurately and adequately, appropriately transmitting the divine message to the audience, whether it was the message the audience wanted to hear or not. However, the biblical image demonstrates that he was unable to influence the outcome, something that the DAPT may contradict in v. 9. “Alles im Allem wird Bileam [in der Bibel] als Mantiker gezeichnet, der die Zukunft vorhersagen kann. Er ist aber kein effektiver Zauberer, der durch seine Tricks den Gotteswillen verändern kann.”181 And: “In diesem Sinne ist die Bileamerzählung an einer Korrektur eines falschen Gottes- und Prophetenbildes interessiert. Eine Manipulation Gottes ist nach der Bileamerzählung zu keinem Zeitpunkt durch einen geübten Propheten möglich, da sich Gott für die üblen Zwecke der Menschen nicht einspannen lässt. Segen und Fluch können folglich nicht von Mantikern kontrolliert werden. Vielmehr ist beides an den Willen Gottes zurückgebunden. Ein wahrer Prophet kann demnach all das und nur das mitteilen, was ihm tatsächlich von Gott offenbart worden ist.”182

The biblical tradition focused on the deity YHWH, who had already taken on some characteristics of El. From this core, the other biblical traditions developed around Balaam. The DAPT presents Balaam as the messenger of El’s vision, which incorporates a message from an assembly of gods to another goddess. The more or less common geographical and linguistic background of the oldest witnesses commends a geographical background for the character Balaam (and any historical figure upon whom he might have been based) in Aramean contexts, perhaps even in occupied Transjordan. The origin of the figure, based both on biblical and Transjordanian sources could be from the area around Damascus. That would mean that Balaam was transposed to Transjordan. From there the story of Balaam was adapted to fit into its biblical context, which continued to develop over time into the final text we find before us now.

180

Contra Kaiser, “Balaam Son of Beor.”, who not only regarded his living as a certainty, but even went so far as to regard all of the disparate biblical material about him as biographically relevant to the description of his person. 181 Gaß, Menschliches Handeln, 4. 182 Gaß, Menschliches Handeln, 4.

Chapter 7

Conclusions and Impetus for Further Research Over the course of this study, a number of results crystallized. We traced Balaam’s development from the ninth or eighth century BCE into late antiquity. We identified this figure’s most likely cultural background in the region around Damascus. We noticed how attitudes about this figure changed over time, leaving distinct impressions about him in different biblical texts and even within the same pericope. Various groups described him and diverse aspects of his character over time. This process continued well beyond the literary history of the Bible, i.e., beyond what anyone would traditionally recognize as an Endtext. Rather, even the various textual traditions indicate continued fascination with this figure and manifold attitudes about him and what he did. To this we will return at the conclusion. When regarding the history of research in Chapter One, a few important points piqued our interest through the discussion of the models for the literary history of the Balaam periope and its place in the biblical compositions. This study demonstrated some weaknesses with contemporary models for the literary history of the Pentateuch, Hexateuch, and Enneateuch. Basically, three problems stand in the background of the issues in the distinct models and their individual iterations. First, to what degree is the story about Balaam uniform? Scholars have offered disparate opinions about this, ranging from entirely uniform to the product of radically disparate redactional undertakings. Many recognized the source-like characteristics of some elements in these chapters. That brings us in turn to the second major issue behind all of these models: to what degree and in what ways does the Balaam narrative relate to its current literary context? Again, the positions vary from its existence as an independent literary piece to its always having been a part of some longer literary composition(s). Finally, how do the Balaam traditions in Numbers 22–24 relate to other traditions about this figure, whether in the Bible or elsewhere? Texts about Balaam are disparate. These differences become apparent even when look at different manuscripts and translations of the material. The manuscript tradition presents a number of variants in passages that deal with Balaam. However, with careful consideration, one is able to reconstruct an older version of the Balaam texts that stood behind the distinct traditions. Naturally, Numbers 22–24, the most extensive text about Balaam, required the most ef-

362

Conclusions and Impetus for Further Research

fort. But others, particularly Deuteronomy 23:4–6 and Joshua 24:9–10, demonstrated similar, albeit distinct issues. Over the course of the text-historical examinations, three factors behind the variations in witnesses were recognized: errors, interpretive reworking, and tendencies toward consistency. These appeared within the various traditions and were not simply a matter outside of the Masoretic text. More specifically regarding the figure Balaam, two important aspects appeared in the text-historical discussions in these surveys. First, it became quite apparent that biblical tradents wrestled with characterizations of Balaam for a long time. And these different versions and attitudes left distinct impressions, even in the New Testament in which Matthew 2 seems to favor a text more like G for Numbers 22–24. Second, the various traditions – particularly Smr and G, but also M – continued to develop more consistent texts, and thus data about Balaam, over time. For example, Deut 23:4–6 later incorporated information about Balaam’s heritage more consistent with Num 22:5. Both Smr and M, albeit in distinct variations, attest this increasing consistency in this case. Sometimes, one even recognizes an ancient awareness of these distinct versions, as in Jerome’s translation of Num 24:24, which reflects a combination of variants from G and M. in turn, these distinct readings indicate scribal interpretation and harmonization in their characterization of Balaam. These variant text forms demonstrate editorial and scribal fascination with Balaam well into late antiquity. At the same time, the limited number of variants demonstrate a marked level of textual stability over the centuries, which exegetes should not too quickly overlook or disregard. Yet, the variants also demonstrate processes that could reflect on the theoretically reconstructed literary history that stands behind the versions we know. That is, the textual history appears as an outgrowth of the preceding literary history and is hardly distinguishable from it. Regarding the figure Balaam in biblical literary history, one can distinguish several phases. The oldest story about Balaam remains behind the text found in Numbers 22–24*. We were able to reconstruct it through a detailed literarycritical examination. This oldest source about Balaam in the Bible (S1) presented him as someone commanded to curse the people undertaking the exodus. However, he did not do that, but blessed them instead. This oldest version consisted of a shorter narrative than that currently found in Numbers 22–24, reflected positively on Balaam ben Beor, and transmitted only three supposed oracles. While it is possible that this story stood within a wider narrative context of some exodus–eisodus recounting, it still demonstrates a strikingly independent character. Beyond that, even in its earliest version, the oldest biblical Balaam story looked toward the monarchy, demonstrating that this version presumably developed during that period. The development of Numbers 22–24 did not end with this earliest version, however. Rather various editors supplemented the material about Balaam, including new data and incorporating his story into different narrative frameworks.

Conclusions and Impetus for Further Research

363

The first group of editors to revise the Balaam material (R1) added mostly to the beginning and end of the text, while leaving the middle largely untouched. With the exception of the inclusion of a fourth oracle, most of the additions were small and particular, consisting only of a few words or phrases. These additions are not dissimilar from the kind of distinctions recognized in the text history of this passage. The second editor (R2) evinced a similar methodology, though incorporating material that appears to have been a second source (S2), not about Balaam. It read like the interpolation of extant material, a fragment of a story about Moabites and Midianites – without any recourse to the figure Balaam – seeking to hex the Israelites. Such can be reconstructed from Num 22:1, 3b–4a, 5*, 7a, and 24:1*. Scribes included it mostly in the exposition of the Balaam story, but these editors included a few phrases to better incorporate it and make the whole more consistent, such as 22:3a* and 11*. This second source might have led immediately into some material currently found in Numbers 25, such as v. 9. The evidence of these first editings left traces mostly through the repeated expositions. To this end, one notices that the current text of Numbers 22 begins distinctly in vv. 1, 2, 3a, and 3b–4a. At the same time, one recognizes the distinct material in the episode’s conclusion, namely the interpolation between Num 24:14a and 25. Later editors continued adding material at the conclusion of Balaam’s oracles. This becomes apparent through the repeated introductions of oracular material in Num 24:15, 20, 21, and 23. Elements of these additions appear to be recontexualizations within new compositions, such as the addition of Num 22:2 and 24:14b–19 (R1). These verses incorporated the older Balaam narrative into an early Deuteronomistic composition. Later additions, such as the inclusion of the aforementioned fragment (S2), incorporated this story into a priestly composition (R2). Still other additions to Numbers 22–24 appear to be Fortschreibungen, expansions for their specific context that may or may not have been part of lengthier editorial expansions throughout the whole corpus. Most readily recognizable to this end are the final oracles in Num 24:20–24 and the story of Balaam’s talking donkey in 22:21aβ + 22–35. Nothing in these additions particularly developed the Balaam narrative as part of some Pentateuch, Hexateuch, or Enneateuch composition. The connections to other biblical materials are limited and dispersed throughout the canon. They developed further in terms of consistency with other biblical passages, much as seen in the pericope’s textual history. Nevertheless, the reconstructed literary and redaction history suggested that attitudes about Balaam became increasingly negative over time. While the earliest version of Numbers 22–24 presented him as a truthful seer, the final edition turned him into a laughable figure, not even able to recognize the truth of a scenario as well as his dumb ass. However, one cannot discuss Balaam in the Bible and only focus on Numbers 22–24. Many other texts demonstrate familiarity with this character or its

364

Conclusions and Impetus for Further Research

narrative tradition. Like the editorial additions to Numbers 22–24, though, these texts often demonstrate distinct knowledge or interpretations about Balaam from that of the earliest biblical Balaam narrative. Yet, they demonstrate the same tendency to reflect increasingly negatively on Balaam as a character. For example, Micah 6 and Judges 11 both seem to affirm an image of Balaam (or the narrative more generally, since Balaam makes no appearance in Judges) consistent with that identified in the first redaction of Numbers 22–24. They reflect information and attitudes about him there: still friendly, or at least neutral, towards him. Later texts, like Joshua 24, appear to reinterpret Balaam’s story from Numbers 22–24 in a manner more like the second redaction of Numbers 22–24. Now, for the first time, Balaam really did attempt to curse the Israelites in order to aid Moab. Numbers 22–24 never explicitly states this in any iteration, suggesting that Joshua 24 represents an editorial commentary on or an interpretation of that text. Numbers 31 even evinces connections to two distinct editorial layers identified in Numbers. On the one hand, Num 31:16a describes Balaam as the responsible party for the Midianite seduction of the Israelites. This identification presumes the combination of Numbers 22–25 in a later version of those stories (R2). Numbers 31:16b mentions the “assembly of YHWH”, a later term in Numbers. Numbers 31:8 goes so far as to describe Balaam’s justified demise, anticipating his responsibility for Num 31:16a (R3+). A later scribe even expounded on this: Josh 13:21–22, which appears to borrow from Num 31:8, described Balaam as the mantic (‫)הקוסם‬, a pejorative in the Hebrew Bible. Finally, Deuteronomy 23 (and Nehemiah 13, which relies on it) added Balaam into a legal corpus as grounds for forbidding certain groups from becoming part of the religious assembly. To this end, it reflects the negative aspects attributed to Balaam from Joshua 24, while perhaps incorporating the conglomeration of Ammonites and Moabites from Judges 11 and Genesis 19. Taken together, these observations confirm the literary-critical and redactionhistorical results of our study of Numbers 22–24 and demonstrate the common development of attitudes toward Balaam over time. The increasingly negative characterization of Balaam came to dominate the landscape, as demonstrated specifically in the New Testament, but also in Qumran and the Rabbinic tradition. Outside of the bible, the Deir ʿAlla Plaster Inscription Combination A (DAPT) promotes an image of Balaam not too dissimilar from that of the oldest layer of the Numbers composition in Numbers 22–24. This text also identifies Balaam as someone with a supernatural ability and apparently some special connection to the divine. Just as in Numbers 22–24, Balaam experiences a nocturnal divine visitation in the text from Deir ʿAlla. In spite of their differences, both of these texts appear to stem from a common tradition-historical background. The Aramaic language of the text may affirm the Bible’s original ascription of Aramean heritage to Balaam. Together, that could imply that this figure originally came from that region. The date of the Deir ʿAlla text, which

Conclusions and Impetus for Further Research

365

can be more certainly fixed than the oldest biblical text, speaks to a date around the transition from the ninth to the eighth centuries up to around 750 BCE. That would again affirm both the monarchic background and Israelite (i.e., northern) provenience of the oldest biblical Balaam narrative. In terms of the tradition-historical background of the text, one notes that different groups at different periods ascribed different features to Balaam and used him to different ends. The oldest groups, both those attested in the Hebrew Bible and those at Deir ʿAlla, apparently understood him as someone presenting legitimate prophecy to some group. In one case, the message was apparently one of hope for the target and an implicit rebuke of the patron. The other case, the DAPT, attributes a warning to Balaam of the potentially perverted world. These variations on Balaam may attest his being a polytheist; while this is certain for Deir ʿAlla, it is less so for the Hebrew Bible. Initially, at any rate, Balaam apparently was a person of some respect and the proponent of legitimate prophecy. The first redaction of Numbers appears to demonstrate essentially the same attitude about Balaam, who now even prophesies about the coming of David. Later editors used Balaam in new ways for new ends. For example, they changed his background from Aramean to Mesopotamian. No longer was he a recipient of legitimately prophetic dreams. He was a mantic who undertook forbidden practices. No longer did he refuse to transgress Israel’s blessing, but rather went out to aid Balak in his attempt to destroy them. For this reason, editors determined that it was only legitimate he be brought to justice and killed at Israel’s hand. From that point on, there was no way to rehabilitate this biblical character. One matter about the figure Balaam cannot be answered with any certainty. It remains impossible to determine if some historical figure stood in the background of this biblical and extrabiblical character. Nothing should preclude such. Some historical prophetic figure Balaam could have inspired these literary iterations of Balaam, just as a historical Caesar inspired different literary images of Caesar. At the same time, that must remain in the realm of speculation. The Bible’s attribution of Balaam to the era before the existence of Israel speaks to a more mythological background for this figure. At least in the ninth or eighth century, the biblical authors demonstrated no qualms about casting him that far back in Israel’s supposed history. In this capacity, the DAPT unfortunately proffers no help, as it provides no date for the supposed delivery of Balaam’s reported vision. At any rate, the possibility remains open that Balaam was a historical figure not entirely like Isaiah or Jeremiah, though this cannot be affirmed with any degree of certainty. With these observations, we can return to the beginning of this study: Who was Balaam and what did he do? No single answer fulfills this question; it depends on who you ask and when. The authors who initially described him

366

Conclusions and Impetus for Further Research

regarded him as a figure of special means, of supernatural ability. He transmitted divine speech and engaged with deities. He have been Aramean or stemmed from Aramean literary tradition. His words were true and evoked response. They merited transmission over centuries, at least in the case of the Bible. Later editors did not favor him as much, but still regarded him as the prophet of David or even of some coming Messiah. He might then have been Mesopotamian. Through his trickery, he led the Israelites astray. Thus, he merited death at the hands of the object of his supposed scorn. Other scribes mocked him as the seer with less ability than his beast of burden, the divine messenger whose message was changed by the divinity, proving that it was not accurate. Balaam thus became different things over the course of the transmission of data about him. The most secure things we can say about him are that his name was Balaam, his father was Beor, he may have been Aramean, and he probably had some preternatural ability. From that, all of the legends were born. What impetus does this study provide for further research? Certainly future survey will provide further nuance to the data presented here regarding the figure Balaam. That is no different than what this study did regarding its predecessors. More study is required in the reception history of the figure Balaam, who features often in art, in addition to Rabbinic literature. Both of these aspects merit further study on how they reflect on this character. In terms of biblical literature, this study suggested that a new model for the development of the Pentateuch, Hexateuch, and Enneateuch is necessary. Perhaps we should attempt such a model with a new heuristic, just as proposed here regarding the epithets used for the group undertaking the exodus. Distinct schools of theological discourse apparently relied reliably on distinct terminology to this end. That could aid in further reconstruction. Additionally, renewed interest in the narrative logic of larger portions remains necessary to determine how the pieces developed. The oldest Balaam story in the Bible, probably even from the time of the monarchy, already presupposed a connection to some exodus legend and pointed toward the arising of an Israelite monarchy. That places it firmly in a tradition that one could identify with the Enneateuch (though the status of works like Genesis or Judges in any such composition would have to be justified and further explored). Whether this implies that portions of Exodus–2 Kings were part of one composition from the outset or were perhaps something like volumes or individual scrolls in a contiguous narrative, still needs to be explored. The extent of the editorial undertakings in Numbers and elsewhere, most prominently in Exodus and Joshua, could be attempted with the same criteria established here. Taken together, these observations suggest that one cannot too easily dismiss all of the material in Numbers as a late bridge between some exodus story and the Deuteornomistic history. Rather, Numbers from its inception or earliest editing seems to have been a part of these compositional and editorial processes. The precise extent of what each of these entailed must be further examined. That will be the object of another study.

Bibliography Bibliography Abramsky, S. “The Qenites [Hebrew].” EI 3 (April 1954): 116–24. Achenbach, Reinhard. Die Vollendung der Tora. Studien zur Redaktionsgeschichte des Numeribuches im Kontext von Hexateuch und Pentateuch. BZABR, vol. 3. Wiesbaden: Harrassowitz, 2003. Aejmelaeus, Anneli. “Translation Technique and the Intention of the Translator.” In VII Congress of the International Organization for Septuagint and Cognate Studies, Leuven 1989, edited by Claude E. Cox, 23–36. Atlanta, GA: Scholars Press, 1991. Albertz, Rainer. “Das Buch Numeri jenseits der Quellentheorie. Eine Redaktionsgeschichte von Num 20–24 (Teil I).” ZAW 123, no. 2 (2011): 171–83. –. “Das Buch Numeri jenseits der Quellentheorie. Eine Redaktionsgeschichte von Num 20– 24 (Teil II).” ZAW 123 (2011): 336–47. Albertz, Rainer, and Rüdiger Schmitt. Family and Household Religion in Ancient Israel and the Levant. Winona Lake, IN: Eisenbrauns, 2012. Albright, William Foxwell. “The Home of Balaam.” JAOS 35 (1915): 386–90. –. “Islam and the Religions of the Ancient Orient.” JAOS 60, no. 3 (September 1940): 283– 301. –. “The Names Shaddai and Abram.” JBL 54 (1935): 180–93. –. “The Oracles of Balaam.” JBL 63 (1944): 207–33. –. “Some Important Recent Discoveries: Alphabetic Origins and the Idrimi Statue.” BASOR 118 (April 1950): 11–20. Albright, William Foxwell, and Christopher S. Mann. Matthew. Introduction, Translation, and Notes. AncB, vol. 26. New York: Doubleday, 1971. Alexandre, Monique. Jésus (Josué). La Bible d’Alexandrie, vol. 6. Paris: Cerf, 1996. Allegro, John Marco. “The Meaning of the Phrase šeṯūm hāʿayin in Num. XXIV 3, 15.” VT 3, no. 1 (January 1953): 78–79. –, ed. Qumran Cave 4.I: (4Q158–186). DJD, vol. 5. Oxford: Clarendon Press, 1968. Ashley, Timothy R. The Book of Numbers. NICOT. Grand Rapids, MI; Cambridge, UK: William B. Eerdmans, 1993. Auld, Graeme A. “Samuel, Numbers, and the Yahwist-Question.” In Abschied vom Jahwisten. Die Komposition des Hexateuch in der jüngsten Diskussion, edited by Jan Christian Gertz, Konrad Schmid, and Markus Witte, 233–46. Berlin; New York: de Gruyter, 2002. Aune, David Edward. Revelation 1–5. WBC, vol. 52A. Waco: Word Books, 1997. Ausloos, Hans. “The Septuagint’s Rendering of Hebrew Toponyms as an Indication of the Translation Technique of the Book of Numbers.” In Textual Criticism and Dead Sea Scrolls Studies in Honour of Julia Trebolle Barrera. Florilegium Complutense, edited by Andrés Piquer Otero, Pablo A. Torijano Morales. JSJS, 35–50. Leiden; Boston: Brill, 2012. Baden, Joel S. The Composition of the Pentateuch: Renewing the Documentary Hypothesis. New Haven; London: Yale University Press, 2012.

368

Bibliography

–. “Identifying the Original Stratum of P: Theoretical and Practical Considerations.” In The Strata of the Priestly Writings: Contemporary Debate and Future Directions, Sarah Shectman and Joel S. Baden, 13–29. Zürich: Theologischer Verlag Zürich, 2009. –. J, E, and the Redaction of the Pentateuch. FAT. Tübingen: Mohr Siebeck, 2009. Baentsch, Bruno. Exodus, Levitikus, Numeri. HKAT, vol. 1.2. Göttingen: Vandenhoeck & Ruprecht, 1903. Barré, Michael L. “The Portrait of Balaam in Numbers 22–24.” Int 51 (1997): 254–66. Bartelmus, Rüdiger. “Von Eselinnen mit Durchblick und blinden Sehern. Numeri 22,20–35 als Musterbeispiel narrativer Theologie im Alten Testament.” TZ 61, no. 1 (2005): 27– 43. Baskin, Judith R. “Origen on Balaam: The Dilemma of the Unworthy Prophet.” VC 37, no. 1 (March 1983): 22–35. Bauckham, Richard. Jude, 2 Peter. WBC, vol. 50. Waco: Word Books, 1983. Baumgarten, Joseph M., ed. Qumran Cave 4.XIII: The Damascus Document (4Q266–273). DJD, vol. 18. Oxford: Clarendon Press, 1996. Bewer, Julius A. “The Literary Problems of the Balaam Story in Numb., Chaps. 22–24.” AmJT 9, no. 2 (April 1905): 238–62. Beyerle, Stefan. “‘A Star Shall Come Out of Jacob’: A Critical Evaluation of the Balaam Oracle in the Context of Jewish Revolts in Roman Times.” In The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, edited by George H. van Kooten and Jacques van Ruiten. TBN, vol. 11, 163–88. Leiden: Brill, 2008. Biale, David. “The God with Breasts: El Shaddai in the Bible.” HR 21, no. 3 (February 1982): 240–56. Bigg, Charles. A Critical and Exegetical Commentary on the Epistles of St. Peter and St. Jude. ICC. Edinburgh: Clark, 1946. Blount, Brian K. Revelation. A Commentary. NTL. Louisville, KY: Westminster John Knox Press, 2009. Blum, Erhard. “Die altaramäischen Wandinschriften vom Tell Deir ʿAlla und ihr institutioneller Kontext.” In Metatexte. Erzählungen von schrifttragenden Artefakten in der alttestamentlichen und mittelalterlichen Literatur, edited by Friedrich-Emanuel Focken and Michael R. Ott, 21–52. Berlin: de Gruyter, 2016. –. “Die aramäischen Wandinschriften von Tell Deir ʿAlla.” In Texte aus der Umwelt des Alten Testaments. Neue Folge Band 8. Weisheitstexte, Mythen und Epen, edited by Bernd Janowski and Daniel Schwemer, 459–74. Gütersloh: Gütersloher Verlagshaus, 2015. –. “Issues and Problems in the Contemporary Debate Regarding the Priestly Writings.” In The Strata of the Priestly Writings: Contemporary Debate and Future Directions, Sarah Shectman and Joel S. Baden, 31–44. Zürich: Theologischer Verlag Zürich, 2009. –. “Die Kombination I der Wandinschrift vom Tell Deir ‘Alla: Vorschläge zur Rekonstruktion mit historisch-kritischen Anmerkungen.” In Berührungspunkte: Studien zur Sozialund Religionsgeschichte Israels und seiner Umwelt. Festschrift für Rainer Albertz zu seinem 65. Geburtstag, edited by Ingo Kottsieper, Rüdiger Schmitt, and Jakob Wöhrle. AOAT, vol. 350, 573–601. Münster: Ugarit-Verlag, 2008. –. Studien zur Komposition des Pentateuch. BZAW, vol. 189. Berlin: de Gruyter, 1990. –. “‘Verstehst du dich nicht auf die Schreibkunst..?’ Ein weisheitlicher Dialog über Vergänglichkeit und Verantwortung: Kombination II der Wandinschrift vom Tell Deir ‘Alla.” In Was ist der Mensch, dass du seiner gedenkst (Psalm 8,5). Aspekte einer theologischen Anthropologie. Festschrift für Bernd Janowski zum 65. Geburtstag, edited by Michaela Bauks, Kathrin Liess, and Peter Riede, 33–53. Neukirchen-Vluyn: Neukirchener Verlag, 2008.

Bibliography

369

Bogaert, Pierre-Maurice. “Heshbon entre Moab et Ammon: La finale ajoutée à l’oracle sur Moab en Jr 48,45–47 TM`.” In Interpreting Translation: Studies on the LXX and Ezekiel in Honour of Johan Lust, edited by Florentino García Martínez and Marc Vervenne. BETL, vol. 192, 45–54. Leuven: University Press, 2005. Borgen, Peder. “‘There Shall Come Forth a Man’: Reflections on Messianic Ideas in Philo.” In The Messiah: Developments in Earliest Judaism and Christianity, edited by James H. Charlesworth, 341–61. Minneapolis: Fortress Press, 1992. Brichto, Herbert Chanan. The Problem of “Curse” in the Hebrew Bible. JBL Monograph Series, vol. 13. Philadelphia: Society of Biblical Literature and Exegesis, 1963. Broshi, Magen, ed. The Damascus Document Reconsidered. Jerusalem: The Israel Exploration Society, 1992. –, ed. Qumran Cave 4.XIV: Parabiblical Texts, Part 2. DJD, vol. 19. Oxford: Clarendon Press, 1995. Budd, Philip J. Numbers. WBC, vol. 5. Waco: Word Books, 1984. Burrows, Eric. The Oracles of Jacob and Balaam. Bellarmine Series, vol. 3. London: Burns, Oates & Washbourne, 1938. Caquot, André, and André Lemaire. “Les Textes Araméens de Deir ʿAlla.” Syria 54 (1977): 189–208. Carr, David M. The Formation of the Hebrew Bible: A New Reconstruction. New York: Oxford University Press, 2011. –. “Method in Determining the Dependence of Biblical on Non-Biblical Texts.” In Subtle Citation, Allusion, and Translation in the Hebrew Bible, edited by Ziony Zevit, 41–53. Sheffield; Bristol, CT: Equinox, 2017. Charles, Robert Henry. A Critical and Exegetical Commentary on the Revelation of St. John, Volume 1. ICC. Edinburgh: Clark, 1920. Coats, George W. “Balaam: Sinner or Saint?” BR 18 (1973): 1–9. –. “Balaam: Sinner or Saint?” In Saga, Legend, Tale, Novella, Fable. Narrative Forms in Old Testament Literature, vol. 35, edited by George W. Coats. JSOTSup, vol. 35, 56–62. Sheffield: JSOT Press, 1985. Collins, John J. “Messianism and Exegetical Tradition. The Evidence of the LXX Pentateuch.” In The Septuagint and Messianism, edited by Michael A. Knibb, 129–49. Leuven: University Press, 2006. Collins, Marilyn Frances. Messianic Interpretation of the Balaam Oracles. University Microfilms International Dissertation Service. Ann Arbor, Mich., 1990. Cook, Edward M. Dictionary of Qumran Aramaic. Winona Lake, IN: Eisenbrauns, 2015. Cross, Frank Moore. Canaanite Myth and Hebrew Epic. Essays in the History of the Religion of Israel. Cambridge, MA: Harvard University Press, 1973. Daiches, Samuel. “Balaam – A Babylonian Bārū.” In Hilprecht Anniversary Volume, 60–70. Leipzig: J.C. Hinrichs, 1909. Davies, T. Witton. Magic, Divination, and Demonology Among the Hebrews and Their Neighbours Including an Examination of Biblical References and of the Biblical Terms. 1898. New York: Ktav Publishing House, Inc., 1969. Davies, William D., and Dale C. Allison. A Critical and Exegetical Commentary on the Gospel According to Saint Matthew, Volume One. ICC. Edinburgh: Clark, 1988. Delcor, Mathias. “Balaʿam Patorah, intrete des songes au pays d’Ammon d’apres Nombres 22:5. Les temoignages épigraphiques paralleles.” Semitica 32 (1982): 89–91. –. “Le texte de Deir ʿAllā et les oracles bibliques de Balaʿam.” In Congress Volume, Vienna 1980. VTSup, vol. 32, 52–73. Leiden: Brill, 1981.

370

Bibliography

Dicou, Bert. Edom, Israel’s Brother and Antagonist. The Role of Edom in Biblical Prophecy and Story. JSOTSup, vol. 169. Sheffield: JSOT Press, 1994. Dijkstra, Meinert. “Is Balaam Also Among the Prophets?” JBL 114 (1995): 43–64. Dillmann, August. Die Bücher Numeri, Deuteronomium und Josua. Kurzgefasstes exegetisches Handbuch zum Alten Testament. Leipzig: Verlag von S. Hirzel, 1886. Donelson, Lewis R. I & II Peter and Jude. NTL. Louisville, KY: Westminster John Knox Press, 2010. Donner, Herbert. “Balaam pseudopropheta.” In Beiträge zur Alttestamentlichen Theologie: Festschrift für Walther Zimmerli zum 70. Geburtstag, edited by Herbert Donner, Robert Hanhart and Rudolf Smend, 112–23. Göttingen: Vandenhoeck & Ruprecht, 1977. Donner, Herbert, and Wolfgang Röllig. Kanaanäische und aramäische Inschriften. Wiesbaden: Harrasowitz Verlag, 2002. Dorival, Gilles. Les Nombres. La Bible d’Alexandrie, vol. 4. Paris: Cerf, 1994. Dorival, Gilles, Marguerite Harl, and Olivier Munnich. La Bible grecque des Septante. Du judaïsme hellénistique au christianisme ancien. Initiations au christianisme ancien. Paris: Cerf, 1988. Douglas, Mary. “Balaam’s Place in the Book of Numbers.” Man (N.S.) 28, no. 3 (September 1993): 411–30. Dozeman, Thomas B. “The Midianites in the Formation of the Book of Numbers.” In The Books of Leviticus and Numbers, Thomas Römer, 261–84. Leuven: Peeters, 2008. Driver, Samuel Rolles. A Critical and Exegetical Commentary on Deuteronomy. ICC. Edinburgh: T&T Clark, 1986. Duncan, J. Garrow. “Notes on the Sites of Succoth and Penuel as Bearing Upon the Routes of Gideon and Jacob (Concluded).” PEQ 59, no. 4 (October 1927): 188–91. –. “Notes on the Sites of Succoth and Penuel as Bearing Upon the Routes of Gideon and Jacob.” PEQ 59, no. 2 (April 1927): 89–96. Ebach, Ruth. Das Fremde und das Eigene. Die Fremdendarstellungen des Deuteronomiums im Kontext israelitischer Identitätskonstruktionen. BZAW, vol. 471. Berlin; New York: de Gruyter, 2014. Ego, Beate, Armin Lange, Hermann Lichtenberger, and Kristin De Troyer, eds. “Minor Prophets.” In Biblia Qumranica, edited by Beate Ego, Armin Lange, Hermann Lichtenberger, and Kristin De Troyer. Leiden; Boston: Brill, 2005. Eissfeldt, Otto. “El and Yahweh.” JSS 1 (1956): 25–37. –. Hexateuch-Synopse. Die Erzählung der fünf Bücher Mose und des Buches Josua mit dem Anfange des Richterbuches in ihre vier Quellen zerlegt und in deutscher Übersetzung dargeboten samt einer in Einleitung und Anmerkungen gegebenen Begründung. 1922. Darmstadt: Wissenschaftliche Buchgesellschaft, 1973. –. “Die Komposition der Bileam-Erzählung. Eine Nachprüfung von Rudolphs Beitrag zur Hexateuchkritik.” ZAW 57 (1939): 212–41. –. “Die Komposition der Bileam-Erzählung. Eine Nachprüfung von Rudolphs Beitrag zur Hexateuchkritik.” In Kleine Schriften. Zweiter Band, edited by Rudolf Sellheim and Fritz Maass, 199–226. Tübingen: Mohr Siebeck, 1963. –. “‘Mein Gott’ im Alten Testament.” ZAW 61 (1945–48): 3–16. –. “‘Mein Gott’ im Alten Testament.” In Kleine Schriften. Dritter Band, edited by Rudolf Sellheim and Fritz Maass, 35–47. Tübingen: Mohr Siebeck, 1966. –. “Sinai-Erzählung und Bileam-Sprüche.” HUCA 32 (1961): 179–90. –. “Sinai-Erzählung und Bileam-Sprüche.” In Kleine Schriften. Vierter Band, edited by Rudolf Sellheim and Fritz Maass, 21–31. Tübingen: Mohr Siebeck), 1968.

Bibliography

371

–. “El und Jahwe.” In Kleine Schriften. Dritter Band, edited by Rudolf Sellheim and Fritz Maass, 388–97. Tübingen: Mohr Siebeck, 1966. –. “Wahrsagung im Alten Testament.” In La divination en Mésopotamie ancienne et dans les régions voisines. XIVe Rencontre Assyriologique Internationale (Strasbourg, 2 – 6 juillet 1965), 141–46. Paris: Presses Université de France, 1966. –. “Wahrsagung im Alten Testament.” In Kleine Schriften. Vierter Band, edited by Rudolf Sellheim and Fritz Maass, 271–75. Tübingen: Mohr Siebeck, 1968. Elitzur, Yoel. “Naḥal Zered in the Bible and the Baraita de-Teḥumin.” PEQ 145, no. 2 (2013): 108–18. Elnes, Eric E., and Patrick D. Miller. “Elyon ‫עליון‬.” In Dictionary of Deities and Demons in the Bible. Second Extensively Revised Edition, edited by Karel van der Toorn, Bob Becking, and Pieter W. van der Horst, 293–99. Leiden; Boston; Cologne: Brill, 1999. Fabry, Heinz-Josef, and Ulrich Dahmen, ed. ṭhr–str. Vol. 2 of Theologisches Wörterbuch zu den Qumrantexten. Stuttgart: Kohlhammer, 2013. Fabry, Heinz-Josef. “‫נָ ָחשׁ‬.” In Theologisches Wörterbuch zum Alten Testament, vol. 5 ‫–מרר‬ ‫עזב‬, ed. Gerhard Johannes Botterweck, Helmer Ringgren, and Heinz-Josef Fabry, 384– 97. Stuttgart: Kohlhammer, 1986. Fabry, Heinz-Josef, and Ulrich Dahmen. “‫פשׁר‬.” In Theologisches Wörterbuch zum Alten Testament, vol. 6 ‫קום–עזז‬, ed. Heinz-Josef Fabry and Helmer Ringgren, 810–16. Stuttgart; Berlin; Cologne: Kohlhammer, 1989. Fernández Marcos, Natalio. The Septuagint in Context. Introduction to the Greek Version of the Bible. Translated by Wilfred G.E. Watson. Atlanta: SBL Press, 2000. Fiedler, Peter. Das Matthäusevangelium. Theologischer Kommentar zum Neuen Testament, vol. 1. Stuttgart: Kohlhammer, 2006. Finkelstein, Jacob Joel. “Mesopotamia.” JNES 21 (1962): 73–92. Fishbane, Michael. Biblical Interpretation in Ancient Israel. Oxford: Clarendon, 1985. Fistill, Ulrich. Israel und das Ostjordanland. ÖBS, vol. 30. Frankfurt a.M.: Peter Lang, 2007. Ford, Josephine Massynberde. Revelation. Introduction, Translation, and Commentary. AncB, vol. 38. Garden City, NY: Doubleday, 1975. Fornberg, Tord. “Balaam and 2 Peter 2:15: ‘They Have Followed in the Steps of Balaam’ (Jude 11).” In The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, edited by George H. van Kooten and Jacques van Ruiten. TBN, vol. 11, 265– 74. Leiden: Brill, 2008. Forsling, Josef. Composite Artistry in the Book of Numbers. A Study in Biblical Narrative Conventions. Studia Theologica Holmiensia, vol. 22. Åbo, Finland: Åbo Akademis förlag, 2013. France, Richard T. The Gospel of Matthew. NICNT. Grand Rapids, MI: Eerdmans, 2007. Frankel, David. “The Deuteronomic Protrayal of Balaam.” VT 46, no. 1 (January 1996): 30– 42. Frankel, Zacharias. Ueber den Einfluss der palästinischen Exegese auf die alexandrinische Hermeneutik. Leipzig: Barth, 1851. Frankemölle, Hubert. 1. Petrusbrief, 2 Petrusbrief, Judasbrief. NEchtB, vol. 18–20. Würzburg: Echter-Verlag, 1987. Franken, Hendricus Jacobus. “Deir ʿAlla Re-Visited.” In The Balaam Text from Deir ʿAlla Re-Evaluated: Proceedings of the International Symposium Held at Leiden, 21–24 August 1989, ed. Jacob Hoftijzer, Gerrit van der Kooij, 3–15. Leiden; New York; Copenhagen; Cologne: Brill, 1991. –. Excavations at Tell Deir ʿAllā, Volume 1: A Stratigraphical and Analytical Study of the Early Iron Age Pottery. DMOA, vol. 16. Leiden: Brill, 1969.

372

Bibliography

–. “The Stratigraphic Context of the Clay Tablets Found at Deir ʿAlla.” PEQ 96 (1964): 73– 78. –. “Texts from the Persian Period from Tell Deir ‘Allā.” VT 17 (1967): 480–81. Freedman, David Noel, ed. The Leningrad Codex: A Facsimile Edition. General editor David Noel Freedman, managing editor Astrid B. Beck, associate editors Bruce E. Zuckerman and Marilyn J. Lundberg, publication editor James A. Sanders. Grand Rapids, MI: Eerdmans, 1998. Frevel, Christian. Mit Blick auf das Land die Schöpfung erinnern. Zum Ende der Priestergrundschrift. Herders Biblische Studien, vol. 23. Freiburg im Breisgau: Herder, 2000. Friedman, Richard Elliott. The Bible with Sources Revealed: A New View Into the Five Books of Moses. San Francisco: HarperSanFrancisco, 2003. –. Commentary on the Torah with a New English Translation and the Hebrew Text. 2001. San Francisco: Harper Collins, 2003. Frisch, Amos. “The Story of Balaam’s She-Ass (Numbers 22:21–35): A New Literary Insight.” HS 56 (2015): 103–13. Fritz, Volkmar. “Jahwe und El in den vorpriesterschriftlichen Geschichtswerken.” In “Wer ist wie Du, HERR, unter den Göttern?” Studien zur Theologie und Religionsgeschichte Israels für Otto Kaiser zum 70. Geburtstag, edited by Ingo Kottsiepter, Jürger van Oorschot, Diethard Römheld, and Harald Martin Wahl, 111–26. Göttingen: Vandenhoeck & Ruprecht, 1994. Fuhs, Hans F. Sehen und Schauen. Die Wurzel ḥzh im Alten Orient und im Alten Testemant. Ein Beitrag zum prophetischen Offenbarungsempfang. FB. Würzburg: Echter Verlag, 1978. Galil, Gershon. “The Boundaries of Aram-Damascus in the 9th–8th Centuries BCE.” In Studies in Historical Geography and Biblical Historiography Presented to Zecharia Kallai, edited by Gershon Galil and Moshe Weinfeld, 35–41. Leiden; Boston, Cologne: Brill, 2000. Gall, August Freiherr von. “Zusammensetzung und Herkunft der Bileam-Perikope in Num. 22–24.” In Festgruss Bernhard Stade zur Feier seiner 25jährigen Wirksamkeit als Professor, edited by Wilhelm Diehl, Richard Drescher, Karl Eger, August Freiherr von Gall, Erwin Preuschen, and Heinrich Weinel, 1–47. Giessen: Ricker, 1900. Galling, Kurt. Die Bücher der Chronik, Esra, Nehemia übersetzt und erklärt. ATD, vol. 12. Berlin: Evangelische Verlagsanstalt, 1958. Garbini, Giovanni. “L’iscrizione di Balaam Bar-Beor.” Hen 1 (1979): 166–88. García Martínez, Florentino. “Balaam in the Dead Sea Scrolls.” In The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity and Islam, edited by George H. van Kooten and Jacques van Ruiten. TBN, vol. 11, 71–82. Leiden; Boston: Brill, 2008. –. “Two Messianic Figures in the Qumran Texts.” In Current Research and Technological Developments on the Dead Sea Scrolls, edited by Donald W. Parry and Stephen D. Ricks, 14–40. Leiden: Brill, 1996. Gaß, Erasmus. “Bileam (AT).” Das wissenschaftliche Bibellexikon im Internet. http://www.bibelwissenschaft.de/stichwort/15381/. –. Menschliches Handeln und Sprechen im Horizont Gottes. Aufsätze zur biblischen Theologie. FAT. Tübingen: Mohr Siebeck, 2015. –. Die Moabiter – Geschichte und Kultur eines ostjordanischen Volkes im 1. Jahrtausend v.Chr. ADPV, vol. 38. Wiesbaden: Harrassowitz Verlag, 2009. Gelston, Anthony. The Twelve Minor Prophets. BHQ, vol. 13. Stuttgart: Deutsche Bibelgesellschaft, 2010.

Bibliography

373

Gerstenberger, Erhard S. “Boten, Engel, Hypostasen: Die Kommunikation Gottes mit den Menschen.” In Gott und Mensch im Dialog. Festschrift für Otto Kaiser zum 80. Geburtstag, edited by Markus Witte, 139–54. Berlin: de Gruyter, 2004. Gesenius, Wilhelm, R.D. Meyer, Herbert Donner, and Johannes Renz. Hebräisches und Aramäisches Handwörterbuch über das Alte Testament: Gesamtausgabe. 18. Berlin; Heidelberg: Springer, 2013. al-Ghul, Omar. “The Question of the Homeland of Balaam Again: The Contribution of the Arabic Sources.” WO 36 (2006): 94–103. Gnuse, Robert K. “Redefining the Elohist.” JBL 119, no. 2 (Summer 2000): 201–20. Görg, Manfred. “Šaddaj – Ehrenrettung einer Etymologie.” BN 16 (1981): 13–15. –. “Die ‘Heimat’ Bileams.” BN 1 (1976): 24–28. Graupner, Axel. Der Elohist: Gegenwart und Wirksamkeit des transzendenten Gottes in der Geschichte. WMANT, vol. 97. Neukirchen-Vluyn: Neukirchener Verlag, 2002. Gray, George Buchanan. A Critical and Exegetical Commentary on Numbers. Impression from 1986. ICC. Edinburgh: T&T Clark, 1903. Green, William Scott. “Introduction: Messiah in Judaism: Rethinking the Question.” In Judaisms and Their Messiahs at the Turn of the Christian Era, edited by Jacob Neusner, William Scott Green, and Ernest S. Frerichs, 1–13. Cambridge: University Press, 1987. Greene, John T. Balaam and his interpreters. A hermeneutical history of the Balaam traditions. BJS, vol. 244. Atlanta: Scholars Press, 1992. Greenfield, Jonas C. “Philological Observations on the Deir ʿAlla Inscription.” In The Balaam Text from Deir ʿAlla Re-Evaluated: Proceedings of the International Symposium Held at Leiden, 21–24 August 1989, ed. Jacob Hoftijzer, Gerrit van der Kooij, 109–1120. Leiden; New York; Copenhagen; Cologne: Brill, 1991. Greenspoon, Leonard J. Textual Studies in the Book of Joshua. HSM, vol. 28. Chico, CA: Scholars Press, 1983. Greßmann, Hugo. Mose und seine Zeit. Ein Kommentar zu den Mose-Sagen. FRLANT, vol. 18. Göttingen: Vandenhoeck & Ruprecht, 1913. Gross, Walter. Bileam: Literar- und formkritische Untersuchung der Prosa in Num 22–24. SANT. Munich: Kösel-Verlag, 1974. –. Richter. HThKAT. Freiburg im Breisgau; Basel; Vienna: Herder, 2009. Guyot, Gilmore H. “Balaam.” CBQ 3, no. 3 (July 1941): 235–42. –. “The Prophecy of Balaam.” CBQ 2, no. 4 (October 1940): 330–40. Gzella, Holger. “Deir ʿAllā.” In Encyclopedia of Hebrew Language and Linguistics, vol. 1, edited by Geoffrey Khan, 691–93. Leiden; Boston: Brill, 2013. Hackett, Jo Ann. The Balaam Text from Deir ‘Allā. HSM. Chico, Calif.: Scholars Press, 1980. –. “Response to Baruch Levine and André Lemaire.” In The Balaam Text from Deir ʿAlla Re-Evaluated: Proceedings of the International Symposium Held at Leiden, 21–24 August 1989, ed. Jacob Hoftijzer, Gerrit van der Kooij, 73–84. Leiden; New York; Copenhagen; Cologne: Brill, 1991. –. “Some Observations on the Balaam Tradition at Deir ʿAllā.” BA 49, no. 4 (December 1986): 216–22. Hagner, Donald A. Matthew 1–13. WBC, vol. 33A. Dallas: Word Books, 1999. Halpern, Baruch. “Dialect Distribution in Canaan and the Deir Alla Inscriptions.” In ‘Working with No Data’. Semitic and Egyptian Studies Presented to Thomas O. Lambdin, edited by David M. Golomb, 119–39. Winona Lake, IN: Eisenbrauns, 1987. Hayward, Robert. “Balaam’s Prophecies as Interpreted by Philo and the Aramaic Targums of the Pentateuch.” In New Heaven and New Earth. Prophecy and the Millennium. Essays

374

Bibliography

in Honour of Anthony Gelston, edited by Peter J. Harland and Robert Hayward, 19–36. Leiden; Boston; Cologne: Brill, 1999. Heltzer, Michael. “‘True’ and ‘False’ Prophecies [Hebrew].” EI 23 (1993): 74–77. Hepner, Gershon. “The Mockery of Kings and Prophets. The Balaam Narrative Contains an Implied Critique of Moses.” RB 118 (April 2011): 180–85. Herrmann, Wolfgang. “El ‫אל‬.” In Dictionary of Deities and Demons in the Bible. Second Extensively Revised Edition, edited by Karel van der Toorn, Bob Becking, and Pieter W. van der Horst, 274–80. Leiden; Boston; Cologne: Brill, 1999. Hertog, Cornelius G. den. “Jesus / Josue/ Das Buch Josua.” In Septuaginta Deutsch: Erläuterungen und Kommentare Band I. Genesis bis Makkabäer, edited by Martin Karrer and Wolfgang Kraus, 605–56. Stuttgart: Deutsche Bibelgesellschaft, 2011. –. “Studien zur griechischen Übersetzung des Buches Josua.” Giessen: Universität Giessen, 1996. Herzberg, Hans Wilhelm. Die Bücher Josua, Richter, Ruth übersetzt und erklärt. ATD, vol. 9. Göttingen: Vandenhoeck & Ruprecht, 1965. Hoftijzer, Jacob. “The Prophet Balaam in a 6th Century Aramaic Inscription.” BA 39, no. 1 (March 1976): 11–17. Hoftijzer, Jacob, and Gerrit van der Kooij, eds. Aramaic Texts from Deir ‘Alla. DMOA, vol. 19. Leiden: Brill, 1976. Hoftijzer, Jacob, and Karel Jongeling. Dictionary of the North-West Semitic Inscriptions. HdO. Erste Abteilung, Der Nahe und Mittlere Osten, vol. 1–2. Leiden; New York: E.J. Brill, 1995. Holzinger, Heinrich. Numeri. KHC, vol. 4. Tübingen; Leipzig: J.C.B. Mohr (Paul Siebeck), 1903. Horbury, William. “Monarchy and Messianism in the Greek Pentateuch.” In The Septuagint and Messianism, edited by Michael A. Knibb, 79–128. Leuven: University Press, 2006. Houtman, Alberdina, and Harry Sysling. “Balaam’s Fourth Oracle (Numbers 24:15–19) According to the Aramaic Targums.” In The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, edited by George H. van Kooten and Jacques van Ruiten. TBN, vol. 11, 189–211. Leiden: Brill, 2008. Hurowitz, Victor (Avigdor). “The Expression ûqsāmîm beyādām (Numbers 22:7) in Light of Divinatory Practices from Mari.” HS 33 (1992): 5–15. –. “What Did Balak’s Emissaries Bring to Balaam? [Hebrew].” EI 23 (1993): 83–86. Ibrahim, Moawiyah M., and Gerrit van der Kooij. “The Archaeology of Deir ʿAlla Phase IX.” In The Balaam Text from Deir ʿAlla Re-Evaluated: Proceedings of the International Symposium Held at Leiden, 21–24 August 1989, ed. Jacob Hoftijzer, Gerrit van der Kooij, 16–29. Leiden; New York; Copenhagen; Cologne: Brill, 1991. Jastram, Nathan. “The Text of 4QNumb.” In The Madrid Qumran Congress. Proceedings of the International Congress on the Dead Sea Scrolls. Madrid 18–21 March, 1991, edited by Julio Trebolle Barrera and Luis Vegas Montaner. STDJ, vol. XI, 177–98. Leiden; New York; Cologne: Brill, 1992. Jastrow, Marcus. Dictionary of the Targumim, the Talmud Babli and Yerushalmi, and the Midrashic Literature. Compiled by Marcus Jastrow. New York: Title Publishing, 1943. Jenks, Alan W. The Elohist and North Israelite Traditions. SBLMS, vol. 22. Missoula, MT: Scholars Press, 1977. Jenni, Ernst. “Faktitiv und Kausativ von ‫‘ אבד‬zugrunde gehen’.” In Hebräische Wortforschung. Festschrift zum 80. Geburtstag von Walter Baumgartner, edited by Benedikt Hartmann, Ernst Jenni, E.Y. Kutscher, Victor Maag, I.L. Seeligmann, and Rudolf Smend, 143–57. Leiden: Brill, 1967.

Bibliography

375

Kafafi, Zeidan, and Gerrit van der Kooij. “Tell Dēr ‘Allā During the Transition from Late Bronze Age to Iron Age.” ZDPV 129, no. 2 (2013): 121–31. Kaiser, Walter C., Jr. “Balaam Son of Beor in Light of Deir ʿAllā and Scripture. Saint or Soothsayer?” In “Go to the Land I Will Show You.” FS Dwight W. Young, edited by Joseph E. Coleson and Victor Matthews, 95–106. Winona Lake, IN: Eisenbrauns, 1996. Kallner-Amiran, David Horst. “A Revised Earthquake-Catalogue of Palestine.” IEJ 1 (1950–51): 223–46. –. “A Revised Earthquake-Catalogue of Palestine.” IEJ 2 (1952): 48–62. Kaufman, Stephen A. “Aramaic.” In The Semitic Languages, ed. Robert Hetzron, 114–30. New York: Routledge, 1997. Keil, Carl Friedrich. Leviticus, Numeri und Deuteronomium. Biblischer Commentar über die Bücher Mose’s. Leipzig: Dörffling & Franke, 1870. Kellenberger, Edgar. “Jahwes unerwarteter Widerstand gegen seinen Beauftragten. Erwägungen zur Episode von Bileams Eselin (Num 22,22–35).” ThZ 45 (1989): 69–72. Kepper, Martina, and Markus Witte. “Job / Das Buch Ijob / Hiob.” In Septuaginta Deutsch: Erläuterungen und Kommentare Band II. Psalmen bis Daniel, edited by Martin Karrer and Wolfgang Kraus, 2041–2126. Stuttgart: Deutsche Bibelgesellschaft, 2011. Kessler, Rainer. Micha. HThKAT. Freiburg im Breisgau; Basel; Vienna: Herder, 2000. Knauf, Ernst Axel. “El Šaddai.” BN 16 (1981): 20–26. –. Midian: Untersuchungen zur Geschichte Palästinas und Nordarabiens am Ende des 2. Jahrtausends v.Chr. ADPV. Wiesbaden: Harrassowitz, 1988. –. “Shadday ‫שׁדי‬.” In Dictionary of Deities and Demons in the Bible. Second Extensively Revised Edition, edited by Karel van der Toorn, Bob Becking, and Pieter W. van der Horst, 749–53. Leiden; Boston; Cologne: Brill, 1999. –. “War ‘Biblisch-Hebräisch’ eine Sprache? Empirische Gesichtspunkte zur linguistischen Annäherung an die Sprache der althebräischen Literatur.” ZAH 3 (1990): 11–23. –. “Welches ‘Israel’ Bileam sah.” In Wort und Stein. Studien zur Theologie und Archäologie. Festschrift für Udo Worschech, edited by Friedbert Ninow. Frankfurt am Main; Berlin; Bern; Brussels; New York; Oxford; Vienna: Peter Lang, 2003. Knibb, Michael A. “The Septuagint and Messianism. Problems and Issues.” In The Septuagint and Messianism, edited by Michael A. Knibb, 3–19. Leuven: University Press, 2006. Knierim, Rolf P. “The Book of Numbers.” In Die Hebräische Bibel und ihre zweifache Nachgeschichte: Festschrift für Rolf Rendtorff zum 65. Geburtstag, edited by Erhard Blum, Christian Macholz, and Ekkehard Stegemann, 155–63. Neukirchen-Vluyn: Neukirchener Verlag, 1990. Kooij, Arie van der. “Perspectives on the Study of the Septuagint. Who Are the Translators?” In Perspectives in the Study of the Old Testament and Early Judaism. A Symposium in Honour of Adam S. Van der Woude on the Occassion of His 70th Birthday, edited by Florentino García Martínez and Ed Noort. VTS, 214–29. Leiden; Boston; Cologne: Brill, 1998. Kooij, Gerrit van der, and Moawiyah M. Ibrahim, eds. Picking up the Threads… A Continuing Review of Excavations at Deir Alla, Jordan. Leiden: University of Leiden Archaeological Centre, 1989. Kratz, Reinhard G. The Composition of the Narrative Books of the Old Testament. Translated by John Bowden. London; New York: T & T Clark, 2005. –. Die Komposition der erzählenden Bücher des Alten Testaments. Göttingen: Vandenhoeck & Ruprecht, 2000. Kuenen, Abraham. “Bijdragen tot de critiek van Pentateuch en Jozua. X. Bileam.” ThT 18 (1884): 497–540.

376

Bibliography

Lange, Armin. Die Handschriften biblischer Bücher von Qumran und den anderen Fundorten. Vol. 1 of Handbuch der Textfunde vom Toten Meer. Tübingen: Mohr Siebeck, 2009. Largement, René. “Les oracles de Bileʿam et la mantique suméro-akkadienne.” In Mémorial Du Cinquantenaire. 1914–1964, École de langues orientales anciennes de l’Institut catholique de Paris, 37–50. Paris: Bound & Gay, 1964. Layton, Scott C. Archaic Features of Canaanite Personal Names in the Hebrew Bible. HSM. Atlanta: Scholars Press, 1990. –. “Whence Comes Balaam? Num 22,5 Revisited.” Bib 73, no. 1 (1992): 32–61. Lee, John A.L. A Lexical Study of the Septuagint Version of the Pentateuch. SCS. Chico, CA: Scholars Press, 1983. Leemans, Johan. “‘To Bless with a Mouth Bent on Cursing’: Patristic Interpretations of Balaam (Num 24:17).” In The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, edited by George H. van Kooten and Jacques van Ruiten. TBN, vol. 11, 287–99. Leiden: Brill, 2008. Leiter, Nechama. “Assimilation and Dissimilation: Techniques in the LXX of the Book of Balaam.” Text 12 (1985): 79–95. Lemaire, Andre. “Les inscriptions de Deir ʿAlla et al littérature aramée antique.” CRAI 129, no. 2 (1985): 270–85. –. “Les Inscriptions sur Plâtre de Deir ʿAlla Det Leur Signification Historique et Culturelle.” In The Balaam Text from Deir ʿAlla Re-Evaluated: Proceedings of the International Symposium Held at Leiden, 21–24 August 1989, ed. Jacob Hoftijzer, Gerrit van der Kooij, 33– 57. Leiden; New York; Copenhagen; Cologne: Brill, 1991. Levin, Christoph. Der Jahwist. FRLANT, vol. 157. Göttingen: Vandenhoeck & Ruprecht, 1993. –. “The Yahwist: The Earliest Editor in the Pentateuch.” JBL 126, no. 2 (Summer 2007): 209–30. Levine, Baruch A. Numbers 1–20. Anchor Bible, vol. 4. New York: Doubleday, 1993. –. Numbers 21–36. Anchor Bible, vol. 4A. New Haven: Yale, 2000. –. “The Plaster Inscriptions from Deir ʿAlla: General Interpretation.” In The Balaam Text from Deir ʿAlla Re-Evaluated: Proceedings of the International Symposium Held at Leiden, 21–24 August 1989, ed. Jacob Hoftijzer, Gerrit van der Kooij, 58–72. Leiden; New York; Copenhagen; Cologne: Brill, 1991. Lichtenberger, Hermann. Die Apokalypse. Theologischer Kommentar zum Neuen Testament, vol. 23. Stuttgart: Kohlhammer, 2014. Lindblom, Johannes. Prophecy in Ancient Israel. Oxford: Basil Blackwell, 1962. –. “Zur Frage des kanaanäischen Ursprungs des altisraelitischen Prophetismus.” In Von Ugarit nach Qumran: FS Otto Eissfeldt, edited by Johannes Hempel and Leonhard Rost, 89– 104. Berlin: Töpelmann, 1958. Lipiński, Edward. Studies in Aramaic Inscriptions and Onomastics II. OLA. Leuven: Peeters, 1994. –. “‫ באחים הימים‬Dans Les Textes Préexiliques.” VT 20, no. 4 (1970): 445–50. Liver, Y. “The Image of Balaam in Biblical Tradition [Hebrew].” EI 3 (April 1954): 97–100. Loewenstamm, Samuel E. “The Death of the Upright and the World to Come.” JJS 16 (1965): 183–86. Löhr, Max. “Bileam, Num 22,2–24,25.” AfO 4 (1927): 85–89. Lust, Johan. “Balaam, an Ammonite.” ETL 54 (1978): 60–61.

Bibliography

377

–. “The Greek Version of Balaam’s Third and Fourth Oracles. The ἄνθροπος in Num 24,7 and 17: Messianism and Lexicography.” In VIII Congress of the International Organization for Septuagint and Cognate Studies. Paris 1992, edited by Leonard Greenspoon and Olivier Munnich. SBLSCS, vol. 41, 233–52. Atlanta: Scholars Press, 1995. –. “The Greek Version of Balaam’s Third and Fourth Oracles. The ἄνθροπος in Num 24,7 and 17: Messianism and Lexicography.” In Messianism and the Septuagint: Collected Essays by Johan Lust, edited by Katrin Hauspie. BETL, vol. 178, 69–86. Leuven: Peeters, 2004. –. “Messianism and the Septuagint, with Special Emphasis on the Pentateuch.” In Theologische Probleme der Septuaginta und der hellenistischen Hermeneutik, edited by Henning Graf Reventlow. Veröffentlichungen der Wissenschaftlichen Gesellschaft für Theologie, vol. 11, 26–45. Gütersloh: Chr. Kaiser – Gütersloher, 1997. –. “Messianism and the Septuagint, with Special Emphasis on the Pentateuch.” In Messianism and the Septuagint: Collected Essays by Johan Lust, edited by Katrin Hauspie. BETL, vol. 178, 129–51. Leuven: Peeters, 2004. Lutzky, Harriet. “Ambivalence Toward Balaam.” VT 49, no. 3 (July 1999): 421–25. –. “Shadday as a Goddess Epithet.” VT 48, no. 1 (January 1998): 15–36. Malamat, Abraham. “ʿAmm Leḇāḏāḏ Yiškōn: A Report from Mari and an Oracle of Balaam.” JQR 76, no. 1 (July 1985): 47–50. Marcus, David. Ezra and Nehemiah. BHQ, vol. 20. Stuttgart: Deutsche Bibelgesellschaft, 2006. –. From Balaam to Jonah: Anti-Prophetic Satire in the Hebrew Bible. BJS, vol. 301. Atlanta: Scholars Press, 1995. Margaliot, Meshullam. “Literary, Historical and Religious Aspects of the Balaam Narrative, Numbers 22–24.” In Proceedings of the Tenth World Congress of Jewish Studies, Jerusalem, August 16–24, 1989. Division A: The Bible and Its World, 75–82. Jerusalem: World Union of Jewish Studies, 1990. –. “The Connection of the Balaam Narrative with the Pentateuch.” In Proceedings of the Sixth World Congress of Jewish Studies, Vol. 1, 279–90. Jerusalem: World Union of Jewish Studies, 1977. Margolis, Max L. Joshua 19:39–24:33. Vol. 5 of The Book of Joshua in Greek. According to the Critically Restored Text with an Apparatus Containing the Variants of the Principal Recensions and of the Individual Witnesses. With a preface by Emanuel Tov. Philadelphia: Annenberg Research Institute, 1992. –. “Τετροπωμένους Joshua 11:6.” JBL 33 (1914): 286–89. Marx, Alfred. “A propos de Nombres XXIV 19b.” VT 37, no. 1 (January 1987): 100–104. McCarter, P. Kyle. “The Balaam Texts from Deir ʿAllā: The First Combination.” BASOR 239 (1980): 49–60. –. “The Dialect of the Deir ʿAlla Texts.” In The Balaam Text from Deir ʿAlla Re-Evaluated: Proceedings of the International Symposium Held at Leiden, 21–24 August 1989, ed. Jacob Hoftijzer, Gerrit van der Kooij, 87–99. Leiden; New York; Copenhagen; Cologne: Brill, 1991. McCarthy, Carmel, ed. Deuteronomy. BHQ, vol. 5. Stuttgart: Deutsche Bibelgesellschaft, 2007. Merrill, Selah. “Balaam’s Birthplace.” The Hebrew Student 1, no. 4 (July 1882): 75. –. “Identification of Succoth and Penuel.” BSac 34, no. 136 (October 1877): 742–54. Meyer, Eduard. Die Israeliten und ihre Nachbarstämme. 1906. Darmstadt: Wissenschaftliche Buchgesellschaft, 1967.

378

Bibliography

Milgrom, Jacob. Numbers = [Ba-Midbar]: The Traditional Hebrew Text with the New JPS Translation. Commentary by Jacob Milgrom. The JPS Torah Commentary, vol. 4. Philadelphia: Jewish Publication Society, 1990. Millard, Alan R. “Epigraphic Notes, Aramaic and Hebrew.” PEQ 110–111 (1978–79): 23– 26. –. “In Praise of Ancient Scribes.” BA 45, no. 3 (Summer 1982): 143–53. Miller, J. Maxwell. “The Israelite Journey Through (Around) Moab and Moabite Toponymy.” JBL 108, no. 4 (1989): 577–95. Moatti-Fine, Jacqueline. Jésus (Josué). La Bible d’Alexandrie, vol. 6. Paris: Cerf, 1996. Moberly, R. Walter L. “On Learning to Be a True Prophet. The Story of Balaam and His Ass.” In New Heaven and New Earth. Prophecy and the Millennium. FS Anthony Gelstron, edited by Peter J. Harland and Robert Hayward, 1–17. Leiden: Brill, 1999. Moore, George F. A Critical and Exegetical Commentary on Judges. 1895. ICC. Edinburgh: T & T Clark, 1989. Moore, Michael S. “Another Look at Balaam.” RB 97, no. 3 (July 1990): 359–78. –. The Balaam Traditions: Their Character and Development. SBLDS. Atlanta: Scholars Press, 1990. Mounce, Robert H. The Book of Revelation. NICNT. Grand Rapids, MI: Eerdmans, 1998. Mowinckel, Sigmund. “Der Ursprung der Bilʿamsage.” ZAW 48 (1930): 233–71. Moyer, Clinton J. “Who is the Prophet, and Who the Ass? Role-Reversing Interludes and the Unity of the Balaam Narrative (Numbers 22–24).” JSOT 37 (2012): 167–83. Müller, Hans-Peter. “Die aramäische Inschrift von Deir ʿAlla und die älteren Bileamsprüche.” ZAW 94 (1982): 214–44. –. “Einige alttestamentliche Probleme zur aramäischen Inschrift von Dēr ‘Allā.” ZDPV 94 (1978): 56–67. –. “Die Sprache der Texte von Tell Deir ʿAllā im Kontext der nordwestsemitischen Sprachen mit einigen Erwägungen zum Zusammenhang der schwachen Verbklassen.” ZAH 4 (1991): 1–31. Müller, Reinhard, Juha Pakkala, and Robert Barend ter Haar Romney. Evidence of Editing. Atlanta: SBLPress, 2014. Müller, Reinhard, and Juha Pakkala, eds. Insights Into Editing in the Hebrew Bible and the Ancient Near East. CBET, vol. 84. Leuven; Paris; Bristol, CT: Peeters, 2017. Müller, Walter W. “Altsüdarabische Beiträge zum hebräischen Lexikon.” ZAW 75, no. 3 (1963): 304–16. Muraoka, Takamitsu. A Greek-English Lexicon of the Septuagint. Louvain; Walpole, MA: Peeters, 2009. –. A Greek-Hebrew / Aramaic Two-Way Index to the Septuagint. Louvain; Paris; Walpole, MA: Peeters, 2010. Murtonen, Aimo. “The Use and Meaning of the Words Lebårek and Beråkåh in the Old Testament.” VT 9, no. 2 (April 1959): 158–77. Naveh, Joseph. “The Date of the Deir ʿAllā Inscription in Aramaic Script.” IEJ 17 (1967): 256–58. Nelson, Richard D. Deuteronomy. A Commentary. OTL. Louisville; London: Westminster John Knox Press, 2002. –. Joshua. A Commentary. OTL. Louisville: Westminster John Knox Press, 1997. Neumann, Mathias. “(El) Šadday – A Plea for an Egyptian Derivation of the God and Its Name.” WO 46 (2016): 244–63.

Bibliography

379

Nicklas, Tobias. “Balaam and the Star of the Magi.” In The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, edited by George H. van Kooten and Jacques van Ruiten. TBN, vol. 11, 233–46. Leiden: Brill, 2008. Niehr, Herbert. Aramäischer Aḥiqar. JSHRZ. Gütersloh: Gütersloher Verlagshaus, 2007. Nihan, Christophe. From Priestly Torah to Pentateuch: A Study in the Composition of the Book of Leviticus. FAT, 2. Reihe, vol. 25. Tübingen: Mohr Siebeck, 2007. Nikolsky, Ronit. “Interpret Him as Much as You Want: Balaam in the Babylonian Talmud.” In The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, edited by George H. van Kooten and Jacques van Ruiten. TBN, vol. 11, 213–30. Leiden: Brill, 2008. Nogalski, James. Literary Precursors to the Book of the Twelve. BZAW, vol. 217. Berlin; New York: Walter de Gruyter, 1993. Nöldeke, Theodor. Ueber die Amalekiter und einige andere Nachbarvölker der Israeliten. Göttingen: Dietrich, 1864. –. Untersuchungen zur Kritik des Alten Testaments. Kiel: Schwers’sche Buchhandlung, 1869. Noort, Ed. “Balaam the Villain: The History of Reception of the Balaam Narrative in the Pentateuch and the Former Prophets.” In The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity and Islam, edited by George H. van Kooten and Jacques van Ruiten. TBN, vol. 11, 3–23. Leiden; Boston: Brill, 2008. Notarius, Tania. “Poetic Discourse and the Problem of Verbal Tenses in the Oracles of Balaam.” HS 49 (2008): 55–86. Noth, Martin. Überlieferungsgeschichte des Pentateuch. 3. Auflage, 1948. Darmstadt: Wissenschaftliche Buchgesellschaft, 1966. –. Das vierte Buch Mose: Numeri, übersetzt und erklärt. ATD, vol. 7. Göttingen: Vandenhoeck & Ruprecht, 1966. Olson, Dennis T. Numbers. IBC. Louisville, KY: John Knox Press, 1996. Oriental Institute of the University of Chicago. The Assyrian Dictionary Volume 1: A, Part 2. CAD. Chicago: Oriental Institute, 1968. –. The Assyrian Dictionary Volume 13: Q. CAD. Chicago: Oriental Institute, 1982. –. The Assyrian Dictionary Volume 21: Z. CAD. Chicago: Oriental Institute, 1960. Otto, Eckart. Deuteronomium 1–11, Teilband 1: 1,1–4,43. HThKAT. Freiburg; Basel, Vienna: Herder, 2012. –. Deuteronomium 12,1–23,15. HThKAT. Freiburg; Basel, Vienna: Herder, 2016. Paine, J.A. “Succoth and Penuel not Yet Identified.” BSac 35, no. 139 (July 1878): 481–98. Pákozdy, László M. “Az istennevek használata a Bileámperikópában.” Theologiai Szemle 14 (1938): 160–65. –. von. “Theologische Redaktionsarbeit in der Bileam-Perikope.” In Von Ugarit nach Qumran: FS Otto Eissfeldt, edited by Johannes Hempel and Leonhard Rost, 161–76. Berlin: Töpelmann, 1958. Pardee, Dennis. “The Linguistic Classification of the Deir ʿAlla Text Written on Plaster.” In The Balaam Text from Deir ʿAlla Re-Evaluated: Proceedings of the International Symposium Held at Leiden, 21–24 August 1989, ed. Jacob Hoftijzer, Gerrit van der Kooij, 100–105. Leiden; New York; Copenhagen; Cologne: Brill, 1991. Person, Raymond F., Jr., ed., Robert Rezetko. Empirical Models Challenging Biblical Criticism. AIL, vol. 25. Atlanta: SBL Press, 2016. Phillips, David Lee. The Samaritan Version of the Book of Numbers with Hebrew Variants: A Close Textual Study. Lewiston, NY: Edwin Mellen Press, 2014.

380

Bibliography

Pike, Dana M. “The Book of Numbers at Qumran: Texts and Context.” In Current Research and Technological Developments on the Dead Sea Scrolls, edited by Donald W. Parry and Stephen D. Ricks, 166–93. Leiden; New York; Cologne: Brill, 1996. Pola, Thomas. Die Ursprüngliche Priesterschrift. Beobachtungen Zur Literarkritik und Traditionsgeschichte von Pg. WMANT, vol. 70. Neukirchen-Vluyn: Neukirchener Verlag, 1995. Puech, Émile. “Balaʿam and Deir ʿAlla.” In The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity and Islam, edited by George H. van Kooten and Jacques van Ruiten. TBN, vol. 11, 25–47. Leiden; Boston: Brill, 2008. Rainey, Anson F. Canaanite in the Amarna Tablets: A Linguistic Analysis of the Mixed Dialect Used by Scribes from Canaan Volume 2: Morphosyntactic Analysis of the Verbal System. HdO. Leiden; New York; Cologne: Brill, 1996. Reicke, Bo. The Epistles of James, Peter, and Jude. AncB, vol. 37. New York: Doubleday, 1964. Rendsburg, Gary A. “Aramaic-Like Features in the Pentateuch.” HS 47 (2006): 163–76. Richardson, H. Neil. “Skt (Amos 9:11): ‘Booth’ or ‘Succoth’?” JBL 92, no. 3 (September 1973): 375–81. Ringgren, Helmer. “‫קבב‬.” In Theologisches Wörterbuch zum Alten Testament, vol. 6 ‫–עזז‬ ‫קום‬, ed. Heinz-Josef Fabry and Helmer Ringgren, 1138–39. Stuttgart; Berlin; Cologne: Kohlhammer, 1989. Ritt, Hubert. Offenbarung des Johannes. NEchtB, vol. 21. Würzburg: Echter-Verlag, 1986. Robker, Jonathan Miles. “The Balaam Narrative in the Pentateuch / Hexateuch / Enneateuch.” In Torah and the Book of Numbers, edited by Christian Frevel, Thomas Pola, and Aaron Schart, 334–66. Tübingen: Mohr Siebeck, 2013. –. “Bileam messianisch gelesen?” In Textgeschichte und Theologie, edited by Frank Ueberschaer, Jonathan Miles Robker, and Thomas Wagner. in press. Tübingen: Mohr Siebeck, 2018. –. The Jehu Revolution: A Royal Tradition of the Northern Kingdom and Its Ramifications. BZAW, vol. 435. Berlin: de Gruyter, 2012. –. Review of The Samaritan Version of the Book of Numbers with Hebrew Variants: A Close Textual Study. Antiguo Oriente 12 (2014): 234–40. Rofé, Alexander. “The Book of Balaam” (Numbers 22:2–24:25). A Study in Methods of Criticism and the History of Biblical Literature and Religion. With an Appendix: Balaam in the Deir ʿAlla Inscription. JBS. Jerusalem: Simor Ltd., 1979. Rose, Martin. 5. Mose 12–25. Einführung und Gesetz. ZBK. Zürich: Theologischer Verlag Zürich, 1994. Rösel, Martin, and Christine Schlund. “Arithmoi / Numeri / Das vierte Buch Mose.” In Septuaginta Deutsch: Erläuterungen und Kommentare Band I. Genesis bis Makkabäer, edited by Martin Karrer and Wolfgang Kraus, 431–522. Stuttgart: Deutsche Bibelgesellschaft, 2011. Rösel, Martin. “Jakob, Bileam und der Messias. Messianische Erwartungen in Gen 49 und Num 22–24.” In The Septuagint and Messianism, edited by Michael A. Knibb, 151–75. Leuven: University Press, 2006. –. “Die Septuaginta und der Kult. Interpretationen und Aktualisierungen im Buch Numeri.” In La double transmission du texte biblique. Etudes d’histoire du texte offertes en hommage à Adrian Schenker, edited by Yohanan Goldman and Christoph Uehlinger. OBO, 25–40. Göttingen: Vandenhoeck & Ruprecht, 2001. –. “Die Textüberlieferung des Buches Numeri am Beispiel der Bileamerzählung.” In Sôfer Mahîr. Essays in Honour of Adrian Schenker Offered by Editors of Biblia Hebraica

Bibliography

381

Quinta, edited by Yohanan A. P. Goldman, Arie van der Kooij, and Richard D. Weis, 207–26. Leider; Boston: Brill, 2006. –. “Wie einer vom Propheten zum Veführer wurde. Tradition und Rezeption der Bileamgestalt.” Bib 80 (1999): 506–24. Rost, Leonhard. “Fragen um Bileam.” In Beiträge zur Alttestamentlichen Theologie: Festschrift für Walther Zimmerli zum 70. Geburtstag, edited by Herbert Donner, Robert Hanhart and Rudolf Smend, 377–87. Göttingen: Vandenhoeck & Ruprecht, 1977. Rouillard, Hedwige. “L’anesse de Balaam: Analyse littéraire de Nomb., XXII, 21–35 (fin).” RB 87, no. 2 (April 1980): 211–41. –. “L’anesse de Balaam: Analyse littéraire de Nomb., XXII, 21–35.” RB 87, no. 1 (January 1980): 5–37. –. La péricope de Balaam: la prose et les “oracles”. EBib. Paris: Gabalda, 1985. Rudolph, Wilhelm. Der “Elohist” von Exodus bis Josua. BZAW, vol. 68. Berlin: de Gruyter, 1938. –. “Zum Text des Buches Numeri.” ZAW 52 (1934): 113–20. Ruppert, Lothar. “‫קסם‬.” In Theologisches Wörterbuch zum Alten Testament, vol. 7 ‫שׁכן–קוץ‬, ed. Heinz-Josef Fabry and Helmer Ringgren, 78–84. Stuttgart; Berlin; Cologne: Kohlhammer, 1990. Safren, Jonathan D. “Balaam and Abraham.” VT 38, no. 1 (1988): 105–13. Sals, Ulrike. “Bileam – der lächerliche Falschprophet? Eine Widerlegung.” In Der Bibelkanon in der Bibelauslegung. Methodenreflexionen und Beispielexegesen, edited by Egbert Ballhorn and Georg Steins, 163–88. Stuttgart: Kohlhammer, 2007. Sander, Ruth, and Kerstin Mayerhofer. Retrograde Hebrew and Aramaic Dictionary. Journal of Ancient Judaism Supplements, vol. 1. Göttingen: Vandenhoeck & Ruprecht, 2010. Sasson, Victor. “The Book of the Oracular Visions of Balaam from Deir ʿAlla.” UF 17 (1986): 283–309. –. “Two Unrecognized Terms in the Plaster Texts from Deir ʿAlla.” PEQ 117, no. 2 (July 1985): 102–3. Savran, George W. “Beastly Speech: Intertextuality, Balaam’s Ass and the Garden of Eden.” JSOT 64 (1994): 33–55. Sayce, Archibald Henry. Assyria. Its Princes, Priests and People. London: The Religious Tract Society, 1885. –. “Balaam’s Prophecy (Numbers 24:17–24) and the God Sheth.” Hebraica 4 (1887): 1–6. Schaper, Joachim. Eschatology in the Greek Psalter. WUNT 2. Reihe, vol. 76. Tübingen: Mohr Siebeck, 1995. Scharbert, Josef “‫ארר‬.” In Theologisches Wörterbuch zum Alten Testament, vol. 1 ‫גלה–אב‬, ed. Gerhard Johannes Botterweck and Helmer Ringgren, 437–51. Stuttgart: Kohlhammer, 1973. –. “‫קלל‬.” In Theologisches Wörterbuch zum Alten Testament, vol. 7 ‫שׁכן–קוץ‬, ed. Heinz-Josef Fabry and Helmer Ringgren, 40–49. Stuttgart; Berlin; Cologne: Kohlhammer, 1990. Schart, Aaron. Die Entstehung des Zwölfprophetenbuchs. Neubearbeitung von Amos im Rahmen schriftenübergreifender Redaktionsprozesse. BZAW, vol. 260. Berlin; New York: de Gruyter, 1998. Schenker, Adrian. “Der Ursprung des massoretischen Textes im Licht der literarischen Varianten im Bibeltext.” Textus 23 (2007): 51–67. Schmidt, Ludwig. “Die alttestamentliche Bileamüberlieferung.” BZ 23 (1979): 235–61. –. “Bileam: Vom Seher zum Propheten Jahwes. Die literarischen Schichten der Bileam-Perikope (Num 22–24).” In Gott und Mensch im Dialog. Festschrift für Otto Kaiser zum 80. Geburtstag, edited by Markus Witte, 333–51. Berlin: De Gruyter, 2004.

382

Bibliography

–. Das 4. Buch Mose. Numeri. ATD, vol. 7/2. Göttingen: Vandenhoeck & Ruprecht, 2004. Schmitt, Hans-Christoph. “Der heidnische Mantiker als eschatologischer Jahweprophet. Zum Verständnis Bileams in der Endgestalt von Num 22–24.” In “Wer ist wie Du, HERR, unter den Göttern?” Studien zur Theologie und Religionsgeschichte Israels für Otto Kaiser zum 70. Geburtstag, edited by Ingo Kottsiepter, Jürger van Oorschot, Diethard Römheld, and Harald Martin Wahl, 180–98. Göttingen: Vandenhoeck & Ruprecht, 1994. Schmitt, Rüdiger. Magie im Alten Testament. AOAT, vol. 313. Münster: Ugarit-Verlag, 2004. –. Mantik im Alten Testament. AOAT, vol. 411. Münster: Ugarit-Verlag, 2014. Schnackenburg, Rudolf. Matthäusevangelium, Band 1. NEchtB, vol. 1.1. Würzburg: Echter, 1985. Schnutenhaus, Frank. “Das Kommen und Erscheinen Gottes im Alten Testament.” ZAW 76 (1964): 1–22. Schüle, Andreas. Israels Sohn – Jahwes Prophet: ein Versuch zum Verhältnis von kanonischer Theologie und Religionsgeschichte anhand der Bileam-Perikope (Num 22–24). Altes Testament und Moderne, vol. 17. Münster: Lit, 2001. Schultz, Brian. “The Kittim of Assyria.” RevQ 23, no. 1 (June 2007): 63–77. Schwartz, Baruch J. “Introduction: The Strata of the Priestly Writings and the Revised Relative Dating of P and H.” In The Strata of the Priestly Writings: Contemporary Debate and Future Directions, Sarah Shectman and Joel S. Baden, 1–12. Zürich: Theologischer Verlag Zürich, 2009. Seebass, Horst. “Einige vertrauenswürdige Nachrichten zu Israel Anfängen: Zu den Söhnen Hobabs, Sichon und Bileam im Buch Numeri.” JBL 113 (1994): 577–85. –. Numeri 22,2–36,13. BKAT, vol. 4.3. Neukirchen-Vluyn: Neukirchener, 2007. –. “Zur literarischen Gestalt der Bileam-Perikope.” ZAW 107 (1995): 409–19. Segert, Stanislav. Altaramäische Grammatik mit Bibliographie, Chrestomathie und Glossar. 4th ed. Leipzig: Verlag Enzyklopädie, 1990. Seow, Choon Leong. “West Semitic Sources.” In Prophets and Prophecy in the Ancient Near East, by Martti Nissinen, contributor Choon Leong Seow and Robert K. Ritner. Writings from the Ancient World, 201–18. Atlanta: SBL Press, 2003. Seybold, Klaus. “Das Herrscherbild des Bileamorakels Num. 24,15–19.” TZ 29 (1973): 1– 19. Shemesh, Aharon. “A Note on ‘4Q339’ ‘List of False Prophets’.” RevQ 20, no. 2 (December 2001): 319–20. Simpson, B. F. “The Story of Balaam Reconsidered.” The Old Testament Student 5, no. 3 (November 1885): 125–28. Ska, Jean Louis. “Old and New in the Book of Numbers.” Bib 95, no. 1 (2014): 102–16. –. “Le récit sacerdotal: une ‘histoire sans fin’?” In The Books of Leviticus and Numbers, ed. Thomas Römer. BETL, vol. 215, 631–53. Leuven: Peeters, 2008. Skinner, John. A Critical and Exegetical Commentary on Genesis. Second Edition. 1930. ICC. Edinburgh: T. & T. Clark, 1980. Smend, Rudolf Sr. Die Erzählung des Hexateuch auf ihre Quellen untersucht. Göttingen: Georg Reimer, 1912. Soggin, J. Alberto. Joshua. A Commentary. OTL. London: SCM Press Ltd., 1972. Speiser, Ephraim Avigdor. Genesis. AB, vol. 1. Garden City, NY: Doubleday, 1982. Stackert, Jeffrey. “Distinguishing Innerbiblical Exegesis from Pentateuchal Redaction: Leviticus 26 as a Test Case.” In The Pentateuch. International Perspectives on Current Research, edited by Thomas B. Dozeman, Konrad Schmid, and Baruch J. Schwartz. FAT, vol. 78, 369–86. Tübingen: Mohr Siebeck, 2011.

Bibliography

383

Stebbins, Rufus Phineas. “The Story of Balaam.” The Old Testament Student 4, no. 9 (May 1885): 385–95. Steuernagel, Carl. Lehrbuch der Einleitung in das Alte Testament mit einem Anhang über die Apokryphen und Pseudepigraphen. Sammlung theologischer Lehrbücher. Tübingen: Mohr Siebeck, 1912. Strawn, Brent A. What is Stronger Than a Lion? Leonine Image and Metaphor in the Hebrew Bible and Ancient Near East. OBO, vol. 212. Fribourg; Göttingen: Academic Press and Vandenhoeck & Ruprecht, 2005. Sturdy, John. Numbers. Cambridge Bible (NEB). New York: Cambridge University Press, 1976. Sukenik, Eliezer Lipa, ed. The Dead Sea Scrolls of the Hebrew University. Jerusalem: Magnes Press, 1955. Sutcliffe, Edmund F. “De Unitate Litteraria Num XXII.” Bib 7, no. 1 (1926): 3–39. –. “A Note on Numbers XXII.” Bib 18, no. 4 (1937): 439–42. Tigchelaar, Eibert. “Balaam and Enoch.” In The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, edited by George H. van Kooten and Jacques van Ruiten. TBN, vol. 11, 87–99. Leiden: Brill, 2008. –. “A Qumran Cave 2 Fragment Preserving Part of Numbers 23:5–7[8] (2Q29 1).” In The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, edited by George H. van Kooten and Jacques van Ruiten. TBN, vol. 11, 83–86. Leiden: Brill, 2008. Timm, Stefan. Moab zwischen den Mächten. Studien zu historischen Denkmälern und Texten. ÄAT, vol. 17. Wiesbaden: Harrassowitz, 1989. Torczyner, Harry. “Zu ‫ חנשתי‬Gen. 30,27.” OLZ 20, no. 1 (1917): 10–12. Tosato, Angelo. “The Literary Structure of the First Two Poems of Balaam (Num. XXIII 7– 10, 18–24).” VT 29, no. 1 (January 1979): 98–106. Tov, Emanuel. The Text-Critical Use of the Septuagint in Biblical Research. Third ed. Winona Lake, IN: Eisenbrauns, 2015. –. Textual Criticism of the Hebrew Bible. Third ed. Minneapolis, MN: Fortress, 2012. Tur-Sinai, Naftali Herz. “El Shaddai [Hebrew].” EI 3 (1954): 39–41. Ulrich, Eugene. The Biblical Qumran Scrolls: Transcriptions and Textual Variants. Volume 1: Genesis–Kings. Leiden; Boston: Brill, 2013. Utzschneider, Helmut. “Michaias / Micha.” In Septuaginta Deutsch: Erläuterungen und Kommentare Band II. Psalmen bis Daniel, edited by Martin Karrer and Wolfgang Kraus, 2362–80. Stuttgart: Deutsche Bibelgesellschaft, 2011. –. Micha. ZBK, vol. 24.1. Zürich: Theologischer Verlag Zürich, 2005. Utzschneider, Helmut, and Wolfgang Oswald, Philip Sumpter, trans. Exodus 1–15. IECOT. Stuttgart: Kohlhammer, 2015. Utzschneider, Helmut, and Wolfgang Oswald. Exodus 1–15. IEKAT. Stuttgart: Kohlhammer, 2013. van der Louw, Theo. Transformations in the Septuagint. Towards an Interaction of Septuagint Studies and Translation Studies. CBET, vol. 47. Leuven; Paris; Dudley, MA: Peeters, 2007. van Henten, Jan Willem. “Balaam in Revelation 2:14.” In The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, edited by George H. van Kooten and Jacques van Ruiten. TBN, vol. 11, 247–63. Leiden: Brill, 2008. van Kooten, George H. “Balaam as the Sophist par Excellence in Philo of Alexandria: Philo’s Projection of an Urgent Contemporary Debate Onto Moses’ Pentateuchal Narratives.” In The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and

384

Bibliography

Islam, edited by George H. van Kooten and Jacques van Ruiten. TBN, vol. 11, 131–61. Leiden: Brill, 2008. van Ruiten, Jacques T.A.G.M. “The Rewriting of Numbers 22–24 in Pseudo-Philo, Liber Antiquitatum Biblicarum 18.” In The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam, edited by George H. van Kooten and Jacques van Ruiten. TBN, vol. 11, 101–30. Leiden: Brill, 2008. Van Seters, John. The Life of Moses. The Yahwist as Historian in Exodus-Numbers. Louisville, KY: Westminster/John Knox Press, 1994. Vaulx, Jules de. Les Nombres. Source bibliques. Paris: Gabalda, 1972. Vera Chamaza, Galo W. Die Rolle Moabs in der neuassyrischen Expansionspolitik. AOAT, vol. 321. Münster: Ugarit-Verlag, 2005. Vermes, Geza. Scripture and Tradition in Judaism. Haggadic Studies. Second, Revised Edition. StPB, vol. 4. Leiden: Brill, 1973. Vetter, Dieter. Seherspruch und Segensschilderung. Ausdrucksabsichten und sprachliche Verwirklichungen in den Bileam-Sprüchen von Numeri 23 und 24. Calwer Theologische Monographien, vol. 4. Stuttgart: Calwer, 1974. Vilders, Monique E. “The Stratigraphy and the Pottery of Phase M at Deir ʿAlla and the Date of the Plaster Text.” Levant 24 (1992): 187–200. Volz, Paul. Review of Zusammensetzung und Herkunft der Bileam-Perikope in Num. 22– 24. TLZ 26, no. 14 (1901): 383–85. von Rad, Gerhard. Das erste Buch Mose. Genesis, übersetzt und erklärt. ATD. Göttingen: Vandenhoeck & Ruprecht, 1967. –. Das fünfte Buch Mose. Deuteronomium, übersetzt und erklärt. ATD. Göttingen: Vandenhoeck & Ruprecht, 1968. Vosté, Jacques-M. “Les Oracles de Balaam d’après Mar Išoʿdad de Merw (c. 850).” Bib 29, no. 3 (1948): 169–94. Vuilleleumier, René. “Bileam Zwischen Bibel und Deir ʿAllā.” ThZ 52 (1996): 150–63. Wagner, Siegfried. “Offenbarungsphänomenologische Elemente in den Bileam-Geschichten von Numeri 22–24.” In Ausgewählte Aufsätze zum Alten Testament, edited by Dietmar Mathias, 83–108. Berlin: de Gruyter, 1996. Way, Kenneth C. Donkeys in the Biblical World. Ceremony and Symbol. HACL. Winona Lake, IN: Eisenbrauns, 2011. Weinfeld, Moshe. “The Balaam Oracle in the Deir ‘Alla Inscription.” Shnaton 5–6 (1981– 82): 141–47. Weippert, Manfred. “The Balaam Text from Deir ‘Allā and the Study of the Old Testament.” In The Balaam Text from Deir ‘Allā Re-Evaluated: Proceedings of the International Symposium Held at Leiden, 21–24 August, 1989, edited by Jacob Hoftijzer and Gerrit van der Kooij, 151–84. Leiden: Brill, 1991. –. “Der ‘Bileam’-Text von Tell Dēr ‘Allā und das Alte Testament.” In Jahwe und die anderen Götter, Manfred Weippert, 163–88. Tübingen: Mohr Siebeck, 1997. –. “Erwägungen zur Etymologie des Gottesnamens ʾEl Schaddaj.” ZDMG 111 (1961): 42– 62. Weippert, Manfred, and Helga Weippert. “Die ‘Bileam’-Inschrift von Tell Dēr ‘Allā.” ZDPV 98 (1982): 77–103. –. “Die ‘Bileam’-Inschrift von Tell Dēr ‘Allā.” In Jahwe und die anderen Götter, Manfred Weippert. FAT, vol. 18, 131–61. Tübingen: Mohr Siebeck, 1997. Weise, Uwe. Vom Segnen Israels: eine textpragmatische Untersuchung der Bileam-Erzählung Num 22–24. Textpragmatische Studien zur Literatur- und Kulturgeschichte der Hebräischen Bibel. Gütersloh: Gütersloher Verlagshaus, 2006.

Bibliography

385

Weiser, Artur. Das Buch der zwölf kleinen Propheten I: Die Propheten Hosea, Joel, Amos, Obadja Jona, Micha übersetzt und erklärt. ATD, vol. 24. Göttingen: Vandenhoeck & Ruprecht, 1967. Wellhausen, Julius. Die Composition des Hexateuchs und der historischen Bücher des Alten Testaments. Dritte Auflage. Berlin: Georg Reimer, 1899. Wenning, Robert, and Erich Zenger. “Heiligtum ohne Stadt – Stadt ohne Heiligtum? Anmerkungen zum archäologischen Befund des Tell Dēr ʿAllā.” ZAH 4 (1991): 171–93. Wevers, John William. “The Balaam Narrative According to the Septuagint.” In Lectures et Relectures de la Bible. Festschrift Pierre-Maurice Bogaert, edited by Jean-Marie Auwers, André Wénin, 133–44. Leuven: Peeters, 1999. –. “The Göttingen Pentateuch: Some Post-Partem Reflections.” In VII Congress of the International Organization for Septuagint and Cognate Studies, Leuven 1989, edited by Claude E. Cox, 51–60. Atlanta, GA: Scholars Press, 1991. –. Notes on the Greek Text of Deuteronomy. SBLSCS. Atlanta, Ga.: Scholars Press, 1995. –. Notes on the Greek Text of Numbers. SBLSCS. Atlanta, Ga.: Scholars Press, 1988. Wiklander, Bertil. “‫זעם‬.” In Theologisches Wörterbuch zum Alten Testament, vol. 2 ‫–גלולים‬ ‫חמץ‬, ed. Gerhard Johannes Botterweck and Helmer Ringgren, 621–26. Stuttgart: Kohlhammer, 1977. Wilson, Robert R. Prophecy and Society in Ancient Israel. Philadelphia: Fortress Press, 1980. Witte, Markus. “Der Segen Bileams – eine redaktionsgeschichtliche Problemanzeige zum ‘Jahwisten’ in Num 22–24.” In Abschied vom Jahwisten. Die Komposition des Hexateuch in der jüngsten Diskussion, edited by Jan Christian Gertz, Konrad Schmid, and Markus Witte. BZAW, vol. 315, 191–213. Berlin; New York: de Gruyter, 2002. Wöhrle, Jakob. Die frühen Sammlungen des Zwölfprophetenbuches. Entstehung und Komposition. BZAW, vol. 360. Berlin; New York: de Gruyter, 2006. Wolff, Hans Walter. Dodekapropheton 4. Micha. BKAT, vol. XIV/4. Neukirchen-Vluyn: Neukirchener Verlag, 1982. Wolters, Al. “The Balaamites of Deir ʿAlla as Aramean Deportees.” HUCA 59 (1988): 101– 13. Worschech, Udo. Die Beziehungen Moabs zu Israel und Ägypten in der Eisenzeit. ÄAT, vol. 18. Wiesbaden: Harrassowitz, 1990. Yahuda, Abraham Shalom. “The Name of Balaam’s Homeland.” JBL 64 (1945): 547–51. Younger, K. Lawson, Jr. A Political History of the Arameans. From Their Origins to the End of Their Polities. ABS, vol. 13. Atlanta: SBL Press, 2016. Zadok, Ran. On West Semites in Babylonia During the Chaldean and Achaemenian Periods. Jerusalem: H. J. & Z. Wanaarta & Tel Aviv University, 1978. Zevit, Ziony. “Scratched Silver and Painted Walls: Can We Date Biblical Texts Archaeologically?” HS 48 (2007): 23–37. Ziegert, Carsten. Diaspora als Wüstenzeit. Übersetzungswissenschaftliche und theologische Aspekte des griechischen Numeribuches. BZAW, vol. 480. Berlin: de Gruyter, 2015. Zobel, Hans-Jürgen. “Bileam-Lieder und Bileam-Erzählung.” In Die hebräische Bibel und ihre zweifache Nachgeschichte: Festschrift für Rolf Rendtorff zum 65. Geburtstag, edited by Erhard Blum, Christian Macholz, and Ekkehard Stegemann, 141–54. NeukirchenVluyn: Neukirchener Verlag, 1990.

 

Source Index Hebrew Bible / Old Testament Genesis 1 1:26 1:28 2–3 4 5 6 6:6 8:20 10:1–4 10:1 10:21–22 11:27 12:1–3 12:3 12:7–8 13:4 13:16 13:18 14:18–20 14:22 15:5 17:1 19 19:30–38 19:37 22:9 24:25–26 25:16 26:25 27:29 27:42 28:3 28:14 32:32 32:36

201 95 95 201 202 203 261 82, 120 122, 155 202 202 202 199 63, 198 60, 141 122 122 80, 120 122 356 356 161 353 261 224, 227, 312 199 122 199 210 122 141, 198 161 353 80 90 308

35:11 35:37 36 36:12 36:16 36:32–33 37:35 38:27–30 40–41 40:17 41:50–52 43:14 48:3 49:10 49:25

353 122 202 179, 201 179 207–209 161 201 320 356 201 353 353 92–93, 122, 124 353

Exodus 1:9 1:12 3:6 3:18 6:3 10:5 13:20 17:1 17:8–16 18:8 19:2 19:8 20:24–25 32:5–6 33:8

139 135, 137–139, 144 161 80 353 140 134 134 202 82 134 100 122 155 161

Leviticus 25:43 25:46 25:53

82, 95 82, 95 82, 95

388

Source Index

26:19

95

Numbers 2:1 4:18 5:2–3 7:32 9:17–18 9:18 9:22 10:12 10:19 10:29–32 10:36 11 12:4 12:8 12:10 13:28 13:30 13:31 14:9 14:12 14:15 14:24 14:30 16:33 17:27 18:2 18:20 20–21 20:14–21 20:14 20:18–20 20:19 20:22 21 21:1–3 21:4–9 21:4 21:10–13 21:11 21:13 21:16 21:17–18 21:18 21:20 21:21–31 21:29–30

353 92, 177 178 208 178 83 178 133, 178 353 198 83 198 244 161 213 210 94 210 213 210 210 94 178 100 100 92, 177–178 94 242 175, 198 82 198 198 133 124, 172, 197 198 136 133 133–134 134 134 136 198 210 134 198 100

21:21–22:1 21:21–35 21:21–32 21:21–31 21:23–35 21:32–35 21:32–22:1 21:32 21:33–35 21:35 22:1–5 22:1–4 22:1–3 22:1–2 22:1

22:2–4 22:2

22:3–4 22:3

22:4–6 22:4–5 22:4

22:5–6 22:5

22:6

67 133, 184 229 134–135, 177 46 133–134 135 132–133, 135, 138 133, 135, 164, 178 94, 170, 201 11 16, 138 177 40, 67 21, 27, 35, 46, 51, 66–67, 131, 133– 139, 155, 171, 173– 174, 177–178, 184 46 12, 15–16, 27, 29, 39–40, 47, 59, 66– 67, 132, 134–135, 137–138, 141, 143, 161, 174, 177 40, 139, 155, 171 12, 25, 46, 59, 67, 131–132, 134–138, 141, 143–144, 173– 174, 177–178, 184, 197 168 166 12, 16, 27–28, 39– 40, 46–47, 59, 118, 124, 132, 134–138, 140, 142–143, 155, 174, 177, 184, 197, 214 73–74, 137 12, 28, 51–52, 59, 74–75, 118, 120, 122, 132, 136–140, 142–144, 156, 162, 171, 174, 184, 197, 220–221, 227–228, 239, 260, 309–311, 317, 320–321 120, 132, 136–137, 141–143, 160, 162,

Source Index

22:7–8 22:7

22:8–9 22:8 22:9–12 22:9 22:10 22:11–1 22:11

22:12

22:13 22:15 22:16–18 22:16 22:17 22:18

22:19 22:20

22:21

22:22–35

22:22 22:25 22:26 22:28–30 22:28 22:29 22:31

165, 174–175, 183, 198, 321–324 129 40, 120, 132, 136, 142–144, 155–156, 171, 174, 177, 214, 260, 324–326 350 347, 349 349 122, 349 76, 118, 122, 124 136, 144 12, 76, 122, 124, 132, 137, 140–142, 144, 160, 174, 197, 239, 326–328 60, 132, 141–142, 144, 160, 165, 174, 222, 321–324 118, 142, 144 16–17 142 143, 301 136, 142, 160, 326– 328 118, 120, 122–124, 142, 148–149, 260, 347 16–17 16, 79, 122, 132, 148–149, 155–156, 174, 348–349 14, 16, 28, 31, 39, 67, 79, 144–149, 174–175, 347 11–14, 16–17, 27– 29, 39, 46–47, 49, 56, 62, 66–68, 79, 132, 145–149, 174– 175, 259, 291, 347 16, 132, 146, 347– 348 16–17 16–17 147 348 124 122, 348

22:32 22:33 22:35 22:36–23:26 22:36–41 22:36 22:37 22:38 22:39 22:41 23:1–2 23:1 23:2 23:3–5 23:3 23:4–5 23:4 23:5 23:6 23:7–10 23:7–8 23:7–8 23:7

23:8–9 23:8 23:9 23:10

23:11 23:12 23:13–14 23:13 23:14

389 77–78, 120, 122, 124 118, 124 14, 39, 120, 146– 149 149 149–150 30, 132, 150 149–150 142, 148–149 120, 122, 149–150 122, 124, 136, 153 132, 153–155, 174 120, 122, 155 118, 123–124, 151, 153–155, 328–330 151 79, 101, 118, 122, 124–125, 153 122 120, 132, 151, 174, 328–330 142, 148–149, 163 120, 122, 153, 157 159–160, 175, 221 330–331 168 15, 139, 142, 156– 157, 160, 163, 168, 170, 174, 182, 257, 304, 312–315, 317– 318, 321–324 170 142, 163, 165, 326– 328, 351–352 89–91, 124, 136, 168, 182, 210 62, 80–81, 96, 118, 120, 122, 124, 141, 158, 160, 163, 168, 170, 174, 182 63, 142, 157, 229, 326–328 63, 120, 142, 144, 148, 149 132, 153 63, 118, 142, 326– 328 122, 153–154, 174

390 23:15–16 23:15 23:16 23:17 23:18–24 23:18–20 23:18

23:19–20 23:19 23:20 23:21–22 23:21

23:22–23 23:22

23:23 23:24 23:25–26 23:25 23:26 23:27–24:24 23:27–28 23:27 23:29–30 23:29 23:30 24 24:1–2 24:1 24:2

24:3–9 24:3–4

Source Index 151 79, 118, 120, 122, 124, 153 122, 142, 148–149 118, 120, 153, 157, 347 160–162 168, 170, 175 81–82, 122, 124, 156–157, 160, 168, 182 162 85, 118, 120, 161, 163, 351 82, 118, 125, 161, 235, 239, 315 352 63, 82–83, 118, 122, 124, 161, 168, 170, 174–175, 351– 352 174, 351 12, 118, 121, 124, 132, 156, 161, 351– 352 161, 168, 324–326, 331–334, 352 136, 161, 168, 170, 175 132, 158 11, 142, 157–158, 229, 326–328 39, 148–149, 158 32 153 142, 326–328 153, 155, 174 132, 155 153–155 261 132, 135 22, 124, 174, 184, 199, 302, 331–334 92, 120, 147, 157, 173, 177–178, 200, 334–336 162–163 132

24:3 24:4

24:5–6 24:5 24:6–7 24:6 24:7–9 24:7

24:8

24:9

24:10–14 24:10 24:11–13 24:11–12 24:11 24:13

24:14–24 24:14–19 24:14

24:15–24 24:15–19 24:15 24:16

157, 162, 168, 170, 174, 336 42, 53, 118–119, 122, 124, 162, 168, 170, 174, 182, 337– 339, 351, 353–356 48 42, 162–163, 165, 168, 170, 174–175 125 42, 50, 83, 119, 123–124, 162, 170 168, 170, 175 48, 60, 64, 83–86, 88, 93, 97, 101, 119, 121, 123–125, 162, 167, 175, 182, 210 12, 119, 121, 124, 132, 156, 158, 161– 162, 168, 171, 174, 177, 182, 197, 201, 210, 351–352 41, 162, 167, 170, 174–175, 183, 198, 311, 315, 320–324 173 142, 157, 174, 229, 326–328 144 158 11, 222, 347 49, 119–120, 123– 124, 148–149, 260, 347 46, 68, 139, 164, 167, 174–175, 357 144, 164–165, 177– 178 17, 39–40, 49, 119, 122, 124, 132, 136, 139, 149, 157–158, 174, 246 163, 165, 167, 354 165–166, 170, 175 163–164, 168, 336 53, 119, 122, 124, 163–164, 337–340, 351–356

Source Index 24:17–19 24:17

24:18–19 24:18 24:19–24 24:19

24:20–23 24:20–22 24:20–21 24:20

24:21–24 24:21–22 24:21 24:22 24:23–24

24:23

24:24

24:25

25

25:1–5 25:1 25:6 25:15 25:16–18 25:18 26:54 27:11

168, 175, 199, 202 60, 89–92, 94–95, 101, 119, 122, 124– 126, 160, 162–163, 165–166, 255–259 124, 160, 163–165 94–95, 119, 163 170–171 95–96, 100, 119, 164, 166–167, 170– 171, 178–179, 201 124 170 179 96, 100, 119, 132, 163–166, 168, 170, 175, 177–179, 202, 210 166, 179 97, 164, 166, 178– 180 132, 163–164, 168, 170 96–97, 119, 163 163–164, 166, 170– 171, 179–180, 184, 203, 205 97, 119, 122, 124, 132, 163–164, 168, 170, 352 96, 99–101, 119, 122, 125, 163, 166, 179, 352 11, 21, 40, 49, 132, 149, 157–158, 173– 174, 209, 216 136, 177, 183–184, 200–201, 214–216, 232, 261, 263 198, 216, 263 51, 245–246 134, 216 210, 217 40 210, 214, 217 94 94

31

31:2–3 31:7 31:8 31:9 31:16

31:32 31:49 31:53 32:33 33 35:34 36:3 36:4 36:7 36: 9 36:12 Deuteronomy 2:5 2:9 2:12 2:19 2:26–3:11 2:34 3:1–3 3:3 3:20 18:10–14 20:1 20:5 22:13–23:1 22:24 23 23:2–9 23:2–3 23:2 23:3 23:4–7 23:4–6 23:4 23:5–7 23:5–6

391 40, 136, 177, 183– 184, 200–201, 239– 240, 243, 261–262 216 209, 217 209–212, 215–217, 229–233, 316 93, 209, 217 209, 212–217, 232, 253, 261, 263, 316, 348 93, 216 245 93 92, 177–178 134 178 92, 178 94 94 94 94

94 94 94 94 133 170 133, 138, 178 170 94 232 216 216 224 248 13, 239, 242–243 218, 225 225 223, 225 223, 225, 248 122, 218, 253 221–222, 227, 233, 240, 246, 312 219, 223–225, 248 224–225 225, 238, 348

392 23:5

23:6 23:7 23:8–9 23:8 23:9 23:10 24:9 25:17 26:7 26:19 27:15–26 28:1 28:16–19 32:8 32:36 34 Joshua 1:15 2:1 3:1 4:19 10:13 11:8 12:6–7 13:21–22 13:21 13:22 13:23 13:27 13:30–31 16:5 24 24:2 24:7 24:8 24:9–10 24:9

24:10 24:11

Source Index 74–75, 219–222, 225–226, 247–248, 313–314, 317, 349– 350 221–222, 226, 237, 249, 340–342 224–225 218, 224–225, 227 224–225 223–225 224 248 248 82 356 142 356 142 356 161 201

94 245 245 245 62–63 234 94 229–233, 239 230–231 211, 230–232, 302, 316, 324–326 232 231–232 232 356 241–243 349 235 234, 236 233–242, 253 226–227, 234–235, 238–239, 249, 340– 342, 349 226, 234–240, 249 234–235

Judges 5:11 6:5 7:12 9:33 11 11:16 11:17 11:25 14:1–4 14:4 21:17

244 140 140 90 227, 241–243, 312 241 241 241–243, 253 39 46 94

Ruth 1:20 4:18–22

353 248

First Samuel 12:7 15 15:29 18:15 22:3–4

244 88, 183 161 137 248

Second Samuel 1:6 1:18 8:11–12 13:22 23:39 22:14 25:1

80 63 175 248 177 356 64

First Kings 5:4 5:30 9:8 9:23

96 96 356 96

Second Kings 5 10:32–33 15:35 16 18:17

316 278 356 155 356

First Chronicles 1:43–44 3:16–17

207–209 203

393

Source Index 18:11 34:1

175 308

Second Chronicles 8:10 96 Ezra 3:2 Nehemiah 9:32 12:42–43 13:1–3 13:1 13:2–3 13:2

203

13:3 13:27

82 249 227, 246–249, 253 246–249 249 221, 248–249, 318, 340–342 249 249

Job 2:3 2:11 37:20

236–237 100 307

Psalms 68:15 68:22 72 91:1 110 117:1 119:52 135:14 Proverbs 5:15

353 93–94, 119, 124– 126, 177, 201 95–96, 124 353 95–96, 124 210 161 161

85

Qoheleth 8:1

320

Isaiah 14:12–15 33:15 40:15 45:24 55:2 60:2

166 244 85–86, 125 244 82 79

61:8 66:13

82 161

Jeremiah 3:24 5:5 6:4 6:22 28:35 48:45

82 100 83 95 82 93–94, 119, 126

Lamentations 2:8 3:65

100 82

Ezekiel 1:24 5:13 23:29 31:3–10 32:7–8 34:4 38–39

353 161 82 48 166 83 88, 121

Daniel 8:9–10 11:25 11:30

166 95 99, 125–126, 179

Amos 1:1 2:2 5:26

274 83, 124, 126 257

Jonah 3:10

82, 120

Micah 4:10 5:3 6:1–2 6:3–5 6:3 6:4 6:5

Haggai 1:1

243 243 348 243–245 348 244 243–246, 253, 318, 348–349

203

394

Source Index

New Testament Matthew 2 2:1–12 2:1–2 2:2 2:17–18 2:23

253, 258 256 91 90, 256–257 258 258

Second Peter 2:2 2:4–5 2:5 2:6–8 2:7

261 261 261 261 261

2:15–16 2:16 2:18

253, 259–261 259 261

Jude 6 7 9 11 14–15

261 261 261 253, 256–263 261

Revelation 2:14

253, 262–263

Apocrypha Jesus Sirach 34:18

50:27 320

320

 

Author Index Achenbach, R. 57–66, 93, 102, 117, 135–138, 141, 145, 155, 165–167, 339, 355 Aemelaeus, A. 71 Albertz, R 66–67, 321 Albright, W. F. 75, 97–98, 102, 111– 114, 116–117, 134, 165, 220 Ashley, T. R. 10, 135, 147, 305, 325, 328 Auld, G. A. 225 Ausloos, H. 71, 75, 92

Finkelstein, J. J. 317 Fistill, U. 134, 138–139 Forsling, J. 198, 200 France, R. T. 257 Frankel, D. 222–223, 237–238, 267, 269 Frankel, Z. 71, 104 Franken, H. J. 4, 271, 274, 354 Friedman, R. E. 28–29, 34, 158 Frisch, A. 146 Fritz, V. 352

Baden, J. 34–37 Barré, M. L. 74, 147, 171, 309, 349 Bartelmus, R. 147 Bewer, J. 84, 98, 108, 144, 150 Biale, D. 361, 354 Blum, E. 9, 16, 40–41, 205, 272–275, 277–278, 282–287, 292–293, 298, 327, 342 Budd, P. J. 38, 135–136, 173

Gall, A. 19–20, 90, 96, 99, 110, 134, 154, 159, 185 Garbini, G. 273, 275 Gaß, E. 306, 335, 337–338, 349, 360 Al-Ghul, O. 75, 312 Graupner, A. 29–32 Gray, G. B. 21–22, 34, 114, 208–209, 211, 213, 314, 317, 328 Green, W. S. 258 Greene, J. T. 217, 246, 328 Greenfield, J. C. 289, 291, 299 Greenspoon, L. J. 229–230 Greßmann, H. 22–23, 139, 143, 146 Gross, W. 10–11, 14, 18–19, 40, 46, 134–135, 146, 153, 171

Caquot, A. 273, 283, 284, 287 Carr, D. 73, 301 Cross, F. M. 351, 354 Daiches, S. 3, 328–329 Delcor, M. 304–305 Dillmann, A. 78, 213, 217, 224–225, 238 Donner, H. 225, 252–253, 268 Dorival, G. 80, 108, 210–213 Douglas, M. 145 Dozeman, T. 216, Ebach, R. 23 Eissfeldt, O. 23–26, 319, 347, 350 Fiedler, P. 257

Hackett, J. A. 278, 282–285, 287, 290, 327 Herrmann, W. 351 Hoftijzer, J. 271–272, 284, 287, 355 Holzinger, H. 18, 20–21, 147, 159, 334 Jastram, N. 353 Kaiser, W. C. 319, 360 Keil, C. F. 88, 314 Kellenberger, E. 145 Kessler, R. 245–246

396

Author Index

Knauf, A. E. 251–252 Kooij, G. van der 4, 271 Kratz, R. G. 50–52, 140 Layton, S. C. 307, 312 Lee, J. A. L. 70 Lemaire, A. 273–275, 283–284, 287, 293 Levin, C. 35, 41–46, 61, 134, 239, 303, 309 Levine, B. 27–29, 85, 110, 118, 134, 213, 215–216, 265 Lipiński, E. 273–274, 276, 282–285 Lust, J. 88, 102, 312 Lutzky, H. 354 Margaliot, M. 12–14 Margolis, M. L. 230–231 McCarter, P. K. 282, 284, 287, 290, 296 Milgrom, J. 30, 83, 85, 140, 143–144, 146–147, 153–154, 157, 169, 173, 179–180, 240, 266, 314, 329, 338 Millard, A. R. 289, 293, 296 Moatti-Fine, J. 232 Moore, G. F. 242 Moore, M. S. 320, 328, 333, 342, 355, 358 Mowinckel, S. 139, 208, 314, 359 Moyer, C. J. 145 Müller, H.-P. 277, 284, 309, 355 Muraoka, T. 236 Murtonen, A. 321–322 Naveh, J. 274 Nelson, R. D. 223, 231, 237–238 Nicklas, T. 356–357 Nöldeke, T. 18, 135–136, 202, 208 Noort, E. 253 Notarius, T. 167–168 Noth, M. 14, 25–26, 77–78, 153, 159, 209, 216 Olson, D. T. 330 Otto, E. 220, 223–224, 228 Pákozky, L. 10, 157, 173 Pardee, D. 276 Pike, D. M. 254–255

Puech, É. 146 Rendsburg, G. 182 Richardson, H. N. 271 Rofé, A. 11–12, 17, 40, 78, 237, 284, 295 Rösel, M. 74, 76–77, 80, 83–84, 120, 211–213, 237, 275, 334 Rost, L. 46, 154, 328 Rouillard, H. 38–40, 46–47, 148, 220 Rudolph, W. 18, 95, 97, 107, 136, 139– 140, 208, 334 Ruppert, L. 324 Safren, J. D. 146 Sals, U. 176 Sasson, V. 276, 285, 290, 292, 294 Savran, G. W. 201 Sayce, A. H. 3–4, 99, 117, 164–165 Schenker, A. 125 Schlund, C. 74, 76, 84, 120, 211–213, 237 Schmidt, L. 31–34, 138, 232, 239, 252 Schmitt, H.-C. 128, 147, 164, 167, 169–170, 181–183 Schmitt, R. 321, 323–324, 326, 330, 332, 338 Schnackenburg, R. 256 Schnutenhaus, F. 320 Schüle, A. 14–15, 20, 75 Seebass, H. 30, 34, 75–78, 83, 88, 95– 97, 102–104, 107–108, 113–116, 205, 213, 215 Segert, S. 279, 282–283 Seow, C. L. 290 Seybold, K. 166 Smend, R., Sr. 139–140, 167, 215 Soggin, J. A. 229, 239 Speiser, E. A. 208 Steuernagel, C. 138 Sturdy, J. 135 Sutcliffe, E. F. 10, 311 Timm, S. 58 Tosato, A. 158 Tov, E. 69–70, 74, 79, 99, 106, 122, 214 Tur-Sinai, N. H. 353

Author Index Ulrich, E. 84, 102, 109 Van Seters, J. 46–50, 141, 144, 151, 156, 181, 275, 302–303 Vaulx, J. de 27 Vermes, G. 82, 85, 88, 91, 112, 260, 266 Vetter, D. 90, 98, 100, 110–112, 114, 118 Volz, P. 108 von Rad, G. 208, 223 Vuilleleumier, R. 269 Wagner, S. 325, 328, 336–339, 359 Way, K. 129, 148, 291 Weippert, H. 273, 284, 355 Weippert, M. 273, 284–285, 351, 353, 355 Weise, U. 15–17

397

Weiser, A. 245–246 Wellhausen, J. 19, 21–22, 30, 33, 90, 100, 116, 149–150, 171 Wenning, R. 274–275, 293, 298, 304 Wevers, J. W. 3, 71, 73, 75, 83, 86, 89, 91, 96–100, 102, 104, 120, 210–211, 219–221, 336 Wilson, R. R. 184 Witte, M. 52–57, 66, 159, 162–163 Wolff, H. W. 245–246 Wolters, A. 297 Yahuda, A. S. 311, 314 Younger, K. L, Jr. 221, 278, 312, 317 Zevit, Z. 305 Zenger, E. 274–275, 293, 298, 304 Ziegert, C. 71–72, 83–84, 87, 112 Zobel, H.-J. 157, 163, 165

 

Subject Index 1QIsaa 161 1QM (see “War Scroll”) 1QSb 255 4Q175 (see “4QTestimonia”) 4Q23 (see “4QLev–Numa”) 4Q266 (see “Damascus Document”) 4Q27 (see “4QNumb”) 4Q339 256 4QLev–Numa 69, 74, 254 4QNumb 69, 74, 76, 78–79, 83, 204, 254 4QTestimonia 90, 255 Aaron 155, 243–245 Abram / Abraham 61, 63, 198–199, 201 Adad / Hadad 284 Agag 53, 64, 167, 169, 181, 183 Ahaz 155 Akkadian 3, 45, 288, 321–322, 324– 325, 327–328, 330, 341, 353 Altars 132, 151–152, 154–156, 328– 329, 346 Amalek(ites) 64, 165, 168, 170, 175– 176, 179, 201–202 Amau 311–312 Ammon(ites) 35, 74–75, 122, 203, 218–219, 222–228, 240–242, 246, 248–250, 265, 275, 278, 309, 311– 312, 314–316, 318, 332, 364 Amorites 46, 132, 134, 138, 177, 197– 198, 229, 241–243, 251, 264 Aquila 220 Arabia 359 Arabic (lang.) 276, 285, 287, 307, 312, 324, 330–331, 336 Aram(eans) 5, 63, 182–183, 186, 198, 206, 208, 216, 220–222, 228, 278– 279, 296, 298, 300, 303–305, 307– 308, 312–319, 346, 358–361, 364– 366

Aramaic (lang.) 182–183, 275–278, 281–288, 297–298, 300–301, 303– 304, 307, 313, 319–320, 331, 336, 341, 343, 350, 353, 355, 359–360, 364 Assyria 97–98, 100–101, 179, 202, 331 Augury 324–325, 331–334, 344–346 Baal 306, 357 Babylon 321, 325 Balak 16, 39, 41–43, 45–47, 56, 59, 63, 73, 120, 122–124, 130–132, 134– 135, 137, 141–145, 147, 149–150, 152–158, 167, 169, 173–176, 181– 183, 207–208, 224, 226–227, 229, 234–235, 239–246, 250–251, 257, 262–265, 269, 302, 311, 314–316, 321, 323, 326–330, 342, 344, 346– 348 Bashan 59, 124, 133, 135, 138, 177, 242 Bela 207–208 Beor 207–208, 211–212, 217–219, 228, 233–234, 238–239, 244–245, 249– 250, 259–260, 265, 275, 288, 300, 303, 307–308, 344, 358, 360, 362, 366 Bless / Blessing 42, 44–45, 49–50, 55– 57, 59, 61, 64, 125, 132, 141–142, 144, 158, 160, 162, 167, 169, 172– 173, 175, 181–183, 198–199, 218– 219, 221–222, 227, 229, 235, 239– 241, 243, 249, 264–265, 302, 311, 315, 321, 323, 327, 330–331, 341, 362, 365 Bronze Age 271 Cain 36, 117, 130, 170, 202, 261–262, 322 (see also “Kenites”)

Subject Index Canaanite (lang.) 276–278, 282, 297, 304, 341 Canaanite(s) 39, 180, 198, 307, 351 (see also “Kenites” and “Cain”) Cisjordan 5, 44, 264, 310, 315 Codex Leningradensis (ML) 43, 52, 69, 73, 75, 118, 151, 211, 311–312 Codex Sinaiticus 259 Codex Vaticanus 259 Conquest (see “Eisodos”) Covenant Code 122 Curse / Cursing 33, 35, 42, 44, 47, 49, 55–57, 59, 61, 132, 141–142, 144, 158, 160, 162, 167, 169, 173–175, 181, 183, 198–199, 219, 221–223, 229, 239–241, 243, 249, 264, 267, 302, 311, 315, 321–323, 326–328, 330–331, 340–342, 344–346, 348, 358, 362–363, 365 Damascus (Iron Age Kingdom; see “Aram(eans”) Damascus Document (CD, 4Q266) 90, 254–256 Daniel 321, 339 DAPT 4, 8, 41–42, 46, 51, 68, 271– 305, 307, 314, 318–321, 327–328, 335, 337–338, 342–344, 346–347, 350, 352–353, 355–360, 364–365 David 166, 176–177, 183, 199, 205, 248, 256, 320, 329, 335, 356, 365– 366 Dittography 76, 84, 97, 99, 109, 111, 221, 235 Divination 176 Divine Nomenclature 12, 18, 20–21, 24, 29–31, 37, 76, 78, 120, 123–124, 131, 181, 301, 345, 347–357 Diviner (see “Mantic”) Documentary Hypothesis (see “Urkundenhypothese”) Donkey Story 11–14, 16–17, 27–29, 36, 39, 41, 44, 46, 47, 49, 56, 62, 66, 68, 76, 78–79, 129, 132, 145–147, 149, 156, 175–176, 181, 184, 201, 259–260, 291, 338–339, 347–348, 363, 366 Dream 320–321, 338, 345–346, 364– 365

399

Dtn / Dtr 5, 7, 9, 37, 39, 40–41, 56–58, 61, 94, 142, 161, 178, 181, 199–202, 204–205, 238, 244, 246, 248, 251, 269, 324, 332, 341, 363, 366 Earthquake 271, 274–275, 301 Edom(ites) 36, 53, 65–66, 92–93, 164– 165, 167–170, 175, 179, 183, 198– 202, 204–205, 207–208, 218–219, 224–225, 227, 241, 248, 308, 314, 359 Egypt(ians) 140, 156, 158, 175, 197, 206, 209, 218–219, 224–225, 227, 241–242, 244, 248, 264, 290, 311, 321, 351–354 Eighth Century bce 182–183, 274–275, 278, 293, 304–305, 313, 361, 365 Eisodus Narrative 5, 7, 12, 51, 56–58, 61, 206, 243–245, 264, 362 El 288, 301, 303, 336–337, 339, 342, 345, 350–354, 356–357, 360 Elyon 301, 339–340, 345, 352, 354, 356–357 Enneateuch 5, 7–9, 61, 65, 134, 180, 185, 199, 201 Epigraphy 4–5, 8, 68, 271–305 Esau (See “Edom(ites)”) Eschatology 53–54, 56, 87, 121 Euphrates 96, 139, 309–310, 317 Exile 49, 200–201, 203, 205 Exodus Narrative 7, 28, 47, 50–53, 56– 57, 135–136, 140, 156, 159, 166, 169, 172, 175, 178, 197–198, 203, 206, 209, 215–216, 219, 222, 224– 225, 228–229, 241–242, 244–245, 248–249, 251, 264, 351–352, 362, 366 Exorcist 328 Ezekiel 50 Flood Narrative 202 Fortschreibung 6, 25, 28, 30, 38, 52– 53, 63, 66, 147, 156, 180, 184–185, 201–203, 224–225, 227, 363 Genre 146 Gilead 276 Gilgal 243–245, 250, 264 Gog 97–98, 121, 123

400

Subject Index

Greed 260–262, 266 Greeks 185 H, Holiness School 95, 269 Hamath 296, 343–344, 346 Haplography 75, 77–78, 86, 106, 113– 115, 140, 213, 235 Hasmonean 93, 206 Hazael 278, 298, 304 Hebrew (lang.) 276 Hebrews 100 Hellenistic Period 37, 46, 58, 69, 166, 185, 201, 204–206 Heshbon 231 Hexapla 98, 219, 221, 235 Hexateuch 2, 6–9, 23, 58–59, 61–66, 134, 141, 180, 185, 201 Hezekiah 203, 329 Hobab 198 Idrimi 311 Irenaeus 91 Iron Age 271 Israel (name for exodus group) 14–16, 20, 41, 51, 73, 120, 135, 138, 141, 143, 155, 158–165, 167–169, 172, 175, 178, 180–183, 185–186, 196– 199, 201–202, 206, 214, 219, 222, 239–240, 242, 246, 250–251, 267, 269, 324, 327, 341–342 Israel (see “Northern Kingdom”) Israelites (name for exodus group) 51, 133–134, 136, 138, 143–144, 172– 173, 177–178, 183–184, 196, 198, 200, 209, 212, 214–217, 219, 222, 225, 229, 231, 239, 250, 364 Jacob 61, 96, 121, 158, 160–165, 167– 169, 198–199, 203–204, 352–353 Jazer 133, 135, 138, 177 Jehuide 206 Jephthah 227, 241–243 Jericho 51 Jerome 362 Jerusalem 96 Joseph 320–321, 332, 335, 345, 353 Josephus 1, 263, 265–267 Josiah 205 Judah 65, 93, 183, 200–202

Kenites 117, 179, 202 Kingship (see “Monarchy”) Kittim 98–99, 101–202 Korah 261–262 Mantic 13, 64, 143, 171, 173, 176, 186, 208, 215, 230, 232, 252, 259–260, 265, 272, 302, 316–317, 319, 324– 326, 331, 342–343, 360, 364–365 Masoretes 70, 81, 89, 211, 315 Masoretic Text 7, 9, 48–49, 63, 69– 101, 118–119, 121–126, 132, 162– 163, 174, 178, 181, 204, 211–214, 218–221, 230–231, 234–238, 241, 243–244, 248–249, 254, 259, 309– 310, 312, 317, 336, 347–348, 362 Mesha 41–42 Mesopotamia 27, 45, 50, 75, 219–222, 228, 265, 303–304, 309–310, 312– 314, 317–318, 329–330, 355, 358, 365–366 Messianism 1, 86–87, 90–92, 94, 121, 209, 255–258, 267, 366 Midian(ites) 40, 46, 132, 136–138, 140, 142–143, 171–172, 176–177, 183– 184, 200, 214–217, 222, 228–229, 231–233, 250–251, 263–264, 316, 318, 334, 348, 363–364 Miriam 243–245 Moab(ites) 5, 12, 16, 35–36, 39, 41–42, 44, 51, 58, 63, 66, 73, 83, 91–93, 120, 125, 128, 131–138, 140–143, 162, 165, 167–173, 175, 177, 179, 181–186, 197–198, 200–206, 214– 215, 217–219, 222–229, 239, 241– 246, 248–250, 265–266, 303, 305, 310–312, 314–316, 319, 321, 334, 341, 358, 363–364 Moabite (lang.) 276, 278 Monarchic Period 5, 7, 9, 51, 182–183, 200, 313–314, 362, 365–366 Monarchy 39, 61, 167, 175, 181–183, 186, 203–206, 265, 269, 362 Monolith Inscription 309 Monotheism 50 Moses 243–246 Naaman 316

Subject Index Narratology 16, 30, 45, 47, 79, 128– 131, 139, 146, 151–171, 196 Necromancy 325 Neo-Babylonian Period 166, 205 New Testament 6, 8, 90–91, 170, 207, 253–254, 259–260, 265–267, 364 Ninth Century bce 278, 293, 304–305, 313, 361, 365 Noah 155, 202–203 Northern Kingdom 7, 121, 182–183, 305, 313–314, 365 Offering(s) 151–155, 328–330, 346 Og 35, 46, 97–98, 122, 124, 133, 138, 164, 170, 177, 184, 232, 242, 264 Omens 171 Omride 206 Oracle(s) 28–30, 32–33, 39–40, 42, 45, 47, 48–49, 52–55, 59, 61–65, 129– 132, 141, 149–171, 173–175, 181, 183, 185, 198–199, 204–205, 254– 256, 258, 268, 299, 301, 315, 324– 326, 331–340, 345, 349, 353–355, 357, 362–363 P, priestly 7, 9, 21, 23, 35–37, 40–41, 46, 56, 61, 66–67, 95, 135, 196, 200–202, 204–205, 217, 251, 269, 329 Paleo-Hebrew 75–76, 85–86, 95 Pentateuch 6–9, 18–20, 26, 33, 37–38, 40–41, 43, 59, 62–67, 71, 133–134, 141, 180, 182, 185–186, 201, 228 People (name for exodus group) 14, 136–138, 140–141, 143–144, 155, 160, 162–163, 168–169, 175, 181, 183, 196–197, 216, 222, 244, 267, 323–324, 327, 362 Peor 26, 168–169, 198, 212, 214–215, 217, 232, 251, 263–264, 267 Persia 100 Persian Period 58, 166, 184, 206, 274 Peshitta 69, 74–76, 79, 82, 90–91, 93– 95, 118, 220–221, 227–228, 309– 310, 312, 318, 320–321, 336, 345 Pethor 3, 74–75, 132, 139, 143, 220– 222, 309–310, 314, 316–318, 321 Philo 1, 91, 266 Phoenician 327

401

Pisgah 36, 150, 168–169 Pitrû (see “Pethor”) Prayer 330 Pre-Smr (see “Smr”) Priest 328–329, 346 Prophet / Prophecy 13, 30–31, 33, 46, 258–259, 268, 272, 319–320, 325, 335–338, 349, 360, 365–366 Proto-M 20, 81–82, 84, 85, 89, 91, 93– 94, 96, 99, 119, 123, 211, 220, 231, 238, 256, 259, 318 Punic 327 Qedem 314–316, 318 Qumran 1, 8–9, 53, 69, 72, 76, 93, 118, 121, 170, 204, 253–256, 258–259, 267, 286, 321, 364 Qur’an 312 R1 144, 162, 169–170, 172, 174–175, 177–178, 180, 183, 197–199, 202– 204, 242–243, 246, 251, 253, 264, 267, 269, 311, 315–316, 319, 322, 326, 333, 336–340, 344, 352, 363– 365 R2 144, 156, 172–173, 175, 180, 183, 200–201, 205, 215–217, 250–251, 253, 264, 267, 316, 363–364 R3 180 R3+ 156, 169–171, 175, 180–181, 200– 201, 217, 228, 251–253, 264, 267, 269, 316, 329, 336, 338–339, 352, 364 R4 180 R5 180 Rabbinic Biblical Exegesis 259–261, 265–267, 364, 366 Roman Period 6, 7, 9, 20, 69, 92, 204, 206 Romans 98 Rubrum 288–289, 293 S1 143–144, 168–170, 175, 180, 182– 183, 195–196, 251, 253, 316, 327, 331, 335, 338–339, 344–345, 351– 352, 354, 362 S2 143–144, 156, 171–173, 180, 184, 200–201, 205, 215, 326, 363 Sacrifice (see “Offerings”)

402

Subject Index

Samaritan Torah 7, 9, 20, 35, 53, 69, 72, 74–83, 88, 93–100, 119–123, 163, 181, 204, 220, 227–228, 254, 309, 312, 317–318, 348, 362 Samuel 176 Saul 176, 181, 183, 206, 335 Second Isaiah 50, 125 Seer 288, 320, 343–344, 349, 363, 366 Seir 53 Seleucid 45 Septuagint Translator 94–95, 100–101, 120–121, 236, 244, 338 Septuagint Vorlage 70, 72, 79–86, 89– 91, 93–97, 99–101, 119, 121, 123, 126, 211–214, 219–220, 231, 234, 236–238, 243–244, 257, 317 Septuagint 7, 9, 20, 37, 53, 69–73, 75– 85, 87–101, 118–126, 153, 161, 163, 174, 178–179, 181, 184, 204, 208, 211–214, 216, 219–221, 223–225, 230–231, 234–239, 241, 243–244, 248–249, 256–257, 259, 265, 308– 309, 311–313, 315, 317–318, 322, 333, 336, 338, 348, 362 Seventh Century bce 198 Shaddai 162, 303, 336–337, 345, 352– 357 Shaddayin 280–281, 284, 290, 300– 301, 303, 350, 355–357 Shalmaneser III 309 Shamash / Shemesh 284–285, 290, 301, 350, 355–356 Sheol 284 Shittim 51, 244–246, 250, 264, 336 Sihon 133, 138, 170, 197, 229, 232, 241, 243, 264 Sixth Century bce 198 Solomon 203–204 Source Criticism (see “Urkundenhypothese”) Stele 271 Structure 128–133, 148–171, 294–295, 299–301 Sukkoth (see also “Tell Deir ʿAlla”) 271

Symmachus 91–92 Syntax 55, 70, 74, 76, 139, 162–163, 171, 213, 234, 279, 281–282, 289, 294, 355 Syria 45 Syriac 282, 330 Syrohexapla 79 Targum 1, 69, 82, 90–93, 95, 98, 100, 119, 181, 219–220, 309 Tel Dan Inscription 276–277 Tell Deir ʿAlla 4, 208, 271–305, 313 Temple Scroll 254 Terah 199 Testament of Moses 261 Textual Criticism 31, 37, 43–44, 48, 49–50, 52–53, 196 Theocratic 7, 65, 269 Transjordan 4–5, 178, 183, 229, 231, 245, 264, 278–279, 293, 296, 302– 305, 310, 312, 315, 319, 358–360 Ugarit / Ugaritic 284, 327 Uriah 155 Urkundenhypothese 2, 5,10, 14, 17–38, 46, 53, 56, 68, 77, 128, 131, 133, 185, 196 Vetus Latina 69, 91, 93, 118, 237, 238 Vision / Visionary (see also “Seer”) 271, 287–295, 299–302, 305, 320, 337–339, 342–346, 353, 356–358 Vulgate 69, 74–76, 78, 82–83, 90–91, 93, 98–100, 118–119, 219–221, 227–228, 260, 309, 312, 317–318, 320–321, 325, 333, 336, 345, 362 War Scroll 255–256 Wiederaufnahme 14, 39, 44, 52, 147, 212, 292 Zakkur 343, 346 Zerubbabel 203 Zeus 357 Zion 96