An Inductive Method for Reconstructing the Biblical Text : Illustrated by an Analysis of 1 Samuel 3

AN INDUCTIVE METHOD FOR RECONSTRUCTING THE BIBLICAL TEXT Illustrated by an Analysis of 1 Samuel 3

JAMES R. ADAIR, JR.

To my Parents, Rob and Vela Adair, Y mis Suegros, Al y Cecilia Sierra

TABLE OF CONTENTS PREFACE

. . . . . . . . . . . . . . . . . . . . . . . . . .

i

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

1

Research Problem . . . . . . . . . . . . . . . . . . . . Definition of Terms . . . . . . . . . . . . . . . . . . . . Outline of This Study . . . . . . . . . . . . . . . . . . .

1 6 7

1. DEVELOPMENT OF A MODEL FOR ANALYSIS OF THE WITNESSES . .

9

Chapter

Literary Analysis before Textual Analysis Textual Analysis before Literary Analysis Genealogical Method . . . . . . . Determining the Translation Technique . The Script of the Vorlagen . . . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

9 13 16 21 27

2. RECONSTRUCTING ARCHETYPICAL SECONDARY WITNESSES . . . 30 Method for Reconstruction . . . . . . Location of Textual Data . . . . . . . Presentation of Reconstructed Witnesses . . Septuagint . . . . . . . . Peshitta . . . . . . . . . Targum . . . . . . . . . Vulgate . . . . . . . . . Aquila . . . . . . . . . . Symmachus . . . . . . . . Theodotion . . . . . . . . Other Readings Attributed to ´ The Lucianic Recension . . . . The Hexaplaric Recension . . . Other Possible Hebrew Readings .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

30 36 37 38 53 56 60 65 65 66 66 67 69 71

3. GRAMMATICAL/STYLISTIC CHARACTERISTICS OF THE SECONDARY WITNESSES . . . . . . . . . . . . . . 72 Septuagint . . . . . . . . . . . . . . . . Limitations of Greek for Rendering Hebrew . Partial Translation Technique . . . . . . Elimination of Variants . . . . . . . . Peshitta . . . . . . . . . . . . . . . . . Limitations of Syriac for Rendering Hebrew . Partial Translation Technique . . . . . . Elimination of Variants . . . . . . . . Targum . . . . . . . . . . . . . . . . . Limitations of Aramaic for Rendering Hebrew Partial Translation Technique . . . . . . Elimination of Variants . . . . . . . . Vulgate . . . . . . . . . . . . . . . . . Limitations of Latin for Rendering Hebrew . Partial Translation Technique . . . . . . Elimination of Variants . . . . . . . . Evaluation of Partial Secondary Witnesses . . . . . Aquila . . . . . . . . . . . . . . iv

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. 73 . 73 . 77 . 85 .105 .105 .108 .113 .128 .128 .129 .132 .143 .143 .145 .150 .161 .161

Symmachus . . . . . . . . Theodotion . . . . . . . . Other Readings Attributed to ´ The Lucianic Recension . . . . The Hexaplaric Recension . . . Other Possible Hebrew Readings .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

.162 .162 .162 .162 .163 .163

4. LITERARY/THEOLOGICAL CHARACTERISTICS OF THE SECONDARY WITNESSES . . . . . . . . . . . . . .164 Septuagint . . . . . . . . . . . . . . . . . . Relevant Literary, Historical, and Theological Data Present in the Translation . . . . . . . Final Translation Technique . . . . . . . . Elimination of Variants . . . . . . . . . . Peshitta . . . . . . . . . . . . . . . . . . . Relevant Literary, Historical, and Theological Data Present in the Translation . . . . . . . Final Translation Technique . . . . . . . . Elimination of Variants . . . . . . . . . . Targum . . . . . . . . . . . . . . . . . . . Relevant Literary, Historical, and Theological Data Present in the Translation . . . . . . . Final Translation Technique . . . . . . . . Elimination of Variants . . . . . . . . . . Vulgate . . . . . . . . . . . . . . . . . . . Relevant Literary, Historical, and Theological Data Present in the Translation . . . . . . . Final Translation Technique . . . . . . . . Elimination of Variants . . . . . . . . . . Evaluation of Partial Secondary Witnesses . . . . . . . The Lucianic Recension . . . . . . . . . . The Hexaplaric Recension . . . . . . . . . Other Possible Hebrew Readings . . . . . . . Conclusion . . . . . . . . . . . . . . . . . .

. . . . .170 . . . .

. . . .

. . . .

. . . .

.170 .172 .173 .177

. . . .

. . . .

. . . .

. . . .

.177 .178 .179 .180

. . . .

. . . .

. . . .

. . . .

.180 .181 .181 .182

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

.182 .184 .184 .187 .187 .188 .188 .188

5. CONSIDERATION OF VARIANTS FROM PRIMARY WITNESSES . . .189 Variants in Masoretic Manuscripts . . . . . . . . . . . . Important Manuscripts as Identified by Goshen-Gottstein Kethib-Qere . . . . . . . . . . . . . . . . Tiqqune Sopherim . . . . . . . . . . . . . . Variants in 4QSama . . . . . . . . . . . . . . . . . Conclusions: Significant Variants in 1 Samuel 3 . . . . . . . Septuagint . . . . . . . . . . . . . . . . Peshitta . . . . . . . . . . . . . . . . . Targum . . . . . . . . . . . . . . . . . Vulgate . . . . . . . . . . . . . . . . . The Lucianic Recension . . . . . . . . . . . . The Hexaplaric Recension . . . . . . . . . . . Other Possible Hebrew Readings . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

.190 .190 .192 .192 .192 .194 .194 .196 .197 .197 .197 .197 .198

6. RETROVERSIONS OF THE SECONDARY WITNESSES . . . . . . .201 Developing a Methodology for Retroverting Translations . . . Proposed Methodologies for Retroverting Translations Orthography . . . . . . . . . . . . . . . Conclusions concerning Methodology . . . . . . v

. . . .

. . . .

. . . .

.201 .201 .212 .216

Retroversions of the Secondary and Partial Secondary Witnesses Septuagint . . . . . . . . . . . . . . . Peshitta . . . . . . . . . . . . . . . . Targum . . . . . . . . . . . . . . . . Vulgate . . . . . . . . . . . . . . . . The Lucianic Recension . . . . . . . . . . . The Hexaplaric Recension . . . . . . . . . . Other Possible Hebrew Readings . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

.217 .217 .225 .228 .229 .229 .231 .233 .234

7. EVALUATION OF THE SIGNIFICANT VARIANTS . . . . . . . . .235 The Problem of Multiple Editions . . . . . . . . . . . Conjectural Emendations . . . . . . . . . . . . . . Guidelines for Evaluating Variants . . . . . . . . . . . Evaluations of the Variants . . . . . . . . . . . . . Variants That Pertain to the Question of Multiple Literary Editions 8. A CRITICAL EDITION OF 1 SAMUEL 3

. . . . .

. . . . .

. . . . .

.237 .240 .241 .243 .268

. . . . . . . . . . . .274

CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . .281 Appendix 1. RECONSTRUCTED ARCHETYPICAL TEXTS OF SECONDARY AND PARTIAL SECONDARY WITNESSES . . . . . . . . . . . .283 Septuagint . . . . . Targum . . . . . . The Lucianic Recension The Hexaplaric Recension

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

.283 .284 .284 .286

2. ADD-OMS AND PRELIMINARY TRANSLATION TECHNIQUE TABLES .288 Septuagint . . . . . . . . . Add-oms . . . . . . Consistency . . . . . Segmentation . . . . Word Order . . . . . Quantitative Representation Peshitta . . . . . . . . . . Add-oms . . . . . . Consistency . . . . . Segmentation . . . . Word Order . . . . . Quantitative Representation Targum . . . . . . . . . . Add-oms . . . . . . Consistency . . . . . Segmentation . . . . Word Order . . . . . Quantitative Representation Vulgate . . . . . . . . . . Add-oms . . . . . . Consistency . . . . . Segmentation . . . . Word Order . . . . . Quantitative Representation

vi

. . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . .

.288 .288 .289 .295 .296 .296 .297 .297 .299 .304 .304 .305 .307 .307 .307 .313 .313 .313 .314 .314 .316 .322 .322 .322

3. FINAL TRANSLATION TECHNIQUE TABLES Septuagint . . . . . . . . . Consistency . . . . . Segmentation . . . . Word Order . . . . . Quantitative Representation Peshitta . . . . . . . . . . Consistency . . . . . Segmentation . . . . Word Order . . . . . Quantitative Representation Targum . . . . . . . . . . Consistency . . . . . Segmentation . . . . Word Order . . . . . Quantitative Representation Vulgate . . . . . . . . . . Consistency . . . . . Segmentation . . . . Word Order . . . . . Quantitative Representation

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . .325 . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

.325 .325 .331 .332 .332 .332 .332 .338 .338 .338 .340 .340 .346 .346 .346 .347 .347 .353 .354 .354

4. ALGORITHM FOR CALCULATING THE DEVIATION FACTOR . . . .356 SELECTED BIBLIOGRAPHY

. . . . . . . . . . . . . . . . . . .359

vii

PREFACE The recovery of the original text of the Old Testament (OT) is an impossible task. All the original manuscripts (mss) have long been destroyed, as have the first, second, third, and many subsequent generations of copies. The oldest more or less complete witnesses to the original text are translations into other languages, and the oldest Hebrew mss of any significant size date from more than three centuries after the authors put down their pens. Of the witnesses that survive, none present satisfactory readings in every location, and there are some crucial places throughout the witnesses where none have convincing readings. There is even good reason to question the existence of an “original text.” If, as many scholars believe, the text of most OT books grew over time through the work of redactors and scribes (not to mention earlier written sources used by the primary authors), more than one “original text” may exist for many books. In spite of the problems, however, the task of textual criticism is of continuing importance. The greatest advances in the field of textual criticism of biblical books have come in the New Testament (NT) field, beginning with the late nineteenth century work of people like Tischendorf, Westcott, and Hort.1 Several factors have led to more textual work being done in the NT than in the OT: the availability of multiple early mss in the original language; the chronological proximity of many of these mss to the autographs; the diversity of readings among these mss; the existence of several versions and patristic witnesses to the text within five hundred years of the autographs; the relative simplicity of composition of many NT books in comparison with OT books (the gospels excepted); ms discoveries and acquisitions in the late nineteenth century, including Codices Sinaiticus, Vaticanus, and Bezae, Egyptian papyri, and the Gnostic mss from Nag Hammadi; and theologically motivated decisions by Christian scholars to do NT work. The difficulties of doing textual criticism on the OT, however, do not suggest the impossibility of progress, 1 Pier Giorgio Borbone, “La critica del testo e l’Antico Testamento ebraico,” review of Critique

textuelle de l'Ancien Testament, vol. 1, Josué, Juges, Ruth, Samuel, Rois, Chroniques, Esdras, Néhémie, Esther, by Dominique Barthélemy, in Rivista di storia e letteratura religiosa 20 (1984): 251, decries the retarded development of textual criticism of the Old Testament as compared with that of the New Testament, especially the lack of a critical Hebrew text (italics original): Il Nuovo Testamento è studiato e letto in edizioni critiche, condotte con metodo filologico; anzi, si può dire che proprio l’approfondimento dei problemi relativi all’edizione dei testi neotestamentari ha promosso in maniera determinante lo sviluppo del metodo della filologia classica. Diversa è invece la situazione dell’AT: mentre esistono edizioni critiche delle tradizioni greca e latina (i LXX e la Vulgata), ed il “Peshitta Institute” di Leida sta lavorando ad un’edizione della tradizione siriaca che, se proprio non si può dir ‘critica’ in senso tecnico, è un importante passo avanti rispetto alla situazione precedente, per il testo ebraico ci si accontenta di stampare un textus receptus.

ii

and renewed interest in the text of the OT has surfaced in the last forty years or so, especially since the discovery of the mss in the Judean desert. Impetus for further study has also come from the discoveries of texts from Ugarit and Ebla and from methodological advances proposed in the past twenty or thirty years, particularly by members of groups like the International Organization for Septuagint and Cognate Studies and the International Organization for the Study of the Old Testament. It is in the light of this newfound interest in the text of the OT and in the belief that methodologies exist or can be found to illuminate more fully previous states of the text that this study is undertaken. The present work is a combination and slight revision of my Ph.D. dissertation, undertaken at Southwestern Baptist Theological Seminary (“A Methodology for Determining the Textual Variants Which Are Relevant for Reconstructing the Original Text of the Old Testament: A Case Study of 1 Samuel 3”), and my M.A. thesis from the University of Stellenbosch (“Reconstructing 1 Samuel Chapter 3”). This work was largely finished in 1992, and few revisions to the bibliography have been made, other than to note my own studies which grew out of this one but were published later. The process of combining and revising my dissertation and thesis has been an onagain, off-again affair, and the calls of job, family life, and other academic interests have turned what I once thought would be a quick process into a five-year odyssey. I have been encouraged along the way by many people, without whom this book might not have seen the light of day. Special thanks go to Johann Cook of the University of Stellenbosch, who encouraged both my work on this book and my dream of creating an electronic journal, TC: A Journal of Biblical Textual Criticism, on whose editorial board he has served since the journal’s inception in 1996. My wife Rosa has been very supportive of my efforts to produce this book, and my daughters Danielle and Elise are a constant joy, tangible reminders of the goodness of God. Finally, my education, my interest in learning, and even my very survival I owe to my parents, Rob and Vela Adair. Though they don’t read Hebrew or Greek, they keep a copy of my dissertation on a coffee table in their living room to show to friends who visit. When I see it there, I’m reminded of how much of who I am is due to them. It is to them that this book is dedicated.

iii

INTRODUCTION Research Problem The text-tradition of the OT that has played the most important role in modern exegetical, historical, literary, and theological studies, as well as that which forms the basis for most translations, is the Masoretic Text (MT). MT owes its distinction to a variety of factors: (1) it is part of the text-tradition used almost exclusively by the Jews since about 100 C. E.; (2) it is the only complete witness in the original languages to the entirety of the Jewish Scriptures (the Protestant OT) that is extant; (3) it was the only Hebrew witness to the text known to scholars before the rediscovery of the Samaritan Pentateuch in 1616 and the ms discoveries of the past century (e.g., the mss of the Judean desert and the Cairo Geniza);1 (4) it was the text Luther and other reformers turned to after their rejection of the Latin Vulgate (V).2 MT, then, has occupied and continues to occupy a privileged place among witnesses to the text of the OT. In the books of Samuel and Kings, however, scholars have long accused MT of poor quality in many places, and they have turned to the Septuagint (LXX) and other witnesses as valuable tools for understanding the text. Several recent studies have analyzed MT (or some other witness) with little or no regard for other text-traditions. James A. Sanders makes the valuable observation that each text-tradition was viewed by its tradents as the text of Scripture, and thus it was understood without reference to other texts. Thus, modern interpreters can benefit from an understanding of the meaning(s) present in a single witness.3 Nevertheless, the goal of textual criticism is “to produce a text as close as possible to the original,”4 and this process 1 Cf. Interpreter’s Dictionary of the Bible [hereafter IDB], s.v. “Samaritan Pentateuch,” by Bleddyn J. Roberts; Shemaryahu Talmon, “The Old Testament Text,” in The Cambridge History of the Bible, ed. P. R. Ackroyd and C. F. Evans, (Cambridge: Cambridge University Press, 1970) 1: 159-99; Paul E. Kahle, The Cairo Geniza, 2d ed. (Oxford: Basil Blackwell, 1959). 2 For a synopsis of the debate over the value of MT from the Middle Ages to the early twentieth

century, see Dominique Barthélemy, Critique textuelle de l’Ancien Testament, vol. 1, Josué, Juges, Ruth, Samuel, Rois, Chroniques, Esdras, Néhémie, Esther, Orbis Biblicus et Orientalis, no. 50/1 (Fribourg: Éditions Universitaires, 1982; Göttingen: Vandenhoeck & Ruprecht, 1982), *1-*40. 3 “You must learn to respect each of the witnesses before you doubt them”; lecture delivered at the University of Stellenbosch, South Africa, 17 August 1989. Cf. also Sanders, From Sacred Story to Sacred Text (Philadelphia: Fortress Press, 1987), 127-51, esp. p. 134, “All versions are to some extent relevant to the communities for which translated”; Hans Joachim Stoebe, Das erste Buch Samuelis, Kommentar zum Alten Testament, vol. 8, pt. 1 (Gütersloh: Gütersloher Verlagshaus Gerd Mohn, 1973), 31: “[Der Exeget] muß sogar jede Rezension zunächst für sich genommen und so weit wie möglich aus sich verstanden werden.” 4 Paul Maas, Textual Criticism, trans. Barbara Flower (Oxford: Clarendon Press, 1958), 1.

1

2

necessarily involves the reconstruction of a presumed “original text.” The fact that the reconstructed text is not exactly the same as the autograph should not dissuade the textual critic, nor should the assertion that the new “eclectic” text does not represent the text used in any one community deter him or her.5 Much text-critical work has focused on a specific verse or word without regard for the larger context. For example, many scholars have proposed emending the pointing of in Isa. 40:6 MT to , partially on the basis of the LXX reading .6 However, the LXX reading here does not clearly indicate that the translator had a different understanding of the Vorlage (nor a different Vorlage), for a priest, rather than a prophet, is speaking in verses 1 and 2, and the translator may have placed the words of the second half of the verse in the mouth of the priest rather than the prophet.7 Furthermore, the translation technique of LXX Isaiah has often been described as free or even paraphrastic.8 The textual critic must be especially cautious about assuming that an apparent variant in a free translation indicates a difference in the Vorlage. 9 5 Of course, the textual critic may choose to reconstruct the text used by a particular community at

a particular time rather than to reconstruct the original archetype. But the charge that a “new” text has been created which has no historical worth is simply invalid. The textual critic’s goal, by definition, is to attempt to recreate earlier, non-extant forms of the text, not to reproduce a form of the text that was actually used and interpreted by a specific community (though the latter goal is legitimate both in itself and as an intermediate step toward the recovery of a more ancient text). Since the transmission history of the OT is so complex, many have sought to reproduce a form of the text other than the original. For example, the Committee for the Textual Analysis of the Hebrew Old Testament, working under the auspices of the United Bible Societies, has identified four possible texts toward which their committee could work: (1) the original oral or written form; (2) the most primitive form attested by extant witnesses; (3) the normative Jewish text after 70 C. E.; (4) MT as determined in the ninth or tenth century. With the exception of Proverbs, Jeremiah, and Ezekiel, they have sought to reconstruct the text as described in the second stage above; see Barthélemy, Critique textuelle, 1:*69. The OT textual critic must also be aware of the overlap between textual criticism and literary criticism when dealing with the early stages of the transmission of the text, since some traditions continued to grow after their form at one stage was recorded. Cf. Shemaryahu Talmon, “The Textual Study of the Bible—A New Outlook,” in Qumran and the History of the Biblical Text, ed. Frank Moore Cross and Shemaryahu Talmon (Cambridge: Harvard University Press, 1975), 33236; Emanuel Tov, “Some Sequence Differences between the MT and LXX and Their Ramifications for the Literary Criticism of the Bible,” Journal of Northwest Semitic Languages 13 (1987): 151-60. 6 For example, D. Winton Thomas in K. Elliger and W. Rudolph, eds. Biblica Hebraica Stuttgartensia, 2d ed. (Stuttgart: Deutsche Bibelgesellschaft, 1983) [hereafter BHS] and Rudolf Kittel in Rudolf Kittel, ed. Biblia Hebraica, 3d ed. (Stuttgart: Privegierte Württembergische Bibelanstalt, 1937) [hereafter BHK]. 7 Sanders, Lecture. 8 Emanuel Tov, The Text-Critical Use of the Septuagint in Biblical Research, Jerusalem Biblical Studies, no. 3 (Jerusalem: Simor, 1981), 63; chapters 36-39, however, are relatively literal; ibid., 51. See also Isac Leo Seeligmann, The Septuagint Version of Isaiah: A Discussion of Its Problems (Leiden: E. J. Brill, 1948), passim. 9 In a discussion of the history of the indiscriminate use of textual variants, M. H. Goshen-

Gottstein remarks sardonically, “Textual critics were at liberty to plough through the collations [of Hebrew mss] and to dig up what they thought useful for their purpose. They have done so ever since”; M. H. Goshen-Gottstein, “Hebrew Biblical Manuscripts: Their History and Their Place in the HUBP Edition,” Biblica 48 (1967): 253.

3

Some scholars, while accepting the validity of textual criticism, believe that it should be done only after each of the distinct text-traditions has been examined from a literary standpoint for consistency and “fit” in the context.10 However, this approach seems to confuse the roles of textual critic, literary critic, and interpreter. Though he or she should consider the literary qualities (or lack thereof) and internal consistency of a textual witness, the textual critic is not primarily interested in an aesthetic evaluation of the various witnesses. The fact that a witness “makes sense” as it stands does not imply its originality, much less its independence from the stream of tradition.11 Moreover, the meaning of a particular reading as enshrined in a “real” text-tradition is not necessarily that of the archetype of the tradition. Many factors, both intentional and unintentional, can produce variant readings, each of which makes perfectly good sense in the context. Other scholars contend that one must begin with analysis of the text before literary analysis is of any value. For example, Emanuel Tov says, “It is absolutely mandatory to start the analysis with the textual evidence.”12 He points particularly to the necessity of determining a witness’s translation technique before evaluating its readings, and he sets forth five criteria for determining the relative literalness of the translation.13 Once the translation technique of each witness is determined, the probability of accurate retroversion into Hebrew (or other source language) can be determined to a large degree, and an approximate retroversion can be produced. Tov also discusses the problems of retroverting translated material.14 A final approach to textual criticism is the genealogical approach, which attempts to reconstruct a tree that depicts the origin of the different extant states of the text as found in the various witnesses. Vinton A. Dearing, an English professor who has edited the text of 10 Cf., e.g., David W. Gooding, “An Approach to the Literary and Textual Problems in the DavidGoliath Story,” in The Story of David and Goliath, by Dominique Barthélemy et al., Orbis Biblicus et Orientalis, no. 73 (Fribourg: Éditions Universitaires, 1986; Göttingen: Vandenhoeck & Ruprecht, 1986), 59-82; cf. p. 83, “The final question is one of textual criticism” (italics mine). See also Stanley D. Walters, “Hannah and Anna: The Greek and Hebrew Texts of 1 Samuel 1,” Journal of Biblical Literature 107 (1988): 385-412. 11 Contra S. D. Walters, “Hannah and Anna,” 410: “Internal coherence and consistency must have

at least equal weight with genetic consideration.” It is not a question of the value of this evidence, but of its relative importance in the text-critical task. Does LXX’s version of Samuel really represent a tradition entirely separate from that of MT, and was there no single original text behind the two traditions? (cf. pp. 410-11). 12 “The Story of David and Goliath in the MT and LXX,” in David and Goliath, by Barthélemy et al., 130. He details his arguments on pp. 130-34. 13 Tov, Text-Critical Use, 54-60. See below, chap. 1, for an elaboration. 14 Ibid., 97-158. For an analysis of some of the problems involved in retroverting one version of a biblical book back into the language of its Vorlage, see John Russiano Miles, Retroversion and Text Criticism: The Predictability of Syntax in an Ancient Translation from Greek to Ethiopic, Septuagint and Cognate Studies, no. 17 (Chico, CA: Scholars Press, 1985).

4

various English authors’ works, provides the most detailed explanation of this method.15 He also propounds an algorithm for determining the tree, and he discusses the use of computers in analyzing the data. The genealogical approach is extremely important in determining the origin and relationship of the various extant witnesses to classical texts and other works which exist in relatively few distinct states,16 and it is this approach which is advocated in books dealing with the textual criticism of classical works.17 The construction of stemmata (genealogical trees) for works which have hundreds or thousands of extant witnesses (such as NT mss), however, is highly problematical, and the issue of how to deal with translations (a vital matter when dealing with the witnesses to the OT) is not directly addressed by this model. In view of the diversity of approaches, it is clear that no unanimity exists concerning the proper methodology for the study of the text of the OT. Therefore, some model needs to be designed to evaluate accurately the various text-traditions. Each of these competing models will be evaluated further in the first chapter of the present study, and a working model will be developed which will then be applied to the data from 1 Samuel 3. No study of the text of Samuel would be possible without the benefit of critical texts of the various witnesses. In the seventeenth century Louis Cappel contributed the earliest study of the OT text in modern times, producing a list of variant readings and suggested emendations.18 In the following century, a fresh collation of all known Hebrew mss was produced by B. Kennicott and supplemented by G. B. de Rossi. In 1875, Frederick Field published the fragments of Origen’s Hexapla preserved in various mss, versions, and church fathers. 19 The latter half of the nineteenth century and first decades of the twentieth century also saw the maturation of Septuagintal textual studies in the work of Paul de Lagarde and Alfred Rahlfs. The two great modern editions of LXX are the Cambridge Septuagint (1906 onward, presented as a diplomatic text, usually based on Codex Vaticanus [B] or Codex Alexandrinus [A], with variants indicated in the lower margin) and the Göttingen Septuagint (1931 onward, presented as a critical, eclectic text, 15 Vinton Adams Dearing, Manual of Textual Analysis (Berkeley: University of California Press, 1959). His method is updated and revised extensively in his later book, Principles and Practice of Textual Analysis (Berkeley: University of California Press, 1974). 16 A “state” is defined as a distinct form of a text which differs from all other forms in some way.

Thus, in the modern era, an entire printing of a work, since all copies are identical, constitutes only one state. 17 For example, Maas, Textual Criticism. 18 Louis Cappel, Commentarii et notae criticae in Vetus Testamentum, Amsterdam: Jacobus

Cappellus, 1689. Cappel’s study deals with the text of the entire OT. 19 Benjamin Kennicott, ed., Vetus Testamentum Hebraicum cum variis lectionibus, 2 vols. (Oxford: Clarendon Press, 1776-80); Giovani Bernardo de Rossi, Variae lectiones Veteris Testamenti, 4 vols. (n.p.: Parma, 1784-88); Frederick Field, ed., Origenis Hexaplorum, 2 vols. (Oxford: Oxford University Press, 1875; repr., Hildesheim: Georg Olms, 1964).

5

attempting to reconstruct the earliest possible form of LXX). Neither is yet complete, but, between the two, most of the books of the OT are represented (Samuel appears in the Cambridge Septuagint).20 Rahlfs produced a complete manual edition of LXX, but it is not limited, like the Cambridge Septuagint, to the presentation of one ms, but rather presents an eclectic text based primarily on the three important uncials B, S (Codex Sinaiticus), and A. It is thus a kind of intermediary between the Cambridge and Göttingen philosophies.21 The Targum Jonathan (T) includes the targum to Samuel, and a critical edition has been produced by Alexander Sperber.22 The Peshit@ta Institute in Leiden is currently preparing a critical edition of the Syriac Peshitta (P), presented as a diplomatic text, based in Samuel on ms 7a1 (Codex Ambrosianus).23 The Benedictine Order is in the process of producing a critical edition of V, on which the manual edition of the Würtemberg Bible Society is based.24 The final primary or secondary witness (see below, p. 7) to the text of 1 Samuel is 4QSama, a ms from the Judean desert, published in preliminary form by Frank Moore Cross.25 Several studies dealing with the text of 1 Samuel in particular have appeared in the last hundred years or so. The commentaries by Otto Thenius, Julius Wellhausen, and Henry Preserved Smith in the nineteenth century, and by S. R. Driver, P. Kyle McCarter, 20 Alan England Brooke, Norman McLean, and Henry St. John Thackeray, eds., The Old Testament in Greek (London: Cambridge University Press, 1906-); J. Ziegler et al., eds., Septuaginta: Vetus Testamentum Graecum auctoritate Academiae Scientarum Gottingensis editum (Göttingen: Vandenhoeck & Ruprecht, 1931-). 21 Alfred Rahlfs, ed., Septuaginta: id est Vetus Testamentum graece iuxta LXX interpretes, Editio

Minor (Stuttgart: Deutsche Bibelgesellschaft, 1979), lv. Cf. Peter Walters, The Text of the Septuagint: Its Corruptions and Their Emendation, ed. David W. Gooding (Cambridge: Cambridge University Press, 1973), 11. It must be pointed out, however, that Rahlfs was one of the architects of the Göttingen edition and that his manual edition was intended as a preliminary edition for use by clergy and students; Rahlfs, Septuaginta, lv. 22 Alexander Sperber, ed., The Bible in Aramaic, 4 vols. in 5 parts (Leiden: E. J. Brill, 19591973). The abbreviation “T” will be used of the targums in general and of Targum Jonathan in particular when it is clear from the context to which it is a reference. In ambiguous contexts, the abbreviation will be avoided. 23 The Old Testament in Syriac According to the Peshi a Version, edited on behalf of the

International Organization for the Study of the Old Testament by the Peshit@t a Institute, Leiden (Leiden: E. J. Brill, 1972-). 24 The Benedictine edition is Biblia Sacra iuxta Latina Vulgatum versionem ad codicem fidem iussu Pii XII, ed. Cura et studio monachorum abbatiae pontificiae Sancti Hieronymi in urbe ordinis Sancti Benedicti (Rome: Typis Poliglottis Vaticanus, 1926-); it is referred to in the text as VR and in the footnotes as Biblia Sacra Romana. The manual edition is Bonifatius Fischer et al., eds., Biblia Sacra iuxta vulgatam versionem, 2d ed., 2 vols. (Stuttgart: Württembergische Bibelanstalt, 1975); it is referred to in the text as VS and in the footnotes as Biblia Sacra Stuttgartensia. The latter edition is more than a mere abbreviation of the larger work. The editors have examined the text of previous editions and their apparatuses, altering the text found in the larger work when deemed necessary. 25 Frank Moore Cross, “A New Qumran Biblical Fragment Related to the Original Hebrew

Underlying the Septuagint,” Bulletin of the American Schools of Oriental Research 132 (1953): 15-26.

6

and Ralph Klein in the twentieth century, deal extensively with the text.26 Eugene Ulrich’s study of the relationships among 4QSama, MT, LXX, and Josephus supports Cross’s theory of local texts as it applies to Samuel, detailing the general convergence of the texts of 4QSama, LXX, and Josephus as opposed to MT. Stephen Pisano’s 1982 doctoral dissertation deals with the additions and omissions in the Samuel texts of MT, LXX, and Qumran. Finally, the work of the Committee for the Textual Analysis of the Hebrew Old Testament in analyzing many significant variants in Samuel provides important data for the study of the text.27 In spite of the great amount of work done in the area of the text of 1 Samuel, however, it remains necessary to determine those textual variants that are truly significant for the recovery of the oldest possible Hebrew text. Developing a methodology by which to determine which variants are significant (Chapter 1), applying this methodology to 1 Samuel 3 (Chapters 2-5), and using the information generated to reconstruct the earliest possible text of 1 Samuel 3 (Chapters 6-8) will be the goals of this study. Definition of Terms For the purposes of this study, the term “significant variant” will refer to those variants which have some probability of representing a Hebrew Vorlage different from the base text, MT, where the variation is not purely orthographic.28 The choice of MT as the base text does not imply that it is superior to other witnesses; it is used as a base because it is the only complete witness in Hebrew. The term “(textual) witness” refers to an extant textual tradition, either complete or incomplete, whether Hebrew or some other language. It may refer to an individual ms, or it may be used to represent the presumed textual tradition lying behind a group of related mss (e.g., LXX or Lucianic mss [LXXL]). 26 Otto Thenius, Die Bücher Samuels, 2d ed., Kurzgefasstes Exegetisches Handbuch zum Alten

Testament, no. 4 (Leipzig: S. Hirzel, 1864); Julius Wellhausen, Der Text der Bücher Samuelis (Göttingen: Vandenhoeck & Ruprecht, 1871); Henry Preserved Smith, Critical and Exegetical Commentary on the Books of Samuel, International Critical Commentary (Edinburgh: T. & T. Clark, 1899); S. R. Driver, Notes on the Hebrew Text and the Topography of the Books of Samuel, 2d ed. (Oxford: Clarendon Press, 1913; repr., Oxford: Oxford University Press, 1960); P. Kyle McCarter, Jr., I Samuel, The Anchor Bible, ed. William F. Albright and David Noel Freedman, vol. 8 (Garden City, NY: Doubleday & Company, 1980); Ralph W. Klein, 1 Samuel, Word Biblical Commentary, vol. 10, gen. eds. David A. Hubbard and Glenn W. Barker, Old Testament editor John D. W. Watts (Waco: Word Books, 1983). 27 Eugene Charles Ulrich, Jr., The Qumran Text of Samuel and Josephus, Harvard Semitic Monographs, no. 19 (Missoula, MT: Scholars Press, 1978); Stephen Pisano, Additions or Omissions in the Books of Samuel: The Significant Pluses and Minuses in the Massoretic, LXX and Qumran Texts, Orbis Biblicus et Orientalis, no. 57 (Fribourg: Univeritätsverlag, 1984; Göttingen: Vandenhoeck & Ruprecht, 1984); Barthélemy, Critique textuelle, vol. 1. See also Frank Moore Cross, “The Evolution of a Theory of Local Texts,” in Qumran and the History of the Biblical Text, ed. Frank Moore Cross and Shemaryahu Talmon, pp. 306-20 (Cambridge: Harvard University Press, 1975). 28 Variants which are not significant will be termed “nonsignificant” rather than “insignificant,”

because of the broader connotations of the latter term. Some of the nonsignificant variants may not be insignificant in terms of understanding the meaning of the text or the historical development of the text.

7

In the latter case, it is equivalent to the term “text-tradition.” A “(textual) variant” within a tradition is a reading that deviates from that of another witness within the same tradition. Thus, the readings of the base text and of a divergent witness are both considered variants. More generally, where a unit of variation exists, one may speak of a variant as either the reading of MT or that which represents a Hebrew Vorlage different from MT. Variants that entail differences in length between various witnesses, ranging from one lexeme to many lexemes, are often called additions/omissions or plusses/minuses. However, this terminology is not satisfactory, because it implicitly gives preference to the base text, normally MT. In this study, all such variants will be termed “add-oms.” 29 The terms addition/omission and plus/minus, when used, will refer to deviations from the presumed archetype, not the base text. Those variants that do not differ significantly in length from the readings of other witnesses may be divided into substitutions and transpositions. All three of these types of variants are termed simple variants; complex variants are composed of any combination of simple variants. The witnesses to the text of Samuel may be divided into three groups.30 The “primary text-traditions” are those written in the original language, Hebrew (i.e., MT and Qumran mss). The “secondary text-traditions” are those translated directly from the Hebrew (i.e., LXX, P, T, V). “Tertiary text-traditions” are translations of secondary text-traditions (e.g., Old Latin [it], Coptic [cop], Armenian [arm], Ethiopic [eth], Syrohexapla [syh]). An attempt will be made to isolate all independent witnesses to a Hebrew text. For example, Lucianic witnesses, even when clearly secondary for determining the text of the Old Greek LXX, may be witnesses to an independent Hebrew textual tradition in places. Thus, all text-traditions which have been corrected at some point by a Hebrew text may also be considered secondary traditions at those points. Such texttraditions will be called “partial secondary witnesses.” Outline of This Study The first chapter of this study will examine in more detail the various models suggested above for doing textual criticism as they apply to the text of 1 Samuel. A model for textual analysis will be developed which will then be used to analyze 1 Samuel 3 in Hebrew, Greek, Syriac, Aramaic, and Latin. Chapter 2 will discuss an abbreviated method of reconstructing archetypical secondary witnesses for use in the following chapters. Chapters 3 and 4 will analyze the various secondary witnesses from 29 Adapted from Dearing, Principles and Practice, 25, where “add-omission” is used. 30 For diagrams detailing the relationships among the various witnesses, see F. E. Deist, Towards

the Text of the Old Testament, trans. W. K. Winckler (Pretoria: N.G. Kerkboekhandel Transvaal, 1978), 236; Ralph W. Klein, Textual Criticism of the Old Testament: The Septuagint after Qumran, Guides to Biblical Scholarship, Old Testament Series (Philadelphia: Fortress Press, 1974), 71; and Talmon, “Old Testament Text,” 195.

8

lexical/grammatical/stylistic and literary/theological points of view, respectively. Chapter 5 will briefly list other extant variants, gleaned from primary witnesses (variants in the Masoretic mss and in 4QSama), summarize the search for significant variants, and list all those that have been uncovered. The sixth chapter will discuss the methodology of retroversion from one language into another and will then attempt to retrovert, in whole or in part, the secondary and partial secondary witnesses into Hebrew. Chapter 7 will take all the Hebrew variants (both originally Hebrew and retroverted readings) that have been determined to be significant, evaluate them, and attempt to arrive at a conclusion concerning the earliest reading. Finally, Chapter 8 will present the results of the study in the form of a critical Hebrew text of 1 Samuel 3, including critical apparatuses.

CHAPTER 1 DEVELOPMENT OF A MODEL FOR ANALYSIS OF THE WITNESSES If one is to avoid an arbitrary or haphazard approach to textual criticism, it is necessary to have some sort of a model to follow. The problem with many attempts at textual analysis is that they are done without paying sufficient attention to matters such as translation technique, larger context, and literary characteristics of the witness. Such shortfalls are always a danger with “textual notes,” whether based on philological or textual data.1 To say this is not to deny the helpfulness of such observations in many cases but rather to remind scholars that brief, random observations must be tested with methodological soundness. The three approaches delineated in the Introduction will be discussed and evaluated in greater detail, and the insights gained from them will be used to prepare a method for evaluating the text of Samuel. Literary Analysis before Textual Analysis Those who believe that literary analysis2 should be done before textual analysis fall into two broad groups: those who do so for theological reasons and those who do so for purely literary reasons. A growing number of scholars is becoming interested in examining MT and other traditions independently of the others, viewing them as traditional texts that 1 Cf. James Barr, Comparative Philology and the Text of the Old Testament (Oxford: Oxford University Press, 1968; repr., with additions and corrections, Winona Lake, IN: Eisenbrauns, 1987), 304: “The basic assumption, that study of the relations between the Semitic languages may further the understanding of the Hebrew Bible, is incontrovertible. The trouble has not lain in comparative scholarship, but in poor judgement in its application, and in failure to see and follow out some of the general linguistic questions which are already implied in the primary use of comparative method.” Although referring only to the misuse of linguistic data, a similar statement could also be made of the misuse of textual data. Cf. also P. A. H. de Boer’s summary of H. S. Nyberg’s approach to the text: “The recension should first be determined, then the interpretation taken in hand and the amendment postponed as long as possible…”; P. A. H. de Boer, Research into the Text of I Samuel i-xvi: A Contribution to the Study of the Books of Samuel (Amsterdam: H. J. Paris, 1938), 9. 2 The term literary analysis as used in this context is usually not the older source criticism (though

the arguments are similar for this type of literary criticism) but rather an attempt to analyze the existing form of the text and discern its meaning(s). Scholars who employ such forms of literary criticism do so in a variety of ways, including reading the Bible as pure literature, structuralism, rhetorical criticism, close reading, and deconstruction. The issue here is not whether any or all of these methods are legitimate but rather whether the search for an earlier form of the text is best served by using literary analysis from the beginning. See, for example, David Robertson, The Old Testament and the Literary Critic, Guides to Biblical Scholarship, Old Testament Series (Philadelphia: Fortress Press, 1977), passim; Robert C. Culley, “Exploring New Directions,” in The Hebrew Bible and Its Modern Interpreters, ed. Douglas A. Knight and Gene M. Tucker (Philadelphia: Fortress Press, 1985; Decatur, GA: Scholars Press, 1985), 167-200.

9

10

were actually used by a worshiping community.3 An example of this approach may be seen in several recent translations of the Bible that, in the apocryphal or deuterocanonical section, render the entire book of Esther from the Greek rather than presenting only the pluses to the Hebrew in a confused and disjointed manner, as many earlier translations did. Implicit in this editorial decision is the recognition that the Greek form of Esther in its totality has served various communities, both past (Greek speaking Jews and Christians) and present (Roman Catholics and Eastern Orthodox), as sacred text.4 As mentioned above, the use of textual criticism to determine a text other than the oldest (i.e., one actually used by a specific community) is a legitimate modification of the more traditional goal of searching for the earliest text. Such a choice, though, is a theological, rather than a textcritical, choice.5 For this reason, the textual critic has no basis for contesting the prior choice of a particular literary tradition before textual criticism is done. However, some scholars, for nontheological reasons, believe that literary analysis should always precede textual analysis. In the process of examining the divergent traditions of the story of David and Goliath found in MT and LXX, D. W. Gooding demonstrates his preference for doing literary analysis before textual analysis.6 He particularly stresses the need to examine the extant form of a text to see what it says, implying that if it makes sense as it stands, textual modifications are unnecessary. Note, for example, the following arguments. 3 Sanders, Sacred Story, 134. Cf. Ferdinand E. Deist, “Is die Massoretiese Teks die Ou Testament?” Skrif en kerk 10 (1989): 9-20, especially 16-19. 4 The inclusion of two forms of Esther, however, is based primarily on the acceptance of the two

forms by various modern, rather than ancient, religious groups. If the use of different forms of a book by various ancient communities were included as a criterion for translating more than one form, variant forms of such books as Jeremiah, Proverbs, and Job would be included on the basis of the differences between the Hebrew and (Old) Greek texts, and two forms of Judges and Daniel (three forms?) would also be translated on the basis of greatly divergent Greek manuscripts, as Sanders has in fact suggested. 5 Others would claim that the presumed “original” text is the only one that carries divine authority, so it is the one that should be sought. This, too, is a theological choice. (The textual critic should avoid the equation “original text” = “correct text”; cf. Peter D. Miscall, 1 Samuel: A Literary Reading, Indiana Studies in Biblical Literature [Bloomington: Indiana University Press, 1986], viii.) However, the search for earlier forms of the text need not be theologically motivated; it can also be a purely text-critical exercise, unfettered by theological presumptions of the superiority of earlier forms of the text. Because textual criticism is not a theological discipline, it cannot tell the contemporary community which ancient witness, extant or reconstructed, is best for modern translation. 6 Gooding, “Approach to the Problems,” 63: “Let us consider first the context of the book as a

whole.” Twenty pages later he says, “The final question is one of textual criticism,” ibid., p. 83. Cf. also the approach of Dominique Barthélemy, “Trois niveaux d’analyse (a propos de David et Goliath),” in David and Goliath, by Barthélemy et al., 47, who begins his detailed discussion of the story by saying, Si nous envisageons ces trois récits [i.e., 1 Sam 16:1-13 (A); 16:14-23 (B); 17:12ff. (C)] comme des essais de légitimation des prétentions de David à la royauté, le récit A est celui qui atteint ce résultat de la manière la plus efficace. Si les auteurs des récits B et C l’avaient connu, ils auraient pu renoncer à leurs essais moins convaincants. Il est donc vraisemblable que A est le dernier-né des trois récits. On a de cela un indice complémentaire dans le fait que les autres récits ignorent cette onction de David enfant.

11 [Critics should not try] to claim to prove irreconcilable discrepancy, and therefore multiple authorship, by selecting a detailed feature from one context, contrasting it with a detailed feature from another context, without first carefully examining the place each feature holds and the function it performs within the thought-flow of its own particular movement. . . . To understand a narrative from a literary point of view, we must first listen to the narrative as it stands, trying to see where and how each part fits into the thought-flow of the whole. In other words we must initially give the narrative the benefit of the doubt. . . . Only after a sustained and sympathetic attempt at making sense of the narrative as it stands should we reluctantly conclude that the narrative is an irreconcilably discrepant hotch-potch.7 Several points may be made in response. First, it must be said that scholars like Gooding who are once again examining the final form of a biblical text to see what its message is as it stands are filling a needed and much neglected void in earlier critical scholarship, which tended to focus on earlier, hypothetical forms of the text and ignore the present form. Having said that, however, the question remains: should this examination of the final form affect text-critical judgments? Just because the final form of a text “makes sense” as it stands, does this imply that it is the original form?8 To say that it does is to agree that any authors or redactors that might have used earlier sources must indeed have used a “scissors and paste” approach in their composition. If, in fact, the author of the final form was truly an author in some sense, it is quite possible that he or she would have smoothed over more obvious difficulties or discrepancies. Thus, an understandable final form does not imply either verbal proximity to the original form or unitary authorship. Second, while it is true that critics in the past have sometimes exaggerated differences of detail between different parts of a biblical text, it remains true that discrepancies and notable differences of presentation do provide both textual and literary critics with clues pointing to deviations from a unitary tradition, whether textual or literary. The phrases “benefit of the doubt,” “sympathetic attempt,” and “reluctantly conclude” seem more appropriately applied to persons standing trial than to critical evaluation of textual data.9 While one must guard against assuming multiple authorship or textual disruption before examining the data, so must one avoid either simplistic or contrived explanations of apparently incongruous data that unfairly prejudice the case in favor of 7 Gooding, “Approach to the Problems,” 58. 8 Cf., e.g., the approach of Gnuse to “retain the Massoretic text when it seems logical”; Robert

Karl Gnuse, The Dream Theophany of Samuel: Its Structure in Relation to Ancient Near Eastern Dreams and Its Theological Significance (Lanham, MD: University Press of America, 1984), 121. See also Stoebe, Erste Buch Samuelis, 25-32; P. A. H. de Boer, I Samuel i-xvi, passim. De Boer also produced the textual notes for Samuel in BHS. 9 Indeed, conservative critics have often used the Anglo-American legal tradition of “innocent until proven guilty” as an analogy to suggest that scholars should respect traditional views, sometimes including acceptance of MT readings, unless overwhelming evidence proves them to be untenable (e.g., Robert Dick Wilson, A Scientific Investigation of the Old Testament [Philadelphia: The Sunday School Times Company, 1926], 19-20). However, the reason for the legal dictum is protection of the innocent, even if injustice is sometimes the result. Since no one is at risk of injury in scholarly discussions of the text, favoritism of traditional views is unwarranted. Of course, Gooding’s views of the development of the text do not correspond to those of these conservative critics, but his use of language at this point is similar.

12

unitary authorship or faithfulness to the textual tradition. Third, when Gooding says, “To understand a narrative from a literary point of view, we must first listen to the narrative as it stands,” he makes a valid point. The exegete should examine the final form of a text. However, though the textual critic does do exegesis, to interpret is not the goal of his work but rather a means to an end, namely, discovering an earlier form of the text. An important difference exists between examining a text “from a literary point of view” and doing so from a textual point of view. The two must not be confused. Stanley D. Walters also believes that literary analysis is primary, and he uses an excerpt from 1 Samuel 1 as an example. His primary contention is that the narratives of Samuel’s birth as recorded in MT and LXX (specifically, ms B) are “discrete narratives, each with its own Tendenz.” 10 He interprets the two versions of the story as they stand, and he compares them with other biblical texts with similar vocabulary or motifs. He notes that “it is a signal advantage of the comparative method that we can see features in each story that could otherwise easily have been missed.”11 The proposed interpretations are interesting, even ingenious at times. However, his understanding of the text-critical task requires scrutiny. He contends that scholars from the time of Thenius have often tried to rewrite the text rather than explain it, and the result has been the creation of various forms of the text that probably never existed. Even the discovery of the Qumran mss, revealing the existence of Hebrew texts closer to LXX than MT, has confused rather than helped the situation. In light of this situation, he says, “we must return to the commentator’s primary task, that of explaining the text.”12 It is at this point that an important distinction must be made. While it may be that a commentator’s primary task is to explain the text as it is (though this assertion in itself is debatable), it is certainly not the textual critic’s task. Again, though the textual critic may in fact create a text that never existed in all its details, it is the nature of the text-critical task to reconstruct hypothetical texts. The amount and nature of the data will determine the chances of success in each case, but the discovery of more material (such as the Qumran mss) enhances rather than hinders the attempt. Literary analysis is important for the text-critical task, as well as for the exegetical task. Examination of the larger literary context may well clear up apparent discrepancies or reveal certain stylistic or theological tendencies in a particular witness or within the tradition as a whole. Nevertheless, it seems clear that for the textual critic, literary analysis must accompany, not supersede or supplant, a thorough examination of all existing textual data.13 10 S. D. Walters, “Hannah and Anna,” 409. 11 Ibid., 410. 12 Ibid. 13 Cf. also Miscall, who says that “text-critical study should be preceded by extended readings,

13 Textual Analysis before Literary Analysis In The Story of David and Goliath, both Johan Lust and Emanuel Tov say that textual analysis is primary. Lust, in his response to the initial papers by the other contributors, states this point clearly. While agreeing that textual and literary analysis are both important, he says, A close reading of the four papers strengthened my conviction that the data and questions of textual criticism should be handled first. To a certain extent, they present facts, providing a sound basis for further more hypothetical theories. The differences between the Hebrew and the Greek texts should be noted carefully, including both pluses and minuses. . . . Once these data [Hebrew and Greek mss, Qumran, patristic evidence] are listed and checked, one should ask whether the difference between the Greek and the Hebrew texts are intentional or unintentional and whether they are due to a scribe, an editor or a translator.14 Tov likewise affirms the necessity of examining the textual data first: For the story of David and Goliath we possess two main sources, MT and the LXX, and these are by implication textual sources, witnessing that abstract entity which we call the biblical text. In our discussion we must attempt to approach these sources as impartially as possible. This means by implication that we should not speak first about the literary problems of the story nor about its textual difficulties, but we should first assess the value of the LXX because the LXX is one of the sources which contain the story. . . . The starting point of the discussion of the Greek version of David and Goliath should be the translation technique of the Greek translator. . . .15 Thus, both Lust and Tov stress the importance of beginning textual criticism with an examination of the text. Tov particularly stresses the importance of determining the translation technique of LXX. It is at this point that a comparison between the text-critical methodology employed in the OT and the NT will be helpful. It has already been noted that many Greek mss the NT exist that were written within the first few centuries of the original compositions.16 In contrast, few if any Hebrew mss date from so close a time to the originals, and the vast majority come from the medieval period, more than one thousand years after composition. Perhaps more importantly, the thousands of mss of the Greek NT contain a rich variety of readings in almost every verse, and at least two, perhaps three or four, major text types can be distinguished among the mss, in addition to numerous other wherever possible, of each version”; Miscall, 1 Samuel, viii. 14 Johan Lust, “Second Thoughts on David and Goliath,” in David and Goliath, by Barthélemy et

al., 87. 15 Emanuel Tov, “Response by E. Tov,” in David and Goliath, by Barthélemy et al., 92-93 (italics original). Cf. also Natalio Fernández Marcos, Introducción a las versiones griegas de la biblia, Textos y estudios “Cardinal Cisneros”, no. 23 (Madrid: Instituto “Arias Montano” Consejo Superior de Investigaciones Científicas, 1979), 25: “El análysis de las técnicas de traducción de cada libro o de cada unidad de traducción ha de preceder a todo estudio de sintaxis.” 16 Kurt and Barbara Aland list forty-five NT mss that date prior to the fourth century; Kurt Aland

and Barbara Aland, The Text of the New Testament, trans. Erroll F. Rhodes (Grand Rapids: William B. Eerdmans, 1987; Leiden: E. J. Brill, 1987), 57.

14

early readings.17 Medieval Hebrew mss, on the other hand, represent one text type and contain almost no variants that might reflect a competing text.18 OT textual criticism, then, is much more dependent on witnesses that are translations than is NT textual criticism. Whereas NT textual critics can almost ignore the versional evidence, since almost every variant within a version with any claim to originality also occurs in at least one Greek ms,19 OT textual critics cannot. In fact, versions of the OT, especially LXX, have long been used to provide textual evidence, though not always in a consistent manner. One desideratum of developing a methodology for OT textual criticism is to determine how to use versional alongside Hebrew evidence. Though NT textual critics have employed various approaches in the past, the majority today use one of two methods: rigorous eclecticism or rational (modified) eclecticism.20 Rigorous eclecticism, as employed, for example, by G. D. Kilpatrick and J. K. Elliott, focuses exclusively on internal evidence in evaluating variant readings. External evidence such as age, geographical setting, quantity of witnesses, and distribution among text types are considered of little or no importance in determining the correct reading. Instead, the critic examines the author’s style and vocabulary to determine what he was more likely to have said. He or she also attempts to discern among the readings which one was most likely to have given rise to the others.21 The scholar who uses rational eclecticism does consider internal evidence, but he or she also takes into 17 Aland and Aland, Text of NT, 56-67; Bruce M. Metzger, The Text of the New Testament: Its Transmission, Corruption, and Restoration, 2d ed. (New York: Oxford University Press, 1968), 211-19. Cf. also Bruce M. Metzger, A Textual Commentary on the Greek New Testament, corrected ed. (London: United Bible Societies, 1975), xv-xx. 18 Goshen-Gottstein, “Hebrew Biblical Manuscripts,” 287. He identifies Kennicott’s manuscripts

70, 89, 174, and 187 as having several readings that differ from the main stream of MT in 1 Samuel. The variations between the Ben Asher and the Ben Naphtali mss are minimal (only eight variations in the consonantal text); cf. Ernst Würthwein, The Text of the Old Testament: An Introduction to the Biblia Hebraica, trans. Erroll F. Rhodes (Grand Rapids: William B. Eerdmans, 1979), 24-25. 19 Cf. the statement of the Alands: “It must be emphasized that the value of the early versions for establishing the original Greek text and for the history of the text has frequently been misconceived, i.e., they have been considerably overrated”; Aland and Aland, Text of NT, 182. They say later on the same page that the versions are useful in “confirming the identity of the regional or provincial text where they were produced,” but this is mere confirmation in most cases of the text found in Greek mss. See also p. 275, “The primary authority for a critical textual decision lies with the Greek manuscript tradition, with the versions and the Fathers serving no more than a supplementary and corroborative function . . . .” (italics mine). Even those scholars who might see somewhat more text-critical value in the NT versions would probably still attribute far less importance to them than most OT textual critics do to LXX or even P. 20 For a brief overview of these methods, see James A. Brooks, “The Text of the New Testament

and Biblical Authority,” Southwestern Journal of Theology 34 (1992): 19-20. 21 “Each reading must be judged on its intrinsic merits and not by weighing manuscripts, whatever we may mean by weighing”; G. D. Kilpatrick, “Literary Fashion and the Transmission of Texts in the Graeco-Roman World,” chap. in The Principles and Practice of New Testament Textual Criticism: Collected Essays of G. D. Kilpatrick, ed. J. K. Elliott (Leuven: University Press, 1990), 63. Cf. also other chapters in the same volume: “Atticism and the Text of the Greek New Testament” and “Eclecticism and Atticism.”

15

consideration the external factors mentioned above. When external and internal evidence conflict, the scholar may choose to follow the lead of either the one or the other, depending on the text and the variants in question (though it is often the external evidence that predominates). This approach is that used by the Alands, Metzger, and many others.22 It is a compromise between the exclusive use of internal evidence in rigorous eclecticism and the exclusive use of external evidence in genealogical methods. Whereas most NT textual critics use a form of rational eclecticism, often stressing external evidence over internal evidence, many OT textual critics tend to put more stress on the internal evidence. Various reasons for this preference for internal criteria may be offered. First, OT witnesses have not been divided into distinct text types that are recognized by all scholars in the field. It would be particularly useful if all the mss of LXX were grouped by text type.23 Second, the age and textual history of some witnesses, particularly P, are disputed, so scholars’ opinions of the value of the witnesses vary. Third, since they have not been reared in the school of Westcott and Hort, OT textual critics have less tradition pulling them toward an extensive use of external criteria. One group of OT textual critics that puts more emphasis on external evidence than others is that heavily influenced by Cross’s local text theory.24 Cross’s theory has influenced many other OT textual critics to a lesser extent, so that they acknowledge some of Cross’s arguments, but are not convinced of all the details.25 22 Rational eclecticism is based largely on the work of Westcott and Hort (Brooke Foss Westcott and Fenton John Anthony Hort, The New Testament in the Original Greek [New York: Harper & Bros., 1882]) and the local text theory of B. H. Streeter (B. H. Streeter, The Four Gospels: A Study of Origins [London: Macmillan & Co., 1924]). A summary of this approach may be found, with some variations, in Metzger, Text of NT, 209-11, and in Aland and Aland, Text of NT, 275-77. Note particularly the Alands’ method of relating external and internal evidence (excerpted from a list of twelve basic rules of textual criticism) (Aland and Aland, Text of NT, 275):

2. Only the reading which best satisfies the requirements of both external and internal criteria can be original. 3. Criticism of the text must always begin from the evidence of the manuscript tradition and only afterward turn to a consideration of internal criteria. 4. Internal criteria . . . can never be the sole basis for a critical decision, especially when it stands in opposition to the external evidence. 23 Cf. the classification of Samuel LXX mss offered by Bo Johnson, Die hexaplarische Rezension des 1. Samuelbuches der Septuaginta, Studia Theologica Lundensia, no. 22 (Lund: CWK Gleerup, 1963), 19. 24 Frank Moore Cross, “Theory of Local Texts.” Cross’s theory is itself an amplification of

previous work, particularly that of Dominique Barthélemy, “Redécouverte d’un chaînon manquant de l’histoire de la Septante,” Revue biblique 60 (1953): 18-29; idem, Les devanciers d’Aquila, Supplements to Vetus Testamentum, no. 10 (Leiden: E. J. Brill, 1963). See also R. W. Klein, Textual Criticism, 69-73; Ulrich, Qumran Text of Samuel, 4-9. 25 See, for example, Talmon, “Textual Study of the Bible,” 323: “[Cross’s] theory, which is yet in the process of being revised and further refined by Cross and his students, appears to have attracted surprisingly little comment either from European scholars, with few exceptions, or for that matter, Israeli students of the Bible text.” Cf. also Emanuel Tov, “A Modern Textual Outlook Based on the Qumran Scrolls,” Hebrew Union College Annual 53 (1982): 11-27, who argues for a scenario more complicated

16

One other area in which the practice of NT textual criticism might shed some light on OT methodology is the statistical analysis of the data. Many NT scholars advocate the use of statistical analysis to delineate text types and smaller groups of related texts within text types. One of the oldest approaches is called the Claremont Profile Method, a method which uses selected passages within a given corpus to test similarity of readings among mss. The goal of this approach is to generate a group profile, that is, a list of readings characteristic to one particular group of mss (shared by at least two-thirds of them). An alternative, though similar, method appears in the work of Larry Hurtado on the so-called Caesarean text. Yet another variation of this theme is found in the dissertation of Mark Dunn.26 The importance for OT textual criticism of these attempts to classify the witnesses should be apparent. Proposed text types, families, and groups of witnesses to the OT text are based more on random observation than on rigorous statistical analysis of the data. Such methods as these could easily be applied to witnesses in a single language (e.g., to Greek witnesses or Syriac witnesses). No method has yet been developed to apply these methods across the language barrier, but such a method is needed. The advantage of statistical studies like these is that OT textual critics would be able to use external evidence with a greater degree of certainty concerning the relationship of witnesses and their relative value in determining earlier stages of the text.27 In conclusion, it is crucial for the textual critic to engage in textual analysis prior to literary or theological analysis. This analysis should ideally include a clear understanding of both the textual history of each witness (external evidence) and the translation technique of each versional witness (internal evidence). After textual analysis has been done, further internal evidence may be gathered from literary and theological analyses of the sources. Genealogical Method Before describing the proposed model for analysis that will be used in this study, a brief discussion of another method is in order. Textual criticism of classical texts arose during the Renaissance. Since relatively few mss of most classical works are extant, it is usually possible to construct a stemma (or tree), graphically showing the relationships of each extant ms to all the others and to the archetype. Reconstruction of the archetype than Cross’s local text theory. 26 Paul Robert McReynolds, “The Claremont Profile Method and the Grouping of Byzantine New

Testament Manuscripts,” Ph.D. diss., Claremont Graduate School, 1968, 3-8; Larry W. Hurtado, TextCritical Methodology and the Pre-Caesarean Text: Codex W in the Gospel of Mark, Studies and Documents, no. 43 (Grand Rapids: William B. Eerdmans, 1981), 10-12; Mark R. Dunn, “An Examination of the Textual Character of Codex Ephraemi Rescriptus (C, 04) in the Four Gospels,” Ph.D. diss., Southwestern Baptist Theological Seminary, 1990. 27 For a more detailed comparison of OT and NT textual criticism, as commonly practiced, see James R. Adair, “Old and New in Textual Criticism: Similarities, Differences, and Prospects for Cooperation,” TC: A Journal of Biblical Textual Criticism 1 (1996).

17

requires three stages: constructing the stemma (recensio), examining the readings presumed for the archetype to see if they may all be considered original (examinatio), and attempting to reconstruct original readings where they no longer exist (divinatio). 28 The stemma is constructed by comparing extant witnesses to one another and isolating shared readings that diverge from readings of the other mss. Trivial divergences that could have been corrected easily by an intelligent scribe (e.g., misspellings or nonsensical errors in word division) are ignored. Once extant mss are grouped into the smallest possible groups, common ancestors, or hyparchetypes, are constructed by comparing the readings of all the independent witnesses to that hyparchetype. If only two mss are descended from a hyparchetype, the reading of the hyparchetype for each unit of variation must be chosen on some grounds (usually internal evidence), or, if neither seems original, a conjecture may be suggested. If three or more mss are descended from a hyparchetype, the reading shared by more than one independent ms, or by one ms and the hyparchetype in the adjacent branch of the tree, is the reading of the hyparchetype (this rule is based on the assumption that, other than trivial or purely mechanical errors, two scribes would not make the same error in any particular place). The process is continued up the tree until the archetype itself is reconstructed.29 It is clear that this method makes several assumptions, not all of which may be valid for the textual criticism of the OT. One assumption that should at least be questioned is that two independent copyists would not make the same non-trivial or non-mechanical error. This assumption does sound intuitively likely, but the possibility of two scribes making the same error cannot absolutely be ruled out. Moreover, the identification of trivial or purely mechanical errors is not as straightforward as it might seem. Is the presence or absence of an article or a conjunction trivial? Is a change in tense or number trivial? Also, some purely mechanical errors may in fact be of a sort that two scribes would not make independently. A more troubling assumption is the idea that a ms was copied from only one other ms. The fact of the matter is that scribes copying OT mss frequently used more than one exemplar when creating a new ms. At other times, scribes would correct a ms, either their own or an older one, from a ms different from the original exemplar. But using more than one exemplar for a ms destroys the simple tree concept and seriously complicates the methodology for determining the relationships among the mss. Clearly the simple genealogical method that works so well with a few classical mss needs revision in order to accommodate the much more complicated textual history of OT mss. Probably the most thorough attempt to account for the possibility of contamination in the simple tree is the work of Vinton A. Dearing. Dearing’s method is rigorous, 28 Maas, Textual Criticism, 1-19. 29 Ibid., 2-9. See especially the diagram on p. 5.

18

allowing little room for subjective selection among variants.30 His method has seven steps: (1) decide on the states31 to analyze; (2) decide on variations to analyze; (3) rewrite any variations that would introduce rings into the tree; (4) find any terminal groups not in the simple variations; (5) connect the terminal groups; (6) locate the archetype; (7) emend the archetype, if necessary.32 This fascinating and thorough approach is quite precise in its methodology, covering every possible textual variation and leaving little room for human variation in its application. Nevertheless, it is open to criticism, both on the philosophical level and on the level of applicability to the text of the OT. M. P. Weitzman begins his critique of Dearing’s book, Principles and Practice of Textual Analysis, with a quotation from A. E. Housman, a textual critic of the early twentieth century: “Textual criticism is not a branch of mathematics, nor indeed an exact science at all.”33 Immediately, then, a philosophical difference emerges: is it even possible to use methods such as Dearing proposes to do textual criticism? In fact, Dearing’s method is not nearly as exact as one might be led to suppose from some of his statements. For example, he lists at least twenty-five rules for recognizing what he calls “directional variants,” variants that indicate which state explains the origin of the others. These rules, however, are of varying degrees of certainty, as he himself admits, with some (“harsher reading,” “reading that smells of a gloss”) quite subjective.34 The fact that the genealogical method works well with works extant in much smaller numbers than OT mss indicates that it could very well be helpful in OT textual criticism as well. Thus, to deny the genealogical approach a hearing on the grounds that textual criticism is an art rather than a science is not justified. A second critique of Weitzman’s, namely, that Dearing’s method is too complex, even though he provides a sample computer program for analyzing the data, may have had some validity in 1977. However, advances in technology greatly reduce the force of this argument. With the help of an algorithm, which Dearing provides, one can easily design a 30 Dearing, Principles and Practice, 83: “Textual analysis, having absolute rules, is not an art.”

Cf. also p. 58, “Textual analysis is a logic engine, like a computer.” He also gives three advantages of the genealogical method over the eclectic method: (1) in the genealogical method, a fact about the text (i.e., the genealogy) is determined; (2) in most cases, the preferable variant is determined without resort to eclectic methods; (3) it is possible to reconstruct other states of the text besides the archetype (ibid., 8). He goes so far as to say that the textual analyst describes not what was (the actual transmission of mss) but what is (the relationship among the extant states of the text), so he cannot be wrong (ibid., 19). 31 For a definition of a state of the text, see above, p. 4, n. 16. 32 Dearing, Principles and Practice, 21, where these steps are listed. He elaborates on each of them

in great detail on pp. 21-83. 33 A. E. Housman, “The Application of Thought to Textual Criticism,” chap. in Selected Prose, ed. J. Carter (Cambridge: Oxford University Press, 1961), 132; quoted in M. P. Weitzman, review of Principles and Practice of Textual Analysis, by Vinton A. Dearing, in Vetus Testamentum 27 (1977), 225. 34 Dearing, Principles and Practice, 44-54. Of course, the seventh step, emending the archetype, is

highly subjective, but it lies outside of textual analysis proper (as he defines it), steps three through six.

19

computer program to do the work of comparing the states and generating a tree (steps three through six). The most time consuming part of the process, once the program itself is written, is data input; but considering that Dearing and Weitzman were considering using punch cards (before the advent of the affordable personal computer), today’s data input is much faster. Still faster methods of data input may be achieved through the use of optical scanners, touch-sensitive screens or mouse technology, and improved data entry algorithms. However, data entry will probably remain the most difficult part of the process for the foreseeable future. A more telling criticism, indeed, the argument that alone threatens the entire validity of Dearing’s method, is the observation that the genealogical method deals with the transmission of the text as though each state were copied from one and only one exemplar, without contamination from others. Weitzman says, If one were dealing with a textual tradition of which each ms (except, of course, the original) had been copied from a single exemplar, without the scribe introducing any reading from a subsidiary source (e.g. by consulting a second ms, or by adopting variants which may have been recorded in the margins of his exemplar), these tasks would be simple enough. . . . But in a textual tradition wherein subsidiary sources were utilised to a significant extent, albeit by no more than a minority of the copyists involved in the whole process of transmission, all neat and exact methods have been found, at least until now, to break down.35 Dearing, of course, is aware that contamination can occur in the process of transmission, and his third step, rewriting the variations that would introduce rings, is designed to deal with this problem. Rings occur when two mss share readings not found in every intervening ms in the tree. Dearing’s solution to this problem is simply to discard some of the conflicting readings by treating them as through they did not conflict. After analyzing the relationships among the variants and the mss, he breaks the ring at the point of its weakest link, where there are the fewest agreements between different states. In fact, he breaks all the links if more than one are equally the weakest. He justifies this action by saying, “Breaking the weakest instead of some other standard connection reflects the principle of parsimony. The fewest possible readings are treated as something different from what they really are.”36 He further differentiates the work of the textual critic from that of the bibliographer. Because the textual critic analyzes extant states of the text 35 Weitzman, review, 226-27 (italics original). Weitzman downplays the amount of contamination in OT mss, but in fact, contamination from other branches of the genealogical tree is quite common, especially among LXX mss. In particular, hexaplaric readings, introduced into Greek mss from Hebrew mss, infect a large number of LXX mss, even those whose scribes used only one (hexaplaric) exemplar. 36 Dearing, Principles and Practice, 88. On pp. 16-19, Dearing attempts to assume the mentality

of a scribe copying a text and using more than one source, claiming that, from the scribe’s point of view, the result is not conflation. He supports this conclusion by distinguishing between constructing a genealogy of the states of a text (the work of a textual analyst) and constructing a genealogy of the records themselves (the work of a bibliographer). He says, “In short, from the point of view of one who deals with states of a message instead of with its records, and the textual analyst is such a one, conflation does not exist at all” (p. 17).

20

(results) and the bibliographer with the presumed transmission of the text (causes), the textual critic defines the weakest agreement between states in terms of number of agreements, whereas the bibliographer defines the weakest agreement as that having the least “weight,” or intrinsic probability.37 His arguments, however, fail to convince that his method yields accurate results, either in archetypical readings or in accurate trees. Weitzman constructs four hypothetical mss descended from a lost archetype, introducing a degree of contamination. After applying all of Dearing’s rules to the hypothetical mss, he arrives at the correct solution in 56% of the cases and at the wrong solution in 44% of the cases.38 These results challenge Dearing’s contention that textual analysis can be done mechanically, with little or no critical judgment. In Dearing’s response to Weitzman, he points out that, given the data, the tree predicted by his method explains the data just as well as the tree presupposed by Weitzman. Less convincing, however, is his contention that the failure of his method to predict the correct archetypical readings in 44% of the cases is no reflection on the usefulness of the method. He says, With real problems, one either knows the facts in some way, in which case no method of reconstructing them is required, or one doesn’t, in which case the best one can do is reason correctly from all the evidence, on the basis of axioms that seem (and one hopes will also seem to one’s fellow-critics) to be satisfactory.39 However, contrary to Dearing’s opinion, the results of a method indicate whether the theory on which it is based is true, not vice versa. Of course, if the method yields results that are better than other methods, even though not perfect, it may still be preferred. The application of traditional methods to the hypothetical data, however, belies Dearing’s claim that his method is superior. Without attempting to construct a genealogical tree, a textual critic using traditional methods of subjective selection based on internal evidence (i.e., eclecticism) would select 25 correct readings (the “x errors”), where the critic could determine on the basis of comparing the readings to one another that one was original. Of the other 25 errors, assuming a 50% chance of picking the right reading from two equally possible readings, the critic would select 12 correct readings, yielding a total of 37 correct readings out of 50, or 74%, a result considerably better than Dearing’s 56%. Consequently, Dearing’s method cannot be expected to be helpful in evaluating the text of the OT. Unless or until an improvement is made in the genealogical method that will deal more effectively with the problem of contamination, it must be set aside. 37 Ibid., 89. 38 Weitzman, review, 231. An alternate interpretation of Dearing’s rules yields the following results: correct 44%, incorrect 36%, indeterminate 20%. The errors are of two types: “x errors” that are clearly not original (and can be determined not to be so on the basis of internal evidence) and “+ errors” that cannot easily be distinguished from the correct readings. 39 Vinton Adams Dearing, “Textual Analysis: A Consideration of Some Questions Raised by

M. P. Weitzman,” Vetus Testamentum 29 (1979): 358-59.

21 Determining the Translation Technique Having examined various proposed methodologies for OT textual criticism, the time has come for developing the methodology to be used in evaluating the text of 1 Samuel 3. To summarize the previous discussion, both textual analysis and literary/theological analysis are important for determining earlier forms of the text, but textual analysis should have priority. Since no genealogical approach has yet been developed that is helpful in the textual criticism of the OT, it can have no place in the present methodology. Also, no particular theory of the relationship among the various extant witnesses (e.g., Cross’s local text theory) will be assumed. Tov has repeatedly insisted that the first step of textual criticism, at least as far as the versions are concerned, is the determination of the translation technique,40 and the methodology to be tested in this study follows Tov in this regard. Scholars often characterize a particular translation with terms such as “literal,” “fairly literal,” “free,” and so forth. While these labels may characterize certain aspects of the translation, they are not sufficiently nuanced to be of much help to the textual critic who wants to know how the translator rendered certain types of constructions in order to use that information to determine whether a particular reading is likely to reflect a different Vorlage. After all, it is entirely possible that a translator could render every Hebrew word with a single word in the target language, yet be relatively indifferent about word order. Would such a translation be literal or not? The answer is that it would be literal in regard to lexical choice but not in regard to word order. The need to push beyond such general terms as “literal” and “free” is apparent.41 In order to quantify these terms and to provide information that is more useful to the textual critic, a methodology consisting of two major phases is proposed. The first major phase of the methodology involves an analysis of the lexical and grammatical characteristics of each witness in order to understand better the translation technique. Prior to analyzing these characteristics, however, each target language must be examined to determine its limitations for rendering Hebrew.42 If these limitations are not 40 Tov, “Response,” 92-93; idem, Text-Critical Use, 50-52; idem, “The Composition of 1 Samuel 16-18 in the Light of the Septuagint Version,” in Empirical Models for Biblical Criticism, ed. Jeffrey H. Tigay (Philadelphia: University of Pennsylvania Press, 1985), 100-102. 41 Cf. Benjamin G. Wright, “The Quantitative Representation of Elements: Evaluating

‘Literalism’ in the LXX,” in VI Congress of the International Organization for Septuagint and Cognate Studies, ed. Claude E. Cox (Atlanta: Scholars Press, 1987), 312-13. 42 Cf. J. W. Wevers, “The Use of Versions for Text Criticism: The Septuagint,” in La Septuaginta en la investigación contemporánea (V congreso de la IOSCS), ed. Natalio Fernández Marcos, Textos y estudios “Cardinal Cisneros” de la Biblia Políglota Matritense, no. 34 (Madrid: Instituto “Arias Montano” Consejo Superior de Investigaciones Científicas, 1985), 15: “Before a version can be used textcritically the nature and limitations of the target language in contrast to those of the source language must be thoroughly understood”; see also p. 20: “Before a translation can be properly used in the text criticism of the text of a source language, one must fully understand just how and from what points of view this translation was done by a particular translator.” Models for such an analysis of the various languages may be found in the various articles found in Fernández Marcos, La Septuaginta, and in Bruce M. Metzger,

22

recognized, some differences in representation may wrongly be thought to be deviations from a literal rendering. Grammatical impossibilities (e.g., a Latin article), grammatical improbabilities (e.g., rendering the Hebrew in Greek), and stylistic characteristics of the target language (e.g., the Syriac anticipatory pronoun) must all be noted. Any rendering in the target language that meets standards for literalness that the translator may have thought to be imposed by the language itself (or at least by proper use of the language) should not be counted against the literalness of the translation, though all such differences should be noted. Tov suggests four criteria for determining the relative literalness of a particular witness’s translation technique: consistency, the representation of the constituents of compound words in the source language by individual equivalents in the target language, word order, and quantitative representation.43 Each of these criteria will be examined in turn, and certain modifications or amplifications will be suggested. The result will be a somewhat more detailed and nuanced approach to determining the translation technique. By consistency, Tov refers to the translators’ tendency to render every occurrence of a given word in the source language by the same word in the target language. This tendency toward consistent representation of words is called stereotyping, and Tov notes that stereotyping was probably the rule rather than the exception from the beginning. However, he also says, “It has yet to be examined which types of words and elements were rendered stereotypically, in which books and under what circumstances.”44 It is one of the goals of this study to undertake just that examination for 1 Samuel 3. One type of consistent representation commonly found in the secondary witnesses in question is Hebraism. Hebraisms are renderings in the target language that are unusual or even unidiomatic, but occur because the translator has attempted to render his Vorlage literally. Because they preserve some structure of the source language not usually found in the target language, Hebraisms can be useful tools for reconstructing the Vorlage. However, it was also possible for the translators to have written Hebraisms that did not reflect a particular text before them, simply because they thought or spoke the target language in terms of structures from the source language.45 Another type of consistent representation that The Early Versions of the New Testament: Their Origin, Transmission and Limitations (Oxford: Clarendon Press, 1977). 43 Tov, Text-Critical Use, 54-60. He adds that a fifth criterion, linguistic adequacy of lexical choices, cannot be described statistically and is highly subjective, so it cannot be used profitably in the analysis of translation units. 44 Ibid., 55. 45 It is common for a people who learn a new language to fuse it with aspects of their own language. See the discussion of pidgin and creole languages in Encyclopædia Britannica, 1985 ed., Macropædia, s.v. “Languages of the World,” 803-5. Cf. also the comment by Conybeare and Stock, that the Greek of the OT (and NT) “is so deeply affected by Semitic influence as often to be hardly Greek at all,

23

sometimes occurs in the versions of the OT is etymologizing, the tendency to render words derived from one Hebrew root (presumably) by words derived from a single root in the target language. All of the examples Tov gives of consistent representation are lexical in nature. However, translators often rendered certain grammatical structures in the source language (e.g., stem, inflection, and person in Hebrew) consistently by particular grammatical structures in the target language (e.g., voice, tense, and person in Greek). Another aspect of consistency that should be measured is whether the target language regularly rendered a certain class of words in one language by a particular class in the other, for example, participles by participles. An examination of the grammatical and syntactical correspondences between MT and the secondary witnesses is another aspect of consistency that should be investigated. In the proposed methodology, the lexical consistency of (1) verbs, nouns, and adjectives; (2) adverbs, prepositions, and particles; and (3) conjunctions will be tallied separately in a variety of ways.46 First, the number of different Hebrew words occurring more than once will be compared with the number of different words in the target language occurring more than once. Second, the total number of Hebrew words occurring more than once will be compared with the sum of the most frequently used words that correspond to each of these Hebrew words. Third, the factor measuring deviation from complete consistency will be figured.47 The final aspect of lexical consistency to be examined will be a measure of the level of etymologizing apparent in the translation. The first element of grammatical consistency to be measured is the percentage of one class of Hebrew words (verbs, nouns and adjectives, pronouns) represented by the same class of words in the target language. Next, various aspects of verbs, nouns and adjectives, and pronouns will be evaluated separately and compared to their corresponding words in the target language. For verbs, the Hebrew inflection, stem, person, and number will be compared with the corresponding grammatical structures in the target language to but rather Hebrew in disguise”; F. C. Conybeare and St. George Stock, Grammar of Septuagint Greek (Boston: Ginn & Co., 1905; repr., Peabody, MA: Hendrickson, 1988), 21. This statement is an exaggeration, but it does show that the Jews often spoke a form of Greek influenced by their Semitic linguistic background. 46 Proper nouns will be excluded from the evaluation of lexical consistency and from the grammatical analyses of nouns, with two exceptions. First, the proper name , since it is equivalent in most contexts to and / , which are common nouns, will be included in the evaluation of lexical consistency. Second, all proper nouns will be included in the analysis of the use (i.e., grammatical function) of Hebrew nouns and adjectives in the sentence. 47 The algorithm for arriving at the deviation factor will be explained more fully in Chapter 3, and

in detail in Appendix 4. The deviation factor itself is a numerical representation of the deviation from absolute consistency, i.e., one-to-one correspondence, with a deviation factor of zero indicating absolute consistency. The deviation factor is a number that is useful primarily in comparison with the deviation factors of other elements in the same version or of similar elements in another version.

24

determine consistency of grammatical representation. For nouns and adjectives, Hebrew gender, number, syntactical use in the sentence, articles, and definiteness will be compared. For pronouns, the Hebrew gender, number, and syntactical use in the sentences will be compared. The level of consistency in each category will be indicated by the deviation factor. Tov’s second criterion for determining the translation technique is the representation of the constituents of words in the source language by individual equivalents in the target language, also called segmentation. Because translators were interested in being faithful to the text before them, they were often meticulous in rendering each element of a compound Hebrew word (including such things as articles, prepositions, and suffixes) with a corresponding element in the target language, even in cases where such a rendering was not idiomatic. Consistency in the segmentation will be measured by the ratio of the number of Hebrew compound words rendered exactly to the total number of compounds.48 The third aspect of translation technique which Tov discusses is word order. Unlike Hebrew, which relies heavily on the position of words within a sentence to determine their function, inflected languages (e.g., Greek and Latin) are capable of great freedom in word placement without altering the basic meaning of a sentence. Whether the translators took advantage of the abilities of their languages or whether they limited themselves to the order present in their Hebrew Vorlagen is an important measure of translation technique. Of course, variations in word order forced by the grammar of the target language (e.g., postpositive conjunctions) do not count against consistency of word order. The total number of Hebrew semantic units that are represented in MT will be compared with the number of unforced deviations from the Hebrew word order in the translation. The fourth criterion for determining translation technique, quantitative representation, refers to the tendency of some translators to render every element of their Vorlage without omitting superfluous or redundant elements and without adding clarifying words. The target language might put certain restrictions on the translator, so that a one-toone correspondence is impossible; however, grammatically imposed strictures (e.g., absence of an article in Latin) will not be counted against consistency. To measure consistency in quantitative representation, it is not sufficient to compare the total number of semantic units in MT and in the version, since one addition and one omission in the version would yield a number identical to that of MT. Instead, the total number of differences 48 Wright notes that the representation of the constituents of a word by elements in the target language “is actually a sub-category of quantitative representation [the fourth criterion, below] because it involves a type of one-to-one representation albeit a very specific type”; Wright, “Quantitative Representation,” 316. Cf. also Barr’s discussion of segmentation and his examples of extremes from the version of Aquila; James Barr, The Typology of Literalism in Ancient Biblical Translations, Mitteilungen des Septuaginta-Unternehmens, no. 15 (Göttingen: Vandenhoeck & Ruprecht, 1979), 26-29.

25

between MT and the version will be calculated and then compared with the total number of Hebrew semantic units to produce a ratio. One other note about the grammatical analysis is necessary at this point. It will become apparent when comparing MT with each version that certain add-oms occur.49 The question of whether to incorporate add-oms in the statistics is important, for a considerable number of add-oms could significantly alter the picture of the translation technique. It is clear that add-oms should play a part in determining the translation technique. On the other hand, it is difficult to compare something to nothing, and calculations of consistency that tried to take add-oms into account might well produce less than meaningful results. Therefore, add-oms will be omitted from the computation of statistics in the first three areas (consistency, segmentation, and word order). However, they will be counted in the measure of quantitative representation. Once the initial statistics for each of the categories in a particular version being investigated have been figured, the process of refining those statistics begins. The first statistics will give only a rough idea of the translation technique, because certain trends in the translation not apparent at the beginning of the investigation will become evident. Each new discovery will lead to a modification of the previously calculated partial translation technique, until finally all the grammatical data will be accounted for. Then, using the partial translation technique as determined to that point, variants in the version that do not reflect a different Vorlage, either because they are nonreproducible in the target language or because they are theoretically reproducible but are not consistently rendered, will be eliminated.50 All other variants that are not purely orthographic will be considered significant, because they have some probability of reflecting a Vorlage different from MT. The second major phase in the proposed model involves determining the literary 49 For a definition of add-om, see above, p. 7. 50 The variants which are eliminated correspond for the most part to the non-variants that Tov discusses in Text-Critical Use, 217-28. Tov’s pseudo-variants, discussed on pp. 228-40, require special comment. He defines pseudo-variants as readings that “presumably were not found in the translator’s Vorlage, but existed only in his mind” (p. 228, italics his). The problem with this definition (as he notes, pp. 228-29; cf. pp. 140-41) is that distinguishing real variants from pseudo-variants is usually subjective. After all, if the translator could mistake one word for another, so could a Hebrew scribe, and there is no way of convincingly demonstrating at what point the mistake occurred, whether in the transmission of the Hebrew or of the versional text. Cf., e.g., P. Kyle McCarter, Jr., Textual Criticism: Recovering the Text of the Hebrew Bible, Guides to Biblical Scholarship, Old Testament Series (Philadelphia: Fortress Press, 1986), 26-57, passim. An exception to this contention is the case of demonstrably inner-version (e.g., inner-Greek) variants, which, of course, could only occur in the transmission of the versional text (cf. Tov, Text-Critical Use, 237-40). It is undoubtedly true that the translators sometimes read their Hebrew Vorlagen incorrectly, thus creating variants that never existed in a Hebrew ms. However, it seems methodologically preferable, except in the case of inner-version corruptions, to treat what might be pseudovariants in exactly the same way as other variants. The decision to eliminate them on the grounds that they probably never occurred in Hebrew is best left to the final stage of the process of textual criticism of the Hebrew text, namely, choosing the best reading from the various Hebrew and retroverted readings (see below, chap. 7).

26

and theological characteristics of the witness. This process will require an examination of the larger context to determine theological tendencies and other changes probably related to literary, cultural, or historical circumstances. If a witness is determined to have a specific tendency (e.g., a tendency to harmonize parallel passages), variants that exhibit this tendency cannot be considered likely indicators of a different Vorlage. An examination of various literary and theological studies of Samuel will be particularly helpful in this analysis. Numerous recent commentaries and articles deal with the larger literary shape of 1 Samuel in a particular tradition (especially MT and LXX), and others, though not dealing specifically with 1 Samuel, use a literary approach that might fruitfully be applied to the text. Literary approaches to be investigated include close reading, rhetorical criticism, structuralism, narrative analysis, deconstruction, and analysis of comparative structures. Literary analyses that borrow from several different approaches will also be consulted. An examination of the theological themes identified in 1 Samuel will be studied for the light they might be able to shed on the state of the text. Finally, any historical or archaeological data that might be relevant to the meaning of the text will be analyzed. The purpose of investigating these different approaches is to determine whether further characteristics of the translators, such as cultural or theological biases, should be indicated in a full description of the translation technique of each of the versions. Once as many characteristics of this nature have been identified as seem probable, the partial translation technique will be adjusted again, and the result will be the final translation technique. On the basis of the new information gathered in the second phase of study, other variants will be eliminated which probably do not reflect a different Vorlage. After the process of eliminating variants has been completed, all the remaining variants may be considered significant. Since Hebrew variants that are not purely orthographic are significant by definition (see above, p. 6), all the variants found in those Masoretic mss that possibly preserve pre-Masoretic readings51 and all the variants found in 51 As identified by Goshen-Gottstein; see above, p. 16, n. 18. Though all such variants will be

listed, Goshen-Gottstein’s warnings concerning the use of variants from medieval mss must be stressed. In his article “Hebrew Biblical Manuscripts,” he makes the following statements: “Among all the MSS and fragments known so far there is not even one the deviations of which can be significantly connected with any non-Massoretic tradition. We possess no medieval manuscript which, on the strength of its readings, may be termed ‘valuable’ or be worthy of our attention more than any other” (p. 277); “For those books of the Bible which have already been checked the results gained for Isaiah are paralleled, and there is no reason to assume that further investigations of MSS and fragments from other books are going to enable us to uphold the notion of a ‘valuable’ medieval MS” (pp. 282-83); “The possibility of infiltration of extraMassoretical variants into the Massoretical ‘central current’ is so negligible that for all practical purposes it may be disregarded” (p. 286). Nevertheless, “the analysis of readings such as in the Book of Kings may at least justify the mentioning of the possibility that different results may be obtained for different books (or parts) of the Bible and that, accordingly, we may have to reckon with different ‘breadths’ of the ‘central current’ and different strengths of the ‘trickle’ from the side” (p. 287). (It is at this point that he lists the four mss that “stand out” from the rest in Samuel.) For example, he finds only seven variants in medieval mss of Isaiah that “might be possibly more than the result of harmonization” or other scribal habits, and only one of these “makes one really prick up one’s ears.” It will be in the light of these cautions that

27 4QSama, the only other primary witness to the text of 1 Samuel 3, may simply be listed. Once all the significant variants have been determined, the process of retroversion and evaluation of variants may begin. Before beginning the analysis of the secondary witnesses, however, the text of the secondary witnesses themselves must be established, and one other preliminary matter requires some discussion, namely, the script of the Vorlagen of the various Samuel translations. The Script of the Vorlagen Although it is often assumed that the Vorlagen of the secondary versions of Samuel were written in the later square script, this assumption needs further discussion and demonstration, particularly in the case of LXX. Rabbinic tradition says that Ezra and the other returnees from Babylonia introduced the square (“Assyrian”) script into Palestine in the sixth and fifth centuries B. C. E. (b. Sanh. 21b-22a; j. Meg. 1.9, etc.), whereupon it replaced the Old Hebrew script then in use. However, evidence from Qumran, coupled with evidence already known (script of the Samaritan Pentateuch, numismatic evidence from the Hasmonean and Herodian periods, revival of the older script during the time of Bar Kokhba), necessitates a new look at the issue. Shemaryahu Talmon concludes on the basis of all the data, especially the existence of several scrolls from Qumran written in the Old Hebrew script, that “the ancient Hebrew script retained a lease on life until after the destruction of the Second Temple and the cessation of Jewish political sovereignty.”52 Qumran data demonstrates that the Old Hebrew script was sometimes used for transmitting biblical material in the late Second Temple period. Most of the mss in Old Hebrew script are from the Pentateuch; however, one Old Hebrew ms of Job does exist (4QpaleoJobc). 53 It is probable that Samuel was composed, and subsequently transmitted, in the Old Hebrew script.54 However, S. R. Driver concludes that the Vorlagen used by the translators of most of the books in LXX (presumably including Samuel) were written in an earlier form of the square script.55 In order to demonstrate that the Vorlage of LXX variants from medieval mss will be listed. 52 Shemaryahu Talmon, “The Ancient Hebrew Alphabet and Biblical Text Criticism,” in Mélanges bibliques et orientaux en l’honneur de M. Mathias Delcor, ed. A. Coquot, S. Légasse, and M. Tardieu, Alter Orient und Altes Testament, no. 215 (Kevelaer: Verlag Butzon & Bercker, 1985; Neukirchen-Vluyn: Neukirchener Verlag, 1985), 391. See his discussion of examples of the alternation of and in the Old Hebrew script, ibid., 393-401, and of the alternation of and , idem, “The Ancient Hebrew Alphabet and Biblical Text Criticism,” in Mélanges Dominique Barthélemy: Études bibliques offertes a l’occasion de son 60e anniversaire, ed. Pierre Cassetti, Othmar Keel, and Adrian Schenker, Orbis Biblicus et Orientalis, no. 38 (Fribourg: Éditions Universitaires, 1981; Göttingen: Vandenhoeck & Ruprecht, 1981), 504-22. 53 Tov also mentions 4QpaleoJoshpara, containing parts of Joshua 21; Emanuel Tov, Textual

Criticism of the Hebrew Bible (Philadelphia: Fortress Press, 1992), 105, n. 79. 54 Cf. ibid., 220. 55 S. R. Driver, Notes on the Books of Samuel, lxiv. Cf. the similar conclusion of Johann

28

Samuel was written in Old Hebrew, it would not be sufficient to point to evidence of confusion that can occur only in Old Hebrew, since this confusion could have occurred during the process of transmission of the Hebrew text that led to the Vorlage. Instead, it would be necessary to show that errors resulting from graphic similarity which only exist in the square script (e.g., and , and , and ) do not appear with any great frequency in the translation. (Furthermore, the confusion of and , that could only have occurred in square script, must be ignored if matres lectionis are involved.) However, such graphic confusion based on the square script does seem to play a major role in Samuel (cf. 3:2, 6, 11), so the Vorlage used by the translators of LXX (as well as those used by the other translators) does appear to have been written in square script. The ingenious, but discredited, transcription theory of Franz Wutz may be mentioned briefly at this point. In several studies produced early in the current century, Wutz propounded the theory that the translators of LXX worked from texts in which the Hebrew had been transcribed in Greek characters, checking the actual Hebrew text only occasionally.56 Though he won several influential early supporters, most notably Rudolf Kittel and Paul Kahle, support quickly faded, and Wutz himself seems to have abandoned the view in later years. Johann Fischer, who supported the transcription theory at least in part, pointed out one of the flaws in Wutz’s theory, namely, that the frequency of such graphic errors as for , which do not resemble one another in Greek, tend to contradict any reliance on mss written with Greek characters.57 One last matter of a methodological nature needs to be mentioned in connection with work on the script of the Vorlage of the translations, a matter that applies equally to attempts to trace the early history of the Hebrew text itself. Many studies, including those of Wutz and Johann Fischer, have based much of their analysis on the evidence provided by the transmission of proper names, on the grounds that since the names are transcribed rather than translated, they should provide a stable interface between Hebrew and the target Fischer for the Pentateuch: “Wie ersichtlich, steht das A[lphabet] der LXX-V[orlage] zwischen neuaramäischer Schrift und Quadratschrift und man wird es als neuaramäisches Alphabet mit starker Neigung zur Quadratschrift bezeichnen dürfen”; Johann Fischer, Das Alphabet der LXX-Vorlage im Pentateuch: Eine textkritische Studie, Alttestamentliche Abhandlungen, vol. 10, no. 2 (Münster: Aschendorffschen Verlagsbuchhandlung, 1924), 93. However, cf. Driver’s comment, “The confusion of and , and and (which cannot be explained from the old character) is in the Pent. so uncommon that it may be due to accidental causes: the books in which it is frequent can only have been translated after the change of character had been effected; the Pent., as tradition states, may have been translated earlier”; S. R. Driver, Notes on the Books of Samuel, lxiv, n. 3. 56 Franz Wutz, Die Transkriptionen von der Septuaginta bis zu Hieronymus, pt. 1, Beiträge zur

Wissenschaft vom Alten Testament, n.s., no. 9, Texte und Untersuchungen zur vormasoretischen Grammatik des Hebräischen, no. 2 (Berlin: W. Kohlhammer, 1925), 5: “Da wir weiterhin beweisen können, dass nur in ganz wenigen Fällen ein hebräischer Konsonantentext eingesehen wurde, so sind selbst die hebräischen Verlesungen nicht direkt mit der Septuaginta in Zusammenhang zu bringen, sondern mit ihrer Transkriptionsvorlage.” 57 Johann Fischer, Alphabet der LXX-Vorlage, 95.

29

language. However, the evidence suggests that the opposite is rather the case. Names, particularly lists of names like those found in Joshua, Chronicles, Ezra, and Nehemiah, are subject to deviation on a grand scale. The problem in using proper names to determine script of the Vorlage is that, unless they refer to well-known individuals, they are meaningless collections of letters, subject to uncontrolled variation, unlike other words, whose variation is controlled to some extent by the context of the sentence. Thus, common nouns, verbs, and adjectives are better suited for studies on script than are proper names.58 Having determined that the square script apparently lay behind all of the secondary versions of 1 Samuel, the archetypical secondary witnesses may now be reconstructed. This is the task of the second chapter of this study. 58 Cf. S. R. Driver, Notes on the Books of Samuel, lxiv, n. 3.

CHAPTER 2 RECONSTRUCTING ARCHETYPICAL SECONDARY WITNESSES Method for Reconstruction Ideally, the steps of the model just described would be applied to each particular versional tradition to determine as nearly as possible the original text. However, that task is beyond the scope of the present study, so traditional methods alone will be employed to determine a standard text for each tradition. This step is particularly important for those versions that do not have critical texts (e.g., the Cambridge edition of LXX, which prints a diplomatic text), but the variants in those traditions with critical texts will also be checked. First, though, a brief textual history of the four secondary witnesses will be given in order to ascertain the relationships of these witnesses to one another and to MT.1 LXX has the most complex textual history of the secondary witnesses. It also exhibits both the greatest number of witnesses and the greatest number of variants within the tradition. The term LXX is often used rather loosely. Though it originally referred to a particular translation of the Pentateuch, it was extended to include the Greek translation of the other books in the OT. Herein lies the problem, for not one, but many different Greek versions existed in the ancient world. Though Paul Kahle proposed the theory that various Greek “targums,” in use among different Jewish communities, lay behind the main LXX tradition,2 the theory that won the day was that of Paul de Lagarde, who proposed a single original translation of the Hebrew, from which the different Greek traditions developed.3 1 Johann Cook, “Die pluraliteit van ou-testamentiese tekste en eksegetiese metodologie,” paper

presented at the annual meeting of Die Ou-Testamentiese Werkgemeenskap van Suid-Afrika, 1988, pp. 1-2, stresses the importance of having a grasp of the history of the text before textual criticism (or exegesis) is undertaken: ['n tekshistoriese perspektiewe] kan enersyds die gevolg wees van bepaalde teoretiese uitgangspunte waarvolgens die tekskritiek, die wetenskaplike dissipline war oa gerig is op die ontstaans- en teksgeskiedenis van bepaade tekste, tot bepaalde voorarbeid beperk word. . . . Andersyds, let 'n oordrewe klem op die finale vorm van die teks (die sg. strukturele metode) en/of op die uitweking van die teks op die ontvanger van die boodskap van die teks (resepsie-kritiek) tot 'n onderwaardering van die tekstuele sy van die teks. 2 Paul E. Kahle, Cairo Geniza, 132-79. His first statement of his theory was in idem,

“Untersuchungen zur Geshichte des Pentateuchtextes,” Theologische Studien und Kritiken 88 (1915): 399439. 3 For a discussion of Lagarde’s views, see Sidney Jellicoe, The Septuagint and Modern Study (Oxford: Oxford University Press, 1968; repr., Winona Lake, IN: Eisenbrauns, 1989), 5-9. It is better to

30

31

Lagarde believed that three recensions of LXX existed by the third century C. E.— associated with the names of Origen, Lucian, and Hesychius—and that all extant mss of LXX present mixed texts, preserving readings from all of these recensions. Many modern scholars question the existence of a Hesychian text, though they accept the Origenic (Hexaplaric) and Lucianic texts. The focus of many modern discussions of the history of LXX is the recovery of the Old Greek (OG) text, the original LXX that lies behind all existing mss. Though no single ms is believed to contain a perfect OG text, those that show little evidence of Hexaplaric or Lucianic readings are believed to be fairly good representatives of it. A problem arises, however, in the text of the four books of Kingdoms in LXX (i.e., Samuel and Kings in MT). Henry St. John Thackeray demonstrated that the text of the major uncial mss was not uniform throughout these books, but represented a mixture of two different Greek versions.4 He divided the books into the following sections: (1 Kingdoms), (2 Kgdms 1:1-9:13), (2 Kgdms 10:1-3 Kgdms 2:11), (3 Kgdms 2:12-21:29), and (3 Kgdms 22:1-4 Kgdms 25:30). Sections , , and he identified as authentic OG sections. Sections and represented another version with different characteristics.5 Barthélemy built on Thackeray’s theory with evidence from a ms found at Nah@al H9ever by suggesting that the sections and do not contain a Greek translation independent of OG but rather a revision of OG, called the kaige recension.6 He later modified his view somewhat under the influence of studies by Robert A. Kraft and Sebastian Brock, suggesting that another layer of revision, which is sometimes called proto-Lucian, probably underlay the kaige recension. The kaige text in turn was the basis for the versions of Aquila ( ´), Symmachus ( ´), and Theodotion ( ´).7 As its name indicates, proto-Lucian is a layer of the text that many scholars find underneath the Lucianic speak of original translations of LXX, since different translators were responsible for the various books. 4 Henry St. John Thackeray, The Septuagint and Jewish Worship, 2d ed. (London: Oxford

University Press, 1923). 5 R. W. Klein, Textual Criticism, 25. The use of the LXX title Kingdoms instead of the Hebrew Samuel and Kings explains the names Thackeray gave to the sections; the Greek letters represent their numerical values, e.g., contains parts of 2 and 3 Kingdoms. Thackeray dates the Old Greek sections to the late second century B. C. E. and the kaige sections to sometime after the turn of the era; Henry St. John Thackeray, A Grammar of the Old Testament in Greek, vol. 1, Introduction, Orthography and Accidence (Cambridge: Cambridge University Press, 1909; reprint, Hildesheim: Georg Olms, 1978), ix. 6 Barthélemy, Devanciers d’Aquila, 126-27. The kaige recension is sometimes called proto-

Theodotion or kaige-Theodotion; ibid., 144-57. 7 Idem, “Les problèmes textuels de 2 Sam 11,2 - 1 Rois 2,11 reconsidérés à la lumière de certaines critiques des ‘Devanciers d’Aquila,’” in 1972 Proceedings, ed. Kraft, 28. The reviews of his book by Kraft and Brock are: Robert A. Kraft, review of Les devanciers d’Aquila, by Dominique Barthélemy, in Gnomon 37 (1965): 474-83; Sebastian P. Brock, “Lucian Redivivus: Some Reflections on Barthélemy’s Les devanciers d’Aquila,” Studia Evangelica 5 (1968): 176-81. This view is further developed in the commentaries by McCarter and Klein, as well as in numerous other studies.

32

readings, itself a revision of OG.8 The importance of these finds to the present study is that they show the importance of the Lucianic text-tradition (mss boc2 e2 in the Cambridge edition) as an independent witness to an early form of the Hebrew text. This summary of the history of the Greek text has been necessarily brief, since what is of primary importance here is the relationship of the various Greek witnesses to their Hebrew Vorlagen.9 Aside from OG itself, (proto-)Lucianic readings that are revisions of OG toward a Hebrew text will be important, especially since it has been shown that this Hebrew text had substantial differences from the later MT.10 Similarly, though Origen (LXXO), ´, ´, and ´ all revised their Greek texts toward a Hebrew text very similar to MT, they will be examined to see if their respective Hebrew exemplars varied at all from it. All the readings from Greek witnesses outside the OG tradition will be considered partial secondary witnesses, since they are not independent translations of their Hebrew exemplars. Unlike the textual history of LXX, on which scholars exhibit general agreement, that of P is hotly debated. The date, provenance, and translators of P are all contested issues. These issues, however, are closely related to one another. Scholars who see a Jewish origin for P believe that it was first used in the province of Adiabene, on the border of the Roman Empire, in the middle of the first Christian century. The rulers of Adiabene had been converted to Judaism about 40 C. E. and needed a translation of the Bible. Other scholars posit a Jewish-Christian origin for P, also in Adiabene, but somewhat later, when missionaries first reached the area with the gospel. Still others place the origin of the version in Edessa, the most important city in Syria (outside of Antioch, which was largely Greek). Some Syriac traditions point to a Christian origin for P in Edessa. The discovery of two Old Syriac gospels has also raised the issue of whether P in the OT also had an Old Syriac predecessor.11 The resolution of these issues also involves identifying the type of text from which 8 Natalio Fernández Marcos, “The Lucianic Text in the Book of Kingdoms,” in De Septuaginta:

Studies in Honor of John William Wevers on His Sixty-Fifth Birthday, ed. A. Pietersma and C. Cox (Mississauga, ON: Benben, 1984), 166-71; Sebastian P. Brock, “The Recensions of the Septuagint Version of 1 Samuel,” D.Phil. diss., Oxford University, 1966. Tov expresses doubts that a separate proto-Lucian can be distinguished from Lucian, believing instead that Lucian itself is a direct revision of OG; Emanuel Tov, “Lucian and Proto-Lucian,” Revue biblique 83 (1976): 51-54. 9 For a fuller discussion of the Greek text, see James Donald Shenkel, Chronology and Recensional Development in the Greek Text of Kings, Harvard Semitic Monographs, no. 1 (Cambridge: Harvard University Press, 1968), 5-21. 10 Ulrich, Qumran Text of Samuel, 257-59. 11 IDB, Supplementary Volume [hereafter IDBS], s.v. “Syriac Versions,” by A. Vööbus, 848-49; Bleddyn J. Roberts, The Old Testament Text and Versions (Cardiff: University of Wales, 1951), 217-23; Svenskt Bibliskt Uppslagsverk, s.v. “Bibeln,” by C.-M. Edsman, col. 256; ibid., s.v. “Gamla Testamentet,” by Ivan Engnell, col. 654; J. B. Segal, Edessa: The Blessed City (Oxford: Clarendon Press, 1970), 42-43, 165.

33

P was translated. Several scholars have noted parallels between P and T, both in language (Western Aramaic elements embedded in the Eastern Aramaic version) and content (certain shared deviations from other witnesses), and have concluded that P was originally transcribed into Syriac from a Western Aramaic Palestinian targum. Subsequently, the text was brought more and more into conformity with the proto-MT tradition. However, not all scholars have accepted this reconstruction of P’s textual history. M. D. Koster, in his study of P in Exodus, comes to the conclusion that P was originally translated from a Hebrew text similar to MT; it was only later that targumic additions were made to the text. Both M. J. Mulder and Johann Cook support Koster’s position, explaining that similarities between P and T might come from a shared Jewish exegetical tradition, rather than direct influence of T on P. Another theory, which stands in the middle of these two, is proposed by Alexander Sperber. He believes that a real textual connection between P and Targum Onkelos does exist, but rather than seeing P as a revision of a Palestinian targum stripped of its paraphrases, he says that both P and Onkelos derive from a common ancestor, and the characteristic targumic paraphrases are later additions to the targum tradition.12 The question must be said to be still unresolved, particularly with regard to 1 Samuel, since recent studies have focused on the Pentateuch. One other note about the text of P must be addressed, namely, its relationship with LXX. The text of P has often been denigrated as being of little value to the textual critic. Since readings shared with LXX have been seen as the result of the influence of LXX on P, readings in P have only been considered valuable in conjunction with LXX. However, Cook points out that though LXX did influence P to some extent, its influence has been greatly overestimated in the past.13 Mulder goes even farther, asserting the essential independence of LXX and P.14 In this study, each reading of P will be examined individually in order to determine its relationship with other versions and, more importantly, the likelihood that it is a witness to an independent Hebrew reading in certain places. 12 A. Vööbus, “Der Einfluss des altpalästinischen Targums in der Textgeschichte der Peschitta des Alten Testament,” Le muséon 68 (1955): 215-18; Paul E. Kahle, Cairo Geniza, 272-73; M. D. Koster, The Peshi a of Exodus: The Development of Its Text in the Course of Fifteen Centuries, Studia Semitica Neerlandica, no. 19 (Assen: Van Gorcum, 1977); idem, “Which Came First: The Chicken or the Egg? The Development of the Text of the Peshit@t a of Genesis and Exodus in the Light of Recent Studies,” in The Peshi a: Its Early Text and History. Papers Read at the Peshi a Symposium Held at Leiden 30-31 August 1985, ed. P. B. Dirksen and M. J. Mulder, 147-68, Monographs of the Peshit@t a Institute, Leiden, no. 4 (Leiden: E. J. Brill, 1988), 99-126; M. J. Mulder, “The Use of the Peshit@t a in Textual Criticism,” in La Septuaginta, ed. Fernández Marcos, 52; Johann Cook, “The Composition of the Peshit@t a Version of the Old Testament (Pentateuch),” in The Peshi a: Its Early Text and History. Papers Read at the Peshi a Symposium Held at Leiden 30-31 August 1985, ed. P. B. Dirksen and M. J. Mulder, 147-68, Monographs of the Peshit@t a Institute, Leiden, no. 4 (Leiden: E. J. Brill, 1988), 153-59; Sperber, Bible in Aramaic, 4b:409-17. 13 Johann Cook, “Composition of the Peshit@t a,” 159. 14 Mulder, “Use of the Peshit@t a,” 53.

34

The various targums have long traditions lying behind their commitment to parchment or papyrus. Oral renderings of the OT in Aramaic probably originated in the synagogue at a time when the general populace no longer understood Hebrew.15 The meturgemanîn (translators) were careful to preserve traditional interpretations of the texts, but, as is clear from a comparison of the targums to the Pentateuch (Onkelos, PseudoJonathan, Neofiti, the Fragmentary targums, and the fragments from the Cairo Geniza), conflicting renderings did develop in different places and at different times. The state of affairs with Targum Jonathan is somewhat different from that of the targums to the Pentateuch. It is the only Aramaic targum preserved in the prophets (except for a number of fragments and glosses), though many scholars believe that Palestinian targums of this material once existed as in the Pentateuch. Jonathan, like Onkelos, attained its final form in Babylonia and served as the official targum of the books it contained. The wording of Jonathan was definitively established sometime in the fifth century after a long and complex history. 16 Though many of the traditions underlying Jonathan antedate the choice of the proto-Masoretic tradition as the official text for all Judaism about 100 C. E., it was revised toward that official tradition, so that few variations from MT remain. Because of this revision, though, any variations that do remain will be important if it can be shown that they probably reflect a Hebrew Vorlage different from MT. The textual history of V is probably the least complicated and controversial of all the secondary witnesses, though it is not as simple as it might appear at first glance. Jerome was commissioned by Pope Damasus I to produce a Latin version of the Bible about 382. His translation proceeded in two steps. First, he revised existing Old Latin texts according to the Greek text of Origen. Dissatisfied with this initial effort, he then began to translate the entire OT directly from the Hebrew, and the end result was V. Since the Hebrew text Jerome used as the basis for this second translation was very similar to the present MT, it might seem as though the textual history of V were straightforward and that V would be of little use as an independent witness to the text. However, the situation is somewhat more complicated. In the first place, the scribes who transmitted the text of V were often less than faithful copyists, frequently mixing various it readings into their texts (not to mention numerous scribal errors). As a result, comprehensive revisions of the text began before V had supplanted it as the preferred Latin version. The most important early revisions are associated with the names of Cassiodorus (sixth century), Alcuin (eighth century), and Theodulf (eighth century).17 15 On the basis of fragments of Palestinian targums from the Cairo Geniza, Kahle states that a targum existed in Palestine by the second century B. C. E. at the latest; Paul E. Kahle, Cairo Geniza, 207. 16 Würthwein, Text of OT, 78; B. J. Roberts, OT Text and Versions, 207-9. 17 IDB, s.v., “Versions, Ancient,” by Bruce Manning Metzger, 752-53; B. J. Roberts, OT Text and Versions, 259-60; Fischer, Biblia Sacra Stuttgartensia, 1: xx-xxi.

35

As Metzger notes, “the more than eight thousand MSS of the Vulg. known today exhibit the greatest degree of bewildering cross-contamination of textual type.”18 Thus, before V can be consulted, a relatively pristine form of the text must be produced, either by Quentin’s method (see below, p. 66**) or by some other. Another problem is the identity of the sources of the various readings Jerome used in V. Though he attained a good knowledge of Hebrew, he himself admitted in letters to friends that he frequently consulted with rabbis in the vicinity of Bethlehem (where he was working on his translation). This contact accounts for some similarities between V and T. He also had access to Origen’s Hexapla and to mss of the secondary Greek versions, and he often relied on them to clarify a difficulty. Finally, the various it versions (no official it version ever existed) undoubtedly played a strong role in his decisions.19 Thus, the textual affiliations of V are quite complex, and agreements between V and other ancient versions must be considered carefully before one can conclude that a Hebrew text different from MT lies behind a reading in V.20 Nevertheless, a reading in V that supports that of another witness at least testifies to the authenticity of that reading in the fifth century, and an independent reading is of great value. Further research would be necessary to see where and to what extent each of the versions mentioned influenced Jerome’s text. Now that the history of the text of each of the secondary witnesses has been reviewed, a method for reconstructing their texts can be stated briefly. For the text of those witnesses that do not have an eclectic critical text (i.e., LXX, P, and T), each variant given in the critical apparatus of the diplomatic text will be evaluated to see if it is preferable to that of the basic text. This evaluation will be done primarily on intrinsic grounds, though extrinsic grounds will be a factor in some cases. Other editions of the version will also be consulted where appropriate, especially in the case of LXX. The text of V, which has two eclectic critical texts, will be easier, since the variants have already been evaluated. Nevertheless, all the variants in the critical apparatuses will be checked again, and the two critical editions of the text will be compared with one another, to see if any differences in evaluation have been made. Finally, the partial secondary witnesses will be presented in fragmentary form. All attested readings from ´, ´, and ´ will be presented. In the case of LXXO, only those readings that deviate from both MT and OG will be evaluated. The entire text of LXXL will be reconstructed, but only those readings that vary from OG will be considered in Chapter 3. The full analysis applied to the complete secondary witnesses will not be done for the partial secondary witnesses. Instead, their variants will be evaluated in a more 18 Metzger, “Versions, Ancient”, 753. 19 Ibid.; Würthwein, Text of OT, 92-93. 20 Friedrich Stummer, Einführung in die lateinische Bibel: Ein Handbuch für Vorlesungen und

Selbstunterrich, (Paderborn, Germany: F. Schoning, 1928), 123; cited in Würthwein, Text of OT, 93.

36

traditional way.21 Location of Textual Data The primary source for the textual data of LXX will be the text and apparatus of the Cambridge edition, but Rahlfs’s smaller edition will be considered as well, especially since Rahlfs makes a preliminary attempt to create a more eclectic text. The editions of LXX edited by Holmes and Parsons and by Swete will also be checked.22 Variant readings from the daughter versions of LXX will be drawn primarily from the critical apparatus in the Cambridge edition, though several of them will be checked to determine the accuracy of citation; however, only secondary text-traditions will be analyzed in detail. The Leiden edition of P provides a diplomatic text of P, along with a critical apparatus. This will be the text of P used in the study. Sperber’s edition of T presents the text of one ms as the basic text, with other readings in the apparatus, and it will serve as the text of T for this study. Daniel J. Harrington and Anthony J. Saldarini, in the introduction to their translation of T, note the variation present in the different mss of T, saying that “the individual manuscripts of what we call Targum Jonathan tend almost to constitute separate works.”23 Many of these variations, though, are simply haggadic or halakic expansions, so are of little value in indicating the Hebrew Vorlage. V is the only secondary witness published as an eclectic critical text, so the basic text has already been evaluated by scholars and determined to be as close as possible to the original. Thus, little reconstruction need be done, except to evaluate the variants presented in the apparatus and to compare the critical edition from the Abbey of St. Jerome with the Stuttgart version, which does have some variations in both the basic text and in the apparatus. For the partial secondary witnesses, the readings of ´, ´, and ´ will be culled from both Field’s edition and the apparatus of the Cambridge LXX. The readings of 21 The importance of the partial secondary witnesses for reconstructing the text of the OT may be seen in Driver’s comment on the Lucianic text: “Whether these renderings were derived by him from MSS. of the LXX of which all other traces have disappeared, or whether they were based directly upon Hebrew MSS. which had preserved the genuine reading intact, . . . is a matter of subordinate moment: the fact remains that Lucian’s recension contains elements resting ultimately upon Hebrew sources, which enable us to correct, with absolute certainty, corrupt passages of the Massoretic text”; S. R. Driver, Notes on the Books of Samuel, xlix. The absolute certainty of the reconstructions may be questionable, but the importance of LXXL and other partial secondary witnesses cannot be denied. 22 Robert Holmes and James Parsons, eds.,Vetus Testamentum Graecum cum variis lectionibus, 5

vols. (Oxford: Clarendon Press, 1798-1827); Henry Barclay Swete, ed., The Old Testament in Greek According to the Septuagint, 3 vols (Cambridge: Cambridge University Press, 1887-1912). 23 Daniel J. Harrington and Anthony J. Saldarini, Targum Jonathan of the Former Prophets, The Aramaic Bible, vol. 10, ed. Kevin Cathcart, Michael Maher, and Martin McNamara (Wilmington: Michael Glazier, 1987), 2.

37 LXXL will be taken from Lagarde’s edition of the Lucianic text24 and the Cambridge apparatus. LXXO will also be taken from the Cambridge apparatus, particularly from mss Acx. Presentation of Reconstructed Witnesses The reconstructed secondary witnesses will not be presented in their entirety. Instead, only those readings that have a reasonable claim to being original (to the version) and that differ from the collating bases (Cambridge LXX, Leiden P, Sperber’s T, and VR) will be listed, along with an explanation of the reason for the choice. For the partial secondary witnesses, all the fragments of ´, ´, and ´ will be given, along with their corresponding LXX reading. Since some mss refer to these three witnesses as a group or rather vaguely, a separate section will be devoted to listing all those readings that might belong to the three but are not clearly indicated. The readings of LXXL that deviate from LXX will be given in the next section. Finally, all the variants of LXXO from LXX will be listed. The variants will be presented in a form compatible with most textual apparatuses, but a few notes about sigla are in order. The reading of the base text will be given first, followed by the witnesses that support that reading, followed in turn by a right bracket ‘]’. If more than one such reading occurs in the verse, the one in question will be identified by a numeral followed by a small superscript o: ‘1°’, ‘2°’, and so forth. Next, the variant reading(s) will each be given, along with the witnesses that support the reading. Each of the variants following the base reading will be separated from the previous one by a vertical bar ‘|’. Witnesses will be listed in the following order: mss in the base language, versional witnesses, patristic sources. Mss in the base language will be separated from other witnesses by a semicolon in order to avoid possible confusion. Witnesses that support the variant in most respects (or the most important respects) but differ in small details will be enclosed in parentheses ‘()’. Witnesses that are inferred from the list of extant witnesses given in the apparatus of the base text but that are not explicitly listed will be enclosed in braces ‘{}’. Transposition of words from the base text or transposition combined with omission will be signified by numerals representing the variant word order (e.g., 132 signifies that the second word in the base text is the third word in the variant). Next, the symbols ‘+’ and ‘>’ represent an addition and an omission with respect to the base text, respectively. The abbreviations ‘pr’ and ‘post’ mean that the variant precedes or follows the word or phrase given as the base text, respectively. Finally, ‘idem’ indicates a reading that is identical with the base text. 24 Paul de Lagarde, Librorum Veteris Testamenti Canonicorum, vol. 1 (Göttingen: Aedibus Dieterich, Arnold Hoyer, 1883).

38 Septuagint As a general introduction to the evaluation of LXX readings, the following additional LXX resources should be mentioned: Hatch and Redpath’s concordance, Swete’s introduction, and Conybeare and Stock’s grammar. Bo Johnson’s study of hexaplaric mss is also helpful, for he identifies certain family groups within LXX mss of 1 Samuel, namely, B (sometimes associated with mss ya2 ), Acx (hexaplaric mss), boc2 e2 (Lucianic mss), dlpqtz (referred to as d+ [l is lacking in the chapter]), and fmsw (referred to as f+). The other mss (MVaeghijnvb2 , and often ya2 ) do not fall into any clear group.25 In addition to Greek mss and daughter versions, the church fathers Theodoret (Thdt) and Chrysostom (Chr) are sometimes cited. 3:1

{Bya2 Acx boc2 e2 d+ f+ MNaeginvb2 }]

h

Despite only one witness to this reading, it is interesting because it preserves Semitic, not Greek, idiom. However, it seems likely that the scribe was imitating the idiom found in both P and T in the same place, an idiom current in Hebrew as well. The only other two occurrences of in Kingdoms with God as the object are 1 Kgdms 2:11, 18. 2:11 reads , and 2:18 reads . In both cases a circumlocution occurs, so the scribe of h could have been simply conforming his text to the idiom of LXX itself. 3:2

B* c] cop eth syhj

{ya2 } A{x} {boc2 e2 d+ f+} B abMN{aeghinvb2 }; arm

The reading is attested only by the original hand of B and by one fifteenth century ms. It could be argued that should be preferred as the reading different from similar statements in the context (i.e., 3:3, 5 bis, 6, 9) and from MT. Furthermore, a closer examination shows that here refers to Eli, whereas all examples of and related forms refer to Samuel. It is possible that a Hebrew scribe (or the translator) could have misread as . However, the extreme paucity of external support makes the preference of questionable. A scribe transmitting the Greek text could just as easily have misread as . 26 If is 25 Edwin Hatch and Henry A. Redpath, A Condordance to the Septuagint and the Other Greek

Versions of the Old Testament (Including the Apocryphal Books), 3 vols. (Oxford: Clarendon Press, 18971906; repr., Grand Rapids: Baker Book House, 1983) [hereafter HR]; Henry Barclay Swete, An Introduction to the Old Testament in Greek, with an Appendix by Henry St. John Thackeray, rev. Richard Rusden Ottley (Cambridge: Cambridge University Press, 1914; repr., Peabody, MA: Hendrickson, 1989); Conybeare and Stock, Grammar; Bo Johnson, Die hexaplarische Rezension des 1. Samuelbuches der Septuaginta, Studia Theologica Lundensia, no. 22 (Lund: CWK Gleerup, 1963). For a summary of the family groups in LXX, see ibid., 19. In the following pages, LXX evidence for particular readings will be grouped according to Johnson's family groups. Mss y and a2 , when supporting reading in the absence of the other ms or B, will be listed with the ungrouped mss. 26 Cf. a similar error in the Ethiopic of 3:3, where

is apparently read for

.

39

indeed original, the scribes responsible for B and c could well have made the graphic error independently of one another or of any text in the direct genetic line between the two. Thus, the majority reading is to be preferred, with Rahlfs. 3:2

{Ba2 Acx bc2 e2 f+ Mghiv}]

o d+ Naenyb2

At issue in this reading is whether the subject of the verb is Eli, thus requiring a singular verb, or Eli’s eyes, requiring a plural verb. Though the main reading of both MT and V is singular, one ms of each has a plural. Both readings are well attested in Greek mss. It is likely that was influenced by the plural verb in the same verse to become . However, a similar change could have happened in Hebrew just as easily. The agreement of the plural with the mss from MT (ms 187, one of the mss Goshen-Gottstein says perhaps contains some non-Masoretic readings) and V suggests the possibility that the plural did exist in an early Hebrew ms; however, the original Greek was probably . 3:3

{Ba2 Acx dpqt MNaeghnb2 }] +

3:7

Ba2 ] +

boc2 e2 f+ vyz; Thdt

A{cx boc2 e2 d+ f+} MN{aeghinvyb2 }

These two readings need to be considered together, inasmuch as the preferred reading in one place is likely to be the same as in the other. The simple is predominantly an Attic reading, whereas the compound is originally Ionic; however, the latter reading predominates in the koine. 27 is well attested in verse 3 (apparently read by Ba2 Acx dpqt MNaeghnb2 ) but is poorly attested in verse 7. If the normal koine form were original in both places, later Atticizing scribes might have omitted the to create a more Atticistic reading. It is unlikely that later scribes would have corrected an original Attic reading in the direction of the koine, especially in the numbers attested in verse 7. As far as the difference in the attestation patterns between the two verses, two mss (c and x) read rather than in verse 3; since even Attic idiom allowed before an infinitive (both of Homer’s uses are before infinitives), it is possible that Atticistic revisers, after changing to in verse 3, would have considered the of verse 7 perfectly acceptable. Though the difference in the verb is attested in only two mss, other such mss may have exercised their influence on the reading of the preposition in some extant mss. Whichever reading is to be preferred, 27 Walter Bauer, A Greek-English Lexicon of the New Testament and Other Early Christian

Literature, 2d ed., trans. and ed. William F. Arndt, F. Wilbur Gingrich, and Frederick W. Danker (Chicago: University of Chicago Press, 1979) [hereafter BAG], s.v. “ ”; ibid., s.v. “h.” Bauer characterizes the compound form as an “Ionism, very rare in Attic wr., but common in the Koine” in the latter article. Smyth notes that though the compound form occasionally occurs in Homer and Herodotus, it is rare and suspect in Attic Greek; Herbert Weir Smyth, Greek Grammar, ed. Gordon M. Messing (Cambridge: Harvard University Press, 1956), 555.

40

both verses probably originally read the same. Rahlfs’s decision to read in verse 3 and in verse 7 seems to have been due to the combined witness of A and B in the former passage and their split witness in the latter. Nevertheless, it seems preferable in light of both internal and external evidence to read in both verses. 3:3

{Bya2 f+ MNaeghinvb2 }]

Ac | +

x boc2 e2 d+

Since MT here reads , it seems likely that the third variant is a correction to a similar Hebrew text. This supposition is strengthened when one realizes that the mss that support this reading are the Lucianic group (boc2 e2 ) and the group that Johnson calls d+, both of which often correct the Greek text in the direction of the Hebrew.28 The hexaplaric ms x also supports the reading. Harder to account for is the second reading, supported only by the hexaplaric witnesses Ac. Though the addition of supports MT in part, renders in only one other instance out of hundreds of occurrences of in the books of Kingdoms (4 Kgdms 20:18). is rendered much more frequently by . These factors suggest that represents a correction of OG to the reading of a Hebrew text different from MT. The first reading, , is almost certainly the OG reading. 3:4

1° {Bya2 Acx d+ fmw aeinb2 }] pr (M)Nhv; (itv )

g | pr

boc2 e2 s

The second and third readings may be taken together, since the omitted in the second is probably the result of haplography. The first reading is clearly the better attested of the two main readings, containing all of Johnson’s groups except the Lucianic mss, as well as a few mss that do not fit in any particular group. The third reading is attested by two uncials, the Lucianic mss, and a few other mss. It is also supported by a similar reading in an it ms: et astitit ei dominus et clamabat. One could argue that the third reading originated from the first as the result of assimilation to verse 10. The first reading may also claim precedence on the grounds that it is the shorter reading. Scribes might also have considered God’s call of Samuel abrupt, leading them to introduce God’s words with mention of his arrival in the sanctuary. On the other hand, the omission of (or even , cf. verse 10) could be due to parablepsis ( … ). The possibility that the shorter reading represents a correction to a Hebrew text similar to MT seems unlikely, for, though some of the witnesses are hexaplaric (Acx) and others sometimes have hexaplaric readings 28 Johnson says of the group d+, “d+ zeigt gelegentlich, besonders bei Eigennamen, eine Textform, die besser mit MT übereinstimmt, als es bei A cx der Fall ist. Dagegen hat d+ kaum eigene Varianten von hexaplarischem Charakter aufzuweisen.” Johnson, Hexaplarische Rezension, 107.

41

(d+, f+), there is little evidence that B is ever corrected toward a Hebrew text.29 If the longer reading is taken to be original, another factor must be considered, namely, that the subject must also be displaced, since a text ( ) , though perfectly acceptable in Greek, does not represent normal Hebrew idiom, since the subject is too far from the initial verb (cf. verse 10: ). It may be questioned why, if is a harmonizing addition, the subject was not moved forward in the sentence to match verse 10. The answer may be that the Greek scribes, unfamiliar with Hebrew idiom, felt no discomfort at the distance of the subject from the initial verb. A Greek form of the works of Ephraem Syrus, preserved in the apparatus of Holmes and Parsons, provides a form of the text that may also have some bearing on the question of the position of the subject. Ephraem reads . Although “Samuel” cannot be the original subject, its presence is an attempt to clarify the subject of , an indication of the difficulty Ephraem (or the scribes) felt with the dangling verb, so the reading presupposes a text identical with the third reading above. Thus, if a longer reading was original in OG, it was probably not that of the third reading. Since internal evidence is ambiguous, it is perhaps best to rely on the reading that has the most, the oldest, and the most diverse witnesses. Though some doubt must exist because of the possibility of parablepsis, the reading of OG was probably that of B and its congeners. 3:5

{Bya2 A dpqt i}] pr msw Mgv; it | + Nf

3:6

Bya2 i; cop eths] MN{aeghnvb2 }

cx aehnb2 ; arm |

244 | +

3:6

{B*y A dpqt i}] pr

3:9

B A qt i] pr

Acx d+; arm | +

boc2 e2 z {boc2 e2 f+}

cx boc2 e2 z f+ BabMNaeghnva2 b2 ; arm

{ya2 cx boc2 e2 dpz f+} MN{aeghnvb2 }; arm eth

All these variants must be considered together from two perspectives, first looking at the presence or absence of between and in verses 5, 6, and 9; and next looking at the presence or absence of (or ) in each of these verses. First, it may be noted that the majority of the witnesses in each verse include , whereas MT does not have a conjunction between the corresponding words. When one looks at the grouping of witnesses, it is surprising to see that the hexaplaric mss cx consistently include the , with A alone omitting it. The group d+ (except z, which often has Lucianic affinities) omits in the first two instances, but two of its members, dp, apparently join the majority in the third instance. The group f+ and the Lucianic group include in all 29 Johnson, Hexaplarische Rezension, 53: “Zusammenfassend wäre zu sagen, dass B nur in sehr

geringen Umfang Stellen enthält, die auf hexaplarische Korrecturen hindeuten könnten. . . . Hexaplarische Korrecturen können in vereinzelten Fällen eingedrungen sein, doch lassen sich die betreffenden Stellen auch auf andere Weise erklären.”

42

three cases. What does this evidence imply? It is clear from an examination of both Greek and Hebrew mss that the addition or omission of a conjunction is common, so it is impossible to tell for certain whether the change occurred before or after the initial translation into Greek. Though Hebrew uses conjunctions with great frequency, asyndeton in certain constructions is fairly common. Such a construction occurs here, where the first imperative is almost an auxiliary to the main idea expressed in the second verb . 30 However, similar constructions do appear with the conjunction. It may have seemed to a scribe or translator that the addition of a between the imperatives provided a better balance to the that was between the indicatives in the next phrase. Since internal evidence is questionable, it seems best to go with the preponderance of external evidence in this case, which favors the inclusion of the in the first instance and strongly 31 favors its inclusion in the last two instances. With regard to the question of the inclusion or exclusion of or in verses 5 and 6, the shortest text, supported by the group Bya2 in both cases, excludes the terms, while MT includes in verse 6. It is significant that in verse 9, all Greek mss include after , but there is no corresponding Hebrew term in MT. The easiest issue to address is the expression in verse 6, an obvious correction to a Hebrew text similar to MT by the hexaplaric and d+ groups. is a common translational equivalent for , rendering it twenty times in Kingdoms (in A, nineteen times in B), though always in sections that are not kaige. It is not nearly as common as , however, which renders some eight hundred times, in both OG and kaige sections. Thus, is an attempt to yield a text closer to the Hebrew. The short reading in verse 6 is the key to determining the OG reading in verses 5 and 6. Though supported by only a small number of witnesses, both Greek and versional, this reading is almost certainly the OG reading. If original, it is easy to understand that the longer text of the third reading could be an assimilation to verse 9 or to some non-Masoretic Hebrew text that added , but it is difficult to explain the origin of the shorter text if the longer were original. If some scribe felt the need to add or in verse 6, or to assimilate the verse to verse 9, the same process could have occurred in verse 5, where the group d+ joins the witnesses to the shorter text. Therefore, it appears that the shorter text without or is the OG reading in both verses 5 and 6, but in verse 5 could very well reflect a Hebrew text different from MT. 3:6

{Ba2 Acx d+}]

y|

MNaghb2 | +

30 E. Kautzsch, ed., Gesenius’ Hebrew Grammar, trans. A. E. Cowley, 2d ed. (Oxford: Clarendon

Press, 1910), § 120g. 31 Jellicoe, along with many others, believes that the Lucianic witnesses, which consistently

include the in all three verses, sometimes preserve primitive readings which have disappeared from most or all other witnesses, including B. See Jellicoe, LXX and Modern Study, 168-71.

43 i|+ 3:6 3:8 3:8

boc2 e2 f+ env {Bya2 boc2 e2 f+ MNaeghinvb2 }] +

Acx d+

1° {Bya2 Acx boc2 e2 d+ f+ MNaeghinv}] + {Bya2 Mehinvb2 }] pr boc2 e2 Ng

a | pr

Acx | pr

b2 ; arm it d+ | pr

f+ | +

The repetition and variation in the story is evident again in verses 6 and 8. At the beginning of verse 6, MT reads ; it lacks an equivalent for the in verse 8. The first item to note is the two forms of the clause that appear in the third and fifth variants in the first unit of variation. Both of these variants conflate the text by adding a clause closer to the Hebrew text that lay before the scribes, namely, the reading of MT. The result is an unwieldy and redundant sentence that certainly is not original.32 Little doubt can exist that is an addition not present in OG. The next bit of variation involves the position of in verses 6 and 8. In verse 6, does it belong before , after , or after ? Or should it be omitted altogether, with Ba2 ? The hexaplaric and d+ groups place in a position equivalent to that of in MT. Most other witnesses put it only in the first clause, either before or after . The variety of position suggests that is a later addition to the text and that the OG reading is that of Ba2 . 33 In verse 8, each of Johnson’s groups supports a different reading: Bya2 variant one, hexaplaric variant three, d+ variant four, f+ variant five, and Lucianic variant six. (Only those readings in v. 8 that deal with will be discussed here; see below for the other variants in this verse). The base text is equivalent to MT, so there is no question of other groups altering the text toward one with the same reading as MT. The third reading is clearly an inner-Greek error for , so variants three, four, and five may be considered together. As in verse 6, the hexaplaric and d+ groups support the inclusion of after a form of , and they are joined by the f+ group. However, the two possible positions for in the verse, and the likelihood of assimilation to the readings of verse 6, make the reading questionable. The failure of the Lucianic group to support the reading also tends to make one suspect that it is not original. Probably, then, is original in neither verse 6 nor verse 8. It may be noted quickly that the omission of in verse 6 (variant 2) is 32 Ms g omits the second phrase

, but this omission is probably due to

parablepsis rather than a conscious effort to improve the text. 33 There are three main occasions for transposition. The first is when a scribe simply inverts the

order of words or phrases that are adjacent or nearly so. The second is when a unit of material is accidentally omitted by a scribe, only to be added back by another scribe in the wrong place. Cf. Albert Curtis Clark, The Descent of Manuscripts (Oxford: Clarendon Press, 1918; reprint, Norwich: Fletcher and Son, 1969), 256. The third occasion for transposition occurs when an addition is made to the text, often as the result of a scribal gloss in the margin. See Shemaryahu Talmon, “Aspects of the Textual Transmission of the Bible in the Light of Qumran Manuscripts,” Textus 4 (1964): 100-103, who discusses the marginal and interlinear notes in Qumran biblical mss and their role in the production of conflate readings.

44

probably the result of haplography (perhaps omitting the first word of 3:6

3° {Bya2 boe2 f+ MNaeghinvb2 }] pr Acx c2 d+; arm eths

).

44 | pr

The third variant is equivalent to the reading of MT, and it appears in the hexaplaric and d+ groups, plus c 2 and two versions. The real questions are (1) whether is an addition added as an assimilation to verse 8 or a clause omitted by parablepsis ( ... ) and (2) whether the assimilation or parablepsis occurred in Hebrew or in Greek. Verse 6 MT at this place reads , but verse 8 reads (also reflected in ms 44 in v. 6). It is clear that if the longer text of verse 6 is the result of assimilation, it is not perfect assimilation, since is missing. Rather than imperfect assimilation, it seems more likely that (or its Hebrew equivalent) was accidentally omitted. The descriptions of Samuel’s actions the first three times God calls him are all different in detail, but in MT Samuel always does two things. In verses 4 and 5, Samuel says, “Here I am!” and then runs to Eli. In verses 6 and 8, Samuel gets up and then goes to Eli. This balance is destroyed if verse 6 omits his getting up, but if both actions are included, the variety of presentation is preserved by the presence of Samuel’s name a second time in verse 6 ( ). It is difficult to decide where in the process of transmission the parablepsis occurred, in a Greek or a Hebrew ms. However, if the longer reading was present in OG, then disappeared, it apparently left no trace in the ms tradition, since the hexaplaric and d+ groups probably derived the reading from Origen’s correction to the Hebrew. It is possible that a trace might remain in the Lucianic witnesses c2 arm, but it is more likely that these, too, reflect the influence of the Hexapla. It seems probable, then, that the parablepsis was already present in the Hebrew Vorlage of OG, so the first variant represents the OG text. 3:6

{Bya2 Acx d+ i}]

boc2 e2 f+ MNaeghnvb2

The first reading is closer to normal Greek idiom (though the article is usually omitted in classical Greek) than the second.34 occurs twice more in Kingdoms—2 Kgdms 14:29 and 3 Kgdms 19:7—both times for ; both of these occurrences are also adverbial. 2 Kgdms 16:19 is also adverbial and renders in MT as , but the adverbial idea is different: “moreover” rather than “a second time.” 3 Kgdms 9:2 renders , “a second time,” as . A broader look at the word in LXX reveals that the Pentateuch uses ( ) exclusively to render 34 Smyth, Grammar, 288. Cf. also p. 317, where he notes that

numerals as partitive genitives; the use of be idiomatic, at least in classical Greek.

is sometimes used with with an ordinal to represent an adverbial idea does not seem to

45

when it means “a second time,” a total of nine times. On the other hand, the other books almost always render , “a second time,” by . 35 It is interesting that B omits two of these occurrences, Josh 5:2 and 1 Chr 29:22. The only occurrence of to render outside the Pentateuch and Kingdoms is B’s text of Jer 40:1 (Heb 33:1), but A has . It appears, then, that the tendency of LXX outside the Pentateuch is to use , but B differs three out of eight times. Since B sometimes displays a tendency towards Atticism (see above, pp. 39-40, on vs. in verses 3 and 7), and since it elsewhere avoids , it seems more likely that the scribe of B, or some previous scribe, would have changed to the more idiomatic than vice versa. 3:7

{Bya2 MNaeghinvb2 }] pr

Acx boc2 e2 d+ f+ r

occurs frequently in Kingdoms both with and without the article. When the examination is restricted to the accusative , the vast majority of anarthrous cases fall into one of two types. The first common use of without the article is in the expression ( ) x, where x is a pronoun or noun in the genitive case (e.g., [1 Kgdms 7:8], [3 Kgdms 15:30]). On one occasion, 1 Kgdms 5:7, the similar construction occurs. The second common use of the accusative of without the article is in plural references to other gods (e.g., 1 Kgdms 28:13; 3 Kgdms 14:9). Once these cases are eliminated, only thirteen instances of the anarthrous remain, ten of them in 4 Kingdoms (a kaige section). Of eleven occurrences of the accusative of with the article, two are plural references to other gods (1 Kgdms 7:3; 2 Kgdms 5:21) and one is in the expression (3 Kgdms 16:13), though AN and many other witnesses do not have before . Only two cases of with the article occur in kaige sections (2 Kgdms 12:16; 22:7). It appears, then, that, apart from the expression x (genitive), the singular accusative references to in Kingdoms generally use the article in OG sections (six of nine times with the article) and generally do not in kaige sections (ten of twelve without the article, all in 4 Kingdoms). Nevertheless, without the article is still fairly common in OG, occurring in 33% of the cases, 40% if this case is considered anarthrous. Furthermore, one of the other instances of anarthrous is in the same chapter, 1 Kgdms 3:13. It seems best, then, to omit the article here, taking its inclusion as an attempt to conform to the more usual pattern. 35 Josh 5:2; 1 Chr 29:22; Jon 3:1; Hag 2:2; Zech 4:12; Jer 1:13. Dan 2:7 LXX renders the

Aramaic

with

, while Theodotion uses

.

46 3:7

{Bya2 Acx p*qtz}] pr pb? | pr MNaeghnvb2 | pr i | antequam adaperiret ; it

boc2 e2 d f+

The second, fourth, and fifth variants in this unit of variation should be seen as witnesses to the reading of the third variant, , the second and fourth variants being inner-Greek corruptions of the reading. The real textual question, then, is whether the preposition appears once or twice. MT reads before both verbs, so the second could be seen as an attempt to conform the Greek text to one similar to MT. Whereas Hebrew prepositions usually govern only one word,36 Greek prepositions frequently govern more than one object.37 Another reason for the second , then, might be conformity to Hebrew idiom. It is particularly significant that the hexaplaric mss Acx agree with B against MT. Thus, it is probable that the reading of the base text is the original reading of LXX. 3:8

{Bya2 Mehinvb2 }] pr boc2 e2 Ng

a | pr

Acx | pr

d+ | pr

f+ | +

The variants that concern have been dealt with above, but it remains to consider the article and the second . The second is probably an assimilation to the previous two accounts of God’s attempt to communicate with Samuel, verses 4 and 6. In both of those cases, and in the Lucianic witnesses here, is clearly considered to be a vocative. The insertion of the accusative article in variants two and five indicates an understanding that here is the direct object of the infinitive. The article was probably inserted to clarify this point. Therefore, the reading of the base text is to be preferred. 3:9

{Bya2 f+ Maghivb2 }] + Nen; ethvid | + cop | + Acx boc2 e2 pqtz

d | + Heli ei ;

Readings two through five present four different attempts to specify the participants in the action. The longest reading, supported by the hexaplaric, d+ (-d), and Lucianic groups, is the same as MT. Since no reason seems to exist for shortening the text, all of the longer readings can be seen as attempts at specification or at conformity to a Hebrew text similar to MT. The variety of readings also suggests that none of the longer texts is original. 36 But see Kautzsch, ed., Gesenius’ Hebrew Grammar, § 119hh. O’Connor calls this phenomenon prepositional override, citing 1 Sam 15:22 as an example. See Bruce K. Waltke and M. O’Connor, An Introduction to Biblical Hebrew Syntax (Winona Lake, IN: Eisenbrauns, 1990), 222-23; M. O’Connor, Hebrew Verse Structure (Winona Lake, IN: Eisenbrauns, 1980), 310-11. 37 Smyth, Grammar, 369-70.

47 3:9

{Bya2 A dpqt ein}] +

cx | +

boc2 e2 z f+ MNaghvb2 ; Thdt

Two hexaplaric mss again add , almost certainly secondarily. More significant is the longer reading that includes and is supported by many witnesses, including the f+ and Lucianic groups. Although no equivalent for appears in MT, it is conceivable that the reading could have arisen from a Hebrew text whose verb could be read as a participle rather than an imperfect.38 The reading seems to be an attempt to specify the subject of the verb, though in a rather unusual way. A participle used in proximity to a verbal form of the same word is probably a Hebraism, but Greek scribes were certainly capable of creating their own Hebraisms. Since the Lucianic text is known for expansions of this type, it is probable that the variant is secondary and within the Greek tradition, so the first reading is preferable. boc2 e2 d+ f+ AMNaa?cegnvxa; arm etha39

3:9

{Bya2 x a*hi*b2 }] +

3:10

{Bya2 A o dpqt fm*sw MNaeghinb2 }] +

cx bc2 e2 z mbv; arm

These two sets of variants exist because scribes were troubled at the lack of agreement between what Samuel was supposed to say and what he actually said. The fact that many witnesses that support the longer reading in the first unit of variation support the shorter reading in the second unit suggests that scribes attempted to correct the problem either by deleting in the first instance or adding it in the second. On the other hand, might have been added in verse 9 or omitted in verse 10 in order to conform the text to a Hebrew text similar to MT. However, other units of variation indicate that most scribes were more concerned with consistency than conformity to a Hebrew text, especially since few scribes could read Hebrew.40 The hexaplaric witnesses are usually an exception to this rule, but here they are split, indicating the importance of internal consistency to at least two of the scribes (the scribes of cx or their predecessors, though x* does read alone in verse 9). These considerations, plus the stronger external support for the longer reading in verse 9 and the shorter in verse 10, indicate that these readings should be preferred, with Rahlfs. 41 38 The possibility that LXX was translated from Old Hebrew script rather than square script has

been raised by various scholars. If so, the present case may be explained as a confusion of the of the imperfect with a , taken as an article, since the two letters are similar in the older script ( and , respectively). Cf. Shemaryahu Talmon, “Ancient Hebrew Alphabet,” in Mélanges Delcor, 387-402. 39 Thdt reads

; Natalio Fernández Marcos and José Ramón Busto Saiz, Theodoreti Cyrensis quaestiones in Reges et paralipomena, Textos y estudios “Cardinal Cisneros”, no. 32 (Madrid: Instituto “Arias Montano” Consejo Superior de Investigaciones Científicas, 1984), 12. 40 Cf. Wellhausen, Bücher Samuelis, 4-5. 41 Several commentators refer to ms B in v. 9 as though it were the LXX reading. The fact that B

here may not be the original LXX reading illustrates the importance of analyzing the variants within LXX before using one of the mss to reconstruct the original Hebrew text. Cf. P. Paul Dhorme, Les livres de Samuel, Études bibliques (Paris: Librairie Victor Lecoffre, 1910), 43; R. W. Klein, 1 Samuel, 30.

48 3:10

{Bya2 f+ MNaeghinvb2 }] Mmg; it | boc2 e2 | +

d|

cx | A pqtz; arm

MT here reads , which corresponds to the sixth variant, supported by A, d+ (-d), arm. The variants which substitute for the second clearly preserve an attempt to produce a better Greek text. It is interesting that in the only verse where MT reads , most Greek mss do not (cf. vv. 4, 6). Since the addition of can be explained as an attempt to conform to the common Hebrew text, the shorter text (variant one) is preferable to the others. 3:11

Bya2 ] pr

A(cx) MN {boc2 e2 d+ f+ aeghinvb2 }; arm eth

Though the vast majority of mss have , it seems more likely that it was added to make the text closer to a Hebrew text similar to MT (which reads ) than that it was deleted by Bya2 . A similar expression appears in two other passages in LXX: 2 Kings 21:12 and Jer 19:3. Though is found in both these passages, other elements of the formula are different, indicating the lack of a fixed form for this stock phrase. Of course, it is possible that was omitted by haplography, but B elsewhere lacks a tendency to conform to the Hebrew text where others do (cf. v. 6). In addition, the omission of the conjunction is supported by T. Thus, the shorter reading is to be preferred. 3:13 |

Bya2 Nhi(v)b2 ] boc2 e2 z | A| M {f+ aegn}| praedixi ; it

dpqt |

cx

The it reading is not particularly helpful in determining which Greek reading it reflects, so it must be set aside. The reading of A seems to be an error, perhaps an aural confusion of the first variant. The other variants reflect two sets of differences: aorist or perfect tense, and a prefix of - or -. The perfect tense, being used less frequently than the aorist, is more probable because scribes would be more likely to change perfect to aorist than vice versa. External evidence also supports the originality of the perfect. The choice between the prefixes is not so easy. Both prefixes are well represented, though appears in a variety of forms. A survey of and in Kingdoms reveals that is more common then in the OG sections (fiftyseven and twelve times in A, sixty-seven and seven times in B, respectively), but is more common than in the kaige sections (thirty-one and seventeen times in A, thirty-one and twenty-one times in B, respectively). Furthermore, is somewhat more frequent in LXX as a whole. These figures suggest that, in the OG sections especially, scribes would have a greater tendency to replace with than vice versa. B in particular strongly favors in the OG sections. The fact that B keeps here, perhaps along

49

with a predecessor of A, indicates that the first variant, the reading of the base text, is the best. 3:13

{Bya2 cx Maehinb2 }] pr

A boc2 e2 d+ f+ Ngv

MT here reads , with no preposition such as corresponding to . In fact, there seem to be no cases in MT of the verb with a preposition before the personal object (excluding , used of deciding between parties). However, frequently takes after it (in about 16 of 36 cases), particularly when rendering (whether with or without in MT, in about 14 of 25 cases). It seems, then, that was added during the process of transmission of the Greek text and that the shorter reading is preferable. 3:13

2° {Bya2 Acx Mag}] quas ipsi scivit ; itb | + Chrvid | + bc2 e2 z | + e|+

b´ dpqt f+ Nhinvb2 ; o | quas ego scio ; itv

All of the variants that differ from the base text are modeled on the phrase found in MT, . Variants five and seven change the subject of the verb to God (“which I knew”), probably by inner-Greek and/or inner-Latin corruption. The main Lucianic reading, , differs from the majority reading, , because the case of the relative pronoun has been attracted to that of its antecedent, . 42 The additional found in variant six is apparently a corruption of the main Lucianic reading. The second variant is basically equivalent to the majority reading. Thus, all the other variants from the base text are derived from the third variant. It is more likely that the phrase was added to agree with a Hebrew text similar to MT than that it was deleted, since no problem with the reading is evident. Thus, the base text should be accepted as original. 3:13 3:15 3:15

B Ac c2 e2 N] pr

{ya2 x bo d+ f+} M{aeghinvb2 }; Chr

{Bya2 A dp MNaeghinvb2 }] pr Bya2 ] pr

cx boc2 e2 qtz f+

A{cx boc2 e2 d+ f+} MN{aeghinvb2 }

The ease with which scribes added and omitted articles has been mentioned above, so it is difficult to have any certainty with regard to them. However, since articles were present more often than not, lacking evidence of haplography, it seems more likely that a scribe would have added the article than omitted it. Construct relationships like the first case occur 17 times with the article and 5 times without in the chapter, excluding this one, so it is likely that a scribe would have changed the reading of his exemplar in favor of the more common reading. In the second case, the phrase occurs three other times in Samuel, none of which has an article. The addition of the article in this case was a 42 Smyth, Grammar, 567.

50

move in the direction of more articles, which reflected better Greek idiom. Thus, the article should be omitted in the first two cases. The third unit of variation is somewhat different, both in the scantiness of external support for the base text and in the fact that the use of the article with the dative of the direct and indirect objects is a special case. Since most Hebrew names are indeclinable as transliterated into Greek, those forms of the dative that cannot take prepositions (i.e., the pure dative) have no way of indicating case without the article. Since the article is not used in Hebrew, its use in LXX is clearly a semantic indicator inserted by the Greek translators. Of about twelve other cases of proper names used as datives of direct or indirect objects in the first twelve chapters of 1 Samuel, all of them have the article. Thus, it seems probable that the article should be included in the present case as well. 3:17

Ba2 ; cop it] + arm (eth)

A(x) {c boc2 e2 d+ f+} MN{aeghinvyb2 };

The expression ... occurs eleven other times in LXX: ten times in Kingdoms and once in Ruth. In each of these cases the full expression is found; is never omitted. Though it is conceivable that the second part of the phrase could be omitted as being unidiomatic in Greek, none of the mss ever do so in any of the other cases. On the other hand, could have been added to conform the passage to the usual expression and to match the Hebrew, but if the omission is original in LXX, the reason for this uncommon readings would still be unexplained. It seems more likely, then, that the original was omitted by B or one of its predecessors, perhaps because of some graphic similarity between and . 43 3:17

Ba2 A] > cx e2 b2 ; ethvid | pr

{boc 2 d+ f+} MN{abeghinovyc2 }

In this unit of variation, the omission in the second variant is the result of parablepsis. Since more nouns have articles than do not, one’s initial reaction, in spite of the extremely thin external support, might be to exclude the article as a later addition. However, two factors argue against this evaluation. First, MT here reads , so the third variant cannot be an attempt to conform to MT (at least not completely), since is singular and is plural. Second and more importantly, the meaning of is not really appropriate to the context, for the phrase 43 A possible two step sequence of events that could lead to the omission of the phrase is as

follows. First, the scribe might have accidentally written a second time for . The resulting phrase would be . Since this new phrase would make no sense to a second scribe reading the exemplar, the easiest correction he could make would be to omit .

51

means, “from all conceivable words that were spoken,”44 but the context clearly indicates that specific words (i.e., God’s words) are being requested. The article probably dropped out as a result of parablepsis. 3:17

2° {Ba2 Acx dpqt i}]

y | > boc2 e2 z f+ MNaeghnvb2 ; arm cop

MT here reads ; it has nothing that corresponds to . Since (or ) functions as the equivalent of , the longer Greek text found in the first variant is redundant. It is possible that is a simple equivalent of the circumlocution and that the longer Greek text is a conflation of these two equivalents.45 Even so, the question remains whether the conflation occurred subsequent to the original translation or whether it was taken from the Hebrew Vorlage. If the longer text were in the Vorlage of LXX, no good reason exists for the omission of . However, if the shorter text represented OG, later scribes might well have added to bring the Greek closer to a text like MT. It is important to observe that among the mss that support this apparent addition are the hexaplaric mss and group d+, which often has hexaplaric readings. The fact that Ba2 (y) also support the longer reading demonstrates the occasional secondary readings of the group in general, and of B in particular. is probably an inner-Greek variant, but it could also be an independent correction to the same Hebrew text. It seems, then, that the shorter text represented by the third variant has a greater claim to originality. 3:18 d(p)qtz; arm | +

{Bya2 v}] f+ MNaeghinb2

Acx boc2 e2

The second variant, though supported by many witnesses, should be considered an assimilation to a text similar to MT. It is supported by the hexaplaric, d+, and Lucianic groups, all of which either occasionally or frequently correct their Greek texts toward the reading of Hebrew texts. The third variant represents an alternative attempt to specify to whom Samuel was speaking, and so is secondary. This variant presupposes the first, rather than the second, variant as the text which needed correction. Thus, the first variant is to be preferred. 3:21

{Bya2 Acx dpqtz* aein}] 21 MNgb2 ; eth itv | boc2 e2 f(h)msvwza?; arm | et factum est profetam; itb Nothing in 1 Sam 3:21 MT corresponds to this phrase. However, it is possible 44 Smyth, Grammar, 296. 45 For a possible parallel, see 1 Sam 1:23. MT reads, in part,

LXX reads is a simplification of 400, 410-11.

, and . Many commentators contend that , but cf. S. D. Walters, “Hannah and Anna,”

52

that the phrase (or whichever is original) is itself a variant of in 3:20, probably a variant based on a different Hebrew text (or different pointing of the same consonantal text). If so, the in the third variant should be seen as assimilation to the previous verse. The genitive article is a matter of Greek idiom rather than Hebrew Vorlage. 46 Once these matters are settled, all that remains is to determine the original word order. Though the reading of the base text has much stronger external support than the second variant, the third and fourth variants also put the verb before the noun. However, the word order in the third variant may be explained as better Greek style, so the first variant may be preferred on the basis of its external support. 3:21 Magb2 |

Bya2 Nm] boc2 e2 h | domini ; eth itb syhj | A {d+ fsw einv}

x|>c|

A look at the rest of the books of Kingdoms reveals that the phrase ( one of these times) occurs seven times outside this chapter, and the similar phrase occurs four times. No other occurrences of ( ) appear, so it is evident that the genitive construction is much more common than the dative. The question is whether an original ( ) was corrected toward the more common or whether ( ) was corrected toward the of the previous verse. The similarity of the phrases in verses 20 and 21 might suggest that the correction was made toward the of verse 20. However, if the theory is true that the phrase in verse 21 is really a variant of that in verse 20, the two phrases should actually be very similar to one another, and later scribes, without an extant Hebrew text to prevent modifications of OG, would have been more inclined to substitute the more common construction. Thus, the reading of the base text is probably the best. 3:21

3° B boc2 e2 ] > {ya2 } A{cx d+ f+} MN {aeghinvb2 }; arm cop eth it

It is better Greek style to write ... than ... , which is a Hebraism. It is apparently better style in the languages of the various daughter versions of LXX as well. That is omitted for stylistic reasons in these versions is evident from a comparison of the same pattern in verse 20, where the same versions omit the in , but with Greek support only from ea2 . The multitude of Greek witnesses that join the versions in the present case can again be explained by the apparent lack of an extant Hebrew text for the purposes of comparison, since scribes that would have a tendency to correct towards the Hebrew would not be able to. Thus, though the omission of the has many more witnesses supporting it, the reading of B and the Lucianic witnesses should be preferred. 46 Smyth, Grammar, 451.

53 Summary of Deviations from the Base Text 3:2 3:3 3:5 3:6 3:6 3:7 3:9 3:9 3:15 3:17 3:17 3:17

] ]+ ] pr ] ] pr ]+ ] pr ]+ ] pr ]+ ] pr 2°] >

Peshitta The family of mss dependent upon 9a1 displays frequent deviations from the text of 7a1 that are best explained as improvements in the direction of more authentic Syriac style. In addition, several deviations seem to be in the direction of a Hebrew text similar to MT. Because of these observations, deviations in 9a1fam will not be considered apart from other supporting witnesses. A second identifiable group of Syriac mss consists of those mss based on 12a1, referred to as 12a1fam. This group shows a strong affinity for agreement with a third group, the witnesses to the standard Nestorian version (9c1, 10c1, 11c1). The corrector of 8a1 is also a witness to the standard Nestorian version, and restored portions of the ms may also sometimes be Nestorian in orientation.47 3:1

{6h1 6h4 7a1 7h12 7k3 8a1 9c1 9l1 10c1 11c1 12a1}] > 9a1fam

The base text of 7a1 agrees with LXX, whereas the omission of by 9a1fam is in agreement with MT. Since 9a1fam often corrects an earlier Syriac text on the basis of mss similar to MT, the reading of the base text is to be preferred here. 3:1

{6h1 6h4 7a1 7h12 7k3 8a1 9c1 9l1 10c1 11c1}]

9a1fam 12a1fam

Both , an active participle, and , a passive participle or adjective, have the same meaning. It is doubtful that variant one can be explained as assimilation to the Hebrew as in MT; it is more likely that one form developed from the other within Syriac itself. The form is more common in Syriac than , so it is probable that 47 OT in Syriac, VI-VII.

54

later scribes changed the active participle to the more commonly employed passive participle. This conclusion is supported by the rather free nature of many renderings in 12a1fam and especially 9a1fam. 3:3

{6h1 6h4 7a1 7h12 7k3 8a1 9c1 9l1 10c1 11c1 12a1}] 9a1fam

The base text agrees with MT and LXX in reading “the ark of God,” but 9a1fam and other mss read “the ark of the Lord God.” An examination of the thirty-four references to the ark in 1 Samuel 4-748 reveals that P agrees with both MT and LXX fourteen times, P agrees with MT against LXX nine times, P agrees with LXX against MT once, P agrees with neither MT nor LXX when the latter two agree seven times, and all three disagree three times. S. R. Driver, in his commentary on Samuel, says that P often agrees with the Lucianic witnesses when both differ from MT, but no significant agreement with LXXL occurs in these cases.49 An examination of the shorter expressions “ark of God” and “ark of the Lord” sheds some light on the solution to the textual problem. When MT and LXX agree in one or the other of these short expressions (nine times), P agrees with them in every case but one (4:11), and this one case is the only time MT reads instead of . When MT and LXX differ in these short phrases (seven times), P agrees with MT four times, agrees with LXX once, and goes its own way twice. Since MT and LXX are equivalent in the present case, it seems likely that P agreed with them. This deduction is confirmed by the fact that, though the expression “ark of the Lord God of Israel” is frequent in various mss of P (as well as in 6:3 LXX; 6:2 LXXcx), no other occurrence of “ark of the Lord God” without “of Israel” exists in any of the mss. Thus, the longer expression of 9a1fam should be seen as an inner-Syriac variant, and the reading of the base text should be accepted as original. 48 The second MT and LXX occurrence in 5:10 is missing from P by parablepsis, so is excluded

from consideration here. 49 P agrees with both MT and LXX L 14 times, P agrees with MT against LXXL 10 times, P agrees with LXXL against MT 2 times, P agrees with neither MT nor LXXL when the latter two agree 2

times, and all three disagree 6 times. It should also be noted that a shift in vocabulary in P occurs in 6:13. Before 6:13, P consistently uses , derived from LXX’s , to render . Frequent divergences from MT (and LXX) occur in these cases, as already noted. From 6:13 on, however, P renders with , clearly derived from Hebrew. In the 8 cases found from 6:13 on, P agrees with MT in every one of them. Before 6:13, P never agrees with MT more than five times in a row.

55

3:5

3:6

{7a1 8a1 9c1 10c1 11c1 12a1}] 7k3 9a1fam 9l1 ]

6h4 9a1fam | +

6h1 7h12

9a1fam

Though the addition of the pronominal suffix “you” in both verses might be considered a correction toward LXX, 9a1fam shows no particular tendency to correct toward LXX. Rather, it frequently demonstrates an independence in modifying the Syriac text that lay before it, apparently without the support of another exemplar. Thus, the additional may also be seen as an independent rendering. The additional in verse 5, derived from verse 6, is not found in either LXX or most mss of MT (though many LXX mss add after ; see above, p. 41). Kennicott lists two mss, one of them considered important by Goshen-Gottstein (the margin of ms 70), as adding the phrase, and two it mss do as well, according to the apparatus of BHS. Since scribes who corrected their Syriac mss with a Hebrew exemplar seem to have used one very similar to MT, the readings of the divergent Hebrew mss and the it mss perhaps suggest a different Hebrew tradition at this point in the text, one that has almost disappeared from the extant witnesses. On the other hand, the phrase could have arisen in the various witnesses independently by assimilation to the following verse. Since internal evidence is ambiguous, a look at external evidence will prove helpful. The first and third variants contain about the same number of pre-fifteenth century witnesses, but the third contains the earliest ms, 6h1, a sixth (or fifth) century ms. However, 6h4, another sixth century ms, also omits , though it adds . It is significant that half of the witnesses that support the first reading are representatives of the standard Nestorian version (9c1, 10c1, 11c1). It is also relevant that, when 8a1 lost two folios, they were replaced with text in a Nestorian hand. Since elsewhere in the first twelve chapters of 1 Samuel the standard Nestorian witnesses show no tendency to correct towards a Hebrew ms similar to MT, their agreement with 7a1 here suggests that the reading is original. Thus, the third reading, despite the support of 6h1, should be seen as an assimilation to the following verse that occurred early within the course of transmission of the book in Syriac. 3:6

1° {6h1 7a1 7k3 8a1 9a1 9c1 9l1 10c1 11c1 12a1}] +

3:6

2° {6h1 7a1 8a1 9c1 10c1 11c1 12a1}] +

6h4 7h12

6h4 7h12 7k3 9a1fam 9l1

Neither MT nor LXX support the longer readings in either case. The similarity of the ms grouping in the second unit of variation to that of the second unit in the previous set should be noted. The addition of 6h1 to 7a1, 8a1, and the witnesses to the standard Nestorian version makes the ms support of the base text almost overwhelming. The early witness 6h4 does support the second variant in each unit, but the ms elsewhere shows a

56

tendency to amplify the text for greater specificity (cf. the previous set of variants; 2:28 + ; 10:14 + ). On internal grounds, the shorter reading is preferable, since the longer reading is more specific. Thus, both internal and external probability support the readings of the base text in both cases. Summary of Deviations from the Base Text No deviations from the text of 7a1 are considered original in the chapter. Targum Sperber groups the witnesses to T that he uses in five groups. The first group, witnesses with Babylonian vocalization, includes the biblical mss p (the base text) and mwxy. Also included among the witnesses with Babylonian pointing are the haphtaroth mss, jk. The second group, witnesses with Tiberian (or no) vocalization, includes the biblical mss acf. The printed editions dbo make up the third category. Sperber’s fourth category, fragments of various targum mss, he refers to as Fr. Finally, the fifth group of witnesses, rabbinic citations, includes Aruk of R. Nathan (Ar), Yonah ibn Ganah (Gan), Rashi (Ra), and Kimh@i (K). Sperber uses subscripted numerals to refer to various marginal notes, often additional toseftoth, that occur in all the mss (e.g., w 1 ). Sperber divides his critical apparatus into two sections, the first dealing with differences in vocalization and the second with differences in the consonantal text. In the present study, differences in vocalization will be ignored unless they indicate significantly different understandings of the Hebrew text. In particular, when one set of consonants allows more than one set of vowel points (e.g., to indicate perfect or participle), the reading of the base text will be accepted, since it is the consonantal text that is being reconstructed. Of course, differences in punctuation are important indicators of how the text is to be understood and of how the translators understood the underlying Hebrew text. However, unless the difference in vowel points in T could reflect a difference in the underlying Hebrew consonantal text, pointing will be ignored. The history of the development of T raises the issue, discussed above for the Hebrew text, of what text to attempt to construct. Though T was an official targum, it was not a creation ex nihilo but relied on centuries of earlier oral and written renderings, particularly from Palestine. In light of this fact, the focus here will be to construct not the earliest form of any given verse but rather the “official” form. This approach will result in the exclusion from consideration of earlier, even pre-Christian, material (e.g., the marginal readings in Codex Reuchlinianus), but in the light of T’s textual history, this method seems best.

57 3:2

{mp f b}]

wy a do

Both the inclusion and the exclusion of the conjunction are supported by Babylonian and Tiberian witnesses. Though the Masoretic ms L does not have the conjunction, Kennicott lists nine mss that include it, one of which (187) is in GoshenGottstein’s list of important mss. LXX and P include the conjunction as well. It is always difficult to evaluate the inclusion and exclusion of conjunctions, since it was easy for scribes to add or delete them inadvertently. However, because T presupposes a text similar to MT in most cases, and because the conjunction is supported by LXX, P, and some Masoretic mss, it seems likely that it was also original in T and was only later deleted, either accidentally or in an attempt to conform to a Hebrew text like MT. 3:6

{mpwy af do}]

b

3:8

{mpwy a do}]

fb

The lack of a before the infinitive corresponds to the lack of a in the Hebrew Vorlage (also lacking in MT). The weakness of the textual support for the second reading in each instance is evident, especially since b is a printed edition rather than a ms. Also, the tendency of the scribes would normally be to add a , since the form without it was uncommon. Thus, the secondary nature of the forms with is apparent on both external and internal grounds. However, these readings have significance for the next set of variants. 3:7

{p af db}]

mwy o

The first variant is clearly to be taken as an infinitive, as the prefix indicates. The second variant could be an infinitive (as pointed by w) or a noun (as pointed by myo). The Aramaic construction, ( ) , is different from that of both MT and LXX, and is certainly the result of the meturgemanîn, so there is no question of a different Hebrew Vorlage here. The external evidence leans somewhat toward the first variant, since both Babylonian and Tiberian mss support it. As for internal evidence, infinitives with are far more frequent than those without. The is even retained after prepositions (e.g., 3:15; cf. also 10:13; 12:23 bis). In fact, the two occurrences of in 3:6 and 3:8 appear to be the only infinitives without in the first twelve chapters of the book. Lest one suppose that the root has some unique tendency toward the lack of a , an examination of all the occurrences of the infinitive of in T in the former prophets where there is a corresponding infinitive in MT reveals that the only other two occurrences (1 Sam 22:11; 1 Kings 22:13) do have . In two other passages, Judges 18:1 and 2 Kings 5:7, the infinitive of in MT has no corresponding infinitive in T (i.e., some other form of appears). The reason for the omission of the in 3:6, 8 is that no

58

was present in the Hebrew Vorlage. Since the construction found in T varies from that in MT, it is almost certain that the presence or absence of was not based on the Vorlage. Since the occurrence of infinitives with is much more common than those without , one would expect the meturgeman to use the more common form. It might be argued that since is the more difficult form, later scribes might have added the to make the text conform to the normal usage. However, the presence of with infinitives is so overwhelming, even in passages that have no corresponding infinitive in Hebrew (e.g., 1:21; 2:1 ter, 3 bis, 5 bis, 6 bis, etc.), that it seems more likely that the original was dropped by a scribe influenced by the two infinitives without in the preceding and following verses (i.e., ). The difficulty that scribes had in accepting as an infinitive may be reflected in the pointing of myo, which take the word as a noun. 3:9

{mpwy f dbo}]

a

in T generally renders Hebrew , the word found in most mss of MT. However, two mss listed by Kennicott read , the reading reflected by in a. Both and (taken as a temporal conjunction) fit the context equally well. However, is a more common conjunction than , and the scribe of a might have unconsciously replaced the latter with the former. Alternatively, because of graphic similarity, the scribe might have misread as , especially if the in his exemplar were smudged or faded at that point. These considerations, coupled with the preponderance of the external evidence, clearly indicate that the first variant is to be preferred. Nevertheless, it cannot be ruled out that the second variant reflects a correction to a varying Hebrew text. 3:10

{p a do}]

mwy f b

The phrase is a common circumlocution for the Hebrew in T.50 It occurs particularly frequently when God is portrayed in a manner that might be considered irreverent.51 The surprising reading is the simple , for one might have expected the circumlocution. It is easy to suppose that a scribe confronted with the shorter reading would have inserted , probably intentionally. It seems highly improbable that the longer reading, were it original, would have been shortened, regardless of the fact that a modern individual might consider the shorter reading closer to MT. The external evidence also supports the shorter reading somewhat. Though both variants are supported by Babylonian and Tiberian mss and by printed editions, the second variant has stronger 50 Etan Levine, The Aramaic Version of the Bible: Contents and Context, Beihefte zur Zeitschrift für die alttestamentliche Wissenschaft, no. 174 (Berlin: Walter de Gruyter, 1988), 58-59. 51 Scholars frequently explain such substitutions as attempts to avoid anthropomorphism, but, as

Levine notes, many much more blatant anthropomorphisms remain in the targumic texts; Levine, Aramaic Version, 55.

59

Babylonian support. However, it is the internal evidence that is most convincing, and it favors the shorter reading. 3:11 3:11

{p af d}]

mwy bo |

f

{mpwy a dbo}]

f

The external evidence in the first unit of variation favors the second variant, from which the third differs only orthographically. However, it is possible that the before in f in the first unit of variation is simply misplaced from the beginning of in the second unit. If so, f cannot be considered to support a deviation from the base text. Moreover, internal factors do not favor the alternatives to the base text. First, the reading of MT, , is equivalent to variants 2 and 3. Scribes would have been more likely to change the text to conform with the standard Hebrew text than to delete the ( ) . Second, the inclusion of ( ) creates a smoother text, connecting and subordinating the following clause to the preceding one. Third, the absence of in LXX mss Bya2 suggests that a Hebrew text lacking was current at one time (see above, p. 48). Thus, despite the stronger external attestation for the second and third variants, the base text should be preferred on the basis of the internal evidence. 3:16

{mp a bo}]

wy f d

These two variants could be considered to be simply inner-Aramaic variations, since as well as can function as the marker of the definite direct object. However, it is more common for T to render with , so it is possible that reflects a different Hebrew word. In fact, though Codex Leningradensis reads , many Masoretic mss (Kennicott lists twenty-three) read . LXX also supports a Hebrew by its reading, . It seems likely, then, that some Aramaic scribe, feeling that was not quite as literal as might be desired, corrected the text to agree more closely with the dominant Hebrew ms tradition, reflected in most Masoretic mss. The external evidence does not contradict this conclusion, so the first variant should be retained. 3:18

{mp a do}]

wy f b

These two expressions are basically equivalent, but the second is the more exact rendering of the Hebrew found in many Masoretic mss.52 The standard rendering of expressions based on and referring to people in T is the similar Aramaic phrase . This rendering is found consistently in 1 Samuel in all twenty-six cases. When God is the object of the expression (either , , or a pronominal suffix referring to 52 Though L reads the singular , many other mss either read the plural in the consonantal text or have the plural as a kethib-qere variant in the margin.

60

God), the standard rendering is (i.e., , , etc.), occurring all sixty times, excluding this one, in Samuel and Kings. In four of these instances, variants containing appear in one or two mss (2 Sam 11:27 f; 15:26 a; 1 Kings 11:38 yb; 2 Kings 18:3 a). However, these readings are surely secondary. The solidarity of the ms tradition leads to the conclusion that renderings including are either corrections to Hebrew idiom or, perhaps more likely, echoes of a lost Palestinian targum. Nevertheless, the official targum represented by T almost certainly read in all these cases, including the present one. 3:19

{pwy af do}]

b|

w1 | +

m

A look at the whole clause is helpful here: . MT here has . All the other versions support MT here, except L LXX , which adds after . It seems clear that the second and third variants are corrections to MT, and m is a conflation of these two renderings, similar to, but probably not genetically related to, that found in LXXL. All that remains is the reading of the base text, and it is by far the most strongly attested reading. Thus, it should be retained. One other item in the chapter that needs some comment is the form in 3:13. The vowels imply that the form is a peal, meaning “to be dull,” but the context demands a pael, whose meaning corresponds with the Hebrew piel of , meaning “to rebuke, punish.” Levy compounds the difficulty by citing the form as , which is anomalous.53 Because the context requires it, this verb is taken to be a pael in the statistics. Summary of Deviations from the Base Text 3:2 3:10

] ]

Vulgate V has the distinction among all the versions being considered, including the Hebrew, of being the only one to have a critical eclectic text. In fact, as noted above, two critical texts of V exist, the smaller Stuttgart version and the larger Roman version. The production of VR was entrusted to the Benedictine order, and in particular to the supervision of Dom H. Quentin, at different times president of the order and abbot of the 53 He designates it a pael, however; J. Levy, Chaldäisches Wörterbuch über die Targumim und einen grossen Theil des rabbinischen Schriftthums, 3d ed. (Leipzig: G. Engel, 1866; reprint, Cologne: Joseph Melzer Verlag, 1959), s.v. “ , ”. He mistakenly gives the reference to the present verse as 1 Sam 2:13.

61

monastery of St. Jerome. Quentin’s methodology for determining the original text of Jerome is to divide the mss into families. The three great families he discerns in Samuel and Kings are the Hispanic family (C LHX FG TOMB), the Alcuinian family ( RZGVP), and the Theodulfian family ( HSAMG). Three less important groups of mss are the Italian ( BDF), Parisian ( SJM ), and the Ambrosian ( AB) groups. Other mss have mixed texts, but D shares many readings with the Alcuinian family, and E shares readings with the Theodulfian family. The two oldest complete mss, R and A, belong to none of these groups,54 having a text that is often reflected in one or more of them but that is more primitive. When these two mss agree, Quentin says, they manifest Jerome’s text. When they disagree, an agreement with another old ms, C of the Hispanic group, provides Jerome’s text.55 Roberts criticizes this approach as too mechanical, but he notes that Quentin himself and the Benedictines who continued his work do not use this method without critical acumen.56 Because of this criticism, it will be important to compare the text of the Roman edition with that of the Stuttgart edition. When citing mss from a family or subfamily in which each ms has a common Greek letter and different superscripts, the agreement of all the mss with a certain reading will be indicated by giving merely the Greek letter (e.g., = AB). None of the fragmentary mss cited in VR has any readings in 1 Samuel 3. Printed editions are cited as lower case italic characters (i.e., agrelvwsc). Finally, patristic sources are occasionally noted, as follows: Gothic Breviary (Brev.goth.), Gregory the Great (Greg.M.), Isidore (Isid.), Bede (Beda). 3:3

lucerna RA C L* F K* wc] lucernam { Lmg HX G B D E Brev.goth. Greg.M. Isid. Beda

PHK mgI};

Most mss and many patristic sources prefer the accusative lucernam to the nominative lucerna, thereby restructuring verses 2 and 3. The deviation is clearly an innerLatin one, perhaps caused by confusion over whether lucerna had a line, representing m, over the a ( ). Another possible explanation for the variation is that scribes might have considered the construction with the nominative to be difficult and so substituted the accusative. This latter reason supports the nominative as the more difficult reading, as it seems unlikely that a scribe who had the accusative in his exemplar would have changed it to a nominative. Thus, though the preponderance of mss have the accusative, the nominative read by the early mss RAC and a few others is to be preferred. 54 Ferdinand Deist suggests that A may be associated with the work of Cassidorus, mentioned

above, p. 34; Deist, Text of OT, 212. 55 Biblia Sacra Romana, 5:xiv-xv. 56 B. J. Roberts, OT Text and Versions, 261-62.

62 3:4

Samuhel {RA CLXPSB DF JHSAG C W GA PHKI}] Samu[h]elem EJM GB

3:6

Samuhel 1° {RA CLXPSB DF JHSAG C W GA PHKI}] Samu[h]elem EJM GB relvsc

3:8

Samuhel {RA CLXPSB DF JHSAG C W GA PHKI}] Samu[h]elem EJM GB relvsc

3:9

ad Samuhel {RA CLXPSB DFRZGP JHSAG C W PHKI}] Samu[h]elem FV EJM G K relvsc

3:11

ad Samuhel {RA CLXP DFRZGV*P JHSAG C WSJM PHKI}] Samu[h]elem SB FV2 EJM WM G relvsc

Several mss and most editions treat Samuhel as a declinable proper noun, whereas the other witnesses consider it to be indeclinable (or at least they do not distinguish between nominative and accusative).57 Though the declinability of the word is an inner-Latin matter, the syntactic use of the indeclinable form is questionable in the first three units of variation. Whereas Samuhel in verses 9 and 11 are clearly accusative objects of the preposition ad (and thus the phrase ad Samuhel functions as the equivalent of a dative of indirect object), the same word in verses 4, 6, and 8 could be either accusative or vocative. Verse 10, where V reads Samuhel Samuhel as vocatives, must also be considered. MT in verse 4 reads . In verses 6 and 8 is ambiguous, since one would expect the direct object to be preceded by or a preposition;58 it is possible that is vocative.59 in verse 10 are certainly both vocative. LXX clearly assumes a vocative in verses 4 and 6 and a direct object in verse 8 (most mss do not include in verse 10; see above, p. 48). T agrees with MT in all four cases. Only P consistently and clearly renders the word as a direct object in the first three cases, but in verse 10 it reads a double vocative. When the context of V is examined, it seems probable that verse 8 should be taken as a direct object, but verses 4 and 6 could be vocatives. It is interesting that the editions relvsc do not have a variant Samuelem in verse 4 (or, of course, in verse 10). In conclusion, then, it is certain that the indeclinable form of the name is to be preferred in all the units of variation, but the function that Samuhel plays within verses 4 and 6 is debatable. 3:5

et dixit {RA C LH FG TB D RZGVP E H*SAMG* B

JM AB P mg}]

+ ei

57 Most mss are inconsistent in their use of case endings with Samuhel. Cf. v. 16, where almost

all mss (except D) read Samuhelem, and v. 21, where Samuheli and Samuhelis are both found. See Stummer, “Einige Beobachtungen über die Arbeitsweise des Hieronymus bei der Übersetzung des Alten Testaments aus der Hebraica Veritas,” Biblica 10 (1929): 1-30. 58 in Samuel and Kings is usually followed by , , , or , but cf. 1 Kings 22:13; 2 Sam 5:20; 1 Sam 9:24 (though this last passage may be corrupt). 59 Kautzsch, ed., Gesenius' Hebrew Grammar, § 115b, says, “The object after the infinitive

construct must also always be regarded as in the accusative, even when it is not expressly introduced . . . by the nota accusativi - .” This is indeed the case in 1 Kings 22:13. However, a verb like introducing a vocative or direct discourse does not seem to have been considered (cf. 2 Kings 11:14 for an example of direct discourse after infinitive construct).

63 X OM 3:6

H1 G2

dixit {A C X

P*HKI

DF

BD E

S

PHKI}] + ei R

It was a common scribal practice to provide more details when the context might be ambiguous (and often when it was not). Several mss in the first case, and one the of most important mss in the second case, specify to whom Samuel was speaking. However, these additions are clearly inner-Latin and secondary. 3:5

revertere {RA C LH* T* D RZGV H*SAMG BDF* Jmg M P*K}] + et H2 X T2 OMB P E H2 P 2HI F 2 SJ* agrelvsc; Greg.M. Brev.goth.

3:5

et 3° {R C LHX T* D RZGVP E Hmg AMG BDF T2 OMB H*S

SJM AB PHKI}]

>A

The addition and omission of conjunctions in V, as in other versions, was a common scribal practice, and external considerations must of necessity play a larger role in determining the original reading. Nevertheless, internal considerations do have a role to play. In the first unit of variation, a number of mss have a conjunction between the two imperatives. The addition of a conjunction at this point of the text is also found in several LXX mss, and, significantly, the it. However, the absence of the conjunction in any of the most ancient witnesses to V suggests that the presence of the conjunction is secondary, perhaps introduced on the basis of it texts; alternatively, it could have been added for stylistic reasons. The absence of the conjunction in the second unit of variation is not supported by as many witnesses as in the first case, but the important ms A supports the variant, as do important Hispanic and Theodulfian mss. However, no witnesses in any other version support the second variant. It is possible that the omission of the conjunction indicates that, in the mind of at least one scribe, abiit et dormivit properly belonged to the following phrase et adiecit dominus vocare rursum Samuhel. Since the preponderance of external evidence points to the first variant, and since no evidence is compelling enough to suggest preferring the second to the first, the first variant should be accepted. 3:9

ait 1° {RA C B E

PHKI*}] dixit X

D I2

; Greg.M.

These two variants are equivalent, and both aio and dico commonly render . In the first two accounts of God’s calling Samuel (verses 4 and 5), after Samuel runs to Eli, dixit is used. It is possible that later scribes, influenced by these similar passages, changed ait to dixit in order to achieve a greater homogeneity. Otherwise, the change might be considered an unconscious lexical substitution, influenced, no doubt, by the earlier passages. 3:10

loquere R*A C

G2

rw] + domine R2

G* {

X

B D E HSAM

PHKI

64 agelvsc}; Greg.M. The second variant agrees with LXXL, arm, and P. More importantly, it reflects Samuel actually saying what Eli instructed him to say in verse 9. The fact that Samuel did not say exactly what he was told to say in many witnesses (including MT, LXX, it, and T) caused the scribes some consternation, and some of them remedied the situation by adding “the Lord.” The question here is whether or not domine was original with Jerome. The age and importance of mss RAC supporting the first variant outweigh the numerical superiority of the second variant. Furthermore, the intrinsic probability of adding domine to an originally shorter text is high, while the probability of dropping an original domine is low. Though extant it mss agree with MT, their propensity for agreeing with LXXL suggests the possibility that other it mss no longer extant did add domine. If so, the second variant might be a correction to this text. Otherwise, it is an independent “improvement” of the text. 3:15

timebat indicare visionem {A C X S | 213 TO

D E HAMG

PHKI}] 132 R MB

The variable in these variants is the position of indicare. Should it be second as in variant one, third as in variant two, or first as in variant three? The other secondary witnesses agree with the first variant and with MT that indicare should be second in the phrase. The origin of the other variants may be a ms of V that inadvertently omitted indicare when the text was first written. The same scribe or a later one, noticing the mistake, added the word in the margin next to its proper place. Later copyists, however, unable to discern the exact location (although the infinitive in Hebrew generally follows the verb, Latin has no such constraint), inserted it in a variety of places. Notwithstanding the testimony of R, the first variant should be accepted as the preferred reading on the basis of superior external attestation. 3:17

ad te RA C D RZGVP2 M2 G2 P*; Goth.Brev.] dominus ad te X DF aelvsc; Greg.M. | ad te dominus P* E2 HSAM*G* B

B P2HKI

The problem with the first variant is that the subject is unspecified, either in the verse or in the immediately surrounding verses. The implied subject in variant one, in fact, is found in verse 11, six verses earlier. Variants two and three have no such problem, since the subject (dominus) is explicit. MT here also omits the subject, as does LXX, but with an important exception. Whereas the pointing in MT indicates an active verb with an implied subject, LXX has a passive verb, . The passive verb, of course, takes no subject, and so is no problem. LXXL, apparently still troubled by the lack of specificity, reads . P addresses the problem in a manner similar to variant three, . It is

65

unlikely that any direct connection exists between the rendering in P and that in mss of V, but the similar solution does indicate the manner in which scribes in various traditions addressed what they saw as textual difficulties. The variant position of dominus in variants two and three suggests that the word might be secondary to the text, as does the observation that the first variant is the most difficult reading. The fact that mss RAC all agree in the omission, as does the it ms Belsheim, sets the external evidence on the side of the first variant as well. Thus, the first variant should be accepted as original. Summary of Deviations from the Base Text No deviations from the text found in VR (which agrees completely with VS in the chapter) are considered original in chapter three. Aquila When presenting the readings of Aquila and the other partial secondary witnesses, the reading of LXX (or another complete secondary witness, in the case of other possible Hebrew readings) will be given as the collating base for identification purposes. Contested Readings 3:3

´3 witnesses |

] ´1 witness

The reading is reflected in three witnesses, while is found in only one. This latter variant, since it combines two essentially equivalent phrases, seems to be a conflation of ´ and another Greek reading related to ´. The first of these readings, then, should be preferred as the authentic reading of ´. Accepted Readings 1 2 3 4 5 6

3:1

]

´ ]

3:3 3:10 3:13 3:21

] pr

´

´ ]

´ ]

]

´ ´

Symmachus Contested Readings There are no contested readings in the chapter, aside from minor orthographical differences among mss.

66 Accepted Readings 1 2 3 4 5 6 7 8

3:1

]

´ ]

3:3 3:4 3:7

´ ]

]

´

´ ]

´ ]

3:8 3:21

] ]

´ ´ ´

Theodotion Contested Readings There are no contested readings in the chapter. Accepted Readings 1 2 3

3:8 3:13 3:21

Other Readings Attributed to

]

´ ]

]

´ ´

´

No evaluation of the readings will be attempted, since it is not clear that different readings in fact represent the same tradition. Purely orthographical variants will be omitted. To avoid confusion, the abbreviation ´ will be shortened to ´ in the collation. 1 3:1 ] ´ 2 3:2 ] ´ 3 ] ´ 4 3:3 ] ´ 5 ] ´ 6 ] ´ 7 3:4 ] ´ 8 ] ´ 9 3:6 ] ´ 10 ] ´ 11 ] ´ 12 ]+ ´ 13 ]+ ´ 14 3:10 ] ´

67

15 16 17 18 19 20 21 22 23

]

´

3:13

] ] ]

´ ´ ´

] ] 3:14 3:18 3:21

´ ´

] ]

´ ´ ...

] ´

The Lucianic Recension Lagarde’s edition of the Lucianic text has been criticized for various reasons, not least because he failed to include a critical apparatus with his text.60 Furthermore, he sometimes accepted the reading of one Lucianic witness against the combined testimony of the other three. For these reasons, Lagarde’s edition will be used, but all of his readings will not be accepted as being the pure Lucianic text. However, all of Lagarde’s readings that are rejected will be listed in a separate section. Studies on the Lucianic text have shown that the purest witnesses to that form of the text are boc2 e2 . The witness referred to as b is in fact the consensus of b´ and b. Of course, the Lucianic mss frequently agree with other groups of mss, but some individual mss occasionally agree with the main Lucianic mss on occasions in which all others differ. The ms which agrees the most often when others do not is z. In the collation below, the reading of LXX, as determined above, is given first, followed by the reading accepted as Lucianic.61 The Lucianic mss that support the reading will be listed, as will any mss or versions that agree uniquely with LXXL or are joined by only one more witness. If more than five other witnesses agree with the Lucianic reading against LXX, this fact will be indicated by the word “many” in parentheses. The agreement of from three to five mss with LXXL against LXX will be indicated by “few” in parentheses. It should be noted that any Lucianic reading that has no non-Lucianic witnesses listed in the collation as agreeing with it is a uniquely Lucianic reading.62 60 Jellicoe, LXX and Modern Study, 7-9; Alfred Rahlfs, Septuaginta-Studien, 2d ed., vols. 1-3 (Göttingen: Vandenhoeck & Ruprecht, 1965), 3:29. 61 The Lucianic reading will be determined primarily on external grounds, by counting the number

of Lucianic mss that support the reading and taking the majority reading. If the witnesses are split two to two, the reading that differs from the main LXX will be accepted. 62 That is, no other LXX witnesses or daughter versions of LXX support the reading. A unique Lucianic reading may be supported by a witness in another tradition (e.g., P) or by one or more of the minor Greek witnesses.

68 Readings Accepted as Lucianic 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

3:1 3:3

3:4 3:5 3:6 3:7

3:8 3:9 3:10 3:11 3:12 3:13

3:14

3:15

3:16 3:17

]

] boc2 e2 ; Thdt ]+ boc2 e2 (many) 2°] pr boc2 e2 z; arm 1°] pr boc2 e2 (many) 3°] boe2 z ]+ boc2 e2 (many) ]+ boc2 e2 (many) ]+ boc2 e2 (many) ] pr boc2 e2 (many) ] pr boc2 e2 (many) ] boc2 e2 ]+ boc2 e2 Ng ]+ boc2 e2 ]+ boc2 e2 (many) ]+ boc2 e2 (many) ] boc2 e2 ]+ bc2 e2 ] pr boc2 e2 (many) ] boc2 e2 ; eth ] boc2 e2 ] boc2 e2 z ] pr boc2 e2 (many) 2°] + bc2 e2 z ] boc2 e2 (many) ] boc2 e2 ] boc2 e2 ] pr boc2 e2 ] pr boc2 e2 (many) 2°] post boc2 e2 ] boc2 e2 (many) 2°] > boc2 e2 (many) ]+ boc2 e2 z ]+ boc2 e2 z ] post

36

bc2

3:18

1° boc2 e2 ]+ boc2 e2 (many)

69

37 38 39 40 41 42 43 44 45 46 47 48 49

1°] + boc2 e2 z ] boc2 e2 (few) ] pr bc2 e2 3°] > boc2 e2 ]+ boc2 e2 z ] boc2 e2 A ] boc2 e2 Mmg ] boc2 e2 (many) ] boc2 e2 za? 2°] > boc2 e2 ] ] boc2 e2 h (many read 1°] + boc2 e2 gzmg

3:19

3:20 3:21

boc2 e2 (many) )

Readings from Lagarde’s Edition Not Considered to Be True Lucianic Readings 3:2 3:2 3:3 3:7 3:8 3:13 3:13 3:20

b zmg o (many)

] ] ]

(no LXX witnesses)

]+ ]

b; it b Mmg ] b (many) ] pr bo (many) ] idem b´ (many)63

The Hexaplaric Recension The hexaplaric text-tradition arose historically out of Origen’s monumental work in creating the Hexapla. In the fifth column of this work, he included the text of LXX common in his day, but he made some changes. When he found a section in LXX that was not reflected in his Hebrew text in column one, he marked the section with an obelus (÷). When he found a section in the Hebrew not reflected in his text of LXX, he added it, usually from one of the minor Greek versions, and marked the section with an asterisk (Ë). The hexaplaric text arose when scribes copied the fifth column of Origen’s work, either ignoring or not paying careful attention to the asterisks and obeli. Even if the original 63 The reconstructed Lucianic text used in this study may be found in Appendix 1, below, p. ***. Except for minor matters such as orthography, punctuation, versification, and the use of square brackets, it agrees with the Lucianic text reconstructed by Bernard Taylor, The Lucianic Manuscripts of 1 Reigns, vol 1: Majority Text, Harvard Semitic Monographs, no. 50 (Atlanta: Scholars Press, 1992), 9-12, in all respects except one: the text reconstructed in the present study reads in v. 13, whereas Taylor’s text reads [ ] .

70

copyists were faithful to preserve the text-critical marks, later copyists were not. The result was the creation of a hybrid text, based on LXX, but supplemented from other Greek versions closer to the proto-Masoretic text. The importance of hexaplaric readings for textual criticism is two-fold. First, its renderings that vary from OG (especially the pluses) often reflect readings of one or more of the minor Greek versions, which are only preserved in fragments and glosses scattered over several mss. Second, and more important for this study, any renderings that differ from both OG and MT might reflect a correction to a Hebrew text somewhat different than MT. Thus, it is only this latter category of readings that will be considered here. Determining the hexaplaric text is problematic, since most mss and versions contain at least some hexaplaric readings. However, the studies of Bo Johnson help to isolate those witnesses that are particularly likely to preserve hexaplaric readings. In particular, Johnson concludes that the Greek mss Acx and, to a lesser extent, family d+ are the best representatives of the hexaplaric text in 1 Samuel.64 It is important to note that, since the hexaplaric readings are being compared with MT as well as LXX, a reading which reflects only variation from LXX will not be listed, even if it might be the authentic hexaplaric Greek text. This limitation is legitimate in light of the fact that the hexaplaric text-tradition is a partial secondary witness rather than a complete secondary witness. Two other important witnesses to the hexaplaric text in 1 Samuel are arm and syh. 65 Those hexaplaric readings that will be considered in this study are listed below, collated in the same manner as the Lucianic text was above. Accepted Hexaplaric Readings Differing from LXX and MT 1 2 3 4 5 6

3:3 3:8 3:10 3:13

]

Ac ] pr d+ f+; (arm)66 ]+ cx b+; arm (few) ] A; arm (cx b; cop read ] pr A d+ (many) ] cx; armvid itb

)

64 Johnson, Hexaplarische Rezension, 88. 65 Ibid., 88-89. Syh is a translation of the fifth column of the Hexapla, including the diacritical

marks. The textual history of the Armenian version is somewhat complex, having gone through three major stages: translation from Syriac, translation from Greek, and translation from another Greek texttradition. The last stage, when it was translated from a hexaplaric Greek text, is the most evident in the present Armenian text, but remnants of the earlier history of the version remain. These can be observed in agreements between arm and P and between arm and non-hexaplaric Greek texts. See Bo Johnson, Die armenische Bibelübersetzung als hexaplarischer Zeuge im 1. Samuelbuch, Coniectanea Biblica, Old Testament Series, no. 2 (Lund: CWK Gleerup, 1968), 13-17, 158-60. 66 Acx reads instead of the of d+ f+, almost certainly an inner-Greek corruption rather than a rendering of a Hebrew or . Arm, along with b2 and it v , places the equivalent of before the infinitive; cf. Johnson, Armenische Bibelübersetzung, 86.

71

7 8 9 10 11

3:14 3:15 3:21

2°] > cx d+; it ] cx 2°] > cx d ] A d+; syhj eth itb (c omits phrase, x reads ) 3°] > Acx; arm (many)

Rejected Readings from Hexaplaric Witnesses The following list contains those readings that are shared by at least two Greek hexaplaric mss but are not considered the true hexaplaric text. Other instances of witnesses that are often hexaplaric but that have unique or almost unique readings (especially readings found in arm and no other hexaplaric witness) are numerous but are not listed. 3:8 ] cx 3:9 ] cx (many) ]+ cx 3:11 ] pr cx 3:14 ] cx ] cx Ne2 * 3:17 1°] cx b´ 3:21 ] cx fm All these readings are unique to cx, among Hexaplaric witnesses. In each case, the reading of cx is an inner-Greek development from the main LXX text and so is textually inferior. These readings will not be considered further. 3:17 ] > cx b2 e2 ; ethvid The omission of here is probably the result of parablepsis, so this reading should not be considered an authentic hexaplaric reading. Other Possible Hebrew Readings In this final section, other readings from witnesses that might possibly reflect a Hebrew Vorlage different from the reading of MT are given. Only those variants are listed that are not reflected in any other secondary or partial secondary witness. 1 3:2 ] Nadenopqtyzb2 2 3:5 ]+ 6h1 7h12 7k3 9a1fam 9l1 3 3:9 ] a

CHAPTER 3 GRAMMATICAL/STYLISTIC CHARACTERISTICS OF THE SECONDARY WITNESSES Having determined the texts to be used in this study, the stage is now set to analyze the secondary witnesses and compare them with MT. The procedure to be followed is described above in Chapter 1, but it may be summarized briefly at this point. Simply put, the method starts with the mass of readings in the secondary witness that might possibly be considered variants and proceeds to weed out those that have no value for textual criticism, a sieve of Eratosthenes approach. The first readings to be eliminated are those which were likely forced by the nature of the differences between the source language (Hebrew) and the target language in question. Since the limitations of the target languages for rendering Hebrew are discussed at the beginning of the analysis of each version, these readings are never even listed. Those readings that are considered are listed under the category of variation that they represent. Next, a summary of the comparison of the secondary witness with MT is given, with specific statistics for each of Tov’s four categories and many sub-categories within these four. Armed with these findings, a preliminary partial translation technique is described, and significant and nonsignificant variants will be distinguished. Once the variants that are probably reflections of the translator’s rendering of the Hebrew have been determined, the partial translation technique will be recalculated for each category and subcategory, eliminating the significant variants from consideration in order to refine the translation technique. The rationale behind eliminating significant rather than nonsignificant variants is as follows (described in terms of LXX). The preliminary partial translation technique, that is, the one based on MT, is a comparison of the deviations in LXX from MT, which, as described above, is used initially as though it were the Vorlage that lay before the translators. Since it is not identical to their Vorlage, the deviation of LXX from MT will be greater than that between LXX and its actual Vorlage. The variants that are determined to be significant are those that probably reflect a difference in the Vorlage, and, although they do not agree with MT, they presumably agree with their Vorlage, and so should be not counted in the statistics as deviations from a literal translation technique. Whereas the partial translation technique based solely on MT will imply a disproportionately high degree of variation in the translation, the revised partial translation technique will imply a

72

73

disproportionately low degree of variation, since some of the variants considered significant at this stage will be eliminated later in Chapter 4. However, most of the nonsignificant variants will be identified in the present chapter, so the partial translation technique for each secondary version at the end of Chapter 3 should be reasonably close to the final translation technique as determined in Chapter 4. It may occasionally be necessary to repeat the process of recognizing significant variants and recalculating the translation technique for a particular category if the preliminary partial translation technique was skewed a large amount because of significant variants. Septuagint Limitations of Greek for Rendering Hebrew Since Greek is a member of that group of languages known as the Indo-European family of languages, whereas Hebrew is a Semitic language from the larger Afro-Asiatic family,1 it is not surprising to find that syntactic structure in the two is different. Nevertheless, Greek translators of Hebrew texts were capable of rendering almost every detail of their Hebrew Vorlagen, if they so chose. Different translators varied in their degree of conformity to Hebrew style, with Aquila being the most slavishly literal, even rendering by . The nominal system in Greek is a highly inflected one, employing five cases (nominative, genitive,2 dative,3 accusative, and vocative), three genders (masculine, feminine, and neuter), and three numbers (singular, plural, and dual). By contrast, biblical Hebrew relies primarily on word order and the use of prepositions to express the nuances present in the Greek cases,4 and it has only two genders (masculine and feminine), but it does have three numbers. Since the association of gender with inanimate objects, abstract ideas, and so on, is largely arbitrary (as far as can be determined now),5 any significant correlation between the Hebrew gender of a word and the gender of its Greek counterpart 1 Also known as the Hamito-Semitic family; Encyclopædia Britannica, “Languages of the World,”

740. 2 Includes ablative uses. 3 Includes locative and instrumental uses. 4 The morphological changes that occur in nouns in the construct state reflect phonetic and

rhythmical phenomena rather than remnants of case endings. Occasional nominative, genitive, and accusative endings on nouns support a picture of an earlier form of the Hebrew language that had three distinct case endings, like its Proto-Semitic ancestor. See Kautzsch, ed., Gesenius’ Hebrew Grammar, §§ 89-90; Waltke and O’Connor, Introduction, 127-28; Gotthelf Bersträsser, Einführung in die semitischen Sprachen, with an appendix, “Zur Syntax der Sprache von Ugarit,” by Carl Brockelmann (Munich: Max Hueber, 1928), 14-15. 5 For a comparison between the use of gender in Hebrew and in other languages, see Waltke and O’Connor, Introduction, 95-110.

74

seems unlikely, and the data from 1 Samuel supports this supposition. On the other hand, it would have been natural for the translators to have rendered the number of the Hebrew nouns by their Greek equivalents. Not surprisingly, the translators did do so for the most part, with the exception that, since the dual was no longer used in koine Greek, they expressed Hebrew duals with Greek plurals.6 The options open to the Greek translators regarding the use of Greek cases to express various Hebrew constructions were much broader than those concerning the rendering of gender or number. Greek translators, even those concerned with being literal, had the option of rendering prepositional phrases with nouns in the appropriate case rather than rendering both the preposition and the noun.7 Both methods of translation could be considered literal, though the translators themselves might have thought that rendering Hebrew prepositions with Greek ones was a more faithful rendering. However, the multiplicity of available options in Greek for rendering these Hebrew constructions requires that the translation technique be examined from several different angles to determine the translators’ own approach to translating them; labeling the translation as “literal” is not sufficient. Another aspect of the nominal system in Hebrew is the (definite) article. Greek also has an article, so one-to-one correspondence was possible in translation. However, the use of the article in idiomatic Greek often varied from what would be required of a strict one-toone rendering. For example, Greek articles could stand for relative pronouns or with infinitives, something that was not possible in Hebrew. Moreover, two barriers to a strict representation of the Hebrew article by the Greek article existed. The first was the failure of the Hebrew to use an article to identify definite nouns in the construct case; the second was the assimilation of the Hebrew article after an inseparable preposition. Whether the Greek translators would render the articles that they could see in the text or whether they would insert Greek articles for definite Hebrew nouns in which the article did not appear— or whether they would simply be inconsistent—is a matter for investigation. Many differences also exist between the verbal systems of the Hebrew and Greek languages. The Hebrew verb can be classified by stem, inflection, person, gender, and number. Greeks verbs have tense, voice, mood, person, and number. A correlation clearly exists between person and number,8 and it is equally clear that Greek will have to render both masculine and feminine Hebrew forms by one common form, since gender is not indicated in Greek verbs. More difficult is the relationship between stem and inflection in Hebrew and tense, voice, and mood in Greek. 6 Conybeare and Stock, Grammar, 25. 7 For example, with indirect object could be expressed in Greek simply by the dative case, and with a noun could be rendered by a Greek noun in the dative (locative) case. 8 A dual number for verbs does not exist in either Hebrew or Greek.

75

The combination of Greek tense and mood corresponds fairly closely in translation to Hebrew inflection. The debate over the exact meanings of the classical Hebrew perfect and imperfect have raged for years, and no universal consensus has yet been reached. However, the Greek translators often rendered the simple Hebrew perfect with a past tense (aorist, perfect, imperfect,9 or pluperfect) and the simple Hebrew imperfect with a present or future tense (present, future, or future perfect), and least when rendering the verb with the Greek indicative mood. The addition of the waw consecutive to the beginning of the Hebrew verb generally resulted in a reversal of the characterization just given in order to make the Greek rendering match the Hebrew intention. This reversal was not required by the Greek language but rather reflects the translators’ understanding of the Hebrew that lay before them. When rendering conditions, possibilities, and wishes, the Greek translators had available to them moods other than the indicative, namely, the subjunctive and optative moods, and they used these moods frequently. However, when the Greek translators used subjunctive or optative moods, the correlation between Greek tense and Hebrew inflection often disappeared: since the time element of the tense no longer mattered in these moods, the translator would generally choose the tense on the basis of its Aktionsart, whether punctiliar (aorist), durative (present), or perfected (perfect).10 One Greek mood which does have an almost exact parallel in Hebrew is the imperative, and Hebrew imperatives were regularly rendered by Greek imperatives, though the Greek tense could vary. Hebrew stems may be divided into three groups: simple stems (qal, niphal), intensive stems (piel, pual, hithpael), and causative stems (hiphil, hophal).11 These stems may also be classified according to type of action (similar to voice in Greek): active stems (qal, piel, hiphil), passive stems (niphal, pual, hophal), and reflexive stems (niphal, hithpael).12 There is no Greek equivalent to the intensive and causative stems, though the 9 A terminological difficulty exists in comparing Greek or Latin to Hebrew, since the Hebrew

imperfect, often reflecting present or future time, functions quite differently from the Greek and Latin imperfects, which reflect past time. Though other terms are available for the Hebrew inflections—notably the suffix and prefix conjugations, referring to the perfect and the imperfect, respectively—perfect and imperfect are still the most commonly used. Furthermore, the term “conjugation” itself presents terminological difficulties, since Greek and Latin conjugations are merely morphological categories, whereas the two Hebrew conjugations reflect semantic differences. It may be best to refer to the inflections simply as the qtl and yqtl inflections, as is sometimes done. See Waltke and O’Connor, Introduction, 45558. 10 A. T. Robertson, A Grammar of the Greek New Testament in the Light of Historical Research, 4th ed. (Nashville: Broadman, 1934), 824. 11 Hebrew also has various other stems (poel, pilpal, etc.) which are related to these primary

stems. For a discussion of the relation of the Hebrew stems to the proto-Semitic language, see Hans Bauer, Pontus Leander, and Paul Kahle, Historische Grammatik des hebräischen Sprache des Alten Testaments, vol. 1 (Tübingen: Max Niemeyer, 1922; reprint, Hildesheim: Georg Olms, 1962), 279-88. The names of the three groups is traditional and not always particularly helpful, especially in the case of the name “intensive.” For a discussion of the different uses of the various stems, see ibid., 289-94. 12 The niphal was originally strictly a reflexive stem, but it came to be used also as the passive of the qal, replacing an earlier qal passive, of which only remnants remain in the Hebrew Bible (esp. the qal

76

LXX translators did occasionally render Greek verbs as though they were causative.13 A greater correlation exists between the type of action reflected in the Hebrew stem and voice of the verb. Hebrew active and passive stems are generally rendered by the Greek active and passive voices.14 The correlation between Hebrew reflexive stems and the Greek middle voice is much smaller, if not nonexistent. The reasons for this lack of correlation are many. First, the niphal is often used with a purely passive meaning. Second, the meaning of the Greek verb used to render the Hebrew in the reflexive stem may not correspond in the middle voice to the meaning required by the Hebrew. Next, many Greek verbs, whether deponent or not, have what appears to be an active meaning expressed by the middle voice.15 When these verbs are used in the middle voice to translate Hebrew verbs, it is unlikely that a non-active Hebrew stem can be assumed. Finally, in later Greek a blending of the middle and passive voices occurred, so that middle verbs were sometimes used with passive meaning, and vice versa.16 The last verbal forms to be considered, infinitives and participles, exist in both languages, and a precise rendering in Greek of these two forms was possible. The fairly common use of the participle as equivalent to a full verb in Hebrew was rare or perhaps unknown in classical Greek. However, the use of the independent Greek participle for the verb was a permissible, if still uncommon, construction in koine, as demonstrated by the papyri and the New Testament.17 Undoubtedly, many of the instances of independent Greek participles used as verbs in LXX were based on a similar construction in the Hebrew Vorlage. Nevertheless, some of the translators seem to have preferred rendering independent Hebrew participles used as verbs by Greek verbs rather than participles, particularly when such a rendering could be supported by the consonantal text. Thus, the rendering of a Hebrew participle by a Greek verb in certain instances cannot be considered indicative of a differing Hebrew Vorlage, though, of course, it does not rule it out. Greek has only one infinitive whereby to render the Hebrew infinitive absolute and infinitive construct. Moreover, the common addition of the preposition to the Hebrew infinitive construct is not rendered by a Greek preposition, which would be unidiomatic. It is possible that some translators may have rendered the by the genitive of the article with passive participle); cf. C. L. Seow, A Grammar for Biblical Hebrew (Nashville: Abingdon Press, 1987), 250; for a more complete discussion see Bauer, Leander, and Kahle, Grammatik, 285-88. 13 Conybeare and Stock, Grammar, 76-77. 14 Except when the Greek verb used did not have the required form, so substituted another. See

Smyth, Grammar, 218-19. 15 A. T. Robertson, Grammar, 811-13. 16 Conybeare and Stock, Grammar, 75-76; A. T. Robertson, Grammar, 333-34. For other

examples of the use of one voice with the meaning of another in Greek, see Smyth, Grammar, 219-24. 17 A. T. Robertson, Grammar, 944-46. Cf. also Conybeare and Stock, Grammar, 74.

77

an infinitive of purpose. This rendering, however, is not consistent.18 This brief discussion of the similarities and differences between Hebrew and Greek is not complete, and other matters concerning the rendering of the Hebrew text by the translators of LXX are discussed in longer works and articles.19 Partial Translation Technique Based on an analysis of the Greek language in comparison with Hebrew, the following are potentially significant variants in LXX that need further scrutiny. Add-Oms Since an analysis of the first three categories of variants will not include add-oms, they must first be identified. Add-oms, as mentioned above, are variants in which the reading of the base text (MT) is either longer or shorter by at least one semantic unit20 than the corresponding section in the translation (in this case, LXX). In general, the list of addoms will be almost the same as the list of quantitative variants (category four), but in each language exceptions may be made, and certain quantitative variants may not be classified as add-oms; in other words, these exceptional cases would be analyzed in the first three categories of variants. It is sometimes difficult to determine whether a difference between the Hebrew and the Greek should be classified as an add-om or not. In particular, a decision must be made concerning Hebrew particles ( ) and prepositions (especially and ) that are not rendered in LXX in a particular passage. Such particles and prepositions will be included in the add-om list if the Hebrew word in question is usually rendered by some equivalent Greek word. The same consideration will apply as well to the other versions analyzed. In LXX, the omission of , , or will not be considered add-oms, so they will be taken into account when analyzing both representation of Hebrew lexemes by Greek lexemes and quantitative representation. A full list of the add-oms of LXX is given in Appendix 2; it may be compared with the list of quantitative variants given later in this section. 18 Conybeare and Stock, Grammar, 58-59. Cf. the tables in Ilmari Soisalon-Soininen, Die

Infinitive in der Septuaginta, Annales academiæ scientiarum fennicæ, no. 132, 1 (Helsinki: Suomalainen Tiedeakatemia, 1965), 180-85. 19 See especially Conybeare and Stock, Grammar, 25-97; Thackeray, Grammar; Jellicoe, LXX and Modern Study, 314-37; and works listed in the Bibliography by Anneli Aejmelaeus, Francis I. Andersen, Albert Debrunner, Kenneth James Dover, H. S. Gehman, Martin Johannessohn, Max Leopold Margolis, E. Nestle, Alfred Rahlfs, Ilmari Soisalon-Soininen, Raija Sollamo, and Emanuel Tov. 20 A semantic unit is a phrase, word, or part of a word that represents a single concept. Of course,

the most common semantic units are words, but the Hebrew pronominal suffixes and inseparable prepositions are also semantic units, as is the Greek compound .

78 Consistency The first criterion that Tov lays out for evaluating the translation technique of a version is consistency. Here, those variants in LXX that involve consistent rendering in comparison with MT are listed. The columns entitled “LXX” and “MT” describe the nature of difference between the LXX reading and MT. For example, if LXX has a full verb where MT has a participle, the respective columns will read “verb” and “participle.” If the LXX rendering reflects a lexical choice other than the main rendering (if one exists), the LXX column will read “lexeme,” and the MT column will be empty. Since MT does have a tradition of interpretation represented by the vowel points, the complete Masoretic understanding of the consonants will be the initial basis for comparison. However, when variants come to be eliminated, those that reflect a possible alternate understanding of the consonants of MT will also be eliminated. Three common forms that would often be ambiguous without vowel points are (1) words that could be considered participles, infinitives, or various forms of the full verb; (2) nouns with inseparable prepositions, which may or may not include an article that has been assimilated; and (3) verbs with a waw prefix that could be either waw consecutive or waw conjunctive. Also, it should be noted that the kethib form of MT is used as the basis for evaluation, though qere forms will be discussed later. Each variant is numbered individually for reference in later discussions. When more than one variation is associated with a particular word, each one has its own reference number. Ref Variation Septuagint Masoretic Text 1 3:1 periphrasis participle 2 periphrasis participle 3 active niphal 4 3:2 verb participle 5 plural singular 6 lexeme 7 infinitive adjective 8 imperfect imperfect 9 3:3 lexeme 10 infinitive verb 11 verb participle 12 3:5 lexeme 13 3:6 verb infinitive 14 3:7 infinitive verb 15 lexeme 16 infinitive verb

79

17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

3:8 3:9

verb participle lexeme verb participle lexeme 3:10 verb participle 3:11 verb participle article not definite plural singular plural singular 3:12 lexeme lexeme verb infinitive verb infinitive 3:13 perfect perfect w/c verb participle plural singular lexeme word division 3:14 active niphal plural singular 3:15 lexeme present imperfect w/c adverb noun 3:16 lexeme 3:17 participle verb passive piel plural singular participle verb passive piel plural singular 3:18 lexeme 3:19 active hiphil 3:20 plural singular adjective participle Now that the potential variants have been listed, it is time to examine the level of consistency reflected by various aspects of the LXX rendering. First, lexical consistency (tables 1-3) will be measured by counting the different Hebrew words used more than once

80

and comparing that number with the number of Greek words used more than once.21 Next, the number of Hebrew words (occurring more than once) rendered by the primary Greek rendering will be calculated. This statistic will measure the translator’s loyalty to a particular vocabulary item when an equivalent lexical choice is available. Next, the lexical deviation factor (described in the next paragraph) will be computed. Finally, a comparison of Hebrew and Greek roots, rather than words, will be made, in an effort to see if the translators have a tendency toward etymologizing.22 Table 4 is an analysis of how the translators rendered words in a particular Hebrew word class, namely, verbs and verbals, nouns and adjectives, and pronouns. Tables 5-16 analyze the translators’ tendencies in rendering various grammatical categories and syntactic relationships (i.e., for verbs: inflection, stem, person, and number; for nouns and adjectives: gender, number, use in the sentence, articles, and definiteness; for pronouns: gender, number, and use in the sentence).23 The deviation factor is a measure of the deviation from absolute consistency, whether lexical or grammatical, by a translator. Absolute lexical consistency, for example, is defined as the use of one and only one Greek word for each distinct Hebrew word. The phenomenon of using the same Greek word for more than Hebrew word is not considered in the calculations. The algorithm for computing the deviation factor is given in detail in Appendix 4, but its rationale is given here, described in terms of lexical consistency. If a single Hebrew word is always rendered by the same Greek word, the deviation factor for that Hebrew word is zero. If more than one Greek word is used to render a single Hebrew word, the deviation factor is a positive number. Given ten occurrences of a given Hebrew word, it is considered a greater deviation from absolute consistency for three Greek words to be employed in rendering that word than for two to be so employed (all other things being equal). Furthermore, if two Greek words are used to render a single Hebrew word, the deviation factor will be greater if each Greek word appears five times than if one appears nine times and the other only once. It is considered a greater deviation when one case out of five differs from the standard than when one out of ten does. The deviation factor is a modification of a chi-square test, adapted to account for the number of discrete renderings in the target language as well as the total number of deviations from the most 21 The number of Greek words will always be greater than or equal to the number of Hebrew

words, for if a Greek word is used to render more than one Hebrew word, it is counted more than once. 22 Tov identifies the technique of etymologizing, the rendering of all words based on a single Hebrew root (real or imagined) by words based on a single Greek root, as an aspect of consistency calling for special attention; Tov, Text-Critical Use, 57. In table 1, the last column identifies different words that share the same root by assigning a common reference number, given in italics. Roots represented by only one word in the chapter are not explicitly indicated, though they are of course figured in the data. 23 A shortened form of the tables, containing the results of the various computations, is given

below. For an expanded form of the tables, containing the data on which the computed figures are based, see Appendix 2 (the category number here corresponds to the table number in Appendix 2).

81

commonly used rendering. Larger deviation factors indicate more variation from absolute consistency, but it must be stressed again that the numbers themselves are not helpful until they are compared with either the corresponding deviation factor in another secondary witness or other deviation factors within the same witness. Of course, the larger the statistical base that is being analyzed, the more meaningful will be the results. Table 1.—Verbs, Nouns, and Adjectives: Lexemes Heb words (>1x): 31 Gk words: 42 1.35 Gk words/Heb word deviation factor: 0.38 total Heb (>1x): 132 primary Gk: 117 88.6% of Heb words by primary rendering Heb roots (>1x): 29 Gk roots: 42 1.45 Gk roots/Heb root Table 2.—Adverbs, Prepositions, and Particles: Lexemes Heb advs (>1x): 13 Gk advs: 26 2.00 Gk advs/Heb adv total Heb (>1x): 85 primary Gk: 67 78.8% of Heb advs by primary rendering deviation factor: 2.78 Table 3.—Conjunctions: Lexemes Heb conjs (>1x): 2 Gk conjs: 2 1.00 Gk conjs/Heb conj total Heb (>1x): 66 primary Gk: 66 100% of Heb conjs by primary rendering deviation factor: 0.00 Table 4.—Word Classes percentage of verbs/verbals represented by verbs/verbals: 94/95 = 98.9% percentage of verbs represented by verbs: 71/76 = 93.4% percentage of participles represented by participles: 4/12 = 33.3% percentage of infinitives represented by infinitives: 4/7 = 57.1% percentage of nouns/adjs represented by nouns/adjs: 52/55 = 94.5% percentage of pronouns represented by pronouns: 37/38 = 97.4% Table 5.—Verbs: Hebrew Inflection, Greek Tense and Mood deviation factor (discrete tense/mood combinations): 6.68 deviation factor (grouped): 0.97 deviation factor (grouped, without verbals): 0.85 Table 6.—Verbs: Hebrew Stem, Greek Voice deviation factor: 3.57 Table 7.—Verbs: Person deviation factor: 0.00 Table 8.—Verbs: Number deviation factor: 0.12 Table 9.—Nouns and Adjectives: Gender deviation factor: 21.54

82

Table 10.—Nouns and Adjectives: Number deviation factor: 2.29 Table 11.—Nouns and Adjectives: Use vs. Case deviation factor (discrete use/case combinations): 6.45 deviation factor (grouped): 0.37 Table 12.—Nouns and Adjectives: Articles deviation factor: 11.37 deviation factor (without implied articles): 12.23 Table 13.—Nouns and Adjectives: Hebrew Definiteness, Greek Articles deviation factor: 8.04 Table 14.—Pronouns: Gender deviation factor: 1.44 Table 15.—Pronouns: Number deviation factor: 0.06 Table 16.—Pronouns: Use vs. Case deviation factor (discrete use/case combinations): 3.76 deviation factor (grouped): 0.06 Representation of Hebrew Lexemes by Greek Lexemes (Segmentation) Translators often had a tendency to represent compound words in the source language with compound words in the target language. Thus, if a Hebrew word were composed of a preposition and a noun, for example, the Greek translator might have preferred to render the expression with a Greek preposition and noun, even though an adverb would have carried the same meaning. Here are listed all the Hebrew compounds which the Greek translators did not render all of the constituent parts. Only the consonantal text is considered, so articles indicated by pointing alone that are not represented are not listed. Ref Hebrew Compound Greek Rendering 51 3:1 52 3:2 53 3:5 54 3:6 55 3:7 56 3:8 57 58 3:9 59 3:13 60

83

61 3:14 62 3:15 63 3:17 64 3:18 65 3:20 66 3:21 Heb compounds: 43 Gk equivalents: 27

62.8% of Heb compounds rendered exactly

Word Order As noted above, since Greek is an inflected language, its writers and speakers had much greater freedom to vary word order and still maintain the thought of the sentence. Whether the translators of LXX in fact chose to take advantage of this facet of their language, or whether instead they attempted to follow closely the word order of their exemplar, is another aspect of the literalness of the translation. The list below contains the deviations from the Hebrew word order found in LXX. As already stated, Greek postpositive conjunctions that appear as near as the language permits to the equivalent Hebrew conjunction will not be considered variations from the Hebrew word order. The number of variations is calculated by ascertaining the minimum number of shifts of position a Hebrew word (including a compound word) would have to make in order to reflect the word order found in LXX. Ref Number of Variations Greek Variant 67 3:16 3 68 3:19 1 Heb semantic units: 373 variations: 4 Heb word order followed 98.9% of the time Quantitative Representation The tendency of literal translators was to render every element of their Vorlage, without adding or subtracting anything unless it was required by the target language. The following list analyzes the literalness of LXX in terms of quantitative representation, in order to determine the translator’s commitment to render all the elements of the Vorlage, and only those elements. The number of Hebrew semantic units represented by the variation is given, prefixed with a plus if the excess is in Greek and with a minus if the excess is in Hebrew. In the last column, each element of MT not found in LXX is listed in Hebrew, and each element of LXX not found in MT is listed in Greek. As noted above, add-oms will be included in these calculations, but quantitative differences arising from the representation of compound words have already been discussed and will not be included here. Furthermore, it should be noted that in verse 1 is rendered by , and the he locale in verse 19 is rendered by ; these renderings are not considered differences in

84

quantitative representation, since Greek has no exact equivalents for the Hebrew expressions. , the sign of the definite direct object, has no real equivalent in Greek and so cannot be rendered idiomatically (Aquila’s notwithstanding). Finally, articles are considered only when they appear or are implied (e.g., in 3:1 requires that an article be assumed in ) in the consonantal text (or when they would appear in the presumed Hebrew underlying an addition in LXX). In order to avoid begging the question as to which witness (if either) contains an addition to or omission from the original, the excess text is given in whichever language it appears. Ref Number of Variations Variant 69 3:1 +1 1° 70 +2 71 3:2 +1 4° 72 3:3 -1 73 -1 74 3:4 -1 75 +1 76 3:5 +1 77 +1 4° 78 3:6 +1 2° 79 -1 80 +1 2° 81 -3 82 +1 83 -2 84 +1 85 +1 6° 86 3:7 -1 2° 87 3:9 -3 88 +1 2° 89 +1 90 3:10 +1 91 -2 92 3:11 +1 93 -1 94 3:13 -2 95 +2 96 3:15 +4 97 -1

85

98 99 100 101 102 103 104 105

3:16 3:17 3:18 3:21

-2 -1 -2 +3 -2 +1 -5 +31

Heb semantic units: 438 Gk deviations: 88

1° 2°

79.9% agreement

Elimination of Variants The LXX translation of 1 Samuel is often characterized as literal or fairly literal.24 A glance at the preceding tables, however, indicates that this characterization, though perhaps true in general, does not describe certain aspects of the translation technique with precision. Whereas conjunctions are rendered with absolute consistency (i.e., exactly one Greek term per Hebrew term), verbs, nouns, and adjectives, and especially adverbs, prepositions, and particles, are not. The translators were reasonably consistent in rendering a particular class of Hebrew words (verbs/verbals, nouns/adjectives, pronouns) with the equivalent Greek class. However, the percentage of Hebrew verbals rendered by the equivalent class of verbals in Greek is low. The level of grammatical consistency is 24 Thenius, Bücher Samuels, XVII-XXII; S. R. Driver, Notes on the Books of Samuel, lix-lxii; Thackeray, Grammar, 13; Swete, Introduction, 323; Fernández Marcos, Introducción, 25. Cf. also Soisalon-Soininen, Infinitive, 171-72; Raija Sollamo, Renderings of Hebrew Semi-Prepositions in the Septuagint, Annales academiæ scientiarum fennicæ, Dissertationes humanarum litterarum, no. 19 (Helsinki: Suomalainen Tiedeakatemia, 1979), 286. De Boer disputes this claim: “We have already observed that the translators did not have literal rendering as motive for translation, but that the thread of the story was the cause for omissions, additions and differences,” de Boer, I Samuel i-xvi, 51. However, he seems to approach his analysis with a foregone conclusion. Note the following incredible conclusion: “Even though certain passages remain without a clear explanation of why G differs from M, and although some of the proposed explanations are nothing more than probabilities, the independence of the translated story, the agreement with Tg and S and elucidation as the main tendency leave us in no doubt that we in G have to do with the same Hebrew text as the one offered by M” (italics mine).

86

more complicated. The translators are absolutely consistent (deviation factor of 0.00) in rendering the person of verbs. They are very consistent (deviation factor less than 1.00) in rendering the number of verbs, the number of pronouns, the inflection of verbs, the use of nouns and adjectives, and the use of pronouns (when the Greek data is grouped appropriately). They are fairly consistent (deviation factor between 1.00 and 3.00) in rendering the number of nouns and adjectives and the gender of pronouns. They are less consistent (deviation factor between 3.00 and 6.00) in rendering verb stems. They are fairly inconsistent (deviation factor between 6.00 and 12.00) in rendering articles found with nouns and adjectives and the definiteness of nouns and adjectives. Finally, they are very inconsistent (deviation factor greater than 12.00), as expected, in rendering the gender of nouns and adjectives. Apart from the category of consistency, the following general results may be mentioned. The translators did not feel compelled to render each component of Hebrew compounds with a Greek equivalent, choosing in 40% of the cases to translate the compound as a whole. By contrast, Hebrew word order is followed very closely, with LXX deviating only 1.1% of the time, and that in only two units. Finally, an initial look at quantitative representation suggests that the Greek translators frequently varied from the Hebrew text, either adding (fifty-seven times) or omitting (thirty-one times) elements. It must be emphasized that these results are preliminary, based on a rigid application of the principles for determining translation technique discussed in Chapter 1, and assuming that MT is the same as the Vorlage of LXX. As individual variants and groups of variants are weighed, a more nuanced picture of the translation technique will emerge, and many of the above results will have to be modified. Variants Related to Consistency The first aspect of consistency to be examined is lexical consistency, of which twelve variants from MT have been identified in LXX. The identification of a lexical variant that might indicate a varying Hebrew Vorlage is based on two factors: similarity in meaning as reflected in the lexicons and Greek equivalents for the same word elsewhere in the chapter. The first lexical variant is variant 6, where LXX reads for MT . The lexicons list two roots , the first meaning “to be dim, faint, blind,” and the second meaning “to rebuke.” The second root is found only in 3:13, where it is a hapax; the present instance comes from the first root. Of the fifteen probable renderings in LXX of either the verb or the adjective , the only time LXX uses a root related to is here. (There is no single Greek root used to render I that dominates, but the words related to are used six times). and related words are often used to render words related to the Hebrew : twenty-eight of thirty-five total

87

uses of the Greek word, twenty-five of thirty uses, one of one, one of one. The close association between and suggests that some form of appeared in the Vorlage of LXX. On the other hand, the use of (rendering ) of blinded eyes in Gen 48:10 suggests that the translators might have chosen as appropriate in the present case (cf. Isa 33:15; 59:1; Zech 7:11). However, there is enough graphic similarity between and to allow the possibility of textual confusion in the transmission of the Hebrew text, or maybe at the point of translation. The reading of P supports the possibility that the confusion occurred during the transmission of the Hebrew text. Thus, it is likely that the Vorlage of LXX at this point was different from MT, so the variant is significant. In variant 9, (“to be made ready”) renders (“had gone out”). Though the context might stretch the semantic field of enough to make it roughly equivalent to , the former renders the latter only here out of seven occurrences in the OT. renders four other words, three of which mean “to restore” and are used in the context of restoring the temple. Ex 30:8 uses to render the hiphil of , referring to Aaron setting up the lamps in the evening. occurs thirteen times in the qal and ten times in the piel that are rendered in LXX. Twentyone of these occurrences are rendered by , and one by . It is likely that scribes confused an original or with 25 , and since no Hebrew verb with a like meaning appears to be similar enough orthographically to , 26 variant 9 should probably be considered an inner-Greek corruption and not significant. in variants 12 and 18 is the normal translation for in the passage (cf. 3:5, 6) and elsewhere in LXX, including the books of Kingdoms. In fact, apart from one other occurrence in several mss (3 Kgdms 6:12 [Rahlfs’s versification] mss AMN etc.), never translates , while rendering some forty times. It seems that either the translator or the Vorlage changed the verbs in these two instances to match the other occurrences in the immediate context. Since no tendency of the translators to harmonize the LXX of this chapter has been demonstrated yet, and since is an unusual translation of in the book, it is quite possible that the Vorlage itself contained rather than in these verses, so these variants are significant. Variant 15 is a variation between the Hebrew (usually rendered by ) 25 Klostermann suggests that the LXX reading is a corruption of

(Erich Klostermann, Die Bücher Samuelis und der Könige, Kurzgefaßter Kommentar zu den heiligen Schriften Alten und Neuen Testamentes sowie zu den Apokryphen, ed. Hermann Strack and Otto Böckler, vol. 3 [Nördlingen: C. H. Beck, 1887], 11). Cf. the readings of ´ ( ), ´ ( ), ´ ( ). 26 Wellhausen suggests , which he says is implied by Josephus, Ant. 3.8.3 (Bücher Samuelis, 52), but this section is not a reference to Samuel but to the Mosaic regulations concerning the lamps in the tabernacle. Cf. de Boer, I Samuel i-xvi, 62.

88

and the Greek . Though the words are of course equivalent in the sense that they refer to the God of the Israelites, the translators of 1 Samuel are fairly consistent in their renderings of both and . 27 All of the other fifteen occurrences of in the chapter are rendered by . Thus, this variant should be considered significant. Variants 20 and 37 both use a form of the verb to render the Hebrew verb , even though forms of are the more common rendering in the chapter (six of eight occurrences of are rendered by ). However, a check of Hatch and Redpath reveals that both Greek verbs are used frequently to render , and they render few other words (in the books of Kingdoms, renders forty-nine times out of fifty occurrences of ; the numbers for are thirteen of fifteen). In fact, as these figures show, is the more common of the two main renderings outside this chapter. Thus, these variants are not significant. Variants 26, 27, and 40 each reflect Greek prepositions that are different from what one might have expected based on the main renderings of the corresponding Hebrew prepositions. Of the thirteen occurrences of in the chapter, only in verse 12 is it rendered by or (one time each; the usual renderings are [six times] or nothing [five times]). Unlike the other renderings of , carries a somewhat different meaning, and one suspects that it might reflect a Hebrew rather than . 28 The semantic field of , on the other hand, overlaps that of to a large extent. However, a perusal of several passages in 1 Kingdoms (Hatch and Redpath do not give the equivalents for each occurrence) indicates that rarely renders , though it is frequently used for - , - , and - (he locale). The readings of the other versions might suggest a reading in their Vorlagen, but this preposition does not lie behind the LXX reading. It is possible that the preposition or lay before the LXX translators in this passage. could have been lost through simple haplography ( became ) and inserted by a scribe who sensed that a preposition was needed. On the other hand, the inseparable preposition might have been replaced with its near equivalent , perhaps as a result of similarity in pronunciation. Either of these two possibilities could account for the origin of as well. in verse 16 is the only instance in the chapter (of nine possibilities) where an apparent equivalent appears for . However, the whole Greek clause is different from the Hebrew: for . It is probable that rather than lies behind the in LXX. Therefore, variants 26, 27, and 40 are all significant. Variant 33 in verse 13 reflects an ancient scribal correction, one of the tiqqune 27

occurrences of

in LXX reflects in MT about 12 out of 100 times in 1 Samuel. The ratio of in LXX corresponding to in MT is even smaller.

28 So Thenius, Bücher Samuels, 17; McCarter, I Samuel, 96.

89

sopherim. In order to avoid reading that Eli’s sons cursed God ( ), lest the reader join in the blasphemy by juxtaposing these two words, scribes altered the text to read , “they cursed themselves.” LXX preserves the original reading, since obviously reflects and not . 29 The final LXX variant that might reflect a different Hebrew lexeme is number 47, which reads for . Though ( ) ( ) might be considered more literal, the phrase that appears here carries the same meaning as the Hebrew of MT. In fact, the translators of 1 Samuel often did use the expression for (especially in kaige sections), but they did not do so consistently, often substituting the more idiomatic (e.g., 1 Kgdms 11:10; 12:17; 14:36). 30 Therefore, this variant cannot be considered significant. Variant 34, (end of 3:13 LXX) for (beginning of 3:14 MT) seems to reflect a difference in word division and a slight change in spelling: instead of . 31 This variant must have arisen in Hebrew rather than Greek, so it is significant. All the other variants that reflect on the consistency of the LXX translators deal with grammatical differences. Variants 1, 2, 4, 11, 17, 19, 21, 22, and 31 all render Hebrew participles by Greek verbs (or periphrastic constructions in variants 1 and 2). In fact, of the eleven participles rendered by Greek verbs or verbals, only two (3:11, 13) are rendered by participles. When it is remembered that the LXX translators worked from a purely consonantal text, it becomes clear that all of MT’s participles that could be understood, with different pointing, as full verbs (i.e., the qal active participles that function as the main verbs in their respective clauses in MT) were in fact rendered by verbs. The piel participle in 3:13 is rendered with a Greek participle. The piel and niphal participles in 3:1 are rendered as periphrastic constructions, consisting of a form of and a participle. The only qal participle actually rendered by a participle is the one in 3:11, but the construction of the clause (the participle is part of a construct chain) mandates that the word be understood as a participle. Although the translators may have had a different reading tradition than that of the Masoretes concerning many of the participles, the consonantal text lying behind their renderings seems identical to that found in MT, so none of these variants is significant. In variants 41, 42, 44, and 45 in 3:17, the reverse situation occurs: verbs in MT are 29 So BHK and most commentators. R. Althann, “Northwest Semitic Notes on Some Texts in

1 Samuel,” Journal of Northwest Semitic Languages 12 (1984): 28-29, proposes on the basis of Ugaritic evidence that “lhm can be understood as a by-form of l m, parsed as the stative participle of the root ym, ‘to inspire awe,’ preceded by the emphatic lamedh.” However, his analysis is not convincing. 30 See the table in Sollamo, Semiprepositions, 138, and, for more discussion, ibid., 123-55. 31 Wellhausen, Bücher Samuelis, 53. S. R. Driver, Notes on the Books of Samuel, 44, gives

several other examples of the same phenomenon.

90

read as participles in LXX. Furthermore, the participles are passive, whereas the verbs in MT are both piel. Since no subject is explicitly stated (the nearest reference to God speaking is three verses earlier), the LXX translators, if they had the same consonantal text as MT, may have considered the forms pual in both instances, in spite of the infrequency of in this stem. The use of the Greek article as the equivalent of in both cases is an idiomatic Greek rendering of the construction, a construction which has no exact parallel in Hebrew. Thus, some degree of flexibility has already been shown in each phrase. Though it is conceivable that the Vorlage of the translators had participles (with the prefixed ), it seems more likely that the translators in these two cases rendered idiomatic Hebrew expressions with idiomatic Greek ones. Even if, as appears likely, the translators read the verbs as puals rather than piels, there is no indication of any variation from the consonantal MT, so the variants are not significant. On one occasion, variant 50, the translators used an adjective, , to render a Hebrew participle. Though this is the only time in the chapter in which a Hebrew participle is rendered by something other than a verb, the participle functions in the Hebrew clause as a subjective complement (predicate adjective). The niphal participle does function elsewhere as a predicate adjective.32 The Greek rendering is surely the meaning intended by the use of the niphal participle, and it may be that the translators felt that the adjective carried the meaning more faithfully than would a passive participle such as . Thus, the variant should not be seen as significant. Three times in chapter 3 the translators use infinitives to render what are full verbs in MT, namely, variants 10, 14, and 16. In each case, the Hebrew verb follows ; two of the verbs are imperfect (3:3, 7), and one is perfect (3:7).33 The Greek equivalent for is (see above, pp. 39-40), and though LXX has only two occurrences of this preposition, all three Greek infinitives are governed by . Since the use of the infinitive rather than some form of the full verb is required when means “before,”34 these variants cannot be taken to be significant. Variants 13, 28, and 29 have Greek verbs for Hebrew infinitives. In variant 13, LXX reads for . Both the infinitive and the imperfect (usually with waw consecutive) can follow in Hebrew when the meaning is “to do again,” though the infinitive is more common. Idiomatic Greek does not 32 Waltke and O’Connor, Introduction, 619-20. 33 Kautzsch, ed., Gesenius’ Hebrew Grammar, § 107c, says that the perfect should be emended to

an imperfect, since the imperfect regularly follows and since the perfect is coordinated with the imperfect ; so also BHK, most of the commentaries. Of course, the imperfect requires no change in the consonantal text. 34 Smyth, Grammar, 549, 553-55. Soisalon-Soininen, Infinitive, 131, notes that the expression ( ) + full verb is usually rendered by either + infinitive or + infinitive throughout LXX.

91

have a similar construction, and the reading of LXX here is definitely unidiomatic. This consideration alone implies that the translators are following a Hebrew text rather than changing the Hebrew construction, especially since it is the less common Hebrew construction that is reflected in LXX. Thus, variant 13 is significant. Variants 28 and 29 also have Greek verbs for Hebrew infinitives. The two infinitives form an expression that, though idiomatic and fairly common in Hebrew, has no exact parallel in Greek.35 The translators managed to create a reading in that represented both infinitives idiomatically, deriving the person of these verbs from . In view of the nature of the Hebrew construction, the LXX reading seems to effectively capture the flavor of the Hebrew, despite avoiding the normal grammatical parallels, so these two variants are not significant. Variant 7 reads the infinitive for the adjective . Greek idiom certainly requires the use of an infinitive after , but Hebrew idiom would seem to require it, too. It is probable that the letters were meant to stand for the infinitive . 36 However, since Greek idiom requires an infinitive, and since such a change in Hebrew would not change the consonantal text, the variant is not significant. In the case of variant 39, a Greek adverb, , is used to render the Hebrew noun and article . Although none of the other fifty-four Hebrew nouns in the chapter is rendered by an adverb, here functions as an adverbial of time, and as such, is an appropriate translation. In fact, is the regular equivalent for in LXX, so this variant is not significant. The next three variants deal with the rendering of the inflection of Hebrew verbs in a manner contrary to the most common representation. First, variant 8 uses the Greek imperfect (a past tense) to render the Hebrew imperfect (usually rendered by the present or the future tense). The context clearly shows that an event in the past in being described, so one would usually expect a past tense. However, the Hebrew imperfect does not correspond completely with the Greek present or future tenses. In the case at hand, the 35 Though is classified as an infinitive construct by BDB and Holladay, it seems more likely that it should be seen with Kautzsch and Klostermann as an infinitive absolute, as is. Cf. Kautzsch, ed., Gesenius’ Hebrew Grammar, § 113h; Klostermann, Bücher Samuelis, loc. cit; Francis Brown, S. R. Driver, and Charles A. Briggs, A Hebrew and English Lexicon of the Old Testament with an Appendix Containing the Biblical Aramaic (Oxford: Clarendon Press, 1906; reprint, 1951) [hereafter BDB], s.v. “ ”; William L. Holladay, A Concise Hebrew and Aramaic Lexicon of the Old Testament (Grand Rapids: William B. Eerdmans, 1971), s.v. “ .” 36 BDB, s.v. “

”; Holladay, Lexicon, s.v. “ .” Wellhausen says that since no preposition is present, an infinitive would be ungrammatical (Bücher Samuelis, 52), but Smith disputes this contention (H. P. Smith, Books of Samuel, 27). Cf. Kautzsch, ed., Gesenius’ Hebrew Grammar, § 114m). Driver says that the infinitive is more in accord with biblical usage (S. R. Driver, Notes on the Books of Samuel, 42). Cf. Walter Baumgartner and J. J. Stamm, Hebräisches und Aramäisches Lexikon zum Alten Testament, 3d ed., 5 vols. (Leiden: E. J. Brill, 1967-) [hereafter KB3 ], s.v. “ .”

92

imperfect is used to indicate the continuous nature of Eli’s blindness,37 so variant 8 is not significant. Variant 30 has a Greek perfect rendering a Hebrew perfect with a waw consecutive. The Hebrew implies that the message to Eli concerning judgment on his house has not yet been given to him, whereas the Greek says that it has (cf. 1 Sam 2:2736).38 This difference of perspective is reflected in other secondary witnesses as well, so the variant is significant. In variant 38, the present is used to render the imperfect with waw consecutive ( ). Normally, the Greek translators use a past tense to render the Hebrew imperfect with waw consecutive, but the present here could be explained as a historical present. Although the imperfect with waw consecutive is rendered thirty-four times by the aorist and only once by the present in chapter 3, the presence of 151 instances of the historical present in 1 Samuel39 suggests that the ratio in the present chapter is disproportionate, and the variant is probably not significant. Variants 3, 35, and 48 represent deviations from the usual rendering of Hebrew stems by means of Greek voice. In the first two instances, a niphal is represented by a Greek verb in the active voice. The use of the active voice in variant 35 is easily explained by the fact that the verb regularly appears in the niphal when it has an active meaning.40 The active voice would be the one that most accurately represented the meaning of the Hebrew, and this is the one the Greek translators chose. Variant 3 is not quite so simple. As noted above, the participle is part of a periphrastic construction that renders a Hebrew participle. The meaning of the verb is “to separate, distinguish, determine,” and “to command, give 41 orders.” qal has meanings such as “break through, break out, break into, break up,” but it also occasionally means “to spread, become known.” The niphal is cited in BDB as “spread abroad,” but the definition given by Holladay is “to be ordered, 37 Kautzsch, ed., Gesenius’ Hebrew Grammar, § 107b. The Greek translators also emphasized the

continuous nature of his blindness by using the imperfect rather than the aorist. 38 Cf. Hans Wilhelm Hertzberg, Die Samuelbücher, 2d ed., Das Alte Testament Deutsch, ed. Volkmar Herntrich and Artur Weiser, vol. 10 (Göttingen: Vandenhoeck & Ruprecht, 1960), 29. Hertzberg interprets the waw as a waw conjunctive rather than a waw consecutive, but the reason for using this construction rather than the usual waw consecutive with an imperfect to indicate past narrative action is unclear, so it seems better to understand MT as a waw consecutive perfect construction; on this point cf. John Mauchline, 1 and 2 Samuel, New Century Bible, ed. Ronald E. Clements and Matthew Black (London: Oliphants, 1971), 58. The translators of P certainly understood the construction as referring to a future time. 39 Thackeray, Grammar, xx. Cf. also A. T. Robertson, Grammar, 866-69. 40 The niphal in this verb carries reflexive rather than active meaning. Cf. Waltke and O’Connor, Introduction, 391. 41 Henry George Liddell and Robert Scott, A Greek-English Lexicon, rev. Henry Stuart Jones and

Roderick McKenzie (Oxford: Clarendon Press, 1968) [hereafter LS], s.v. “

-

.”

93

orderly.”42 The exact meaning of the Hebrew word is obscure, with commentators divided between the definitions given above (contrast McCarter and Klein).43 The textual question at this point, however, is whether the LXX reading reflects a Hebrew text different from MT. BHK suggests that LXX reflects a reading , to be pointed or (though it prefers either the reading of MT or the qal passive participle to the reading of LXX),44 but BHS does not have any textual comment on the word. It is possible that the nun of the niphal is the result of dittography and that the occurrence of this verb in the niphal stem in postbiblical Hebrew developed from this passage.45 The niphal stem is not frequent enough in the chapter to state conclusively that LXX normally rendered it with a passive verb, at least when it had a passive meaning, though the other three occurrences in the chapter are rendered with passives. However, the methodology only allows the elimination of variants that are probably not significant, so this borderline case will be considered significant at this point. In the case of variant 48, a hiphil is rendered by a verb in the active voice that does not normally have a causative meaning. Greek does not have a causative voice, though it does possess words whose root meanings are causative (cf. 3:12 ; 3:13 ; 3:15 ; 3:17 ). does not usually have a causative meaning. However, it is not uncommon for words in the active voice to have causative meanings in LXX that they do not normally have in nonbiblical Greek.46 Therefore, this variant should not be considered significant. The last two variants among the verbs concern instances in which the Greek verb is plural but the Hebrew verb is singular. Though table 8 on verbal number suggests that the overwhelming majority of Hebrew singulars are rendered by Greek singulars, both variants 46 and 49 involve the word . (Since the noun in variant 43 also figures in the discussion of variant 46, it will be considered here as well.) Greek does not utilize collective nouns to the extent that Hebrew does, and this fact is demonstrated in variant 49, where the Hebrew verb is singular to agree with the singular form of , whereas the Greek verb is plural to agree with the plural idea of . An awareness of the translators’ shift from singular to plural in this verse helps explain the shift in variants 43 42 BDB, s.v. “ ” (BDB says, though, that the text of this passage is dubious); Holladay, Lexicon, s.v. “ .” Holladay follows the suggestion of G. R. Driver, who says on the basis of Assyrian evidence that two different roots exist in the OT. Cf. Godfrey Rolles Driver, “Some Hebrew Roots and Their Meanings,” Journal of Theological Studies 23 (1922): 71-73; idem, “Studies in the Vocabulary of the Old Testament: III,” Journal of Theological Studies 32 (1931): 365. 43 Smith says the word seems to have “no good meaning” (H. P. Smith, Books of Samuel, 27). 44 Cf. also Wellhausen, Bücher Samuelis, 51; McCarter, I Samuel, 95. 45 McCarter, I Samuel, 97; cf. Wilhelm Caspari, Die Samuelbücher, Kommentar zum Alten

Testament, ed. Ernst Sellin, vol. 7 (Leipzig: A. Deichertsche Verlagsbuchhandlung Dr. Werner Scholl, 1926), 53. 46 Conybeare and Stock, Grammar, 76-77.

94

and 46. The construction in verse 17 can mean either “the whole matter” or “every word” (i.e., “everything”); the translators apparently understood the phrase to mean “the whole matter,” to which the translation (“all the words”) is 47 roughly equivalent. Since the participle (in Greek) had to agree with the noun, it was also changed from singular to plural. Thus, the presence of and in these two verses provides an extenuating circumstance in each case that explains the translators’ retreat from their usual method of rendering verbal and nominal number, and the variants are not significant. Five other cases of difference in number in nouns and pronouns appear in the chapter, variants 5, 24, 25, 32, and 36, all of which have a plural in the Greek that corresponds to a singular in MT. In variant 5, the plural corresponds with the qere reading in MT, , and it also agrees with the number of the verb , so the Greek almost certainly reflects a plural Vorlage. However, it would be a mistake to assume that the translator was rendering rather than , since the latter reflects an orthography current until about the fifth century B. C. E. 48 In other words, the translation in LXX could well reflect the reading , so variant 5 is not significant.49 In the cases of variants 24, 32, and 36, table 10 indicates that Hebrew singulars are usually represented by Greek singulars, but there are enough instances of singulars being represented by plurals to question the propriety of calling such variants significant without supporting evidence. In variants 24 and 32, such evidence may be present. In both cases the LXX sentence structure is different from that of MT. In verse 11, MT reads , whereas LXX reads ; the following clause is also different in that LXX lacks anything to render the relative particle. Furthermore, it is possible that the plural ending or the pronominal suffix were omitted through haplography and graphic confusion: or became . The fact that (variant 25) is also plural and that it is the only instance in the chapter in which a singular pronoun in MT is rendered by a plural pronoun in LXX also supports the possibility that the noun in variant 24 was plural in the Vorlage of LXX. In variant 32, a difference in structure also exists: for . This difference in structure is probably related to the tiqqun sopherim in the same verse. The disturbance in the text in verse 13, like that in verse 11, increases the 47 Cf. Smyth, Grammar, 296. “Every word” would require ( )

.

48 Francis I. Andersen and David Noel Freedman, “Another Look at 4QSam b ,” Revue de Qumran

14 (1989): 23-27. For a more extensive discussion of orthography, see below, pp. 212-16. 49 Whether the LXX translators would have known of earlier orthographic practices is ultimately unanswerable, though it seems arrogant to deny the possibility that individual ancient translators would have been aware of different spelling conventions. Even if one suspects that the LXX Vorlage read , the fact that and are probably simply orthographic variations rules out the possibility that the LXX reading is significant as defined in this study.

95

likelihood that the text underlying LXX was different from that of MT. Thus, variants 24, 25, and 32 should all be considered significant at this point. The text around variant 36, on the other hand, shows no signs of textual disturbance, and since the table indicates that it is not uncommon for singulars to be rendered by plurals, this variant cannot be taken as significant. The final variant to be considered under the category of consistency is variant 23, in which stands for , a noun with an article for an indefinite Hebrew noun. A look at the two tables concerning articles and definiteness indicates that the translators were consistent about rendering Hebrew articles, both present and implied in pointing in MT, with Greek articles. However, Hebrew substantives that are definite by virtue of being in construct with a definite substantive (including determination by a pronominal suffix) are not consistently rendered by Greek words with articles. Furthermore, Hebrew nouns without articles are not rendered consistently, though indefinite nouns are usually translated by Greek nouns without articles. However, the lack of overall consistency in rendering both articles and definiteness (note the high deviation factors) casts doubt on the reliability of Greek articles as to the reading of the Vorlage. In addition, the add-om in LXX, if it accurately reflects the Hebrew Vorlage, would eliminate the possibility of a Hebrew article, since a noun cannot have both an article and a pronominal suffix.50 Therefore, variant 23 is not significant. Now that the significant variants dealing with consistency have been determined, the partial translation technique in several categories needs to be recalculated, in order to obtain a more realistic appraisal of the translators’ approach to their task. The first step in the reevaluation process is to eliminate all references to the data contained in the significant variants. For example, since variant 3, rendering a niphal stem with a verb in the active voice, is considered a significant variant, it is deemed probable at this stage of the investigation that the Vorlage of LXX did not have a niphal at this point. Not enough groundwork has yet been laid to allow one to retrovert the LXX reading into Hebrew, so at this stage the references to the niphal and the active voice are simply eliminated from the data. If it is found later that the variant is in fact not significant, the references will be added again to the data. Variant 33 is not represented in the statistical data because the Hebrew is a compound that appears only once with a single equivalent, and Hebrew words must appear twice to be counted for lexical consistency. Similarly, variant 34 is not represented in the data because no category for word division exists. The elimination from the statistical data of significant variants like these two that are not already represented will not affect the calculation of the translation technique. The elimination from the statistics of the significant variants does not end the 50 For a discussion of definiteness in Hebrew nouns, see Kautzsch, ed., Gesenius’ Hebrew

Grammar, §§ 125-27.

96

recalculation of the translation technique. Other factors besides variant Vorlage have also skewed the results so that the translators appear usually somewhat less literal, but sometimes more literal, in their approach than they really were. The other factors in question will be dealt with in the order of the tables. After all these other factors have been noted, the summaries at the bottoms of all the tables that have changed will be listed. First, in representing the Hebrew verbal stem (cf. table 6), the translators were constrained by the Greek language in some of their choices, specifically, those verbs whose active voices were not used (deponents) or were used with a different meaning.51 There are no truly deponent verbs in chapter 3,52 but there are several that correspond to qal or piel verbs whose use in middle or passive voice is required in order to preserve the correct verbal idea: (three times), , (twice), , . In addition, LS indicates that is found more commonly in the middle than in the active, and Bauer notes that in the koine of the NT and early Christian writers, the meaning “to begin” is found only in the middle; the active means “to rule.”53 It is probable, then, that the translators considered it necessary to use the middle in order to convey the proper meaning. Furthermore, as noted above, though and do not require a change in the consonantal text, it is likely that the translators looked at in both places as puals rather than piels. The second table that needs to have some of its data adjusted is table 11, comparing the use of nouns and adjectives in Hebrew with case in Greek. Hebrew grammars often describe Hebrew as though it were an Indo-European language, using terms, such as genitive, that do not correspond exactly to genitives in languages such as Greek or Latin. It seems preferable to describe the function of Hebrew substantives in terms that are appropriate to the language and then attempt to compare it with a language from a different family of languages. The initial analysis evidenced in table 11 shows quite a bit of disparity, as might have been expected, but it is possible to manipulate the data somewhat in order to arrive at a clearer picture of the translators’ understanding of both Hebrew and Greek. First, the dative that corresponds to a Hebrew direct object is in fact a dative of direct object, following a verb of serving,54 so it may be combined with the accusative when the deviation factor is figured, since the dative is required by the language. Second, since the relationship between Hebrew prepositions and objects is manifold, the data from all the Greek oblique cases should be combined, and the result is exact correspondence. Finally, Hebrew participles that function as verbs really have no counterpart in the Greek substantival system, even though they may be represented by Greek participles with 51 Cf. Smyth, Grammar, 218-24 52

in B, v. 2, is deponent, but

53 LS, s.v. “

”; BAG, s.v. “

54 Smyth, Grammar, 339.

has been read instead; see above, pp. 38-39. .”

97

substantival attributes, so they should be eliminated from consideration. The same considerations also apply to table 16, which deals with the use and case of pronouns. In addition, the row labeled “adj” should be grouped with the objects of the prepositions, since all three of the pronouns functioning as adjectives (demonstratives) are in apposition to a noun that is an object of a preposition. Table 14 deals with the gender of pronouns, and it might seem surprising at first glance that the deviation factor is so low (1.44) in comparison with that of the table 9, gender of nouns and adjectives (21.54). However, when one realizes that most of the pronouns (thirty-four of thirty-eight) in the chapter refer to people and that their gender is not a mere grammatical category, the low deviation factor becomes understandable. An alternate deviation factor, with pronouns referring to people excluded, will be calculated for comparison. The changes to the tables that result from the elimination of the significant variants and from the modifications to the data discussed above yield the following results: Table Changes 1 Gk words: 40 1.29 Gk words/Heb word deviation factor: 0.31 total Heb (>1x): 129 primary Gk: 116 90.5% by primary rendering Gk roots: 40 1.38 Gk roots/Heb root 2 Gk advs: 23 1.77 Gk advs/Heb adv total Heb (>1x): 82 81.7% of Heb advs by primary rendering deviation factor: 1.27 4 percentage of verbs/verbals represented by verbs/verbals: 93/94 = 98.9% percentage of infinitives represented by infinitives: 4/6 = 66.7% 5 deviation factor (discrete tense/mood combinations): 6.70 deviation factor (grouped): 0.85 deviation factor (grouped, without verbals): 0.83 6 deviation factor: 3.55 10 deviation factor: 1.33 14 deviation factor (excluding references to people): 5.00 15 deviation factor: 0.00 These results permit a better informed appraisal of the translators’ approach to the text than was possible before the raw data was examined more closely (see above, pp. 8586). The conclusions concerning lexical data and word classes appear sound after a reappraisal of the data. Several modifications must be made, though, to the characterization of grammatical consistency stated earlier. The translators were absolutely consistent in rendering not only the person of verbs, but also the number of pronouns. They are very

98

consistent in rendering the inflection of verbs (grouped), verbal number, grammatical use of nouns and adjectives, and grammatical use of pronouns. They are only a little less consistent in their rendering of the number of nouns and adjectives. They are still fairly inconsistent in their rendering of articles and definiteness, as well as in their rendering of verbal stem. They are still very inconsistent in their rendering of the gender of nouns and adjectives. The rendering of pronominal gender requires special comment. It might have been supposed that the deviation factor would have soared to levels near that of the gender of nouns and adjectives once references to the gender of people were removed, but instead it increased only to 5.00. The reason for this unexpectedly low deviation factor is that only four of the pronouns in the chapter (out of thirty-eight) refer to objects other than people, and the highest deviation factor possible for four items (they are all masculine in Hebrew) put into three categories (masculine, feminine, neuter in Greek) is 8.00, when the items are split 0, 2, 2 and the 0 refers to the category that corresponds most closely to that of the source language, the one that was expected (in this case masculine gender).55 The split in the present case is 0, 3, 1, because there are no masculine pronouns in Greek, three feminine pronouns, and one neuter pronoun. The rendering of the gender of pronouns referring to objects other than people, then, is inconsistent (in the chapter it is consistently something other than what might be expected, but too few examples occur to state conclusively that the rendering of pronominal gender is very inconsistent, though such is probably the case over a larger amount of data). The corollary to this observation is that when the pronouns do refer to people, the rendering in LXX is absolutely consistent, at least in this chapter. Variants in the Representation of Hebrew Lexemes by Greek Lexemes (Segmentation) The fact that only 62.8% of the Hebrew compound words found in MT are rendered exactly in LXX leads one to suspect immediately that none of the variants is significant. Even if one of the variants appeared to have some claim to represent a different Vorlage, reasons other than simply the failure of the Greek to represent every Hebrew element in the compound would have to be present for the variant to be considered significant. Most of the variations in segmentation deal with instances in which the Greek translators failed to render the prepositions or in compounds. Since the omission of any Greek preposition is one of the two main ways the translators chose to render (omission four times, six times, others two times), and since omission was the main way they rendered (eleven of thirteen times), the variations in segmentation that result 55 The zero as part of the data is a special case that is used only when the category in the target language that corresponds most closely to that in the source language is empty, as it is here. See Appendix 4 for details on the calculations.

99

from the omission of these prepositions (i.e., variants 52, 53, 54, 55, 56, 57, 58, 59, 61, 63, 65, and 66) cannot be considered significant. Once these variants are eliminated, only four remain to be considered in this category. The first is variant 51, the representation of by . is, strictly speaking, a compound of the inseparable preposition and the plural construct of . Thus, it would have been possible for the translators to render the constituent parts by such a formula as (cf. 1 Sam 18:16; Eccl 2:26 [bis]; 5:1, 5, etc.). However, developed into a kind of improper preposition in Hebrew, and it is regularly rendered in LXX by and other prepositions and prepositional 56 phrases. Thus, the variant is not significant. In variants 60 and 64, the preposition is not represented in Greek, though it is usually rendered by (twelve of fifteen times in the chapter). The in variant 60 is apparently part of the Hebrew idiom with II piel (though it is a hapax), 57 and, since would be unidiomatic in Greek, its exclusion here should not be considered a significant variant. In variant 64, the Greek rendering is equivalent to the Hebrew found in MT (see the discussion on variant 47, above), and again should not be considered significant. Variant 62 concerns the omission of - before an infinitive. The verb regularly takes before the thing to be feared, even when it is an infinitive clause.58 However, the Greek expression does not need a preposition before the infinitive to clarify the meaning; in fact, such a preposition would be unidiomatic.59 Thus, as was suspected at the outset, none of the variants in segmentation is significant. Therefore, no adjustment to the preliminary partial translation technique for segmentation is necessary. Finally, it should be noted that the phrase in verse 10 was counted as a single unit in the calculations, and was considered to be an exact representation of it. Variants in Word Order In contrast to their lack of concern to render every Hebrew element of compound words, the Greek translators were meticulous in reflecting the word order of their Vorlage to the extent that the language allowed. Thus, any variation from word order probably represents a Hebrew text different from MT. Only two deviations from the word order of MT are found in LXX. The first, variant 67, is found in verse 16, and represents a shift of one compound Hebrew word 56 Sollamo, Semiprepositions, 13-122. 57 KB3 , s.v. “II 58 BDB, s.v. “

.” .”

59 See Smyth, Grammar, 503, for a discussion of the infinitive with verbs of fearing. Cf. also

Soisalon-Soininen, Infinitive, 100-101.

100

three positions. Associated with this apparent shift of is the omission of any word in LXX corresponding to (variant 98, below) and the rendering of with (variant 40, above). It is almost certain that the LXX translators had a text different from MT at this point, so the variant is significant. Variant 68 has for . It is easy to see how a Hebrew scribe could have confused the word order in either direction between and . Since either order is idiomatic Hebrew, this variant is also significant.60 When these two significant variants are eliminated from the statistical data because they are no longer considered variations from the Vorlage, the data related to word order is modified. There are now 369 Hebrew semantic units, 0 variations, and the Hebrew word order is followed 100.0% of the time. Variants in Quantitative Representation A glance at the summary at the bottom of the table dealing with quantitative representation indicates that the Greek translators were not overly concerned with representing every possible aspect of the Hebrew text without addition or omission. A more careful perusal does reveal some trends, though. LXX has fifty-seven semantic units which have no apparent counterpart in MT, and MT has thirty-one semantic units which have none in LXX. If the long addition in LXX in verse 21 (variant 105) is omitted from consideration, however, the excess in MT outweighs the excess in LXX thirty-one to twenty-six. (Variant 105 is assumed to be significant at this point, but it will be examined in detail later.) If the figures are recalculated, omitting the thirty-one words of the long addition in verse 21, the result is an 87.0% agreement between MT and LXX. This figure is a more accurate representation of the translators’ concern to maintain a one-to-one correspondence between their Vorlage and LXX, though it will have to be modified again after an in-depth examination of the data. The first case to be examined, variant 97, has an excess in MT, and it is concerned with the omission of which is followed by a noun functioning as an indirect object. Since an omission is one of the two main ways in which is rendered, this variant cannot be considered significant. Variant 69 concerns the addition of a form of to a participle in LXX, resulting in a periphrastic construction. As noted above (see discussion on variant 1), the LXX translators often rendered Hebrew participles with Greek verbs, including periphrastic verbs, so this variant cannot be considered significant. The next group of cases to be examined are those that, if secondary, could be 60 Of course, the confusion could have arisen at the moment of translation, without the benefit of a different Vorlage, but for methodological reasons, and since the variant is clearly not an inner-Greek corruption, the decisions as to when exactly the variant arose, as well as which order is original, belong to a later stage in the process of evaluating the various extant readings. See above, p. 25, n. 50.

101

classified as explanatory expansions;61 variants 70, 72, 76, 79, 82, 87, 89, 90, 92, 99, 102, and 103 fall into this category. Together, they represent a deviation of sixteen semantic units between MT and LXX. The fact that each version has eight excess semantic units as compared with the other demonstrates that the tendency toward expansion was not limited to the translators of LXX, for it shows eight possible Hebrew expansions in MT. An examination of these variants might reveal that one or two are primary, and their omissions secondary, but the evaluation process belongs to a later stage in the process of reconstructing the text. The important point here is that the LXX translators do not show any tendency toward expanding their text with explanatory elements, so it is likely that most or all of the explanatory expansions in LXX are derived from the Hebrew Vorlage, so all these variants should be considered significant. On five occasions, LXX has a which does not correspond to a Hebrew conjunction (variants 71, 77, 78, 85, and 88). It is important to note that all five additions of conjunctions are in LXX; no additional conjunctions (aside from conjunctions included as part of a larger phrase) appear in MT. On the other hand, the LXX translators show a concern for consistency with conjunctions (see table 3), and this factor, too, must be taken into account. Three of the variants (77, 85, and 88) concern the phrase . It was noted above (pp. 41-42) that the Hebrew phrase in MT displays a common form of Hebrew asyndeton with an auxiliary-like verb and that the translators might have felt that an additional in each instance would have been a better balance for the following phrase. Thus, it is doubtful that these conjunctions reflect conjunctions in the Hebrew Vorlage, so these variants are not significant. Variant 71 has an additional at the beginning of the final clause of verse 2. Since MT usually begins clauses in a narrative with a conjunction, it is somewhat surprising not to find one here. The additional in LXX could either be a reflection of a different Vorlage or an assimilation to usual Hebrew style. The addition or omission of conjunctions is a frequent variant even among Masoretic mss, and it is impossible to determine at this point whether the conjunction lay in the Vorlage or in the translator’s mind, but it seems best to consider the variant significant.62 The final example of an additional is variant 78, which reads for . This variant is related to variant 12 above, and, as noted there, the construction in LXX contains a Hebraism that a literal rendering of MT would not have, namely, for a single verbal idea. The 61 See Barthélemy, ed., Critique textuelle, 1:*72, in the discussion of factor 7: “Dans certains cas, la forme particulière d’un texte est le résultat d’une exégèse spéciale que l’on en donnait. . . . Ou il arrivait aussi qu’ils voulaient un texte qui exprimât plus clairement un sens qui n’en ressortait qu’imparfaitement.” Cf. also the discussion on p. XIX concerning explicitation and amplification. 62 So Alfons Schulz, Die Bücher Samuel, Exegetisches Handbuch zum Alten Testament, ed.

Johannes Nikel, vol. 8 (Münster: Aschendorff, 1919), 57-58.

102

omission of - on the front of could be explained as graphic confusion with the adjacent , perhaps abbreviated as . Variant 78, then, is a significant variant. Another group of variants that share a common feature is variants 75, 80, and 91, each dealing with the number of times Samuel’s name was repeated when God called him. MT in verses 4, 6, 8, and 10 reads ; ; ; and , respectively. Of these, the first is an indirect object (following ), and the last is clearly vocative. The other two could conceivably be interpreted as either accusatives or vocatives, but the lack of before in both cases implies that they are vocatives.63 LXX reads ; ; and in verses 4, 6, and 8, and it omits any reference to Samuel in the corresponding clause in verse 10. The references in verses 4 and 6 are vocatives, but that in verse 8 is probably accusative. Of the four instances cited, MT and LXX agree in only one, verse 8 (and even there they probably differ in their understanding of the syntactic relationship of “Samuel” to the rest of the verse). The fact that neither version is consistently shorter or longer than the other and that neither is identical in all three instances suggests that the variants did not arise from the translators but from their Vorlage. Thus, variants 75, 80, and 91 should all be considered significant. The omission of in variant 74 should be considered in conjunction with variant 75. While it is true that the omission of is usually not significant (see above on variant 97), the similarity between and , coupled with the fact that variant 75 involves an omission, suggests the possibility that the in MT could be a remnant of an earlier . Thus, variant 74 should be considered significant.64 The next variants to be considered are two pairs of apparent substitutional variants: 83 and 84, and 100 and 101. MT in verse 6 reads , and LXX reads , so it seems that and are substitutional equivalents for one another.65 It 63 Wellhausen says that in 3:6 must be vocative, since it lacks a preceding ; Wellhausen, Bücher Samuelis, 52; cf. Kautzsch, ed., Gesenius’ Hebrew Grammar, § 117a. However, there are exceptions to the rule that a definite direct object should be preceded by ; see Waltke and O’Connor, Introduction, 180. Takamitsu Muraoka, Emphatic Words and Structures in Biblical Hebrew (Jerusalem: Magnes Press, 1985; Leiden: E. J. Brill, 1985), 146-58, also notes exceptions, but in his analysis of 1 Samuel 1-8, he finds no examples of accusative proper names without (cf. p. 150). KJV, ASV, NIV, and NRSV read a vocative in 3:6 and an accusative in 3:8. NAB, GNB, and REB apparently read accusatives in both verses, but the accusatives could have been used because they are stylistically preferable in English. Klostermann, Dhorme, Budde, McCarter, and Klein all accept the LXX reading in 3:6, but in 3:8, Dhorme and Budde read accusative, while the other three read vocative; Klostermann, Bücher Samuelis, 11; Dhorme, Livres de Samuel, 43; Karl Budde, Die Bücher Samuel, Kurzer Hand-Commentar zum Alten Testament, ed. Karl Marti, vol. 8 (Tübingen and Leipzig, J. C. B. Mohr [Paul Siebeck], 1902), 27; McCarter, I Samuel, 94; R. W. Klein, 1 Samuel, 29. 64 Cf. Thenius, Bücher Samuels, 16; cf. also Wellhausen, Bücher Samuelis, 52; S. R. Driver,

Notes on the Books of Samuel, 42. 65 Cf. Barthélemy, ed., Critique textuelle, 1:*73-74. After a discussion of conflated readings (factor 13), he says, “En d’autres occasions, le texte subissait une correction, mais on omettait d’ôter du texte la forme primitive qu’on venait de remplacer par la forme corrigée.” Cf. also p. XX.

103

is possible that the variants arose from the graphic similarity of and , perhaps from an original , but the direction of the variant (if in fact it did arise in this way) is unclear. Nevertheless, the likelihood of graphic similarity suggests a Hebrew rather than Greek origin of the variant, so variants 83 and 84 are significant. In the last clause of 3:17, MT reads , whereas LXX reads . The variation between and probably did not arise because of graphic similarity but because the phrase reflected in LXX is cumbersome. It is, however, the type of expression characteristic of Hebrew rather than Greek (cf. 1 Sam 15:14; 25:24; 2 Sam 7:22; 18:12; 22:7, etc.). The tendency of the translators would probably be in the direction of better Greek idiom than a more Hebraic expression if they were the ones who introduced the variant. It is probable, then, that the root of the variant found in LXX lies in the Vorlage, and variants 100 and 101 should be considered significant.66 The next variants to be considered are variants 94 and 95. Though the two variants correspond in position, they are not substitutional variants like the previous cases, since they are not semantically equivalent. It seems that a textual disruption exists here, for it is difficult to see how one reading could have arisen from the other in either Hebrew or Greek, though both make sense. It is possible that was transposed from later in the same verse, but the same transposition could have occurred in Hebrew. In the absence of other data, and in light of the textual disruption, these variants should be considered significant. Variants 81, 96, and 98 have in common that they all represent a compound (twofold) verbal phrase in one version where the other version has a simple (one-fold) verbal phrase. In the first two of these variants, the shorter text is probably the result of parablepsis, but the omission could have occurred as easily in one language as in the other. There is no indication that the translators would have omitted the phrase intentionally, but it is certainly possible that in variant 81 was omitted accidentally in the process 67 of translation. However, variant 96 demonstrates that Hebrew scribes were not immune from such mistakes, so it is impossible to tell whether the presumed omission in variant 81 occurred in the process of transmission (Hebrew or Greek) or translation. Therefore, it should probably be considered significant, and variant 96 should certainly be considered significant. Variant 98 is somewhat different from the other two, inasmuch as it is associated with further textual disruption (see above, variants 40 and 68). The shorter reading in verse 16 is not the result of parablepsis. Instead, MT here looks like the result of the conflation of two variant readings, and . If this is indeed the case, the 66 Cf. Schulz, Bücher Samuel, 65. 67 Contra Budde, Bücher Samuel, 27.

104

shorter LXX version represents an earlier form of the text, one clearly based on a Hebrew Vorlage, since the conflation occurred in Hebrew rather than Greek. Variant 98, then, is significant. Variants 73, 86, and 93 all represent Hebrew texts that are one word longer than LXX. In variant 73, MT reads , and LXX reads . Though one might have expected a rendering like that found in Gen 2:11 ( ), the translators here were content to render both Hebrew words with a single Greek word. The usual relative adverb in classical Greek would have been , but the genitive of the relative pronoun was also used to indicate the adverbial idea of place,68 so this variant is not significant. Variant 86 deals with the failure to render the second in verse 7. This variant has already been discussed somewhat in the previous chapter (p. 46), where it was noted that whereas Hebrew prepositions usually govern only one word, Greek prepositions frequently govern more than one object. It is not unknown, however, for Hebrew prepositions to govern two or more words (cf. p. 46, n. 36). Furthermore, though it is no problem to have a Greek preposition govern multiple nouns, it is also perfectly idiomatic to repeat prepositions. The question to ask is whether the translators show a tendency to modify an acceptable Greek reading that closely reflects the Hebrew text to one that is perhaps somewhat closer to classical style, but further removed from Hebrew idiom. The answer seems to be that the translators show no such tendency. What might be seen as modifications in the renderings of (3:3) and (3:10) would be unidiomatic if translated word for word and so do not apply to the present case. The numerous Hebraisms preserved in LXX (e.g., [3:6], [3:17], [3:17]) show that the translators were not primarily interested in good Greek idiom, though they sought an understandable text. The fact that the hexaplaric mss Acx also fail to render the second supports the idea that the second was not present in all Hebrew texts even in the third Christian century, and it seems probable that the ms from which the translators of Samuel worked omitted it as well. Variant 86 should be considered significant. In verse 11 (variant 93), MT reads , while LXX simply has . Of the other four occurrences of in the chapter, the translators rendered it with a relative pronoun twice and an article (substituting for a relative pronoun) twice. A perusal of other occurrences in the books of Samuel indicates that the relative or its equivalent was the common rendering of , and it seems unlikely that the translators of LXX would have failed to render it had it been in their Vorlage. The omission of the conjunction in T supports the assumption that Hebrew mss existed which omitted the word. Therefore, the variant is significant. 68 Smyth, Grammar, 562; LS, s.v. “ A,B LXX .

, , .” Cf. the rendering of

in Judg 5:27

105

Only two variants dealing with quantitative representation remain, variants 104 and 105 in verse 21. Though one phrase is missing from LXX ( , variant 104), by far the shorter text is MT, which lacks thirty-one semantic units found in LXX (variant 105). Much of the content seems to be a duplication of phrases from verse 20 and even phrases from verse 21 itself. It is probable that part of the difficulty in MT arose from parablepsis: the last word in verse 21 in MT is , and the last word in LXX is . However, simple parablepsis does not account for LXX’s omission of , and the redundancy of the verse suggests an earlier disturbance of the text.69 Much of verse 21 may even be seen as an alternate version of verse 20, though the content is not identical. At this point it is sufficient to point out the strong probability that the disturbance arose in Hebrew rather than in Greek (cf. especially the Hebraism ), so both variants are significant.70 When the variants that probably represent a different Hebrew Vorlage are eliminated from the quantitative representation data, only 7 of the original 88 deviations remain. Of the 438 Hebrew semantic units in the original calculations, 28 are eliminated from consideration at this time, that is, all those significant variants in which the excess was in MT. The result is a 98.3% rate of agreement between MT and LXX, a figure much higher than the preliminary partial translation technique indicated. This figure indicates that the translators of LXX were careful to represent all the words that lay before them in their Vorlage, without adding any, with a few exceptions. It is true that some of variants eliminated as significant may never have existed in Hebrew mss but were rather unintentional mistakes made in the process of translation. However, unintentional variants say nothing about the intention of the translators, which is the basis of translation technique.71 Peshitta Limitations of Syriac for Rendering Hebrew Unlike Greek, Syriac is a Semitic language closely allied with Hebrew; both are representatives of the Northwest branch of the language group. The Northwest (or simply West) Semitic languages, at least the later representatives of the group, are usually divided 69 The exact meaning of in the context of the verse is unclear; cf. Walter Dietrich, David, Saul und die Propheten: Das Verhältnis von Religion und Politik nach den prophetischen Überlieferungen vom frühesten Königtum in Israel, Beiträge zur Wissenschaft vom Alten und Neuen Testament, no. 122 (Stuttgart: W. Kohlhammer, 1987), 119-20. 70 Cf. Barthélemy, ed., Critique textuelle, 1: 151-52, and the commentaries. 71 Of course, the carelessness of the translators is also a factor in translation technique, but in a project as important as the translation of the Hebrew Scriptures into the common language must have been for them, it may probably be assumed that the translators made an effort to minimize careless errors.

106

into two main subgroups: the Canaanite (including Hebrew, Moabite, and Phoenician) and the Aramaic.72 Aramaic itself may be divided into various subgroups on the basis of time and geography.73 Since Syriac is a Semitic language, the Syriac representation of a Hebrew text could be very close indeed to the Vorlage, in both syntax and vocabulary. Nevertheless, significant differences do exist between the two languages, and these must be noted. The nominal system of Syriac shares with Hebrew two genders (masculine and feminine). Because both are Semitic languages, they share many common roots, and it is not surprising that a large number of words sharing common roots also share the gender inherited from a common ancestor. Not all words from common roots have the same gender, however, and Syriac possesses many roots foreign to Biblical Hebrew. Like Greek, Syriac only has two numbers (singular and plural); the dual has fallen into disuse, except for the numbers two and two hundred.74 Thus, Hebrew duals are generally rendered by Syriac plurals. Another difference exists in the number of states available, with two in Hebrew (absolute and construct) and three in Syriac (absolute, construct, and emphatic). The existence of the emphatic state in Syriac highlights another difference, namely, that Syriac lacks an article.75 In Syriac, the absolute state rarely occurs, except in certain constructions.76 Its place is regularly assumed by the emphatic, which is the lexical form in most lexicons. It cannot be assumed, then, that the use of the emphatic state in the Syriac implies any sort of definiteness in the Hebrew Vorlage. The construct case is also 72 The International Standard Bible Encyclopedia [hereafter ISBE], 1988 ed., s.v. “Semites,” by Mario Liverani, trans. William Sanford LaSor, 388; for a concise chart of the Semitic languages, see M. J. Mulder et al., eds., The World of the Bible, trans. Sierd Woudstra, vol. 1 of Bible Handbook, ed. A. S. van der Woude (Grand Rapids: William B. Eerdmans, 1986), 77. For a way of grouping the Semitic languages into two branches rather than three, see Waltke and O’Connor, Introduction, 5. 73 Joseph A. Fitzmyer distinguishes five historical phases of the language: Old Aramaic

(925 B. C. E. to 700 B. C. E.), Official Aramaic (Reichsaramäisch) (700 B. C. E. to 200 B. C. E.), Middle Aramaic (200 B. C. E. to 200 C. E.), Late Aramaic (200 C. E. to 700 C. E.), and Modern Aramaic (still spoken today in isolated locales). Syriac is a representative of the Eastern branch of Late Aramaic. The particular dialect of Syriac present in the Peshitta is Jacobite, a western form of Syriac, as opposed to Nestorian, an eastern form. See Joseph A. Fitzmyer, “The Phases of the Aramaic Language,” chap. in A Wandering Aramean: Collected Aramaic Essays, Society of Biblical Literature Monograph Series, no. 25 (Chico, CA: Scholars Press, 1979), 60-63. 74 Carl Brockelmann, Syrische Grammatik, 6th ed., Porta Linguarum Orientalium, ed. Richard Hartmann, no. 5 (Leipzig: Otto Harrasowitz, 1951), 51n. 75 The characteristic

ending of the emphatic state in Aramaic (including Syriac) may be the remains of a postpositive article. See Brockelmann, Grammatik, 51; Fitzmyer, “Phases,” 66, 69. 76 Robinson lists four types of cases: predicate nominatives, nouns in distributive phrases, nouns after (= “all, every”), and nouns after numerals. The use of the absolute case is not universal in the last three types of cases. See Theodore H. Robinson, Paradigms and Exercises in Syriac Grammar, 4th ed., ed. L. H. Brockington (Oxford: Clarendon Press, 1962), 21. Brockelmann adds proper nouns to the list of uses of the absolute, as well as a few rarer instances of its use; Grammatik, 104-5.

107

used less in Syriac than in Hebrew, its place being taken by the relative particle followed by a noun in the emphatic case.77 Because of these differences between Hebrew and Syriac idiom in regard to the use of states, the textual critic must use caution in analyzing the implications of differences between P and MT. With regard to the verbal system, both Hebrew and Syriac verbs can be classified by stem (or conjugation), inflection, person, gender, and number. Some differences do exist, however, in the use of stem and inflection. Hebrew’s seven stems in three groups are rendered by Syriac’s six stems in three groups: simple stems (peal, ethpeal), intensive stems (pael, ethpael), and extensive (or causative) stems (aphel, ettaphal). There are no distinct reflexive stems in Syriac.78 Despite the differences, a great degree of correlation is possible: for example, between qal and peal, niphal and ethpael, or hiphil and aphel. However, Syriac may employ a verb in one stem to render a Hebrew verb that regularly occurs in another.79 The greatest distinction in the use of inflection between Hebrew and Syriac is the lack of any construct in Syriac corresponding to the Hebrew waw consecutive construction. Thus, the correspondence that normally exists between a Hebrew and Syriac inflection will generally be reversed when the waw consecutive is used in Hebrew. Another difference between the use of inflections in Hebrew and Syriac is that in the latter, the inflections have come to be used more or less as tenses (perfect for past, imperfect for future), perhaps under the influence of Greek. In addition, the participle is used with a personal pronoun as a present tense, and three compound tenses have developed, all of which use forms of the copula . 80 One stylistic characteristic of the Syriac language must be noted, namely, the extensive use of the anticipatory pronoun. A verb that has a noun (either a common or a proper noun, but not a pronoun) for a direct object will often have a third person pronominal suffix attached to the verb that “anticipates” the direct object, agreeing with it in number and gender. Similarly, nouns that could possibly appear in the construct case frequently have an anticipatory pronominal suffix attached to them and are followed by the relative particle and the word that it governs (i.e., what is often called the genitive).81 77 Brockelmann, Grammatik, 105-6. 78 T. H. Robinson, Grammar, 51-53. 79 For example,

in P (peal) often corresponds to

80 T. H. Robinson, Grammar, 53, 59-60. 81 Brockelmann, Grammatik, 115, 106.

(piel) in MT.

108 Partial Translation Technique Add-Oms Since Syriac is a Semitic language, one might suppose that structures in the Hebrew text that translators of LXX found difficult to represent idiomatically would be easier to render in Syriac, and an analysis of the data supports this supposition. The translators of P often render or with equivalent Syriac terms, and they almost always have an equivalent for . However, Syriac idiom does show itself to be different from Hebrew in the frequent replacement of the Hebrew construct chain with a noun in the emphatic state followed by the relative particle and the next word. Such constructions in Syriac will not be considered add-oms, since they are in fact the usual (though not exclusive) rendering of the Hebrew. Another aspect of Syriac idiom that has no equivalent in Hebrew is the use of anticipatory pronominal suffixes before a direct object. These anticipatory pronominal suffixes will not be considered add-oms, either. See Appendix 2 for a full list of P’s addoms. Consistency The following variants in P differ from MT in some way related to consistency. Ref Variation Peshitta Masoretic Text 1 3:1 cont past participle 2 1° preposition DDO 3 3:2 plural singular 4 verb participle 5 plural singular 6 participle adjective 7 cont past imperfect 8 3:3 1° lexeme 9 verb participle 10 3:5 lexeme 11 3:6 lexeme 12 3:7 lexeme 13 3:8 verb participle 14 3:9 imperative perfect w/c 15 3:12 1° lexeme 16 lexeme 17 2° lexeme 18 noun infinitive 19 noun infinitive

109

20 3:13 2° lexeme 21 cont past perfect 22 lexeme 23 periphrasis participle 24 subject direct object 25 3:14 plural singular 26 plural singular 27 3:15 singular plural 28 verb infinitive 29 3:17 2° lexeme 30 lexeme 31 plural singular 32 3° lexeme 33 3:18 plural singular 34 3:19 lexeme 35 lexeme 36 3:20 lexeme 37 3:21 plural singular Table 17.—Verbs, Nouns, and Adjectives: Lexemes Heb words (>1x): 29 Syr words: 35 1.21 Syr words/Heb word total Heb (>1x): 132 primary Syr: 123 93.2% of Heb words by primary rendering deviation factor: 0.26 Heb roots (>1x): 27 Syr roots: 34 1.26 Syr roots/Heb root Table 18.—Adverbs, Prepositions, and Particles: Lexemes Heb advs (>1x): 12 Syr advs: 22 1.83 Syr advs/Heb adv total Heb (>1x): 85 primary Syr: 63 74.1% of Heb advs by primary rendering deviation factor: 3.41 Table 19.—Conjunctions: Lexemes Heb conjs (>1x): 2 Syr conjs: 4 2.00 Syr conjs/Heb conj total Heb (>1x): 64 primary Syr: 61 95.3% of Heb conjs by primary rendering deviation factor: 0.16 Table 20.—Word Classes percentage of verbs/verbals represented by verbs/verbals: 94/96 = 97.9% percentage of verbs represented by verbs: 74/75 = 98.7% percentage of participles represented by participles: 7/12 = 58.3% percentage of infinitives represented by infinitives: 4/7 = 57.1% percentage of nouns/adjs represented by nouns/adjs: 65/70 = 92.9% percentage of pronouns represented by pronouns: 42/42 = 100.0%

110

Table 21.—Verbs: Inflection deviation factor: 1.83 deviation factor (without verbals): 0.42 Table 22.—Verbs: Stem deviation factor: 1.33 Table 23.—Verbs: Person deviation factor: 0.00 Table 24.—Verbs: Gender deviation factor: 0.02 Table 25.—Verbs: Number deviation factor: 0.00 Table 26.—Nouns and Adjectives: Gender deviation factor: 1.38 Table 27.—Nouns and Adjectives: Number deviation factor: 2.03 Table 28.—Nouns and Adjectives: Use deviation factor: 1.09 Table 29.—Nouns and Adjectives: Hebrew Articles, Syriac Emphatic State deviation factor: 6.06 deviation factor (without implied articles): 6.46 Table 30.—Nouns and Adjectives: Definiteness deviation factor: 2.09 Table 31.—Pronouns: Gender deviation factor: 0.05 Table 32.—Pronouns: Number deviation factor: 0.00 Table 33.—Pronouns: Use deviation factor: 8.15 Representation of Hebrew Lexemes by Syriac Lexemes (Segmentation)

38 39 40 41 42 43 44

Ref 3:1 3:5 3:6 3:8

Hebrew Compound

Syriac Rendering -

3:9 3:10

-

111

45 3:13 46 3:15 47 3:20 Heb compounds: 47 Syr equivalents: 36

-

76.6% of Heb compounds rendered exactly

Word Order Ref Number of Variations Syriac Variant 48 3:1 1 49 3:6 2 50 3:11 1 Heb semantic units: 396 variations: 4 Heb word order followed 99.0% of the time Quantitative Representation

51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74

Ref 3:1

3:2 3:3

3:5 3:6

3:7 3:8

Number of Variations +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 -1 -1 +1 +1 -1 +1 +1 +1 +1 +1 -1

Variant 1° - 1° - 1° - 2° - 4° - 1° - 1° - 2° - 2° - 3° - 4° - 5° - 1° - 2° - 1° - 2° -

112

75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112

3:9

3:10

3:11 3:12 3:13

3:14

3:15

3:17

3:18

+1 +1 +1 -1 -1 -1 -1 -3 +1 +1 +1 +1 +1 -1 -1 +1 +1 -1 +2 -1 +1 +1 +1 +1 -1 +1 +1 +1 -1 +1 -1 +2 +1 +1 -1 +2 +1 -1

- 1°

- 1° - 3° 2° - 2° - 1° - 1°

- 2° - 1° - - 1° 1° - 1° - 1° - 2° - 3° 2°

1° 3° 2° -

113

113 -1 114 +1 115 +1 116 +1 117 +1 118 3:19 +1 119 3:20 +1 120 -1 121 3:21 +1 122 -7 123 +1 124 -1 Heb semantic units: 425 Syr deviations: 85

-

- 1°

3° 80.0% agreement

Elimination of Variants Deist characterizes P as follows: “On the whole the Peshitta follows the Massoretic tradition fairly faithfully. This is true especially of the Pentateuch, but also to a large extent of the books of Joshua, Judges, 1 and 2 Samuel and 1 and 2 Kings.”82 This view of P is supported by the fact that the number of variants listed in the section on consistency is 30% shorter than the corresponding LXX list. A comparison of the first three Greek and Syriac tables shows that the lexical consistency of P is similar to that of LXX, being somewhat less consistent in the use of conjunctions and somewhat more consistent in lexical choices involving verbs, nouns, and adjectives. P tends to render words of one class in Hebrew by words of the same class in Syriac, though some deviation is evident in the rendering of nouns and adjectives. A significant amount of variation exists in the rendering of Hebrew verbals by verbals of the same class in Syriac; although the variation in participles is less than that in LXX, the translators of P clearly felt no compunction about rendering a participle or an infinitive by something other than a Syriac participle or infinitive. As for grammatical consistency, the translators were absolutely consistent in rendering the person and number of verbs and the number of pronouns. They were very consistent in their rendering of the inflection of verbs (excluding verbals), verbal gender, and pronominal gender. They were fairly consistent in their rendering of verb stems, gender and number of nouns and adjectives, use of nouns and adjectives in the sentence, and definiteness of nouns and adjectives (though P does use the emphatic state fairly frequently when MT is indefinite). They were less consistent in their use of the emphatic 82 Deist, Text of OT, 145. Emanuel Schwartz says that P is fairly faithful to its original, though it takes many liberties; Emanuel Schwartz, Die syrische Uebersetzung des ersten Buches Samuelis und ihr Verhältniss zu MT., LXX und Trg. (Berlin: H. Itzkowski, 1896), 95.

114

state to render Hebrew articles (especially when the article is lacking in MT), and in the use of pronouns (though if objects of prepositions, which have a deviation factor of 19.13, are excluded, P becomes fairly consistent in this category). In comparison with LXX, P can be characterized as slightly more consistent in most categories related to consistency. In a few cases—namely, gender and use of nouns, rendering of articles and definiteness—P is much more consistent. One would suspect a priori that P would be much closer to MT in regard to both noun gender and use, since it is a cognate language, and the data supports this suspicion. The fact that P is quite a bit more consistent in rendering Hebrew articles and especially in indicating the definiteness of Hebrew nouns is somewhat unexpected, given the propensity of Syriac authors to use the emphatic state in most cases. This fact seems to indicate that the translators did have some inclination to use the emphatic state as an indicator of definiteness in the Hebrew text, but the data shows that it was not an overriding concern. One other surprise in the comparison of P with LXX is the lower deviation factor in LXX for the use of pronouns, when one would have expected the similarity of sentence structure and vocabulary to have made P have the lower deviation factor. The deviation factors of LXX and P are similar if the category of object of preposition is excluded from the Hebrew column, and the use of direct object in P for Hebrew objects of prepositions results from the omission of or with a pronominal suffix in every case. P exhibits a greater tendency than LXX to render compound Hebrew words with compound Syriac words, but one out of four Hebrew compounds still goes unrepresented by a Syriac compound. Next, the translators of P were just as concerned as the translators of LXX to follow the Hebrew word order precisely. Finally, the initial data concerning quantitative representation yields a figure of 80% agreement with MT, a number quite close to the initial LXX agreement. One obvious difference between the two, however, is the lack of the long addition in 3:21 P that is found in LXX. Variants Related to Consistency Having given a general description of the initial data for P, it is time to begin examining specific cases. Variants 8 and 36 may be considered together, since both deal with names of God. In variant 8, renders , where is expected; in variant 36, renders , where is expected. appears only three times in the chapter, and it is rendered by twice and once. appears sixteen times that are rendered, and only here does P have . De Boer notes that renders eight times in 1 Samuel 1-16, and renders six times in those chapters (though he omits the occurrence in 3:20), which, he says, “proves that the names were supposed to be of similar signification,”83 that is, that they were more 83 De Boer, I Samuel i-xvi, 23-24.

115

or less interchangeable. However, 6 occurrences of out of 222 instances of is hardly a trend, though 8 occurrences of out of 72 instances of is somewhat more significant. Nevertheless, the translators clearly show a tendency toward careful rendering of the divine names, and though some of the instances of for might reflect the influence of T (or a similar exegetical tradition), both of these variants should probably be considered significant. In variants 10 and 11, the imperative appears for in MT. A more literal translation of in many contexts would be the cognate , but in the present instances, is used as a kind of auxiliary verb, just as is in MT.84 These variants are different from LXX variants 12 and 18 in that those variants ignored the auxiliary nature of the Hebrew verb. Thus, variants 10 and 11 are not significant, since they are merely an attempt to use idiomatic Syriac to render a Hebrew phrase. The next four variants—variants 12, 15, 17, and 20—all deal with the rendering of prepositions. In the first three cases, P uses to render . The other renderings for in the chapter are (five times), (three times), and nothing (one time). It might have been expected that a literal translation of would be either or , which both have semantic fields comparable to , and, of course, both are frequently used, both in the chapter and elsewhere in P. In addition, is cognate with , and one could perhaps postulate a Hebrew Vorlage of in these three cases, and especially in the last two, which have some support from LXX. However, has a broader range of meanings than its Hebrew counterpart. It is frequently used of motion toward, whether concrete or abstract (as in the phrase, “the word of the Lord came to . . .”), and can even be used to indicate possession ( = ). A look at surrounding chapters reveals that does render in other contexts (1:27; 2:27, 34 bis; 4:21 bis; 5:4; cf. 4:1 and 5:4 for - ; 4:18 for ), in addition to rendering . Thus, though 3:12 may have originally read in both cases, the evidence of P cannot be used to support such a reading, so variants 12, 15, and 17, are not significant. The other prepositional variant, variant 20, involves the rendering of by . The reason for this choice is that appears as part of the phrase , which is equivalent to and is its usual translation. Thus, this variant is not significant. Variants 16, 29, and 32 all concern the translation of by . Syriac has no cognate to , and is obviously cognate to . Furthermore, , which does frequently translate , has a semantic field that is closer to than is. However, the range of meanings in is larger than in , and thus is an appropriate translation for is these three cases.85 The variants, then, are not 84 Cf. J. Payne Smith, ed. A Compendious Syriac Dictionary (Oxford: Clarendon Press, 1903),

s.v. “

”; BDB, s.v. “

“; Kautzsch, ed., Gesenius’ Hebrew Grammar, § 120g.

85 Cf. Holladay, Lexicon, s.v. “

”; ibid., s.v. “

”; J. Payne Smith, Dictionary, s.v.

116

significant. Verse 13, as already noted, is a textually troubled verse. Variants 22 and 24 reflect the difficulties found in MT and presumably in the Vorlage of P. In variant 24, the Hebrew , a direct object (“them, themselves”) if taken as is, is rendered by the Syriac , a subject (“they”). Variant 22 is a change in the meaning of the verb, from “cursing” in Hebrew to “abusing” in Syriac. The Syriac in this clause seems to imply more than just speaking ill of someone and is a reference to the description of Eli’s sons in 2:1217 and their mistreatment of the people. P also adds , “the people,” and the resulting clause is, “because he knew that his sons (they) were abusing the people.” Since both these differences from MT are attempts to make sense of a confused text, they should not be considered significant. In variant 30, (“to be afraid”) stands for (“to conceal”). Later in the same verse (3:17) is translated by its equivalent . The usual Hebrew verb for “to fear,” in its appropriate form, is , but this word bears little graphic similarity to . The Hebrew word that most closely resembles and means something like “to fear” is , “to dread, be in awe of,” and is capable of carrying this meaning. It is unlikely that appeared in the Vorlage of P, since all eight occurrences of in 1 Samuel 1-12 (including 3:15) are translated by . There is a graphic similarity between and in both their Estrangela and Sert@a (t˙Bt and åÍKt) forms, but the similarity is not as close as it is in Hebrew. Furthermore, there are no variants in mss of P that preserve a reading of here, despite its occurrence just a few words later. Therefore, though some doubt must remain as to whether the graphic confusion occurred in Hebrew or in Syriac, the chances are good that it happened during the transmission of the Hebrew text, so the variant is significant. The last lexical variants to be considered are variants 34 and 35, and there is evidence of graphic confusion here as well. For the of MT, P has , presumably reflecting a in the Vorlage. The normal equivalent of in Syriac is . There is some graphic similarity in these forms in both Hebrew and Syriac (Estrangela, not Sert@a script). In either case, two letters would have to have been omitted or altered, but it is easier to suppose that a could have disappeared and a been added than that a and a could both vanish, especially since is the smallest Syriac letter and one of the largest Hebrew letters (and the upper stroke of the is often exaggerated in the mss).86 The Syriac is a valid translation of the Hebrew conjunction at times, but the translators did “

”; ibid., s.v. “

.”

86 One possible scenario in which the transformation could have occurred is that the could have been smudged and lost most of its top half, thus resembling a . A later scribe, seeing the nonsense word (perhaps without the diacritical mark that distinguishes from ), could have conjectured that the first two letters should be interchanged, resulting in . Alternatively, the could have been lost altogether, and a later scribe could have added the initial by conjecture.

117

not avail themselves of other opportunities to translate it in this way (e.g., 3:2, 19; cf. 3:14). However, it is probable that an original was modified to a after the corruption of the verb in order to smooth out the sentence. Since it is probable that the confusion in both variants appeared in Syriac, they are not significant. The first grammatical variants to be considered deal with the representation in P of participles in MT, variants 1, 4, 9, 13, and 23. The translators of P do not at first seem as systematic in their renderings of participles as were the translators of LXX, who consistently use a verb when the Hebrew consonants and context permit that interpretation. On the contrary, they sometimes use a verb and sometimes a participle (cf. table 20). It is important to note, though, that these variants represent all the Hebrew participles that deal with past time; those that deal with the present (i.e., those in 3:11-14) are rendered by Syriac participles, which are the equivalent of the present tense in these cases. Variants 1 and 23 have the Syriac continuous past tense, a periphrastic construction, rather than a simple tense, but since the piel participles in MT clearly refer to past time, the two versions are equivalent. It seems, then, that the Syriac translators were more concerned with idiomatic Syriac renderings that with conformity to the forms of Hebrew verbs. Moreover, the translators might have had traditions that considered some or all of the participles represented by variants 4, 9, 13, and 23 as verbs. As a result of this analysis, it must be concluded that none of these variants is significant. Somewhat similar to the analysis of these participles is the analysis of variants 7 and 21, which concern verbs in MT paralleled by verbs in the continuous past tense in P. In variant 21, MT has a perfect, which P usually renders with a Syriac perfect; only here do the translators use the continuous past. However, it is possible that the translators wanted to emphasize that Eli’s knowledge of his sons’ misdeeds was not limited to isolated incidents but that he was aware of their ongoing sin, or it may be that the continuous past in the next verb influenced this one as well. In either case, the continuous past is appropriate. In variant 7, the continuous past in P reflects an imperfect in MT. The use of the imperfect in Hebrew to indicate a continuous condition has been discussed above (pp. 91-92), so it may suffice to say here that the continuous past is an excellent translation of the Hebrew. Neither of these variants, then, is significant. Variant 14 has an imperative verb in Syriac where MT has a waw consecutive and a perfect. The meaning of the two versions is identical, and P also omits the waw and a previous . There are too few waw consecutive perfects in the chapter to determine any translation pattern in P, but it is probable that the shift from perfect to imperative was occasioned by the omission of the , since the sentence structure had then changed. This omission will be discussed below, but it may be concluded now that variant 14 is not significant. The remaining variants that deal with verbal forms are all concerned with the

118

rendering of Hebrew infinitives. Variants 18 and 19 have the nouns and for the infinitives absolute and . Syriac lacks a distinct infinitive absolute, and its infinitive does not have the same range of use as the Hebrew infinitive absolute. Because of this, the rendering of the these words by nouns is reasonable, since infinitives are verbal nouns. Therefore, these variants are not significant. The other variant to deal with a Hebrew infinitive is variant 28, which, instead of the complementary infinitive of MT (“to declare”), has a relative particle followed by a finite verb (“that he should declare”). Though Syriac can use the infinitive to complete the meaning of the main verb, the construction with the relative particle is equivalent and common.87 It is unlikely that any construction other than that found in MT lay behind the text of P at this point, so this variant is not significant. On several occasions P contains a plural noun or adjective where MT has a singular, namely, variants 3, 5, 25, 26, 31, 33, and 37. Table 11 indicates that eight out of fifty-seven (14.0%) singular nouns or adjectives are rendered by plurals in Syriac.88 This fact, and the deviation factor of 2.03, indicates the definite tendency to render singulars with singulars but also shows a degree of flexibility. As with similar variants in LXX, each variant must be considered in the light of its context and evidence of textual disturbance in the vicinity. Variant 3 is a general temporal phrase describing the time when the call of Samuel occurred, that is, while he was serving the Lord before Eli in Shiloh. The Syriac plural here has no different meaning than the Hebrew singular. It is probable that the translators were influenced by the same expression in 3:1 (plural in both MT and P), so this variant is not significant. Variants 5 and 33 probably render the plural found in the qere of BHS and many mss, and in the kethib of many mss as well. However, as noted above (p. 94), the difference between and in these verses is purely orthographic, so the variants are not significant. Variants 25 and 26 render two Hebrew words for types of sacrifices by plurals, whereas MT has singular in each case. Though the forms found in MT are singular, they clearly contain the idea of multiple sacrifices, and this factor might have influenced the translators, who were interested in stressing the magnitude of the sins of Eli’s sons (i.e., not only could a single sacrifice not expunge their guilt, neither could a multitude of sacrifices). On the other hand, LXX also has a plural, parallel with the second word here, which was not considered significant, and it is possible that reconsideration is in order. However, the translators of LXX also occasionally render Hebrew plurals by singulars, and the influence by LXX on P (either textual or, perhaps more likely, sharing a common 87 J. Payne Smith, Dictionary, s.v. “ ”; cf. T. H. Robinson, Grammar, 16. 88 The idiomatic Syriac rendering of

as a variant, though its data appears in the table.

by

in 3:10 is not counted

119

exegetical tradition89 ) is a possibility. Thus, these variants should probably not be considered significant. In variant 31, corresponds to of MT. It is clear that in MT refers many words, whether the expression be taken to mean “of every word” or “of all the message” (i.e., as a collective noun), so the rendering in P is equivalent. Since no indication of textual difficulties is present in the verse in P, and since no apparent reason for omitting - appears in MT, this variant should not be considered significant. The case is otherwise with the similar variant in variant 37. In order to understand this variant, it is necessary to anticipate the last three variants in quantitative representation, namely, variants 122, 123, and 124. Variant 122 deals with the omission of a phrase found in MT by parablepsis, skipping from to . Variants 123 and 124 concern the presence of the pronominal suffix - in P where appears in MT. It is probable that the pronominal suffix renders Hebrew characters such as or 90 (the significance of these variants is discussed below). The plural indicated by the form of the pronominal suffix (and sey m ) implies an additional in the Vorlage, and such an addition could easily have arisen in Hebrew, either by dittography or by the use of a two-character abbreviation for . On the other hand, the singular in MT can be explained by haplography. The main point to make, however, is that the variation is probably due to a Hebrew rather than the Syriac scribe, so variant 37 is significant. Only one case of a Syriac singular corresponding to a plural in MT is found in 1 Samuel 3: variant 27, which refers to the door(s) leading into the sanctuary. It is possible that the second in was either added or omitted in a Hebrew ms by dittography or haplography. On the other hand, it may be that the translators preferred to read “one door” to correspond to the single curtain leading into the sanctuary of the tabernacle (Ex 26:36), as contrasted with Solomon’s temple (1 Kings 6:31-34). However, the translators apparently have no problem with the much more troubling issue of Samuel sleeping in the sanctuary (cf. the reading of T), so it is hard to imagine great concern over the number of doors. Since graphic confusion, if it occurred, could have happened in Hebrew as easily as in Syriac, this variant should be considered significant. Variant 6 has an active participle in P corresponding to an adjective in MT. The possibility that should be pointed as an infinitive rather than an adjective has already been discussed above (p. 91), but the difference between infinitive and participle remains. 89 Cf. Koster, “Which Came First?,” 123, who discusses the common exegetical traditions of the translators of P and the Palestinian targums to the Pentateuch. Cf. also the discussion in Johann Cook, “Text and Tradition: A Methodological Problem,” Journal of Northwest Semitic Languages 9 (1981), 3-11. 90 Cf. IDBS, s.v. “Abbreviations, Hebrew Texts,” by Michael Fishbane; Godfrey Rolles Driver,

“Abbreviations in the Masoretic Text,” Textus 1 (1960): 112-31; idem, “Once again Abbreviations,” Textus 4 (1964): 76-94.

120

The reason for using a participle in P lies in the previous word, . Whereas cannot be followed by a participle to complete the verbal idea, can.91 Therefore, the participle in P does not imply the presence of a participle in its Vorlage, so the variant is not significant. The last variant in the category of consistency is variant 2, in which P has the preposition where MT has the sign of the definite direct object . The present instance is the only time in the chapter in which is rendered by . is used before only one other time in the chapter: in verse 7, where P uses —a common equivalent—to render . De Boer notes three other places in 1 Samuel 1-16 in which P renders with . This figure in itself is hardly overwhelming, but is also used at times to render prepositions such as , , , and , and it is found in other 92 constructions as well. An important parallel to this usage is found in T, which frequently uses , or a circumlocution involving , to avoid having be the direct object of a verb. It is likely that P here reflects the influence of a Jewish tradition shared by T, though it is clear that P does not utilize this tradition to the same extent as does T. The variant, then, is not significant. Having completed the variants dealing with consistency, the tables relating to this factor need to be adjusted by eliminating all data in them related to significant variants. Only five of the thirty-seven variants have been identified as significant, indicating that the Vorlage of P lies closer to the text of MT than does that of LXX. Only two tables, 17 and 27, are changed by eliminating the significant variants from the data; since Syriac is a Semitic language closely related to Hebrew, no other changes to the tables need to be made to reflect constraints on the translators. The changes resulting from the elimination of significant variants are as follows. Table Changes 17 Syr words: 32 1.10 Syr words/Heb word deviation factor: 0.22 total Heb (>1x): 128 primary Syr: 122 95.3% by primary rendering Syr roots: 31 Syr roots/Heb root: 1.15 27 deviation factor: 1.58 The changes to the tables representing the Peshitta translators’ consistency do not consequentially alter the picture of the translation technique given above. A reevaluation of the data after the significant variants are removed still shows the translators very consistent in making lexical choices, and, though the level of their consistency may be said to have increased somewhat, they remain fairly consistent in rendering the number of nouns and adjectives. 91 BDB, s.v. “III

”; J. Payne Smith, Dictionary, s.v. “

92 De Boer, 1 Samuel i-xvi, 24.

.”

121 Variants in the Representation of Hebrew Lexemes by Syriac Lexemes (Segmentation) Though the percentage of Hebrew compounds rendered exactly in P (76.6%) is higher than that in LXX (62.8%), it is still not very high, and it remains doubtful whether any deviation in this category is significant. A closer examination of the ten variations in segmentation reveals that six of them (variants 39, 40, 42, 43, 45, and 47) are cases of the failure of the Syriac translators to render the prepositions or . Instead, the translators preferred to use a pronominal suffix to attach the object directly to the verb. Though is so rendered only once in the chapter, the one time is the only occasion in which the pronominal object of the preposition could be understood as a direct (rather than indirect) object (following ). The five instances of omitting by joining the pronoun to the verb are examples of the frequency of this kind of translation in P. Therefore, none of these variants is significant. The other variants in this category (variants 38, 41, 44, and 46) may also be easily dismissed. Though is technically a combination of the preposition and the plural construct noun , it is used as the equivalent of a preposition throughout the OT, and P’s translation is entirely appropriate. The use of two words to render is an idiomatic rendering of the Hebrew.93 Similarly, was the translators’ attempt to render the Hebrew phrase . Finally, the rendering of by has already been discussed and dismissed when dealing with variant 28. One may conclude, then, that none of the variants in the category of segmentation is significant. Variants in Word Order Just as the low percentage of agreement concerning compound words throws serious doubt from the beginning on the significance of any of the variants in that category, so the high percentage of agreement in word order leads one to suspect that all of the deviations in P are significant, unless other factors override the translators’ tendency to follow word order. In the case of variant 48, MT reads , whereas P reads . Though it is of course possible that a transposition of adjacent words occurred in Syriac, the similarity of and (a similarity that does not exist between and ) suggests that it is more likely that was accidently omitted in Hebrew, then reinserted in the wrong place later. Variant 48, then, should be considered significant. Though no graphic similarity occurs in variant 49, it is likely that the transposition of the adverb meaning “again” occurred in Hebrew rather than in Syriac, in light of the translators’ propensity of following the Hebrew word order. In fact, the absence of any 93 On the use of cardinal numbers with

128.

in place of ordinals, see T. H. Robinson, Grammar,

122

such adverb in LXX suggests that both and may be later insertions. At any rate, variant 49 is probably significant. The final variant that deals with word order is variant 50. In Hebrew, the subject of the clause, if expressed, usually follows immediately (but cf. 2 Sam 14:7), especially if that subject is a pronoun. It is questionable whether Hebrew idiom allowed a construction such as , which seems to be implied by P.94 Furthermore, the usual word order for the Syriac present tense is participle + personal pronoun, as in this verse (cf. 1 Sam 12:3 P).95 Therefore, it is doubtful that this variant is significant, since it probably arose in Syriac rather than in Hebrew. When variants 48 and 49 are omitted from consideration, the number of Hebrew semantic units becomes 393, the number of variations becomes 1, and Hebrew word order is followed 99.7% of the time in P. Variants in Quantitative Representation The 80.0% agreement between semantic units in P and MT is almost identical to the agreement in the raw data for LXX, and it indicates initially that the translators of P were not overly concerned with matching their Hebrew Vorlage word for word in the translation, at least in certain contexts. The fact that P has a longer text fifty-five times as compared with thirty times for MT (almost a two to one ratio) suggests that the translators were more inclined to add to the text than to subtract from it. It remains to be examined if certain Hebrew constructions in particular lent themselves to modifications toward more idiomatic Syriac style, or if the translators’ deviations from their Vorlage were relatively haphazard. The use of anticipatory pronominal suffixes, either attached to verbs before a direct object or attached to substantives before the relative particle, is common in original Syriac works, and though it is not quite as prevalent in translated Syriac such as Samuel, it remains an important aspect of the translators’ style.96 Sixteen of the quantitative variants are related to this stylistic phenomenon: variants 53, 54, 58, 59, 60, 61, 62, 63, 69, 70, 86, 101, 102, 114, 118, 119. Since this construction has no parallel in Hebrew, these variants are not significant. Somewhat related to this construction is the Syriac preference for an emphatic noun with the relative particle in place of the Hebrew construct case. Variants 98 and 100 fall into this category and should not be considered significant. Six other instances of the use of the relative particle in P in different constructions are present in the chapter: variants 55, 75, 85, 96, 104, and 117. The presence of so 94 Cf. Muraoka, Emphatic Words and Structures, 137-40, who does not include such a construction

in his list of possible uses of

in a sentence.

95 T. H. Robinson, Grammar, 60. 96 Brockelmann, Grammatik, 106; T. H. Robinson, Grammar, 82.

123

many extra relative particles leads one to suspect that they were added as stylistic devices rather than as indicative of a different Vorlage. Variant 75 represents an idiomatic substitution in Syriac for the Hebrew phrase. Variants 85 and 104 substitute the relative plus a finite verb for a verbal form in Hebrew. Variant 117 is forced to add the relative after inserting before . The other two variants simply appear to be additions that seemed stylistically preferable to the translators. Thus, none of these variants presupposes a Vorlage different from MT. The next group of variants to be considered is the periphrastic constructions in P that are not periphrastic in MT. Variants corresponding to variants 51, 57, 90, and 91 have already been considered above under consistency and found to be nonsignificant. Thus, these variants should be considered nonsignificant from a quantitative standpoint as well. Six conjunctions are found in one or the other of MT and P but not in the other (variants 56, 65, 81, 87, 94, 97). Three of these (56, 87, and 97) are present in P but not MT, and the other three (65, 81, 94) are in MT but not P. The translators thus appear to have no single tendency to either add or delete conjunctions. De Boer contends that “the connecting particle is very often not translated,” and he then lists ten examples in 1 Samuel 1-16, as well as seven cases in which the Syriac is added.97 However, these few instances out of about eight hundred occurrences of the conjunction can hardly be called “very often,” and each case must be examined on its own merits.98 Because variant 81 is so closely associated with variant 79, a discussion of its significance is reserved for later. Variant 56 adds a conjunction where MT has , and it is supported in this addition by several Hebrew mss, LXX, and T. The presence of a conjunction at the beginning of a clause is of course standard Hebrew style, so this variant must be considered significant, since a good probability exists that the difference arose in the transmission of the Hebrew text. Similarly, variant 94 lacks a conjunction where MT has . In this case, no Hebrew mss support the omission of the , but V does. However, the support of V here is probably worthless, since the data indicates that V often does not translate (thirteen times), and it also occasionally adds a conjunction (four times). KB3 says that occurs 188 times in the OT, and it lists several examples.99 However, the Masora of BHS indicates that in only five of these instances is it preceded by . Therefore, scribes might have had the tendency to add a were it missing in their exemplar; it is less likely that they 97 De Boer, I Samuel i-xvi, 26 (italics mine). 98 The figure eight hundred assumes about fifty occurrences of

per chapter, as in chapter 3. However, this figure may be somewhat low, since most of chapters 1-16 are longer than chapter 3, so the total number may approach one thousand. 99 KB3 , s.v. “

.”

124

would have omitted the . Lacking other data, it seems best to count this variant as significant, since it could have arisen in Hebrew as easily as in Syriac. The additional in variant 87 is the result of the association of the expression “in that day” with what precedes (v. 11) rather than what follows (v. 12). Either one of the texts of MT or P could have given rise to the variant through graphic confusion. In MT, a between could have dropped out when the eye of the scribe skipped from - to - . In P, the extra could have arisen from an original through dittography. On the whole, the latter seems the more likely possibility, so variant 87 is probably not significant. The other two variants dealing with conjunctions can be explained as the translators’ attempt at an idiomatic rendering. In variant 65, lacks a before the second word because the expression is an idiom with (similar to the English “and he went to sleep”). Furthermore, the lack of a conjunction between the preceding may have influenced the translators. Finally, though Syriac has an equivalent for used as a negative in oaths (i.e., ), the translators in variant 97 have chosen to render it instead with the more obvious negative , adding the conjunction to make the rest of the sentence an indirect statement; Hebrew idiom prefers that the oath formula be a direct statement. Therefore, none of these variants is significant. The Hebrew preposition is rendered by in seven cases and is omitted in translation in the other seven cases. Thus, the omission of a rendering for is a normal equivalent and provides no evidence for the omission of the preposition in the Vorlage of P. Consequentially, variants 64, 68, 78, 89, 92, 112, and 120 must be considered nonsignificant. P also has an extra preposition in four places: variants 66 and 67 and variants 72 and 73. These four variants are actually two occurrences of the same pairs of words, neither of which has a preposition in MT. The first word in each pair is , an infinitive preceded by a preposition. Hebrew infinitives are often preceded by the preposition , and Syriac infinitives likewise are frequently preceded by . In fact, the use of the preposition with the infinitive is probably more prevalent in Syriac. In chapter 3, only two of the five infinitives construct in MT are preceded by , whereas all six infinitives in P are preceded by . Thus, it is probable that the presence of the in P is a stylistic convention rather than evidence of a varying Vorlage. The second word in each pair is , with the initial acting as an indicator of the direct object, similar to in Hebrew. The question of whether the equivalents in MT are to be taken as vocatives or accusatives has been addressed above, but for the translators of P, the words were clearly accusatives, and since the is the normal sign of the accusative in Syriac, no Hebrew Vorlage different from MT can be postulated on the basis of this evidence. None of these variants, then, is significant.

125

In variant 80, the preposition in MT is not represented in P, which reads . The pronominal suffix is attached to the verb as a direct object, much as it is in some of the cases in which MT reads (3:5, 6, 8, 13, 18). It is true that is rendered by some Syriac preposition in every other instance in the chapter, and it is possible that MT read here, which P often omits, the having arisen in the text through dittography. However, the tendency of the translators to attach the direct object directly to the verb is evident, and the number of cases of in the chapter is too few to outweigh the importance of this stylistic tendency. Therefore, lacking further data, this variant should probably be considered nonsignificant. Variants 88, 99, 105, and 113 all have P omit the sign of the definite direct object, . Though Syriac can employ to render , 100 in the present chapter the translators apparently declined to render it as often as they rendered it with . Since the omission of any equivalent is one of the two main renderings of , these variants cannot be considered significant. Variant 103 involves the lack of an equivalent for the preposition before an infinitive. P here does not have an infinitive but rather the relative particle and an imperfect (see above on the discussion of variant 102). Because of this different sentence structure, a preposition could play no grammatical role in P, so its omission is not significant. Several variants in quantitative representation are the result of the attempt of the translators to produce an idiomatic rendering of a Hebrew expression. Variants 74 and 76, the failure to render the preposition in MT and the addition of the word , are closely associated with variant 75, discussed above. As already noted, the translators here used an idiomatic Syriac phrase to render the Hebrew phrase, and the idiom required both the omission of and the addition of , so this variant is not significant. Variants 82 and 83 deal with the rendering of by . This phrase is not exactly equivalent, since the Hebrew means “as frequently beforehand,” while the Syriac means “two times.” Still, the reading of P does seem to reflect the same words as MT, probably influenced by the following . Variant 95 reflects the rendering of the single word (“therefore”) in MT by the two words (“because of this”). Since these phrases are equivalent, it is unlikely that the Vorlage of P had anything different from what is in MT. The addition of the pronoun , functioning as a copula, in variant 107 likewise does not reflect a different Vorlage, since is consistently rendered by the verb rather than the pronominal copula, though both carry the same meaning. The next variant related to idiom is the omission of in variant 109. Though the translators could have inserted some word such as the of T, which is really not 100 Even

Dictionary, s.v. “

can be used, as in Aramaic (cf. Gen 1:1). J. Payne Smith calls this use “archaic”; .”

126

equivalent in meaning, they chose instead to omit the word, without losing anything in the translation. Finally, variant 116 adds , so that the Hebrew “the good” becomes “all that is good,” a phrase that apparently sounded better to the translators. In conclusion, none of these variants that concern idiomatic expressions is significant. Several explanatory additions appear among the quantitative variants, namely, variants 52, 71, 77, 84, 106, 108, 110, 111, 115, and 121. The fact that in every case the excess text appears in P leads one to suspect that the additions occurred either at the point of translation or during the later transmission of the Syriac text. This state of affairs is quite different from the case of LXX, where MT and LXX each had eight semantic units that were classified as explanatory. Some of the variants appear in other traditions, and others are unique to P, but the one-sidedness of these explanatory elements suggests that those variants that are shared with other traditions either appeared independently or arose as a result of contamination from other traditions. Variants 52 and 115 have readings in common with LXX. In variant 52, P adds the description after the mention of Eli. Variant 115 adds the subject to clarify that it was he and not Samuel who was speaking. Scholars have often noted correspondences between P and LXX and have assumed that the translators of P borrowed readings from LXX. In the light of this examination of P, a modification to this assumption seems in order. The translators of P show no tendency to consult other versions in any category other than quantitative representation, so it is probable that they did not do so here, either. The addition of may well be an independent expansion, but the addition of might depend on LXX. If so, the point of borrowing was probably not at the point of translation but rather somewhere in the process of transmission.101 However, the fact that 2:11b P is identical to 3:1 P suggests that LXX may not have been involved at all. Variants 71 and 121 both have an additional after , perhaps to clarify the meaning of the verb. These readings are based on the same construction as in 3:6. The fact that the reading of variant 71 is equivalent to that found in V is probably not important; the additional adhuc of V is most likely an independent phenomenon. Variant 84 adds after , reflecting the instructions Eli gave to Samuel in the previous verse. It has been observed that several Greek mss contain this addition, but 101 If this conclusion is true, and data from more chapters would have to be examined before one could conclude that it is, it might be relevant for the question of the socio-religious origins of P. If the translators did not use LXX but did use traditions found in T, as suggested above in a few places, it is likely that Jews rather than Christians or Jewish Christians were the translators of P. The occasional influence of LXX in the process of transmission would be natural after the transmission of the text were taken over by Christian scribes. This scenario would also seem to suggest that there was no distinct Old Syriac translation at least of Samuel, and perhaps of most or all of the OT. Rabbula’s work, then, would be a revision and standardization of P rather than the creation of a new translation. This hypothesis, however, obviously requires further substantiation.

127

its presence in P is probably based on a desire for internal consistency rather than an attempt at conformity to Greek mss. In variants 77, 108, and 111, explicit subjects have been added in order to clarify the sentences. is the subject in variant 77, and is the subject in the other two variants. The problem of a missing subject in verse 17 has been discussed in another context already (see pp. 89-90), and the translators of P insert the name twice in the verse, both times in Eli’s question, and both times in the phrase . The last two explicatory variants are both prepositional phrases in verse 17: variant 106 adds , and variant 110 adds . The first variant clarifies who was being addressed, and the second completes the elliptical construction . None of these explicatory variants can be considered significant. Variant 79 concerns the failure to render in P, and it is associated with the omission of in variant 81. In variant 81, P substitutes the imperative for the waw consecutive plus perfect , which is equivalent to an imperative in the context (cf. V). There seems to be no explanation for the omission of in Hebrew or Syriac apart from simple haplography. It is true that the omission of the word does not change the meaning of the text, but the translators rendered the similar expression in 3:2, so one would expect them to render here as well. On the other hand, may be an addition in MT, perhaps based on the missing subject about whom Eli is talking, and maybe even influenced by the question of the presence of later in the verse and in the following verse. Though these last possibilities are highly suspect in light of the data, the omission should probably be considered significant, since it could have occurred as easily in Hebrew as in Syriac, and perhaps more easily in Hebrew. If it is significant, then the omission of the in variant 81 must also be considered significant. Variant 93 concerns the addition of in P in verse 13. The troubled nature of this verse resulting from the reading has been discussed above. Whereas the translators of LXX rendered the original , the translators of P attempted to make the best of the text they had, which apparently read like MT. Instead of interpreting this word as the reciprocal object of the verb, they took it to be the subject (see above, p. 116). Since the verb required an object, the translators supplied , based on the narrative in the previous chapter (2:12-17). This variant, then, is not significant. The last three Syriac variants—122, 123, and 124—are related to one another and so must be considered together. The omission of six words found in MT is the result of parablepsis, the scribe’s eye skipping from the first to the second reference to Shiloh. The fact that “Shiloh” is spelled differently in the two places in MT ( and ) might suggest that the parablepsis occurred in P rather than in MT, since both occurrences are spelled the same in P and thus are more likely candidates for this type of error. As noted

128

above in the discussion of variant 37, the pronominal suffix on is probably the result of confusion in the Hebrew tradition between (perhaps abbreviated as ) and . It is unclear from which option the variant arose, but it is more likely to have arisen in Hebrew than in Syriac. Therefore, variants 123 and 124 are probably significant, but variant 122 is not. Only six of the seventy-four quantitative variants in P are classified as significant variants, so the percentage of agreement between P and the presumed Hebrew Vorlage only rises to 81.1%. This figure is significantly lower than that of LXX, and it indicates a relative lack of concern on the part of the translators to render every Hebrew semantic unit without embellishment, at least as compared with the translators of LXX. Even if anticipatory pronouns and associated relative particles are omitted from consideration as so characteristic of Syriac style as to be indispensable in the minds of the translators, the percentage of agreement still only rises to 84.4%. It is clear, then, that the data so far analyzed indicates that the translators of P were somewhat stricter in their renderings of lexical units, certain grammatical categories, and compound words than were the translators of LXX. However, they were less concerned with consistency in regard to quantitative representation. Another striking element of the data so far considered is the far fewer significant variants in P than in LXX, indicating that the Vorlage of P stood closer to MT than did that of LXX. Targum Limitations of Aramaic for Rendering Hebrew Since the Aramaic of Targum Jonathan is simply a dialect of Late Aramaic, like Syriac, little more needs to be said about the differences between Hebrew and Aramaic. One slight terminological difference is that the names of the verbal stems in the western branch of Late Aramaic differ from those of the eastern branch (Syriac) in the initial vowel of the passive stems (i.e., ith- for eth- in all three cases).102 In addition, Aramaic does not employ the anticipatory pronoun so often found in Syriac. Otherwise, the discussion of the limitations of Syriac for rendering Hebrew may be applied to the Aramaic of Targum Jonathan as well.103 102 Gustaf Dalman, Grammatik des jüdisch-palästinischen Aramäisch, 2d ed. (Leipzig: J. C. Hinrichs, 1905), 250. 103 Alejandro Díez Macho says that the Aramaic of Targum Jonathan is fundamentally the same as

that of Targum Onkelos. He quotes the article by M. Z. Kaddari, who describes the Aramaic of Onkelos as follows: “[El arameo] de Onqelos no es lenguaje puramente de traducción, sino lengua independiente, como se desprende de sus formas de ‘determinación,’ de la manera de expresar el genitivo y el complemento directo: se parece, está próximo, al arameo imperial . . ., pero presenta signos de transición al arameo ‘medio’ . . . .” However, Díez Macho notes that the system of matres lectionis shows that the vocalization of Onkelos (and Jonathan) derives from the supralinear system of Babylonia, even in those mss with

129 Partial Translation Technique Add-Oms As is the case with P, the Aramaic translators of T generally render the sign of the definite direct object , and they always render the prepositions and . However, T often substitutes the relative particle followed by a noun in the emphatic state for the Hebrew construct state, as is also frequently done in P. Thus, the addition of the relative in such cases will not be considered an add-om. See Appendix 2 for a list of add-oms in T. Consistency

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Ref 3:1

Variation

3:2 3:3

1° 3°

3:7 3:8 3:10 3:11 3:12

1° 2°

Targum preposition lexeme lexeme participle lexeme lexeme plural infinitive lexeme lexeme lexeme infinitive preposition ithpeel lexeme ithpeel definite verb lexeme lexeme lexeme verb verb

Masoretic Text DDO

adjective

singular adjective

verb DDO qal qal not definite participle

infinitive infinitive

Tiberian vocalization; Alejandro Díez Macho, El Targum: Introducción a las traducciones aramaicas de la biblia, Textos y estudios “Cardenal Cisneros,” no. 21 (Madrid: Consejo Superior de Investigaciones Cientificas, 1982), 72-73. Cf. also p. 93, where he says that Jonathan is a Babylonian revision of an earlier Palestinian targum, and R. Le Déaut, Introduction à la littérature targumique, part 1 (Rome: Pontifical Biblical Institute, 1966), 124-27.

130

24 3:13 perfect perfect w/c 25 preposition DDO 26 plural singular 27 3:14 plural singular 28 plural singular 29 plural singular 30 3:16 preposition DDO 31 3:17 1° lexeme 32 lexeme 33 2° lexeme 34 3:18 lexeme 35 rel clause noun 36 lexeme 37 3:19 prep phrase preposition 38 lexeme 39 3:20 lexeme 40 lexeme 41 rel particle preposition Table 34.—Verbs, Nouns, and Adjectives: Lexemes Heb words (>1x): 31 Aram words: 34 1.10 Aram words/Heb word deviation factor: 0.04 total Heb (>1x): 139 primary Aram: 136 97.8% of Heb words by primary rendering Heb roots (>1x): 29 Aram roots: 34 1.17 Aram roots/Heb root Table 35.—Adverbs, Prepositions, and Particles: Lexemes Heb advs (>1x): 12 Aram advs: 22 1.83 Aram advs/Heb adv total Heb (>1x): 85 primary Aram: 71 83.5% of Heb advs by primary rendering deviation factor: 2.22 Table 36.—Conjunctions: Lexemes Heb conjs (>1x): 2 Aram conjs: 2 1.00 Gk conjs/Heb conj total Heb (>1x): 68 primary Aram: 68 100.0% of Heb conjs by primary rendering deviation factor: 0.00 Table 37.—Word Classes percentage of verbs/verbals represented by verbs/verbals: 98/98 = 100.0% percentage of verbs represented by verbs: 76/77 = 98.7% percentage of participles represented by participles: 11/12 = 91.7% percentage of infinitives represented by infinitives: 5/7 = 71.4% percentage of nouns/adjs represented by nouns/adjs: 65/68 = 95.6% percentage of pronouns represented by pronouns: 42/42 = 100.0%

131

Table 38.—Verbs: Inflection deviation factor: 0.29 deviation factor (without verbals): 0.28 Table 39.—Verbs: Stem deviation factor: 1.02 Table 40.—Verbs: Person deviation factor: 0.00 Table 41.—Verbs: Gender deviation factor: 0.04 Table 42.—Verbs: Number deviation factor: 0.02 Table 43.—Nouns and Adjectives: Gender deviation factor: 1.05 Table 44.—Nouns and Adjectives: Number deviation factor: 0.50 Table 45.—Nouns and Adjectives: Use deviation factor: 1.47 Table 46.—Nouns and Adjectives: Hebrew Articles, Aramaic Emphatic State deviation factor: 1.14 deviation factor (without implied articles): 1.21 Table 47.—Nouns and Adjectives: Definiteness deviation factor: 0.52 Table 48.—Pronouns: Gender deviation factor: 0.00 Table 49.—Pronouns: Number deviation factor: 0.00 Table 50.—Pronouns: Use deviation factor: 0.34 Representation of Hebrew Lexemes by Aramaic Lexemes (Segmentation) Ref Hebrew Compound Aramaic Rendering 42 3:18 43 3:20 Heb compounds: 50 Aram equivalents: 48 96.0% of Heb compounds rendered exactly Word Order There are no deviations from the word order of MT found in T, as the following summary indicates.

132

Heb semantic units: 416 variations: 0

Heb word order followed 100.0% of the time

Quantitative Representation Ref 3:1 3:2

Number of Variations 44 +1 45 +1 46 +1 47 3:3 +2 48 +1 49 +6 50 +1 51 3:7 +1 52 +3 53 +1 54 +1 55 3:8 +2 56 3:11 -1 57 +1 58 3:12 +1 59 3:13 +1 60 3:14 +1 61 3:15 +1 62 +1 63 +1 64 3:18 +1 65 3:19 +1 66 +1 67 +1 68 -2 69 +1 70 3:20 +2 71 3:21 +1 Heb semantic units: 428 Aram deviations: 39

Variant - 1° - 4° - 1° - 4°

-

-

-

90.9% agreement

Elimination of Variants Targum Jonathan is a mixture of literal renderings and midrashic elements and so is difficult to characterize by terms such as literal or non-literal.104 Le Déaut describes it as 104 See Díez Macho, Targum, 12-30.

133

more paraphrastic than Onkelos, but otherwise substantially the same in regard to language and method of translation. Above all, it is a uniform (i.e., consistent) translation.105 The preceding tables and lists of variants demonstrate the mixed nature of the translation. Though the total number of variants is about three-fourths of the total of LXX, the lexical consistency of T in representing verbs, nouns, and adjectives (table 34) is quite a bit greater than that of either LXX or P (cf. especially the deviation factors and the percentage of Hebrew words rendered by their primary renderings). The consistency in rendering Hebrew adverbs, prepositions, and particles by a single Aramaic equivalent (table 35) is comparable to the statistics of LXX and P, but T, like LXX, is absolutely consistent in rendering conjunctions (table 36). T is also more consistent than either LXX or P in rendering words of one class by words of the same class (table 37). The translators of T106 were generally more consistent in rendering syntactic structures than were the translators of LXX or P. Particularly noteworthy is the significantly greater degree of consistency in regard to rendering the number of nouns and adjectives (table 44), Hebrew articles and definiteness (tables 46 and 47), and the use of pronouns. In no category does either LXX or P demonstrate significantly greater consistency. The translators of T were absolutely consistent in rendering the person of verbs and the gender and number of pronouns. They were very consistent in rendering the inflection of verbs (including verbals), the gender and number of verbs, the number and definiteness of nouns and adjectives, and the use of pronouns. They were fairly consistent in every other category: verb stems, gender and use of nouns and adjectives, and Hebrew articles vs. emphatic state. No deviation factor exceeds 1.50. The greatest shift toward consistency in T as compared with LXX and P is in the area of segmentation, where 96.0% of the Hebrew compound words are rendered by comparable Aramaic equivalents, as compared with 62.8% and 76.6% in LXX and P, respectively. No deviations from the Hebrew word order are reflected in T, though both LXX and P also contain few deviations. Another area of significantly greater consistency is in quantitative representation, where T shows 90.9% agreement with MT, as compared with about 80% for the preliminary figures of both LXX and P. The initial impression, then, is that T is more literal in many regards than either LXX or P, but this impression needs to be tested and qualified by a closer examination of the data. Variants Related to Consistency A glance at the list of variants related to consistency reveals that most of these 105 La Déaut, Introduction, 126. 106 The term “translators” here and throughout the section is used for the sake of simplicity, but it refers to any translators, scribes, or editors who played a role in the development of the text from its original oral forms in the synagogues to its final written form.

134

variants are lexical, rather than syntactical, in nature, contrary to the situation in P or LXX. This fact is an indication of the desire of the translators to render certain Hebrew constructions with a single equivalent Aramaic construction. The lexical variants will be examined first. The first verse yields four lexical variants (variants 2, 3, 5, and 6), which may be considered together. These variants reflect the rendering of by , of by , and of by . Variant 2 substitutes a temporal phrase for an improper preposition emphasizing location or relationship. It is also possible that the translators avoided the equivalent preposition because it had already been used of the relationship between Samuel and Yahweh, and they wanted to emphasize that Samuel’s service to Yahweh was more important than his service to Eli. Variant 3 uses , “hidden,” for , “rare,” because corresponds with , “revealed,” in the next clause. In addition, the translators might have wanted to convey the idea that the word of Yahweh, and thus Yahweh himself, was not absent from Israel, but was only awaiting a worthy individual to receive the revelation.107 T generally avoids the idea that God communicates directly, either visibly or audibly, with mere mortals, and the substitution of , “prophecy,” for , “vision,” is designed to avoid the impression that Yahweh himself appeared to Samuel. Thus, none of these variants is significant. Similar to variant 6 is variant 15, which reads a form of instead of the of MT. Rather than saying that Yahweh came, and was thus potentially perceptible to Samuel, T prefers to say that the Yahweh revealed himself, presumably in a way that would not require immediate contact with the divine.108 The reasoning behind this lexical change is the same as in verse 1, so variant 15 is not significant. Variants 9, 11, and 32 all use the abbreviated proper name to render . This rendering might seem to suggest a Hebrew , but in fact T is consistent in rendering in the same way it renders , so the variants are not significant. The next several lexical variants deal with apparent differences in the choice of prepositions. Variants 19 and 20 use to render . These variants are interesting, because LXX in the first instance reads , and P joins T in reading in both places. As in the case of P, one might have expected the translators of T to use either or to render , as they do five and two times in the chapter, respectively. However, as with , the semantic range of is larger than that of , and it can carry the meaning “to, toward.”109 In lieu of more statistical data, and in light of the statistical summary at the 107 Cf. Levine, Aramaic Version, 74. 108 Several mss further remove Yahweh from Samuel by speaking of the “Glory of Yahweh” that

reveals itself. See above, pp. 58-59, and cf. Levine, Aramaic Version, 57-59. 109 Levy, Wörterbuch, s.v. “

.”

135

bottom of table 35,110 it seems likely at this stage of the investigation that the Vorlage of T read just as MT does. It is possible that the Vorlage read in one or both places, of course, but the nature of the Aramaic data does not allow such a claim to be put forth with any degree of confidence.111 Thus, these variants cannot be considered significant. The next two prepositions to be considered are variants 31 and 33, where is rendered twice by . As in English, it is permissible in Hebrew for a person to speak either to ( , ) or with ( ) someone, and little, if any, difference exists between the meanings of the prepositions in such cases.112 Sperber notes that T often uses to render both and when they could be translated “with,” and many of his examples involve one person speaking with (to) another.113 Variants 31 and 33, then, are not significant. Variant 36 involves the rendering of the expression ( ) by . The reason for avoiding the expression “in his eyes” is often taken to be the translators’ preference for substituting an expression that avoids an anthropomorphism, but some recent studies suggest that the rendering is simply a translational equivalent unrelated to the avoidance of anthropomorphism.114 In either case, this variant is not significant. Though not strictly a lexical variant, a similar concern spawned variant 37, which reads , “at his aid,” for MT’s . In this case, the translators wanted to avoid the idea that God was physically with Samuel, an idea further eschewed by the addition of before , so this variant is also nonsignificant. The last lexical variant to deal with prepositions is variant 39, with which variants 40 and 41 are associated. T reads that Samuel was faithful , where MT reads . The intent of this rendering seems to be to shift the emphasis of the verse from the faithfulness of Samuel to the trustworthiness of the word of Yahweh. This type of shift in meaning is related to all those renderings which tend to exalt God 110 Deviation factor 2.22, only 83.5% of Hebrew adverbs, prepositions, and particles rendered by the primary rendering. 111 Thus, the assertions by S. R. Driver (Notes on the Books of Samuel, 43) and McCarter

(I Samuel, 96) that P and T support the reading in the Hebrew must be questioned, since the evidence suggests that they could just as easily have read . Sperber, Bible in Aramaic, 4b:111, lists the readings represented by variant 19 as an example of the indiscriminate use of and in MT. 112 Cf. BDB, s.v. “

.”

113 Sperber, Bible in Aramaic, 4b:105-6. 114 Sperber, Bible in Aramaic, 4b:37, says that “the Targum avoids using Biblical expressions, which speak of God as if being possessed of a body just like a human being, with hands and eyes etc.” However, Michael L. Klein, in a study of the Pentateuchal targums, demonstrates that the phrases and are often rendered by the preposition , even when referring to people; Michael L. Klein, “The Preposition (‘Before’): A Pseudo-Anti-Anthropomorphism in the Targums,” Journal of Theological Studies 30 (1979): 505-7. Levine is certainly right in stating that targumic renderings often taken as anti-anthropomorphic should be understood in light of Jewish concerns for reverence in reference to God rather than Hellenistic conceptions of deity; Levine, Aramaic Version, 55.

136

rather than humans. Therefore, variants 39, 40, and 41 are not significant. The next lexical variant to be considered is variant 21, in which in T corresponds to in MT. Whereas the emphasis in MT is on the complete fulfillment of God’s word to the house of Eli, the emphasis in T seems to be that God will completely destroy his house. 115 Despite this different emphasis, however, the result is the same: Eli’s dynasty will be abruptly and completely ended, just as God has said. It is doubtful that any Hebrew word such as stood in the Vorlage of T, so this variant is not significant. In variants 34 and 35, stands for in MT. Though a rendering such as might be more literal, the expression “what he determines” is roughly equivalent in meaning in the context, and the translators might have felt that their rendering safeguarded the sovereignty of God in making decisions more explicitly than did the reading of the Hebrew text before them.116 Therefore, these variants cannot be considered significant. A similar concern for preserving God’s sovereignty might explain variant 38, where T reads (“was vain”) for (“he let fall”) in MT. The thrust of the clause in both MT and T is that all of Samuel’s prophetic words came to pass. However, to say that the Lord did not let any of Samuel’s words fail could imply that God was at Samuel’s bidding and was obliged to fulfill whatever he said. T corrects this possible misunderstanding by saying that none of Samuel’s words was vain. This manner of stating the issue suggests that Samuel was controlled by God rather than vice versa. The lexical substitution, then, is not significant. The final lexical variant to be considered is variant 10. Whereas MT in verse 3 says that Samuel was sleeping in the temple of the Lord, such an act would violate the regulations prohibiting anyone who was not a priest from entering the temple.117 In order to avoid the possibility that Samuel was guilty of breaking the law, T says that Samuel was sleeping “in the court of the Levites” outside the temple proper. Thus, this variant is not significant. The next several variants deal with the use of an Aramaic word of one class to render a Hebrew word of another class. Variants 1, 13, 25, and 30 all have T using a preposition where MT has the sign of the definite direct object, . The first two in this list involve the use of the preposition (in variant 13 ) in place of where is the direct object in MT. T often avoids constructions in which God is either the subject or direct object of a verb, and one of the most frequent means of changing the 115 Cf. the translation of Harrington and Saldarini, Targum Jonathan, 109, of the final phrase of the verse: “I will consume and destroy.” 116 Cf. Levine, Aramaic Version, 52-54. 117 Ibid., 120. Cf. also bQiddushin 78b.

137

structure of the sentence is by inserting or substituting the preposition . 118 Such a construction in T in no way implies any different construction in the Vorlage different from that found in MT, so these two variants are nonsignificant. Variant 25 has the phrase , in which in T stands in the place of in MT. The reason for this substitution is Aramaic idiom, because when the ithpeel of means “to 119 punish,” the recipients of the punishment are often preceded by . Thus, variant 25 is not significant. Variant 30 is different from the others in that no apparent reason for rendering by exists. It is true that the Aramaic preposition can act as an indicator of a direct object, as in Syriac and late Hebrew. However, since the passage is translation Aramaic rather than original Aramaic, the question is whether the rendering implied if MT is assumed to be the Vorlage of T is consistent with the translation technique found elsewhere in T. A survey of all the occurrences of in the former prophets indicates that in every other instance, T uses to render . Most of the cases in which follows involve one person naming another person or a place (e.g., Judg 1:17; 13:24), a double accusative construction. However, in all six cases in which means “to summon,” T renders with , so this variant should be considered significant.120 In several cases, T uses one verbal form while MT has another. The first such case is variant 12, where T has , “to know,” for , “he knew,” of MT. The reason for the difference is immediately apparent from a glance at the context, for T inserts the verb before , so a complementary infinitive was required. Thus, the variant is not significant. In variant 18, T has the verb, , in place of the participle . 121 The verb is preceded, however, by the relative particle , so the particle plus the verb is the translational equivalent of the participle in MT. It is true that all the other participles found in MT in the chapter are rendered by participles in T, but none of them is part of a construct chain as this participle in MT is. T does not show nearly the propensity of P for converting construct chains into chain of words connected by the relative particle, but including a participle as a substantive in a construct chain seems to have been considered poor style, or perhaps even unidiomatic, for T also renders the similar construction with in 118 M. L. Klein’s study of the Aramaic portion of Daniel shows that the use of

is a sign of reverence, not an avoidance of anthropomorphism, since the king is addressed in the same way as God. Similarly, the targums exhibit dozens of cases in which before a human direct object is rendered by . See M. L. Klein, “The Preposition ,” 502-7. 119 Levy, Wörterbuch, s.v. “

.”

120 The six locations in which T renders

“summon” with 1 Sam 22:11; 2 Sam 13:17; 1 Kings 1:9, 10; 12:20. Josh 21:9 has different construction in T. 121 Ms f alone has a participle.

are Josh 8:34; in MT, but has a

138

2 Kings 21:12 with a relative particle and verb.122 Thus, this variant does not appear to be significant. On two occasions, variants 22 and 23, T has a verb where MT has an infinitive absolute. As has already been noted in the discussion of Syriac variants 18 and 19 above, the Aramaic languages lack a distinct infinitive absolute like Hebrew has, and the Aramaic infinitive does not have the same range of meaning as the Hebrew infinitive absolute. Sperber lists many examples in which T renders infinitives absolute with verbs, including the present verse.123 Since it seems to have been characteristic of T to use a verb to render an infinitive absolute, variants 22 and 23 are not significant. In variant 4, T has the passive participle in place of the adjective in MT. As noted in the discussion of variant 3, the translators apparently chose the word to correspond to later in the same verse. Just as is a passive participle, so also was made a passive participle. This shift in form does not alter the meaning of the phrase, since many adjectives are formed from the passive participle.124 Therefore, variant 4 is not significant. The last variant that deals with different word classes is variant 8, which uses an infinitive where MT has an adjective. The problem with MT’s use of an adjective here has already been discussed, and it was pointed out that the same consonants found in MT could be pointed as an infinitive. Since significant variants are only those that presume a different consonantal text, this variant cannot be considered significant. The only variant dealing with the inflection of the verb is variant 24, where T has a perfect corresponding to a perfect with waw consecutive in MT. Though one would expect T to use an imperfect to render a perfect with waw consecutive, and though the other two instances of this construction in the chapter are so rendered, three examples are not enough to get an idea of the translators’ tendencies. However, the low deviation factor in table 38 suggests a tendency to render inflection consistently, and the possibility that the Vorlage might have been different from MT is supported by the fact that both LXX and V also use past tenses in the same place. Therefore, this variant should be considered significant. On two occasions, reflected by variants 14 and 16, T uses the passive ithpeel stem where MT has the active qal stem. In both cases, the subject of the verb in MT is , and T alters the construction in order to avoid having God the immediate subject of an action that might be perceived as bringing him into direct contact with humans. Thus, concern for reverence of God dictated the shift in verb stem, so the variants are not significant. 122 So also Jer 19:3. Cf. 2 Kings 25:19, where T renders a participial

with

plus a verb. 123 Sperber, Bible in Aramaic, 4b:91. 124 Cf. Dalman, Grammatik, 57.

in a construct chain

139

The next four variants all deal with the use of a plural noun in T where a singular noun appears in MT. Variant 7 reflects the qere reading of Hebrew ms L, as discussed above (p. 94), but since the variation between kethib and qere is merely orthographic, the variant is not significant. Variant 28, with which variant 27, a change in verbal number, must be considered, and variant 26 both employ plural forms of the noun , “sin.” 125 Sperber notes that T frequently renders singular words for sin by plurals, so the plural in variants 26 and 28 probably does not suggest a different Vorlage. Variant 27 then is a modification to the verbal number in order to accommodate the plural noun that is the subject of the verb. Thus, none of these variants is significant. Finally, variant 29 is concerned with the rendering of the singular noun by the plural noun . Table 44 indicates that the translators were generally concerned with a precise rendering of nominal number, but the previous several variants indicate that they could vary the number in certain circumstances. The translators might have been influenced by their rendering of the previous by , which contains a plural. The only other occurrence of the phrase that is translated in Targum Jonathan is in Isa 43:23, where it is also rendered by a plural, . It is likely, then, that the translators felt that a plural rendering was more appropriate in the present case, even if the Vorlage was singular, so variant 29 is probably not significant. 126 The last variant to be considered under the rubric of consistency is variant 17, where a definite in T corresponds to an indefinite in MT. Table 47 indicates that the translators were inclined to be fairly precise in rendering the definiteness of their Vorlage, and a closer examination of the exceptional cases indicates that in all the other cases which deviate from MT, a different construction in T explains the deviation. The renderings of verse 17 would seem to indicate the translators’ care in rendering definiteness: in this verse is rendered , but is rendered twice. However, this façade of precision does not continue throughout the book, at least as far as the word is concerned. Of the nine other renderings of , absolute and without the article, in the book, T has five times and four times. Thus, the translators were inconsistent in their rendering of the definiteness of , so variant 17 is not significant. Only three of the variants related to consistency are significant, so few changes are necessary in the statistical tables on the basis of the discovery of significant variants. Like Syriac, since Aramaic is a Semitic language, no other factors necessitate modifying the tables further. Certain stylistic and theological tendencies have already been noted that affect the literalness of the translation, and it is often difficult in T to separate stylistic from 125 Sperber, Bible in Aramaic, 4b:96-97. 126 Could the tendency to render words for sin as plurals also affect words like

, which are means of removing sin?

and

140

theological tendencies. Since these tendencies have not been investigated in any consistent manner, their effect on the translation technique will be left until the next chapter. The following changes to the summaries at the bottom of the tables should be noted. Table Changes 35 Aram advs: 21 1.75 Aram advs/Heb adv total Heb (>1x): 84 84.5% of Heb words by primary rendering deviation factor: 2.02 38 deviation factor: 0.27 deviation factor (without verbals): 0.26 These changes to the statistical tables are minor in nature, and they in no way change the general perception of the literalness of the translation. Variants in the Representation of Hebrew Lexemes by Aramaic Lexemes (Segmentation) The tendency of the Aramaic translators to render each component of compound Hebrew words is much higher than that found in the either LXX or P, 96.0% as compared with 62.8% and 76.6%. This higher number suggests that the translators felt it important to render compound words as precisely as possible, and deviations from this pattern deserve scrutiny. However, both cases of deviation in T have already been discussed under the category of consistency (variants 36 and 41), and neither was found to be significant. Variants in Word Order T is absolutely consistent in following the word order reflected by MT in chapter 3. Variants in Quantitative Representation As already noted above, the 90.9% agreement between T and MT in quantitative representation is substantially greater than the initial figures of the other versions so far reviewed, reflecting only about half as many deviations from MT as LXX and P have. A separate tabulation of the positive and negative numbers in the list reveals that T’s longer text amounts to thirty-six extra semantic units, while MT has only three extra semantic units. Most of the excess text in T can be accounted for in two ways. First, the translators’ frequent use of the relative particle accounts for an extra ten units in T. Second, T’s theological concern for preserving reverence for God accounts for most of the other excess text. It should be noted that the overlap between chapters 3 and 4 of this study is most clear in T, for no analysis of the style of T is possible without a consideration of the translators’ theological concerns. Whereas the theological concerns of the translators of the other versions are generally subtle, and even obscure at times, many of the concerns of T’s translators are manifest. Since so many quantitative variants in particular revolve around an

141

understanding of T’s theological concerns, those that seem unambiguously expressed in the text and which have been documented in the works of Churgin, Sperber, Levine, and others will be dealt with in the present chapter. The first variants to be addressed are those in which T adds the relative particle alone, namely, variants 44, 48, 50, 54, 57, 62, 64, 66, and 71. In every case, the relative renders part of a construct construction in MT (or what presumably would be a construct construction if the structure of MT were the same as that of T). Since the Hebrew used when Samuel was written did not yet have the particle of later Hebrew, it is clear that no difference in Vorlage can be assumed; the use of is simply a typical Aramaic equivalent for a Hebrew construct. Thus, none of these variants is significant. The next several variants are those in which the longer text of T can probably be explained by reference to the theological concerns of the translators, and since the theological concern is so evident, they may be dealt with rather cursorily. Variants 47 and 49 may be considered together, since they are related. The translators had a problem with Samuel, who was not a priest, sleeping in the temple of the Lord. They solved the problem by having him sleep in the court of the Levites (see above, variant 10), but they preserved MT’s reference to the temple by anticipating the call of Yahweh in the following verse.127 The additional of variant 47 in turn anticipates the reference to the temple later in verse 3, and it also clarifies exactly which lamp is being discussed, so these variants are not significant. The addition of to in variant 61 is a typical rendering of T, and it refers back to variant 47 as well, so it is not significant, either. Variants 51 and 52 are an attempt to avoid saying that Samuel had a direct knowledge of God. Instead of saying that he did not yet know God, T says that he had not yet learned to know instruction about God. The targumic use of has already been mentioned above (variant 13, pp. 136-37); it is frequently used before in T. Variant 55 also contains and so may be included along with variants 51 and 52 as nonsignificant variants. In variants 53 and 63, T inserts the word , “prophecy,” after a reference to something revealed to Samuel (a word in v. 7 and a vision in v. 15). The translators seem to have been concerned again to emphasize that what Samuel heard and saw was not God himself; rather, Samuel received revelation through a prophetic audition or vision. These variants, then, are not significant. These additions may be compared with T’s rendering of in verse 20 as , preceded by (variant 70). Here, too, the translators are apparently emphasizing the prophetic medium through which Samuel received his revelations, so variant 70 is also nonsignificant. Variants 65 and 67 are attempts to preserve the holiness of God by distancing him from human beings, in this case Samuel. The rendering of by has already 127 Cf. Levine, Aramaic Version, 120.

142

been discussed (see above, p. 135), and the addition of before is frequent in 128 T. Thus, these variants should not be considered significant. The next three variants to be considered are probably not theologically motivated but rather simple translational equivalents. Variants 58 and 59 both add the word before , referring to the “house” of Eli. The reason for adding the word is not clear, though perhaps the translators wanted to stress that judgment was coming upon the family rather than the buildings (cf. also 2:32 T). On the other hand, the addition may just be an Aramaic equivalent without special significance (cf. 3:14, ). Either way, the variants are not significant. Variant 60 concerns the addition of , “(holy) offerings,” after , another word for “offerings.” The translators of T used a variety of word to render the Hebrew : the cognate (1 Sam 1:21; 2:19), the word (1 Sam 2:13), and the compound (1 Sam 16:5; Josh 22:26, 28, 29). The variation in T, and particularly the frequent use of the compound , show that this variant is not significant. The remaining variants all have a somewhat more forceful claim to represent a Vorlage different from that found in MT. Variant 45 has an additional in T preceding an infinitive. The two other Hebrew infinitives in the chapter that stand alone and that are rendered by T as infinitives (3:6, 8) also lack a in T, but T renders in 3:15 with , which has both prepositions and preceding the infinitive. An investigation of other infinitives in 1 Samuel reveals that, though the translators usually omit before an infinitive when MT does, they occasionally add it (cf. 1 Sam 17:28), especially after some form of (Dt 2:25, 31; Josh 3:7), as in the present verse. This data suggests that the variant is probably not significant, but that the translators probably understood their Vorlage as an infinitive rather than an adjective. In variant 46, T has an extra conjunction as compared with MT. Table 36 shows that MT and T share fifty-eight occurrences of , and the list of add-oms indicates that T never fails to render a conjunction found in MT. Moreover, variant 46 is the only instance in which a conjunction was added. These statistics indicate that the translators were concerned to render conjunctions as accurately as possible (except when added as part of a theological expansion, as in variant 49). Supporting evidence comes from the versions and from ten Masoretic mss listed by Kennicott, one of which (187) Goshen-Gottstein considers important. The evidence suggests, then, that the Vorlage of T contained a conjunction, so the variant is significant. Variant 56 is the first variant to be considered in which MT has the longer text; it has the particle , and T has nothing corresponding. Table 35 indicates that the translators rendered the other five occurrences of in the chapter with , and a perusal 128 Ibid., 59-60.

143

of other occurrences of in the book shows that was indeed the normal rendering. No obvious graphic similarities in either Hebrew or Aramaic suggest themselves as reasons for accidental omission, though the haplography of a single word or letter is certainly plausible in either language. The failure of many mss of LXX to render the word, however, suggests the likelihood of Hebrew mss that omitted the particle. Thus, variant 56 is probably significant. The last two variants to be considered, numbers 68 and 69, must be treated together. MT says , and T reads . Both versions of this clause carry the same basic meaning, that none of Samuel’s words failed. However, whereas MT uses the colloquialism “fell to the ground,” T substitutes “was vain” for “fell” to make the meaning more obvious. As a result, the translators were also forced to substitute “one” for “to the ground,” in order to have the resulting clause make sense. This explanation of the data seems the most satisfactory one, despite the fact that a presumed bears some graphic similarity to . 129 Thus, the variants are probably not significant. A review of the quantitative variants in T reveals that only two of the thirty-nine deviations are significant, raising the percentage of agreement between MT and T to 91.3%. Thus, in spite of the translators’ theological concerns, they still produced a translation that is more literal than P, though it is less literal is this area than the adjusted figures for LXX. However, when the theological modifications and the use of the relative to render the construct are dropped from consideration, T become extremely literal in quantitative representation, on the order of 98.4% agreement between T and MT, a figure almost identical to the 98.3% agreement in quantitative representation between LXX and MT. These figures suggest that the translators of T were generally concerned with a precise rendering of their Vorlage, but this desire for consistency could be overridden if some theological or haggadic clarification of the text were necessary. For the textual critic, the data suggests that all those variants that do not result from some known theological tendency of the targumists deserve careful scrutiny. Vulgate Limitations of Latin for Rendering Hebrew Latin, like Greek, is an Indo-European language, so it naturally differs from Hebrew in several ways in regard to vocabulary, grammar, and idiom. A highly inflected language, Latin has six cases (nominative, genitive, dative, accusative, ablative, and 129 Cf. the comment by Churgin: “The general underlying principle in the exegesis of T. Jonathan consists in an attempt to render intelligible to the fullest possible degree that which is obscure,” [italics mine]; Pinkhos Churgin, Targum Jonathan to the Prophets, Yale Oriental Series, vol. 14 (New Haven: Yale University Press, 1907), 78.

144

vocative), three genders (masculine, feminine, and neuter), and two numbers (singular and plural).130 Word order, so important in Hebrew, is extremely flexible in Latin, since the case endings on the nouns allow them to placed in many different places within the sentence without altering the meaning. Thus, a close correspondence in word order between Hebrew and Latin would indicate a certain measure of commitment to literalness. Because Hebrew and Latin share almost no common roots, little similarity should be expected in regard to vocabulary. Similarly, it is unlikely that the correlation between the gender of nouns in Hebrew and Latin, unless they reflect true gender, will be any greater than might be expected from random similarity. One significant difference between Latin and Hebrew is the lack of a Latin article. Definiteness in Latin common nouns is usually implied from the context rather than explicitly indicated, though the Latin translators could use demonstrative adjectives to emphasize definiteness, if they so chose.131 However, at the stage of development of the Latin language represented by the Vulgate, extensive use of demonstratives to indicate definiteness is not a common feature, so the absence of a demonstrative to reflect a Hebrew article, for example, should not be seen as a deviation from literalness. A number of differences between Hebrew and Latin also appear in the respective verbal systems. As noted earlier, Hebrew verbs can be classified by stem, inflection, person, gender, and number. Latin verbs can be classified by tense, voice, mood, person, and number. As is the case with Greek, Latin person and number will generally reflect Hebrew person and number, and since gender is not represented in the Latin verb, one common form must be used for both masculine and feminine in Hebrew. Hebrew inflection is represented, to a large extent, by a combination of Latin tense and mood. Though the simple Hebrew perfect does not always represent past time, it is generally rendered in Latin by one of the past tenses (perfect, imperfect, or pluperfect) in the indicative mood. The simple Hebrew imperfect is usually rendered in Latin by a present or future tense (present, future, or future perfect) in the indicative. A tense other than one of the usual ones was often employed by the translators if they felt that the subjunctive mood was more appropriate. Furthermore, the use of the waw consecutive with the Hebrew perfect or imperfect usually reversed the above characterization. Hebrew imperatives were rendered consistently with Latin present imperatives. The division of Hebrew stems into three groups—basic, intensive, and causative— has been discussed above. Latin does not reflect these distinctions, but it does differentiate 130 The Proto-Indo-European dual, reflected in classical Greek, disappeared from the Italic family of languages, including Latin, in prehistoric times; Carl Darling Buck, Comparative Grammar of Greek and Latin (Chicago: University of Chicago Press, 1933), 170-71. 131 In fact, the article that is present today in the Romance languages is descended from the Latin

demonstrative pronoun ille; Frederic M. Wheelock, Latin: An Introductory Course Based on Ancient Authors, 3d ed. (New York: Harper & Row, Barnes & Noble, 1963), 44.

145

between active and passive voice. Thus, active stems (qal, piel, hiphil) are generally represented by a Latin verb in the active voice, and passive stems (niphal, pual, hophal), by a Latin verb in the passive voice. Since there is no Latin reflexive voice, nor is there a middle voice as in Greek, the reflexive stems (niphal, hithpael) have no natural parallel in Latin.132 It might be expected, then, that these stems would not be rendered as consistently as the others. Latin, like Hebrew, has other verbal forms that cannot be classified as finite verbs. The uses of the Latin infinitive correspond fairly closely to those of the Hebrew infinitive, so a fairly high degree of correlation was possible in translation. One common difference, however, is the failure of the Latin translators to render the Hebrew preposition before infinitives. Some correspondence of use also exists between Hebrew and Latin participles, although the use of a Latin participle as a finite verb was not common. Finally, Latin has two other verbal forms that have no equivalent in Hebrew, the verbal nouns called the gerund and the supine. Since their use in Latin corresponds most closely to Hebrew participles and infinitives, they, too, may be be considered literal renderings of these Hebrew verbals. 133 Partial Translation Technique Add-Oms Like Greek, Latin does not need to render the Hebrew sign of the definite direct object or various Hebrew prepositions in order to clarify the meaning of a passage. Of course, it would have been easy for Jerome to use equivalent Latin prepositions, but he did not always choose to do so. In the light of an examination of the primary renderings of the Hebrew prepositions and sign of the definite direct object, the omission of , , , or will not be considered add-oms. See Appendix 2 for a list of add-oms in V. Consistency

1 2 3 4 5

Ref 3:1

3:2

Variation autem ministrabat manifesta ergo iacebat

Vulgate lexeme verb adjective lexeme verb

Masoretic Text participle participle participle

132 The Latin deponent passive, though related to the medio-passive voice in Proto-Indo-European,

from which the Greek middle voice developed, came to be considered a passive form with active meaning by Latin speakers; Buck, Comparative Grammar, 237. 133 Harry E. Wedeck, Third Year Latin, 2d ed., The Heath Latin Series, ed. Wilbert Lester Carr (Boston: D. C. Heath & Co., 1938), 311.

146

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

3:3 3:4 3:6 3:7 3:8

3:9 3:10 3:11

3:12

3:13

oculi caligaverant nec autem dormiebat respondens consurgens respondit neque vocavit consurgens vocaret audit audit facio quicumque audierit adversum super incipiam conplebo praedixi iudicaturus essem in propter agere

3:16

3:17

3:18

respondens praesto sum interrogavit verbis respondit est oculis

plural verb conjunction lexeme verb participle participle lexeme lexeme verb participle verb verb verb verb lexeme verb lexeme lexeme verb verb lexeme perfect periphrasis lexeme lexeme lexeme infinitive participle lexeme lexeme verb lexeme plural passive lexeme verb plural

singular verb phrase negative particle participle verb verb

infinitive verb participle participle participle participle participle

infinitive infinitive perfect w/c participle

participle verb

pronoun singular piel pronoun singular

147

44 3:19 cecedit active hiphil 45 3:20 fidelis adjective participle 46 3:21 appareret verb infinitive 47 iuxta lexeme Table 51.—Verbs, Nouns, and Adjectives: Lexemes Heb words (>1x): 29 Lat words: 45 1.55 Lat words/Heb word deviation factor: 2.13 total Heb (>1x): 134 primary Lat: 108 80.6% of Heb words by primary rendering Heb roots (>1x): 27 Lat roots: 43 1.59 Lat roots/Heb root Table 52.—Adverbs, Prepositions, and Particles: Lexemes Heb advs (>1x): 12 Lat advs: 30 2.50 Lat advs/Heb adv total Heb (>1x): 82 primary Lat: 55 67.1% of Heb advs by primary rendering deviation factor: 5.41 Table 53.—Conjunctions: Lexemes Heb conjs (>1x): 2 Lat conjs: 12 6.00 Lat conjs/Heb conj total Heb (>1x): 55 primary Lat: 37 67.3% of Heb conjs by primary rendering deviation factor: 30.98 Table 54.—Word Classes percentage of verbs/verbals represented by verbs/verbals: 95/97 = 97.9% percentage of verbs represented by verbs: 74/76 = 97.4% percentage of participles represented by participles: 0/12 = 0.0% percentage of infinitives represented by infinitives: 3/7 = 42.9% percentage of nouns/adjs represented by nouns/adjs: 57/68 = 83.8% percentage of pronouns represented by pronouns: 36/40 = 90.0% Table 55.—Verbs: Hebrew Inflection, Latin Tense and Mood deviation factor (discrete tense/mood combinations): 6.15 deviation factor (grouped): 1.53 deviation factor (grouped, without verbals): 0.25 Table 56.—Verbs: Hebrew Stem, Latin Voice deviation factor: 0.63 Table 57.—Verbs: Person deviation factor: 0.00 Table 58.—Verbs: Number deviation factor: 0.02 Table 59.—Nouns and Adjectives: Gender deviation factor: 31.64 Table 60.—Nouns and Adjectives: Number deviation factor: 2.95

148

Table 61.—Nouns and Adjectives: Use vs. Case deviation factor: 13.58 (discrete use/case combinations) deviation factor: 1.02 (grouped) Table 62.—Pronouns: Gender deviation factor: 0.22 Table 63.—Pronouns: Number deviation factor: 0.00 Table 64.—Pronouns: Use vs. Case deviation factor: 9.91 (discrete) deviation factor: 3.44 (grouped) Representation of Hebrew Lexemes by Latin Lexemes (Segmentation) Ref 3:1 3:2 3:5 3:6 3:7 3:8

Hebrew Compound

48 49 50 51 52 53 54 55 56 3:9 57 3:10 58 3:13 59 60 3:14 61 62 63 3:15 64 3:17 65 66 67 3:18 68 3:20 69 Heb compounds: 49 Lat equivalents: 27

Latin Rendering coram videre me me ei tertio me puerum te sicut vocaverat secundo ei eos domui victimis et muneribus indicare me tibi tibi ei propheta Domini 55.1% of Heb compounds rendered exactly

149 Word Order Ref Number of Variations Latin Variant 70 3:17 1 oro te ne Heb semantic units: 372 variations: 1 Heb word order followed 99.7% of the time Quantitative Representation

71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101

Ref 3:2

Number of Variations +1 -1 -1 3:3 -1 -1 +1 3:4 -1 +2 3:5 -1 +1 3:6 -1 -1 +1 +1 +1 3:7 +1 3:8 +1 +1 -1 +1 -1 3:9 -1 +1 -1 +1 -1 3:11 -1 3:12 +1 3:13 -1 +1 +1

Variant factum - 2° - 1° erat - 2° qui respondens - 3° qui - 3° - 5° qui te et 3° fuerat et 2° adhuc - 2° qui - 3° et 3° deinceps - 3° - 1° sum essem quod

150

102 -1 2° 103 +1 indigne 104 -2 105 3:14 -1 - 1° 106 +1 quod 107 +1 eius 108 -1 109 +1 usque 110 3:15 -1 111 3:16 -1 - 3° 112 +2 qui respondens 113 3:17 +1 eum 114 +1 est 1° 115 +1 est 2° 116 +1 te 2° 117 +1 sunt 118 3:18 +1 ille 119 +1 quod 120 +1 est 2° 121 3:20 -1 - 2° 122 +1 esset 123 3:21 +1 fuerat 124 -1 Heb semantic units: 424 Lat deviations: 57 86.6% agreement Elimination of Variants The textual history of a version is important to consider when evaluating the variants exhibited by that version. This statement particularly applies to an evaluation of V, for, as noted above, Jerome made use of both it and LXX, as well as later Greek versions, when producing his translation. In his introduction to the Latin Bible, Friedrich Stummer says that Jerome’s agreements with LXX or the later Greek versions should generally be excluded from consideration, unless further evidence for a deviating Hebrew text exists.134 Nevertheless, it must also be remembered that he was translating from a Hebrew ms, the 134 Friedrich Stummer, Einführung in die lateinische Bibel, 123. He says, in part,

Wo Hieronymus mit der Septuaginta oder den späteren Übersetzern gegen unseren heutigen Masoratext übereinstimmt, scheidet er m. E. überhaupt aus. Denn das beweist höchstens, daß die Septuaginta seiner Zeit oder einer der Späteren so und so las, nicht aber ohne weiteres, daß auch der hebräische Text, der ihm vorlag, von dem unseren verschieden war.

151

character of which is problematical.135 Stummer does not explain what further evidence he might accept as supporting a reading different from MT in the Vorlage of Jerome, but he seems to imply that support from the other versions would constitute such evidence. Surely another type of supporting evidence, however, would be Jerome’s failure to render simple Hebrew vocabulary and sentence structures in a way consistent with his translation technique. It is undoubtedly true that Jerome frequently resorted to LXX or the other Greek versions (especially Symmachus)136 when he faced a Hebrew passage that was difficult to understand. However, his knowledge of Hebrew was surely good enough to enable him to translate “easy” Hebrew without recourse to the Greek. Therefore, variants in V that would otherwise be considered significant will be compared with the extant Greek versions and the OL to see if any influence from these versions is present. A reading in V that agrees with one of these other versions will not be considered significant if the Hebrew at that point in the text is difficult, either lexically or grammatically. However, if the Hebrew would probably not have been the sort that Jerome would have found difficult to translate, the variant will be considered significant, even if it is supported by other versions. It is clear that determining what Jerome would and would not have been able to translate without recourse to the versions is subjective, but it seems best to proceed in this fashion in order to avoid the extremes of including too much or too little. Jerome himself states that his approach to translation is to render “with complete fidelity what stands in the Hebrew,” but not necessarily to create a word for word translation, for “if we follow the syllables, we lose the understanding.”137 Jerome’s skill and originality as a translator are most notable in the historical books, including Samuel, where he follows the Hebrew more closely than in the prophetic books.138 A more precise initial estimate of his translation technique may be gleaned from an analysis of the tables. The first fact to be noticed is the decided propensity for variety in lexical choice indicated by the first three tables.139 The use of Latin verbs to render Hebrew verbs (table 54) is comparable to that in other versions, but V’s rendering of verbals by the same class of verbals is even lower than that of LXX, and none of the participles is rendered by a 135 Deist, Text of OT, 209. 136 “Where the Vulgate exhibits a rendering which deviates alike from the Hebrew text and from the LXX, the clue to its origin will generally be found in one of the other Greek translations, especially in that of Symmachus”; S. R. Driver, Notes on the Books of Samuel, liv; cf. also lxxxi-lxxxii. 137 Jerome, Epistle to Sunnia and Fretela, cited in Robert H. Pfeiffer, Introduction to the Old

Testament (New York: Harper & Bros., 1948), 124. 138 Pfeiffer, Introduction, 124. Cf. also Jerome’s comment in his prologue to the books of Samuel and Kings, cited in VS: “Et cum intellexeris quod antea nesciebas, vel interpretem me aestimato, si gratus es, vel , si ingratus, quamquam mihi omnino conscius non sim mutasse me quippiam de hebraica veritate”; Biblia Sacra Stuttgartensia, 1:365. 139 A comparison of V’s deviation factors of 2.13, 5.41, and 34.65(!) with those of the other

versions highlights this tendency.

152

participle. The percentage of nouns, adjectives, and pronouns rendered by words of the same class is also substantially lower in V than in the other versions. This variety exhibited in both vocabulary and rendering of word class does not appear to the same extent in V’s rendering of grammatical categories. In fact, the deviation factors for the grammatical tables (tables 55 through 64) are quite similar to those of LXX, the other version in an Indo-European language, and are often lower. The rendering of the person of verbs and the number of pronouns is absolutely consistent. V is very consistent in its renderings of the inflection of verbs (when grouped, without verbals), verb stem, the number of verbs, and the gender of pronouns. It is fairly consistent in rendering verbal inflection (grouped, including verbals) and number of nouns and adjectives. V is fairly inconsistent in rendering the use of pronouns. Finally, it is very inconsistent in rendering the use and the gender of nouns and adjectives, as would be expected from the differences between Latin and Hebrew. As in LXX, the Hebrew use most inconsistently rendered among nouns, adjectives, and pronouns is the object of the preposition. In categories other than consistency, the representation of all the elements of compound words is lower in V than in any other version, though it is fairly close to LXX is this regard. V agrees with the other versions in closely following the Hebrew word order, but in the category of quantitative representation, V has more variations from MT than any other version, though its percentage of agreement is not significantly lower than the initial figures of LXX and P. In summary, V is closest to its Indo-European ally in many areas, but its variety of lexical choice is higher than that of any of the other versions. Variants Related to Consistency Eighteen of the forty-seven variants related to consistency are lexical variants, and the variety of renderings in tables 51, 52, and 53 suggests that few, if any, of them are significant. The least consistency in clearly present in the rendering of conjunctions, and variants 1, 4, 8, and 9 deal with the rendering of conjunctions. In light of the inconsistency reflected in the rendering of conjunctions, none of these variants should be considered significant. Variants 23, 24, 30, 31, and 47 are all lexical variants that deal with prepositions. Table 52 shows that V is less consistent in rendering adverbs, prepositions, and particles than might be expected in a version characterized as literal. It has a tendency to use several different words to render the same Hebrew word, even though it often prefers one reading (cf. the renderings of and ). The number of variations from a single main rendering suggests that Jerome was more concerned to produce a smooth Latin translation of these words than to render each Hebrew preposition with a single Latin preposition. In variants 23 and 24, V reads adversum and super, respectively, and MT reads in both places. It is possible that Jerome based his renderings on LXX, which reads and in these

153

two places, but the Hebrew is far from difficult, so he would have had no compelling reason to consult another version. On the other hand, he may have felt that adversum and super better conveyed the meaning of the prepositions in the context. Furthermore, the semantic range of overlaps that of in the books of Samuel, Kings, Jeremiah, and Ezekiel more than in other books.140 Thus, Jerome’s rendering has at least as much probability of being based on inner-Latin considerations as on differences in the Vorlage, so these variants should probably not be considered significant. The other lexical variants that concern prepositions may be dealt with similarly. Variant 30 reads in aeternum for in MT. Though is usually rendered by usque in the chapter, in aeternum, “forever,” is an idiomatic rendering of the prepositional phrase. In variant 31, propter iniquitatem stands for . In or simply a case ending are the usual ways which V renders , but can sometimes have a causative force, as in the present case, and propter is an appropriate rendering in such a case. Variant 47 has iuxta verbum for in MT. It is possible that Jerome read in his Vorlage; on the other hand, iuxta might simply be another example of contextual rendering, much like propter in the previous variant. Lacking further evidence of V’s pattern of rendering the prepositions and , the freedom with which prepositions are rendered in general prohibits concluding that the Vorlage was different from MT. Thus, none of these variants should be considered significant. In variant 14, the particle neque, “and not,” is used to render ; elsewhere, is rendered by necdum and antequam, both of which modify the negative idea by limiting its scope: “not yet, before.” It is unlikely, however, that neque here suggests a pure Hebrew negative ( ), and the fact that it follows necdum in the same verse suggests that the meaning attached to the enclitic dum carries over to neque as well. Variant 14, then, is not significant. Most of the other lexical variants may be dealt with more briefly. Variants 13, 41, and 38 render with respondit (in the first two cases) or interrogavit (in the last case). Though these are not the usual renderings aio or dico, they do adequately and accurately render the words in MT, and table 51 does suggest a tendency toward freedom in choice of vocabulary. Quicumque (variant 21), though not a form of omnis or universus, is a good contextual rendering of . Praedixi for (variant 27) and praesto sum (variants 35 and 36) for are also acceptable contextual renderings, though in each case they avoid the more common translation equivalents. None of these variants is significant. The final lexical variant is agere in variant 32. Unlike the other variants considered to this point, the MT of verse 13 is not straightforward, a fact evidenced by the variety of renderings among the versions. Jerome’s Vorlage seems to have been the same as MT, for he paraphrases the difficult of MT (he shows no 140 BDB, s.v. “

,” note 2.

154

knowledge of the tiqqun sopherim) by eo quod noverat indigne agere filios suos, “because he knew that his sons were acting shamefully.” This rendering, though not strictly literal, certainly captures the import of the clause. Thus, variant 32 is not significant. The next several variants deal with words in one class rendered by words in another. The most common example of this inter-class rendering is when V uses something other than a participle to render a participle in MT, as in variants 2, 3, 5, 10, 17, 18, 19, 20, 22, 29, 33, and 45. Since table 54 indicates that Hebrew participles are regularly rendered by something other than participles in the chapter, none of these variants is significant. Another common shift in word class involves the rendering of an infinitive in MT by a verb in V (variants 15, 25, 26, and 46). Another look at table 54 reveals that less than half of the infinitives in MT are rendered by infinitives in V, so it seems that Jerome was not overly concerned with translating Hebrew infinitives with Latin infinitives. Variants 25 and 26 are special cases, since they render infinitives absolute in MT. The Hebrew construction would not have lent itself to idiomatic Latin renderings with an infinitive, so these variants cannot be considered significant. The other two variants both appear as part of the rendering of a phrase associated with , “and he repeated.” As mentioned above in the discussion of LXX variant 13, can be followed either by an infinitive or by waw and another verb. It is possible, then, that variants 15 and 46 reflect a variant Hebrew text that did not have an infinitive. Variant 46, however, has ut appareret instead of an infinitive, and since this expression is a common equivalent in V, it is not significant.141 Variant 15 reads et adiecit Dominus et vocavit, whereas MT has an infinitive for the second verb. LXX here follows MT, though in 3:6 it reads . It is possible that V here reflects a Hebrew text divergent from MT and all the other versions, but the second et may also be an inner-Latin corruption of ut. In light of this latter possibility, and pending a more thorough examination of V’s rendering of such constructions in a wider context, the evidence of chapter 3 does not support considering variant 15 significant. In variants 11, 12, 16, and 34, V has a participle corresponding to a verb in MT. Variants 12 and 16 are renderings of , and in both cases the participle consurgens transforms an independent clause in Hebrew into a dependent clause in Latin. Variants 11 and 34 render with qui respondens. Though not the exclusive rendering, this sort of transformation from compound to complex sentence is common in V.142 These variants, then, are not significant. 141 W. E. Plater and H. J. White, A Grammar of the Vulgate (Oxford: Clarendon Press, 1926),

23-24. 142 Ibid., 117: “The Double [i.e., compound]Sentence, connected by vav, and constantly found in the Hebrew of the Old Testament, is as a rule skilfully woven into a Complex Sentence of the well-known classical type.” Cf. also p. 127: “In the Vulgate, and especially in the Old Test., qui is constantly used as

155

In two cases, variants 37 and 42, V uses a verb where MT has a pronoun. In each case, the Hebrew text of MT has a nominal clause which V renders by including the proper form of the copula sum. Nominal sentences were certainly possible in Latin, but the use of the copula was increasing in later Latin.143 Thus, these variants are not significant. Closely related to the variations in word class is variant 7, where V uses a verb caligaverant where MT has a verb phrase . Though table 54 indicates a concern for rendering verbs more exactly than either participles or infinitives, the reason for the paraphrastic rendering here is obvious. V uses a single Latin word to convey the meaning found in an idiomatic Hebrew expression whose literal translation would not have been readily understood by Jerome’s non-Semitic audience. Thus, variant 7 is not significant. The differences in the verbal systems of Latin and Hebrew make correlation more difficult than between Hebrew and the other Semitic languages, but V does show more consistency in rendering the grammatical categories of verbs than in rendering either lexemes or certain aspects of word class. Variant 28 has a Latin perfect corresponding to a Hebrew perfect with waw consecutive. Thus, V has God refer to a message he had proclaimed previously (note also the prae prefix), probably in 2:27-36. The third deviation factor listed for table 55 indicates a high degree of consistency in the rendering of Hebrew verbs, so this variant, supported also by LXX and T, should be considered significant. Two variants deal with the rendering of the Hebrew stem in V: variants 40 and 44. In variant 40, V has the passive dicta sunt where MT has the piel (active). As noted in the discussion of LXX variant 45, the lack of an explicit subject in the verse might have led Jerome to read the verb in his Hebrew text as a pual rather than a piel. Although he might just have preferred to render the word as a passive, table 56 indicates a fairly consistent tendency to render Hebrew verbs by their “natural” equivalents. In either case, however, the consonantal text would have been no different from MT, so variant 40 is not significant. Like Greek, Latin lacks a voice to express the causative idea found in the hiphil of . V sometimes uses facio or do as auxiliaries to express the causative idea, but this construction is not used consistently.144 In fact, all other hiphils in the chapter are rendered by the active voice alone, and though none of them has a distinctive causative meaning, it is unlikely that the Vorlage of V read a qal rather than a hiphil. All of the remaining variants deal with variations in the number of nouns. Two of them, variants 6 and 43, have a plural of oculus where MT reads . Since in both cases a connecting link between two sentences.” 143 J. B. Hofmann and Anton Szantyr, Lateinische Syntax und Stilistik, Handbuch der Altertumswissenschaft, division 2, part 2, vol. 2 (Munich: C. H. Beck, 1965), 419-23, especially 41920. 144 Plater and White, Grammar, 23.

156

is apparently simply an orthographic variant of (the qere of MT ms L; see above p. 94), as in the other secondary versions, neither variant can be considered significant. The last variant under the heading of consistency is variant 39, where V reads the plural verbis and MT reads the singular . As in the case of LXX variant 43, Latin idiom probably required the use of the plural rather than the singular, since the Hebrew noun was a collective singular. Thus, variant 39 is not significant. Only one of the variants related to consistency in V can be considered significant, so the initial picture of the translation technique does not change much. Only table 55 is affected, as follows. Table Changes 55 deviation factor (discrete tense/mood combinations): 6.26 deviation factor (grouped): 1.54 deviation factor (grouped, without verbals): 0.23 Variants in the Representation of Hebrew Lexemes by Latin Lexemes (Segmentation) As is the case with LXX and, to a lesser extent, P, V shows little interest in rendering each individual lexeme in compound words. In fact, only 55.1% of the compound words in the chapter are rendered exactly. In particular, prepositions, which are necessary in Hebrew, are superfluous in Latin with its various cases of nouns and adjectives. Some prepositions are rendered, but no consistency appears. Such a low percentage of exact renderings makes it unnecessary to examine individual cases in detail. None of the variants in segmentation appears to be significant. Variants in Word Order In sharp contrast to its lack of concern for rendering all the lexemes in compound words, V shows a penchant for following the Hebrew word order wherever possible, as do each of the other secondary witnesses. Since postpositive conjunctions are not counted as variants in word order if they are as close to the Hebrew conjunction as possible, the only variant in word order is variant 70, which reads oro te ne for , with oro te being the rendering of . The overwhelming concern for following Hebrew word order implies that this variant should be taken as significant, unless Latin usage demands the variation. It seems that such is indeed the case. Since Jerome chooses to use oro te with a subjunctive following ne, the word order of V is apparently necessary for idiomatic Latin,145 so the variant is not significant after all. Variants in Quantitative Representation As is the case with Greek, Latin has no equivalent for the Hebrew particle 145 Hofmann and Szantyr, Syntax und Stilistik, 533-34.

, so

157

the omission of this particle in V will not be considered a variant. Furthermore, Latin also has no means of representing the Hebrew article or state of definiteness (though means to do so did develop in the Romance languages),146 so the omission of articles will be ignored. In the area of quantitative representation, V’s percentage of agreement with MT (86.6%) is between the initial figures of LXX and P, on the one hand, and T, on the other. As with P, most of V’s deviations from the text of MT can be explained as stylistic deviations that bring the text closer to the idiom of the target language. A perusal of the list of quantitative variants indicates that of the fifty-seven deviations, thirty-three are instances of a longer text in V and twenty-four of longer text in MT. Thus, V has a longer text somewhat more frequently than MT does. The individual quantitative variants may now be examined in more detail. On twelve occasions (variants 72, 74, 77, 79, 81, 82, 89, 91, 96, 105, 111, 121), the conjunction in MT has no equivalent in V; in three other cases (variants 85, 87, 93), V has et where MT has no conjunction. Though it is true that the addition or omission of conjunctions is probably the most common variant in extant Masoretic mss, two factors combine to cast doubt on the significance of any of these variants in V. First, the sheer frequency of the variations (particularly the apparent omissions) leads one to infer that Jerome did not consider the omission or, perhaps, the addition of conjunctions as detrimental to his attempt to render the Hebrew into Latin. Second, the large number of different words used to render conjunctions (see table 53) suggests more concern for Latin idiom than for exact correspondence between source and target language in this area. Though one or another of the omissions might have been the result of a difference in Vorlage, the translation technique of V, as described to this point, does not allow one to draw such a conclusion in any specific case. None of the omissions of should be considered significant. The addition of et is less frequent than the omission of , but unless some other factor intervenes, these variants should be disregarded as well. Variants 85 and 93 both occur in similar contexts, in the phrases revertere et dormi and vade et dormi. Though revertere dormi appears in verse 5 without et, it is likely that the inclusion of the conjunction was more in accord with typical Latin idiom. In any case, Jerome might have been following the language of LXX here, so these variants cannot be considered significant. Variant 87 has already been dealt with in the context of the discussion of variant 15, where it was decided that evidence does not allow one to suppose that the variant is significant. Therefore, variant 87 should also be considered nonsignificant. Closely related to these variants are variants 75, 101, 102, and 106, which also deal with the presence or absence of conjunctions. In variant 75, MT reads , and V 146 Cf. ibid., 191-92.

158

reads ubi. The Latin conjunction ubi combines the relative idea of with the local idea of , so no content is lost (cf. LXX), and the variant is not significant. Variant 101 is also an example of the use of two words in one language rendering one in another, but this time the longer text is in V, which uses eo quod to render . The text of this difficult passage has already been discussed above to some extent under variant 32. The reason for the compound conjunction eo quod seems to be to delineate sharply the clause it introduces in Latin; it is not so carefully distinguished from the preceding clause in Hebrew (in fact, Masoretic punctuation associates it closely with the preceding word). Since can sometimes be used as a conjunction,147 V’s rendering can be said to be a literal rendering of the text found in MT, even though the phrasing is different from MT. Thus, variant 101 is not significant. The omission in variant 102 may also be considered here briefly. Since Jerome rearranged the phrasing of the verse, the second in Hebrew became superfluous and was apparently omitted for this reason. In variant 106, V reads the conjunction quod where MT has no conjunction. The effect of this insertion in V is to turn a direct speech into an indirect speech. The looseness with regard to the insertion and omission of conjunctions in V leads one to conclude that this insertion is demanded by good style rather than by a varying Vorlage, as also in P. V frequently uses periphrastic forms of verbs to render Hebrew forms which are not periphrastic, namely, variants 71, 86, 98, 100, 115, 117, and 123. In each case, normal Latin idiom demands the use of a periphrastic form, either to express a passive idea (variants 71,148 86, 117, 123), because the verb was deponent (variants 98, 115), or to express a future idea in the subjunctive (variant 100). Thus, the extra words in V cannot be considered significant.149 On four occasions, V has an extra form of sum which has no counterpart in MT and is not part of a periphrastic construction, namely, variants 76, 114, 120, and 122. In each case, MT has a nominal sentence. Though nominal sentences are permitted in Latin, it seems that, in the present chapter at least, copulative sentences were used instead, since all four instances of Hebrew nominal sentences are translated in this way. Because the addition of the copulative sum is apparently part of the translation technique, these variants are not significant. Another apparent aspect of V’s translation technique is the substitution of hypotaxis for parataxis by means of a clause beginning with qui and the omission of the Hebrew (variants 78, 80, 83, 90, and 112). This type of construction is common in V, especially 147 BDB, s.v. “

.”

148 Cf. Plater and White, Grammar, 118. 149 For a discussion of the form revelatus fuerat rather than revelatus erat, see Hofmann and Szantyr, Syntax und Stilistik, 321.

159

in the OT.150 The additional respondens in variants 78 and 112 could conceivably reflect an additional in the Vorlage (cf. 9:12, 21), but it is more likely that respondens is simply an expansion original with V (cf. 2:16). Thus, none of these variants is significant. Variants 110 and 124 are examples of the omission of the preposition . The reason for this omission is the same as it was for the omission of in the category of segmentation, namely, that Latin has no need to use a preposition to convey the meaning, since it is an inflected language. Thus, these two variants are not significant. The appearance of an extra pronoun in either MT or V is frequent in the chapter, occurring eight times (variants 84, 97, 99, 104, 107, 113, 116, and 118). In three cases (variants 97, 99, and 104) MT has the longer text. Variant 99 is an example of V’s tendency to render verbal participles with full verbs, so, since the pronoun is implied in the verb ending, this variant cannot be considered significant. The end of verse 13, where variant 104 is located, is difficult to render, as has already been noted in the discussion of variant 32. The text of V here, though not strictly literal, does not seem to presume a text different from MT, so the variant is not significant. Syntactic variation or semantic difficulty are not involved in variant 97, where V does not render the pronominal suffix found in MT. It is possible that the pronoun was omitted as not necessary in the context, but, as the two previous variants show, V does not seem to omit pronouns without a good reason. It seems probable that a Hebrew scribe accidentally omitted the , though it is also possible that an original id was omitted in the course of the transmission of V. It is probably best, then, to consider variant 97 significant. In variant 116, the clause oro te is used to render the Hebrew particle , so this variant is not significant. Variant 118 involves the addition of a pronominal subject ille to specify who is speaking. The use of ille to refer to a subject previously mentioned is characteristic of Latin,151 and the insertion of or some form of the demonstrative would not clarify matters in Hebrew, so this variant is not significant, either. Variant 107 needs to be considered alongside variant 108, since they are apparently substitutional variants: whereas V simply reads “his house,” MT reads “the house of Eli.” It is possible that the repetition of “the house of Eli” in the verse was considered redundant and so was modified at the point of translation. However, no other evidence for such a concern appears in the chapter, at least as so far analyzed. Moreover, it is just as possible that “the house of Eli” earlier in the verse also affected the later part of the verse. Thus, these variants should be considered significant. In variant 84, V reads non vocavi te fili mi, where MT reads , and LXX reads . It appears that V is a conflation of the texts found in MT and LXX. As noted in the discussion of LXX variants 83 and 84, and - are probably 150 Plater and White, Grammar, 127. 151 Hofmann and Szantyr, Syntax und Stilistik, 187, 413.

160

substitutional variants that occurred in Hebrew. Since V shows no other signs of conflating MT and LXX in the chapter, it is probable that the conflation occurred in Hebrew, so variant 84 is significant. Variant 113 is the only purely explanatory addition in V in the chapter (excluding substitutional variants 84, 107, and 108). Unlike P, which has ten explanatory additions, all in P itself, or LXX, where both MT and LXX have eight explanatory additions relative to one another, variant 113 is the only one found in V (variant 92, considered below, is the only explanatory addition in MT relative to V). The small number of explanatory forms demonstrates the proximity of the Vorlage of V to MT, but it also indicates that explanatory additions are not typical in V. The only other secondary witness that has this addition is P, but it is probable that the expansions occurred independently in the two traditions. It is possible that the addition occurred in the transmission of V, though no evidence exists that it did (the first hand of one ms omits eum), but it is equally possible that the variant was in the Vorlage of V. In light of these facts, variant 113 should probably be considered significant. As mentioned above, variant 92 is the only explanatory addition in MT relative to V in the chapter. Whereas MT (along with P and T) specifies both the subject and the object of the verb, V does not include the subject “Eli.” A look at LXX is instructive here, for it omits both subject and object (see LXX variant 87). Since V does not agree completely with LXX, it is unlikely that the reading of V is based on LXX here. Nothing in V’s translation technique, as discerned to this point, suggests that such an omission is characteristic of V, so the variant should be considered significant. In variant 73, MT reads ( ) , but V fails to render the second word, reading et oculi eius caligaverant. The expression “began to be dim” is fine Hebrew idiom, but it is likely that Jerome did not believe it would communicate effectively in Latin. V’s rendering “and his eyes had been blind” avoids the foreign idiomatic expression but captures the meaning, so variant 73 should not be considered significant. Variants 88, 95, and 109 are adverbs in V that have no corresponding word in MT. Variant 95, which has an additional deinceps in V, should perhaps be considered alongside variant 94, which has in MT not represented in V. To have the expression in a conditional sentence is typical Hebrew, but it is not idiomatic Latin, and Jerome appears to have omitted for this reason. It may be that he added deinceps to replace the omitted . On the other hand, the three variants listed above may indicate some tendency to add adverbs, though the evidence available is not conclusive. Variant 109 is almost surely an addition original with V, and probably coming from Jerome himself, since usque merely strengthens the following in aeternum. In fact, Jerome probably would not have considered the word an addition, since it adds nothing to the meaning of the verse. Variant 88 is the most difficult of these variants to evaluate, for a floating “again” appears

161

in all the versions under consideration except LXX from time to time (verses 6 [MT, T, LXXL], 8 [P, V, LXXO], 21 [P]). Since verses 6 and 8 represent the second and third of Yahweh’s calls to Samuel, the appropriateness of such adverbs is apparent. The additions could have arisen in either Hebrew or the versions, but the agreement of LXXO with V suggests that variant 88 should probably be considered significant; the other adverbial additions should not. The context of variant 103 has been discussed above, and it was noted that, though Jerome restructures the Hebrew text, it does not appear to have been any different from MT. The same conclusion applies as well to the present variant, for the additional indigne describes his sons’ behavior: they were acting shamefully. Thus, the variant is not significant. In variant 119, V inserts a relative pronoun to make the passage more understandable. Rather than rendering “the good in his eyes” literally, as does LXX, V inserts quod, so that the phrase becomes “what is good in his eyes.” Since the insertion is a matter of idiom rather than difference in Vorlage, the variant cannot be considered significant. All of the quantitative variants have now been evaluated, and only seven of them have been considered significant. Eliminating these variants from the statistics, the percentage of agreement between V and the presumed Vorlage rises to 88.0%. Comparing this figure with those of the other versions, Jerome was apparently somewhat more interested in representing each Hebrew word by a single word in translation than were the translators of P, though the figure does not approach the fidelity to quantitative representation found in LXX or T (after theologically motivated factors are eliminated in T). Evaluation of Partial Secondary Witnesses As was the case in Chapter 2, the partial secondary witnesses will not be dealt with in the same exhaustive manner in which the proper secondary witnesses were. Instead, from the list of accepted readings given there for each of these witnesses, those which might reflect a Hebrew reading different from MT will be selected, with little comment on individual choices. Aquila None of the readings of ´ stem from any Hebrew text different from the consonantal text of MT. However, variant 3 apparently renders instead of , and in 3:13 (variant 5) uses the root I found in verse 2 rather than II . Since these differences do not affect the consonantal text, they will not be considered further.

162 Symmachus The readings attributed to ´ sometimes correct LXX in the direction of MT, and sometimes they are simply different (better) Greek readings. None of them, however, implies a Hebrew reading not found in MT. Theodotion Like the previous two witnesses, none of the readings of ´ implies a Hebrew text different from MT. Other Readings Attributed to

´

The reading attributed to the three translators that has the best claim to represent a variant Hebrew text is in verse 4 (variant 7). It is probable, though, that the reading is a partial harmonization with verse 10. If the reading reflected a true Hebrew variant, one would expect the subject to appear after the first verb, as it does in verse 10. Thus, none of the variants attributed to the three in general can be considered significant. The Lucianic Recension S. R. Driver characterizes the Lucianic recension as employing the substitution of synonyms, double renderings, and renderings that are different from both LXX and MT, and frequently superior to both.152 For the present, whether a reading is superior to MT is not the question, but rather whether it is based on a Hebrew reading independent of MT. One characteristic not mentioned by Driver, but which appears with enough consistency in the chapter to note it, is the tendency to add explanatory words or phrases, especially implied subjects or objects (cf. variants 14, 16, 33, 34, 41).153 Another characteristic noted in LXXL is the tendency toward consistency of expression in similar passages (cf. variants 5, 7, 13, 17, 18). There are a few readings that suggest themselves as significant. Variant 8 is similar to the reading of MT, but it supposes a different position for . Variant 12 suggests a different word order from MT. Variant 13 makes the third call of Samuel resemble the first two (in LXX and LXXL) by adding a second . Variant 20 implies a missing conjunction and a different preposition. Variant 30 moves the second occurrence of to the end of the verse, an odd position, perhaps supporting the omission in the hexaplaric mss. Variant 40 deletes a conjunction, which could have arisen as a result of dittography from the preceding or been deleted by haplography in 152 S. R. Driver, Notes on the Books of Samuel, xlix. 153 S. D. Walters, “Hannah and Anna,” 393, n. 14, notes LXXL’s tendency to insert proper

names.

163

Hebrew. Finally, three variants to the long addition in verse 21 are taken as significant (variants 46, 47, and 48), since they might alter the evaluation of the Hebrew text lying behind the Greek addition.154 One rejected reading that deserves some note is variant 23, which inserts after . The expression is common in the prophets, but almost always as a translation of , and never as a translation of . It is probable that in variant 23 is an inner-Greek corruption based on Septuagintal idiom, and perhaps also on the graphic similarity between epi and egw. The Hexaplaric Recension Since all the hexaplaric readings that agree with either MT or LXX have been omitted in the list in Chapter 2, it is not surprising that a number of the readings that remain are significant. Variant 1 apparently reads for (cf. 1:9). Variant 2 bears witness to a free-floating “again” mentioned above. Variant 3 adds after to make verse 10 conform with verse 9 (unlike LXXL, LXXO shows little tendency towards internal consistency at the expense of fidelity to the Vorlage). Variant 8 seems to read instead of . Variant 9 omits the second occurrence of (cf. L 155 LXX ). Finally, variants 10 and 11 are variants to the long addition of verse 21. It is interesting to note that two exact agreements (LXXO variants 3 and 5) and three apparently related readings (LXXO variants 1, 9, and 10) between LXXL and LXXO do exist among the significant variants. Other Possible Hebrew Readings Since these readings, which reflect variants within the ms tradition of the secondary witnesses themselves, were chosen specifically as those which are probably significant, all three of them will be included. With this note, the evaluation of the lexical, grammatical, and stylistic characteristics of the secondary witnesses comes to a close. The majority of the work toward determining the translation technique of each of the witnesses is finished. All that remains is an analysis of such literary and theological factors as can be determined from a careful study of the chapter as a whole, using various available techniques. 154 Since this passage in LXX (the long addition in 3:21) has been found to be significant, and

since the Greek text here preserves a passage not found in Hebrew, all variants to the text of LXX, including those in the partial secondary witnesses, become significant, just as all Hebrew variants are significant by definition. 155 See the previous note.

CHAPTER 4 LITERARY/THEOLOGICAL CHARACTERISTICS OF THE SECONDARY WITNESSES The interest in literary approaches to narrative portions of Scripture has increased tremendously in recent years. Some of these approaches are purely literary, in the sense that they take the final form of the text (or, more accurately, one of the final forms of the text, usually MT) as the basis for analysis. Other approaches combine literary and theological considerations with information gleaned from historical-critical methods, especially form criticism. The variety of literary methods produces complementary and, at times, contradictory results when applied to specific texts (as do the more traditional disciplines), so care must be exercised when using them. For the present study, it is unnecessary to delve into all the details of narrative research and methodology. The issue at hand is the possibility that one or more of these methods might provide information that will alter the evaluation of specific variants as analyzed in the previous chapter, particularly in borderline cases. It should also be noted that a certain amount of overlap exists between the methodology of Chapter 3 and that of the present chapter. In particular, certain stylistic characteristics of the secondary versions have already been identified and considered in the analysis of some of the variants. For example, P’s penchant for substituting the emphatic case plus the relative particle for MT’s construct constructions is an aspect of the style of the translation. V’s variety in lexical selection, especially of conjunctions, is part of V’s translation technique already noted. The theological concerns of T have also been noted and taken into account. The purpose, then, of the present chapter is not to go over old ground but rather to see if some of the newer approaches to narrative as applied to Samuel can shed additional light on the translation technique of any of the secondary witnesses. One of the difficulties of using the newer literary approaches as an aid to the textual analysis of the chapter has already been hinted at, namely, that fact that most literary analyses begin and end with MT. It is true that some do take LXX into account, but few do a comprehensive study of LXX as a complete text in itself.1 A purely literary analysis of P, T, or V in Samuel does not seem to exist, though theological analyses of T as a whole 1 An exception to this generalization is the article by S. D. Walters, “Hannah and Anna.” Though

it deals only with 1 Samuel 1, the insights that Walters raises concerning the tendencies of the translators of LXX will be analyzed and, if appropriate, applied to chapter 3 as well.

164

165

abound. For these reasons, it seems preferable to begin this chapter with an analysis of 1 Samuel 3 in MT. Some of the insights claimed for MT may shed some light on the Hebrew texts that lay before the various translators of the secondary witnesses. Since the focus of this book is textual criticism, no attempt is made to deal with every literary aspect of 1 Samuel 3 that could be noted. Instead, only that information that might provide insight into one or more textual variants will be considered. During the discussion of literary insights related to MT, variants in the secondary witnesses that will be discussed later at the appropriate time are indicated. First of all, some of the authors who have done extensive literary treatments of all or part of 1 Samuel 3 should be mentioned. Shimon Bar-Efrat, in his book Narrative Art in the Bible, presents an approach to analyzing biblical narrative as a literary work of art.2 Though he does not do a comprehensive analysis of 1 Samuel 3, he does refer to various aspects of the chapter periodically. Furthermore, his method may be extended to other parts of the chapter. Bar-Efrat divides his discussion of biblical narrative into five chapters, which deal respectively with narrator, characters, plot, time and space, and style. Peter D. Miscall does a detailed literary study of the entire book of 1 Samuel. He expresses his dissatisfaction with the historical-critical approach to biblical studies, and he proposes his approach as a possible alternative, though he notes that his work is still preliminary.3 Lyle M. Eslinger is a proponent of the close reading method. In his book Kingship of God in Crisis: A Close Reading of 1 Samuel 1-12, Eslinger defines close reading as an attempt “to uncover and describe the intricate reticular connections that unite the narrative (or poem), making it into a singular entity, however complex or devious its plot may be.”4 Unlike Miscall, Eslinger believes that the results of the historical-critical approach must be acknowledged and taken into account, even in a literary analysis of the text.5 Robert Karl Gnuse’s revised dissertation is entitled The Dream Theophany of Samuel, and it primarily treats 1 Samuel 3, discussing comparative literature from the 2 Shimon Bar-Efrat, Narrative Art in the Bible, trans. Dorothea Shefer-Vanson, Journal for the

Study of the Old Testament Supplement Series, no. 70, Bible and Literature Series, no. 17 (Sheffield: Sheffield Academic Press, Almond Press, 1989), 1. 3 Miscall, 1 Samuel, vii. It is perhaps better to speak of approaches rather than approach in Miscall’s case, since he claims to employ a variety of approaches in his examination of the text (ibid., xvi). Though he does not attempt a thorough deconstructive analysis of 1 Samuel, he does make use of deconstructionist language and tactics to understand the text (ibid., xx-xv). 4 Lyle M. Eslinger, Kingship of God in Crisis: A Close Reading of 1 Samuel 1-12, Bible and

Literature Series, no. 10 (Sheffield: JSOT Press, 1985), 40. 5 Ibid., 42. He says,

The close reading of any biblical text must be carried out in the light of the previous historical-critical readings of the text. We have been cued to the tensions, doublets and varying points of view and their location in the narrative by historical criticism. A close reading will have to describe the contextual role of such phenomena if the hypothesis that the narrative can be read as a unity is to be maintained.

166

ancient Near East and the biblical tradition, then turning to a discussion of the chapter within its context in the book of 1 Samuel.6 Gnuse’s method is thus not purely literary, for it uses historical-critical data in its analysis. Robert Polzin’s Samuel and the Deuteronomist: A Literary Study of the Deuteronomic History deals exclusively with the final (Masoretic) form of the text. Though he makes use of historical-critical insights, he stresses the importance of applying these insights to the final form of the text, not just hypothetical earlier forms of the text. He uses the insights of Mikhail Bakhtin and other Russian formalists to illumine the text of Samuel.7 Finally, Moshe Garsiel uses the method of comparative structures to analyze 1 Samuel in The First Book of Samuel: A Literary Study of Comparative Structures, Analogies and Parallels. He, too, acknowledges the contributions of the various historical-critical theories, but he focuses on the final form of the text, which he believes is “for the most part a work of deliberate artistry which deserves consideration on its own account.”8 A number of articles or short studies on 1 Samuel also exist. Uriel Simon, like Eslinger a proponent of close reading, in his article “Samuel’s Call to Prophecy: Form Criticism with Close Reading,” uses a method similar to that of Bar-Efrat to supplement form-critical insights, though he notes his disagreement with many of the results of form criticism.9 John T. Willis believes that 1 Samuel 1-7 is, for the most part, a unified narrative. He discusses a number of stylistic and structural indicators that tie these chapters together.10 Similarly, Michael Fishbane sees the text of 1 Samuel 3 as more than a simple factual report; it is a narrative whose factual content is affected by literary presentation. He discusses how chiasm, verbal linkage, irony, and musicality affect both the form and the meaning of the text.11 Matitiahu Tsevat focuses his attention on one specific stylistic indicator that he believes demonstrates the unity of 1 Samuel 1-4. Taking his initial observations from Benno Jacob, he investigates the use of patterns of repetition 6 Gnuse, Dream Theophany, passim. 7 Robert Polzin, Samuel and the Deuteronomist: A Literary Study of the Deuteronomic History,

part 2, 1 Samuel (San Francisco: Harper & Row, 1989), 1-17. 8 Moshe Garsiel, The First Book of Samuel: A Literary Study of Comparative Structures, Analogies and Parallels (Jerusalem: Revivim, Rubin Mass, 1990), 34. 9 Uriel Simon, “Samuel’s Call to Prophecy: Form Criticism with Close Reading,” Prooftexts: A

Journal of Jewish Literary History 1 (1981): 119-32, especially 119-21. 10 John T. Willis, “An Anti-Elide Narrative Tradition from a Prophetic Circle at the Ramah Sanctuary.” Journal of Biblical Literature 90 (1971): 288-308; idem, “Cultic Elements in the Story of Samuel’s Birth and Dedication,” Studia Theologica 26 (1972): 33-61; idem, “Samuel Versus Eli, I Sam. 17,” Theologische Zeitschrift 35 (1979): 201-12. 11 Michael Fishbane, “I Samuel 3: Historical Narrative and Narrative Poetics,” in Literary

Interpretations of Biblical Narrative, vol. 2, ed. Kenneth R. R. Gros Louis and James S. Ackerman (Nashville: Abingdon Press, 1982), 191-203.

167

(Abzählungen), particularly patterns involving the number seven.12 Unlike the other authors discussed here, Joseph Bourke engages in a good deal of source-critical and redaction-critical work in his article, “Samuel and the Ark: A Study in Contrasts.” However, he sees 1 Samuel 1-3 as basically a unity, and he uses insights into literary structures and patterns in the text to illuminate its meaning, an approach similar to that of Garsiel. Like Fishbane, he sees the story of Samuel’s childhood as a literary interpretation of the real persons and events, conveying the theme of the “triumph of Good over Evil, worked out in this particular episode in Israel’s history.”13 In addition to the studies mentioned in these two paragraphs, other works, including commentaries, will be examined for the light they might shed on 1 Samuel 3. Eslinger notes the similarity between 2:11 and 3:1, but he also notes what he believes are significant differences. Whereas 2:11 reads , 3:1 reads . Eli is no longer called “the priest” in 3:1, so Samuel is no longer said to serve (the face of) Eli but rather to serve before Eli14 (LXX variant 70; P variants 2, 52; T variants 1, 2). Gnuse says that the phrase “before the lamp went out” in verse 3 indicates that the lamp is not to be identified with the lamp that never went out, referred to in Ex 27:20 and Lev 24:2. Eslinger, on the contrary, does identify these lamps, suggesting that the fact that the lamp is still burning indicates that God had not yet cut off relations with Israel on account of Eli and his sons. Fishbane goes further, seeing in the lamp either a symbol that Eli was not yet dead or blind or that the spiritual illumination of the time, though diminished, was not yet entirely extinguished. Merrill says that the clause is a veiled allusion to the king, since David is called the “lamp of Israel” in 2 Sam 21:1715 (LXX variant 9). Bar-Efrat discusses the use of repetition in narrative, noting that repetition is common in biblical narrative. Though the purpose of repetition varies from case to case, a frequent reason for repeating material is to stress some matter of importance in the story.16 12 Matitiahu Tsevat, “Abzählungen in 1 Samuel 1-4,” in Die Hebräische Bibel und ihre zweifache Nachgeschichte: Festschrift für Rolf Rendtorff zum 65. Geburtstag, ed. Erhard Blum, Christian Macholz, and Ekkehard W. Steggemann (Neukirchen-Vluyn: Neukirchener Verlag, 1990), 207-14. 13 Joseph Bourke, “Samuel and the Ark: A Study in Contrasts,” Dominican Studies 7 (1954): 73-

103, quotation from p. 87. On the same page he says that the narrative of Samuel’s childhood “is a remarkable example of how a Hebrew writer can use literary form to convey meaning.” 14 Eslinger, Kingship, 145-46. 15 Gnuse, Dream Theophany, 153; Eslinger, Kingship, 148-49; Fishbane, “I Samuel 3,” 197

(more accurately, Fishbane sees the text as referring to both of these options; 3:1-2 is a “remarkable bivalent image”); Arthur L. Merrill, “I Sam 1-12: A Traditio-Historical Study,” Ph.D. diss., University of Chicago, 1962; quoted in Gnuse, Dream Theophany, 153-54. Cf. also Polzin, Samuel and the Deuteronomist, 52-53, who draws a parallel between in verse 3 and in 2 Sam 21:17, used with reference to David. 16 Bar-Efrat, Narrative Art, 116.

168

Several variants deal with the presence or absence of an element found elsewhere in the narrative (e.g., P variant 52; T variants 53 and 63; V variant 92), but the application of Bar-Efrat’s insights concerning repetition is unclear in most of these cases. In one particular set of variants, however, the factor of repetition needs to be investigated, namely, those variants that deal with God’s four-fold call of Samuel in verses 4-10. In this context, Schulz’s observation that repetition in Samuel regularly involves variation should be taken into account17 (LXX variants 12, 18, 20, 75, 76, 79, 80, 81, 82, 83, 84, 89, 90, 91; P variants 10, 11, 49, 71). Eslinger comments on another repetition, namely, the repeated in verses 3 and 7. He suggests that the meaning of the two-fold use of in verse 7 is conditioned by the use of in reference to the lamp in verse 3 (LXX variant 86). Willis characterizes yet another repetition—“my son,” in verses 6, 9, and 16—as connectives in the narrative18 (LXX variants 83, 89; V variants 84, 92, 97, 99, 104, 107, 108, 113, 118). Simon discusses Samuel’s failure to mention God’s name in verse 10, comparing it to Eli’s aversion to speaking of God in verse 17. Polzin explains the omission differently. The first three chapters of 1 Samuel is, in a sense, a parable of the first readers’ situations in the postexilic period, when it was no longer possible to converse directly with God. “The situation of a partially absent LORD, one who hears humans and whom humans hear but do not—or cannot—respond to directly, is indicative of these opening chapters, but expressed in a special way in chapter 3: precisely where Samuel has an opportunity directly to speak to the LORD, he is commanded by Eli to express a willingness simply to listen—and even then Samuel unaccountably omits the LORD’s name”19 (LXX variants 42, 45; P variants 84, 108, 111; T variant 32; V variant 40). 17 Alfons Schulz, “Narrative Art in the Books of Samuel,” in Narrative and Novella in Samuel: Studies by Hugo Gressmann and Other Scholars 1906-1923, ed. David M. Gunn, trans. David E. Orton, Journal for the Study of the Old Testament Supplement Series, no. 116, Historic Texts and Interpreters in Biblical Scholarship, no. 9 (Sheffield: Sheffield Academic Press, Almond Press, 1991), 148-51. Cf. also Simon, “Samuel’s Call to Prophecy,” 125-27; Gnuse, Dream Theophany, 154; Polzin, Samuel and the Deuteronomist, 40. Willis sees “clearly defined gradations” in the various calls of Samuel; Willis, “AntiElide Narrative Tradition,” 293-94; cf. Budde, Bücher Samuel, 27. Jacob Licht, Storytelling in the Bible (Jerusalem: Magnes Press for the Hebrew University, 1978), 55, disputes the idea that the subtle variations in the four calls are significant as conveyers of meaning:

Such subtleties, however, should not lead us to the mistaken conclusion that repetitive patterns with variations are built as exercises for ‘close reading.’ The basic and truly relevant feature is the pattern itself, which is not subtle at all; the subtleties, where one does find them, are accidental elaborations. Garsiel classifies the repeated calls of Samuel as an example of a “recurrent structure”; Garsiel, First Book of Samuel, 31. 18 Eslinger, Kingship, 151 (cf. also Fishbane, “I Samuel 3,” 198-99); Willis, “Anti-Elide

Narrative Tradition,” 294. 19 Simon, “Samuel’s Call to Prophecy,” 132, n. 15; Polzin, Samuel and the Deuteronomist, 52. Cf. also Schulz, “Narrative Art,” 148-51.

169

While discussing the points of comparison and contrast between the Eli’s sons and Samuel, Garsiel notes the repeated use of the verb in chapters 2 and 3. The sons of Eli did not know the Lord (2:12), Samuel did not yet know the Lord (3:7), Eli knew his sons’ sins (3:13), and all Israel knew that Samuel had been established as a prophet to the Lord (3:20)20 (LXX variants 94, 95). Bar-Efrat says that flashbacks in the narrative, though relatively rare, are important when they occur. They serve to introduce new characters, explain the actions or utterances of people or the origin of situations, insert information about a character or event outside the main story-line, synchronize events, and provide emphasis. God’s message to Samuel concerning Eli’s family is related to the prophecy of 2:27-36, and it may be that flashback is a factor in 3:13-14. On the other hand, Polzin emphasizes the repeated use of verbs in the present or future tense in verses 11-14, which suggests that “this story from Israel’s past has present meaning for the author’s present audience”21 (LXX variant 30; T variant 24; V variant 28). In his discussion of the use of characters in the narrative, Bar-Efrat notes: The characters’ actions are also the building blocks of the plot, though the protagonists should not be regarded merely as a means for getting the story going. The narrative is concerned not only with the events which occur but also with the people involved.22 He points out that the reference in 3:15 to Samuel opening the doors of the house of the Lord after receiving the vision is a way of characterizing Samuel. “The tremendous event—the revelation of God—does not turn his head (even though ‘the word of the Lord was rare in those days,’ ‘there was no frequent vision’), he does not become conceited but continues fulfilling his duties as usual.”23 Bourke sees a different significance in verse 15b, pointing to it as the beginning of the seventh scene in chapters 1-324 (LXX variant 96). In his discussion of the repetition of words as a stylistic device, Bar-Efrat notes that in 3:17, the same words appear at the beginning and at the end of Eli’s speech to Samuel, forming an envelope. The purpose of such a framework, he says, is to provide emphasis. Alongside this analysis, Miscall points out the repeated use of words related to hearing in 20 Garsiel, First Book of Samuel, 40-41. 21 Bar-Efrat, Narrative Art, 175-79; Polzin, Samuel and the Deuteronomist, 51. 22 Bar-Efrat, Narrative Art, 77. 23 Ibid., 79. Polzin’s view of Samuel is quite different; Polzin, Samuel and the Deuteronomist,

50-51. Cf. Walter Brueggemann, First and Second Samuel, Interpretation: A Bible Commentary for Teaching and Preaching, ed. James Luther Mays, Patrick D. Miller, Jr., and Paul J. Achtemeier (Louisville: John Knox Press, 1990), 25, who sees Samuel as “perfectly responsive (v. 10). Moreover he is fully supported by Eli, who is now dependent upon him (v. 18).” 24 Bourke, “Samuel and the Ark,” 81.

170

verses 11-14, an observation that might be relevant as well in verse 1725 (LXX variants 41, 42, 43, 44, 45, 46, 100, 101; P variant 31; V variants 39, 40). Eslinger discusses the structure of verse 21, noting that MT contains two parallel sentences with a prepositional tag added to the second. This tag, , standing outside the parallelism, is thereby “foregrounded.” Furthermore, “the repetition of ‘in Shiloh’ emphasizes that the renewal through Samuel reaches to the very heart of the damages done by the Elides, for it was in the cult at Shiloh that they carried on their priestly malpractise.” Tsevat, in his examination of patterns of repetition, notes that in the singular, denoting the word of Yahweh, occurs seven times in chapter 3, which is a distinct unit. He believes the significance of this patterned repetition is that it stresses that it is through the word of Yahweh given to him that Samuel becomes a prophet. In Fishbane’s discussion of the chiastic pattern in chapter 3, he mentions a “micro-chiasm” that begins with Samuel in verse 1, followed by a reference to the national situation of divine absence; in verse 21, the divine presence comes to all Israel, because Yahweh was revealed to Samuel at Shiloh, “by the word of the Lord”26 (LXX variants 104, 105). Having discussed some of the aspects of various literary investigations of MT that are related to variants in one or more of the versions, it is time to turn to those versions for further analysis of the variants. Septuagint Relevant Literary, Historical, and Theological Data Present in the Translation27 As mentioned above, most literary studies of 1 Samuel are based entirely or at least predominantly on MT, with perhaps an occasional glance toward LXX. However, in his article “Hannah and Anna: The Greek and Hebrew Texts of 1 Samuel 1,” Stanley D. Walters does attempt to characterize certain aspects of LXX (specifically, ms B, though he notes other LXX traditions/recensions as well) alongside those of MT. Though the article does not deal at all with chapter 3, the conclusions he reaches about the concerns of the translators of LXX for chapter 1 may also apply to chapter 3. Walters does not attribute all the variations between MT and LXX to the Greek translators; instead, he allows for 25 Bar-Efrat, Narrative Art, 216; Miscall, 1 Samuel, 25. 26 Eslinger, Kingship, 159; Tsevat, “Abzählungen,” 210; Fishbane, “I Samuel 3,” 194. 27 Though not as obvious or prevalent as literary and theological concerns, the translators may

have had other concerns which would fall into the broad category of historical data. Such data would include historical references in the text that differ from those of the original (e.g., references to the Greeks and the tower of Babel in Isa 9 and 10 LXX), mention of contemporary customs (or failure to mention outmoded customs), or allusions to the religious beliefs and practices of the translators, the recipients of the translation, or their neighbors. Cf. Staffan Olofsson, The LXX Version: A Guide to the Translation Technique of the Septuagint, Coniectanea Biblica, Old Testament Series, no. 30 (Stockholm: Almqvist & Wiksell, 1990), 1-5.

171

differences in the Vorlage of LXX.28 However, he does see the Greek translators as creative in several ways. In his discussion of in 1:11, he points out that the word is a hapax in LXX and is uncommon even in classical Greek. Since more common equivalents were available, it is apparent that was chosen because it has the connotation of necessity or obligation, something which must or ought to be given.29 He subsequently relates the use of and in 1:28 to in 1:11, noting that the former words connote something lent rather than given.30 The relevance of these insights to the text of chapter 3 is that they demonstrate that the translators did not translate their Vorlage mechanically with little understanding of the text. Though they might not always have captured all the subtleties of the Hebrew text, and though the result does not resemble classical Greek style, the translation they produced is a literary work in its own right with meaning and internal consistency (at least for the most part) of thought. This observation challenges the textual critic to read LXX more carefully for internal clues that might explain apparent variations from MT. It does not, however, negate LXX as an important witness to the common underlying text. Since matters of literary, theological, and historical importance to the translators are much more difficult (if not impossible) to quantify than lexical and grammatical matters, it will be helpful before analyzing the variants from a literary standpoint to have some more detail about the literary and theological concerns of the translators, as identified by various scholars. In addition to Walters, other scholars have also addressed the literary and theological concerns of the translators of LXX. Frankel notes that LXX often has double translations for a single Hebrew word, phrase, or verse. However, many of these doublets come from Hebrew sources, including marginal glosses and translations of both kethib and qere forms. Characteristics of LXX that come primarily from the translators include textual improvements, interpolations from other parts of the OT, rhetorical additions, and transpositions of verses.31 Unfortunately, Frankel does not always distinguish which characteristics are found in which books (e.g., his examples of rhetorical additions come primarily from Proverbs). Swete, in his Introduction to the Old Testament in Greek, lists several characteristics of the translation. The translators frequently add words to clarify the meaning, interpret difficult words or Hebrew technical terms, occasionally avoid 28 Cf. S. D. Walters, “Hannah and Anna,” 394-96. In cases of excess text in either MT or LXX, it is not always clear whether he thinks the variation arose in Greek or was based, at least to an extent, on a divergent Hebrew reading. His desire is primarily to show the relative independence of the two versions of the story found in MT and LXX, not to indicate in every instance whether the difference arose in Hebrew or Greek. 29 Ibid., 399. 30 Ibid., 406-7. 31 Z. Frankel, Vorstudien zu der Septuaginta (Leipzig: Fr. Chr. Wilh. Vogel, 1841), 70-85.

172

anthropomorphism, sometimes follow Jewish halakic or haggadic traditions, employ lexical variety, and exhibit freedom in translating metaphorical phrases and other figures of speech. In addition, the translators also sometimes misunderstand the text or produce a faulty translation as a result of mechanical errors (metathesis, parablepsis, etc.).32 Perhaps the most recent work devoted to the translation technique of LXX is that by Staffan Olofsson. In addition to many of items discussed above, Olofsson notes the translators’ religious and cultural concerns which often surface in what at first appear to be unusual or inconsistent renderings.33 These observations deal mostly with isolated phenomena within the LXX text and do not address inner-LXX matters such as schematization or harmonization. Simon characterizes the text of LXX as smoother and more uniform than MT,34 but the extent of the translators’ tendency toward uniformity in parallel or similar passages requires further investigation. Final Translation Technique The description of the translation technique in this section may be described as “final” in the sense that the literary and theological concerns of the translators are added to the description of the translation technique. However, the evaluation of the variants in the next section may require more modifications to the following description, just as the elimination of variants in the preceding chapter led to a better understanding of the translation technique at that stage. In chapter 3, the translators’ approach to the text was described in some detail, so it may suffice here to summarize those results. The translators were concerned with literalness in most areas for which Greek is suited to deal with Hebrew categories (excluding gender for non-human referents), with the exception of the Hebrew articles and definiteness, which they did not render with the same degree of consistency as the other categories. The translators show little concern for exact renderings of Hebrew compound words, but they are quite concerned to follow the Hebrew word order. They also appear to have refrained from adding or omitting quantitative elements to a great extent. The data from the previous section indicates that, while the translators follow their Hebrew Vorlage quite closely, generally speaking, they do exhibit a measure of freedom within their self-imposed parameters. In particular, they are free to make lexical choices that seem best suited to the context, and they show a concern for theological and cultural 32 Swete, Introduction, 325-30. 33 Olofsson, The LXX Version, 1-5. 34 Simon, “Samuel’s Call to Prophecy,” 131, n. 10. Cf. also Johann Cook’s analysis of Genesis 1, where he finds “a significant harmonizational pattern” in LXX; Johann Cook, “Genesis 1 in the Septuagint as Example of the Problem: Text and Tradition,” Journal of Northwest Semitic Languages 10 (1982): 32.

173

propriety. They also at times seem to have felt the need for greater consistency than was found in the Hebrew. However, this issue will be discussed more fully in the next section. Elimination of Variants Eslinger believes that the differences between 2:11 and 3:1 in MT signify Eli’s downgraded position in respect to Samuel by the time of Yahweh’s call of Samuel.35 However, his observation that Eli is no longer called “the priest” in 3:1 does not apply to LXX. In fact, the text of LXX calls into question all of his observations about these verses, for LXX in 2:11 suggests a slightly different Hebrew Vorlage. Whether or not there is a Vorlage difference in 2:11, the line drawn between and seems artificial; in 2:11 is surely equivalent to in 3:1,36 especially since no separates the mention of Yahweh and Eli (contrary to his statement that Samuel serves both Yahweh and the “face of Eli”). The failure of MT to mention , then, might just be an attempt to avoid exact repetition. The fact that the text of LXX is different in the two verses indicates that the translators were not trying to make the verses identical for the sake of consistency (unlike the translators of P in 3:1), so the addition of “the priest” probably did originate in Hebrew. Thus, the previous evaluation of variant 70 as significant stands. Of the various analyses of the lamp in the phrase “before the lamp had gone out” in 3:3, those of Merrill, Polzin, and Fishbane, that the lamp is a veiled allusion to David or his dynasty (Merrill and Polzin) or to Eli (Fishbane), have some bearing on variant 9. If the lamp does represent David, then LXX’s rendering “before the lamp was made ready” would seem to convey the meaning more clearly than MT’s “before the lamp had gone out,” since David had not yet assumed office—or even been born—and so was in no danger of “going out.” On the other hand, if either Eli (dealing with the situation in Samuel’s day) or David’s dynasty (dealing with the situation in the days of the original readers of Samuel) is in view, the reading of MT is preferable. Though the first option (David) might favor finding the origin of variant 9 in a Hebrew text, the second (Eli or David’s dynasty) does not change the evaluation given in Chapter 3.37 Therefore, 35 Eslinger, Kingship, 145-46. He says,

In 2.11, the lad serves Yahweh and “the face” of Eli, both of which are modified by the accusative particle et. In 3.11, however, Samuel serves only Yahweh, doing so “before” or “in the presence of” (lipnê) Eli. The reason for this change is revealed by the final variant: Eli is no longer called “the priest” in 3.11; he is simply called Eli. Samuel no longer serves Eli “with” Yahweh because Eli has lost the rank and title of priest. Already Eli takes a back seat to Samuel in the priestly service. 36 Cf. BDB, s.v. “

.” Fishbane finds in the similarity of vocabulary in 2:11, 18; 3:1 evidence of continuity, rather than contrast; Fishbane, “I Samuel 3,” 194-95. 37 It is also plausible that the translators were influenced by Ex 30:8, which states that the priest was to set up ( hiphil) the lamps at the time of the evening sacrifice. It seems likely that the prohibition against letting the lamp of the sanctuary go out (Ex 27:20-21; Lev 24:1-4) did influence the translators, who understood the passage to be referring to the same lamp, rather than other lamps used for

174

variant 9 remains nonsignificant. The largest set of variants addressed by the various literary readings of Samuel are those that deal with God’s repeated call of Samuel. A perusal of the four calls indicates variations among each of the calls in both MT and LXX. Willis applies his claim of “clearly defined gradations” in the calls only to the descriptions of God’s actions and words in these verses. However, when one looks at the actions and words of both Samuel and Eli, they, too, show variation from call to call. Simon does see significance in the variation from call to call of Samuel’s and Eli’s words and actions: the first time Samuel runs to Eli; the second time he gets up and walks (indicating less enthusiasm; the extra verb “and he arose” also slows the pace of the narrative to stress Samuel’s reticence); the third time is “nearly identical to the preceding one.” Similarly, Eli’s second response adds the vocative “my son,” demonstrating that Eli perceived Samuel’s lack of haste but understood it.38 Had Simon analyzed the text of LXX rather than MT, he could have strengthened his argument somewhat, for Samuel’s reaction to the third call is not nearly identical to the previous one. Samuel’s response to the second call in LXX does not include the statement that he arose, but the additional does slow the pace of narrative in support of the replacement of by . Only on the third occasion is added, slowing Samuel’s pace even more. Eli’s command to Samuel to go back and lie down and the description of what Samuel does in response (variants 12, 18, 20) is also repetitive. One could argue that LXX shows a tendency toward schematization in its description of these events—since MT uses twice and three times, while LXX uses four times and only once—but it must also be noticed that LXX has in the description of the third call, whereas MT has every time. Thus, LXX is no more uniform than MT in these variants. Finally, concerning the number of times “Samuel” appears in each of the four calls, LXX’s pattern of two, two, one, zero, respectively, in the four calls is not more uniform than MT’s pattern of one, one, one, two (see above, p. 102). The variations among the calls in LXX belie Simon’s claim of “the conspicuous tendency of the Septuagint translators to compare repetitions with one another for the sake of comprehension and clarity,” thus producing a “smoother, uniform version.”39 The lack of uniformity among the calls in LXX supports the earlier contention that the variants are probably based on variations in the Vorlage. 40 illumination at night. ready.”

would then have to be translated “was set up” instead of “was made

38 Simon, “Samuel’s Call to Prophecy,” 126-27. 39 Ibid., 131, n. 10. 40 The first and second accounts of Eli’s response are identical in LXX, since it omits “my son” in the second account (LXX adds on both occasions). However, LXX does add to the third account, where it is missing in MT. It is hard to see that LXX is any more uniform in these verses than MT is, especially in light of the “nearly identical” second and third responses of Samuel. The floating “again” of

175

The plurals in variants 24 and 25 should also be considered in the context of a discussion of repetition and schematization. The difference in sentence structure between MT and LXX has already been noted in the discussion of these variants. On the other hand, the plural in variant 24 could be explained as a schematization by the translators regarding the words and , since they are translated as plurals in three similar contexts (3:17, 18, 19; cf. also the plural in 3:12). Several of these could be explained as the result of the LXX rendering of with the plural , 41 but not in 3:17 (variant 43). The LXX translators do not show a consistent tendency toward schematization in the chapter, but it does appear that they preferred to render Hebrew collective singulars with Greek plurals, especially where and are involved. Thus, a deeper look at the whole chapter suggests that variants 24 and 25 should not be considered significant after all. The other instances of repetition in the chapter probably do not affect earlier analyses of the variants. Whether occurs once (LXX) or twice (MT) in verse 7, its earlier appearance in verse 3 could still be called a conditioning factor. Willis’s statement that “my son” serves as a connective in the narrative is interesting, because he appears to assume the originality of both in verse 6 (variant 83) and in verse 9 (variant 89), in addition to its appearance in verse 16 (in both MT and LXX); neither version has three occurrences of the word/phrase. Thus, he apparently supports the analysis of variant 89 as significant. Neither Willis’s nor Eslinger’s observations affect the evaluations of the variants already given. Polzin’s explanation of the reason Samuel neglected to mention Yahweh when he finally answered his call is interesting, but not convincing. Better is Simon’s comparison with Eli’s aversion to mentioning Yahweh’s name in verse 17, but better yet may be Schulz’s observation that repeated events regularly employ variation.42 Either way, LXX is identical to MT with regard to Samuel’s response. If Simon is right about Eli’s aversion to saying Yahweh’s name (or if the narrator for some other reason wanted to avoid it) the passive voice of the participles in LXX (variants 42 and 45) removes the reader even further from reference to Yahweh, so the earlier evaluation of these variants does not change. Garsiel’s observation that the verb is repeated several times in chapters 2 and 3 is instructive, and his arguments must be considered when attempting to determine the original text of 3:13 (variants 94 and 95). However, the overall difficulty of the verse in both MT and LXX suggests a common disruption in Hebrew behind both versions, so MT (variant 79) is probably independent of thereof) is not affected by the above discussion. 41 See above, pp. 93-94. 42 Schulz, “Narrative Art,” 148.

(variant 82), and its originality in verse 6 (or lack

176

these variants should continue to be considered significant. Bar-Efrat emphasizes the importance of flashbacks in biblical narrative, and 3:13-14 seems to be such a flashback, referring to 2:27-36.43 Polzin’s observation that present or future tense verbs abound in 3:11-14 does not account for in verse 12 in a phrase that refers back to the same verses in chapter 2. Thus, variant 30 continues to be significant. In his discussion of characterization, Bar-Efrat notes that MT in 3:15 implies that the vision Samuel received the night before did not distract him from his duties or cause him to become conceited. The reading of LXX, reflected in variant 96, strengthens this picture of Samuel’s character, for not only does he open the doors of the house of the Lord the next morning, but he arises early to do so. The phrase “and he arose early ” is common in the OT, and it indicates that the person concerned hastened to undertake a certain task.44 Bourke’s division of 1 Samuel 1-3 into scenes would likewise be strengthened if the phrase “and he arose early” is accepted, for he identifies the beginning or end of these scenes on the basis of “conventional phrases of introduction and conclusion,” one of which is “and they arose early” (1:19, scene two).45 This evidence strengthens the evaluation of variant 96 as significant. Bar-Efrat’s comments about the significance of an “envelope” in verse 17 (MT) is a helpful explanation of the phenomenon and deserves careful consideration. However, as already noted, he also discusses variation in repetition as a narrative tool. If words related to hearing are common in verses 11-14 by design, as Miscall believes, examining the text of LXX would reveal another occurrence in verse 17 (variants 100 and 101). These contrasting observations suggest that no conclusive statements concerning the variants in verse 17 can be based on literary analysis alone. Finally, the discussions of Eslinger, Tsevat , and Fishbane on verse 21 need some comment. If one analysis of the “tag” suggests that it is “foregrounded” because it stands outside the parallelism of the verse, another analysis might suggest it was secondary for the same reason. The repetition of “in Shiloh” may indeed be emphatic, but the variation in spelling in MT is puzzling and suggests a Hebrew rather than Greek origin for the variant.46 Tsevat’s observation that singular appears seven times in the chapter does not hold in LXX, since variant 104 omits a phrase that contains the word. However, once this reference is omitted, one could argue that 4:1a belongs to the unit 43 So R. W. Klein, 1 Samuel, 30; cf. Edmund Kalt and Willibald Lauck, eds., Herders

Bibelkommentar: Die Heilige Schrift für das Leben erklärt, vol. 3/1, Die Samuelbücher, by Peter Ketter (Fribourg: Herder & Co., 1940), 28. 44 Bar-Efrat, Narrative Art, 79. On the phrase in Gen 22:3, cf. ibid., 80. 45 Bourke, “Samuel and the Ark,” 81. 46 Tsevat, following Tur-Sinai, suggests reading the word as word of Yahweh; Tsevat, “Abzählungen,” 210, n. 10.

, “whenever he consulted” the

177

containing chapter 3, and the seventh would appear.47 In fact, since in 4:1a refers to the (prophetic) word of Samuel, one could further argue that the word of Yahweh has become identified with the word of Samuel; that is, Samuel has been confirmed as a prophet by the narrator as well as by all Israel. It seems, then, that the contention that this particular group of seven is significant is questionable.48 Fishbane’s micro-chiasm in verses 1 and 21 begins with Samuel, but, as he sees it, it ends with the phrase “because YHWH was revealed to Samuel at Shiloh through the oracle of YHWH.”49 The final phrase “through the oracle of YHWH” could be seen as an intrusion into the chiasm. However, if the phrase is omitted with LXX (variant 104), then the end of the chiasm matches the beginning more exactly.50 Thus, literary analysis does not necessitate change in the evaluations of the variants given earlier. The only variants affected by the application of literary principles are variants 24 and 25, which are no longer considered significant. The elimination of these variants will change tables 10 and 15 slightly, as follows: Table Changes 10 deviation factor: 1.78 15 deviation factor: 0.06 The final translation technique as described in the previous section remains accurate, except that the translators’ concern for consistency between parallel and similar passages, resulting in harmonization and patterns of schematization, seems to have been overstated by Simon, at least as far as 1 Samuel 3 is concerned. Peshitta Relevant Literary, Historical, and Theological Data Present in the Translation Though fewer studies exist of P than of LXX, several characterizations of P are available to the textual critic, in addition to those of Deist and Schwartz, cited in Chapter 3. Driver says that the translation, though not strictly literal, “represents fairly the general 47 Ibid. Tsevat is forced to argue that 4:1a does not belong to the unit containing chapter 3 in

order to maintain his group with seven elements. 48 The issue of whether patterns of repetition that do not seem to be emphasized in the text can be used to demonstrate the cohesion of a unit of text is problematic, since it is easy to find such patterns, particularly if one also looks at numbers like ten, twelve, and forty, as Tsevat suggests. Thus, e.g., appears fourteen times (= 2 x 7) in the chapter; also, if the omissions in P variants 122 and 124 are considered to be the result of parablepsis and graphic confusion, as discussed in Chapter 3, then the word (“Lord”), an unquestionably important word, appears twenty-one times (= 3 x 7) in the chapter. 49 Fishbane, “I Samuel 3,” 194 (italics his). 50 Even if part of variant 105 is original, the chiasm can be maintained. If the first part of the variant ( 1° 2°) is omitted as a doublet of verse 20, the remainder of the verse could be considered original, but connected with the narrative of chapter 4, the loss of the ark, rather than with chapter 3.

178

sense of the original.”51 P has a “Jewish element,” particularly evident in its use of the preposition in a manner similar to that of T. It frequently adds short explicatives to the text, and it occasionally gives paraphrases.52 In his study of P in the Pentateuch, Cook finds that the influence of LXX on P has been overestimated in the past, though some possibility of contact remains. Moreover, most of the connections between P and T can be attributed to a common Jewish exegetical background rather than direct influence.53 As far as original elements in P are concerned, the translators were concerned to produce a coherent text, so they sometimes specified subjects or objects or added other obviously lacking material. They also occasionally harmonized one passage with another, but in general they can be characterized as following their Vorlage carefully.54 The comments concerning the translators’ tendency to add epexegetical elements has been confirmed for 1 Samuel 3 as well by the previous analysis. A comparison of 2:11 with 3:1 (variant 52) also shows some tendency toward harmonization. Bar-Efrat discusses sound and rhythm as stylistic devices in MT,55 and the same applies in P. For example, 3:1-2 uses the similar-sounding words (“rare”) and (“heavy”), and 3:2-3 have (“was extinguished”) and (“lying,” twice) in the same context. These observations suggest that another aspect of the translation is an aesthetic sensitivity. Final Translation Technique The conclusions reached in Chapter 3 concerning the translation technique of P may be summarized as follows. The translators were generally faithful to their Vorlage in rendering most lexical and grammatical categories, though they show some tendency toward more idiomatic Syriac readings in regard to the use of the emphatic state (though not as much as might be anticipated). Although P shows a greater concern for rendering Hebrew compounds exactly than does LXX, only about three out of four are so rendered. Like LXX, P follows the word order of the Hebrew meticulously. Finally, the differences in quantitative representation demonstrate that clarity in the translation was more of a factor than absolute fidelity to the Vorlage. The translators use the Syriac relative particle and constructions with the anticipatory pronoun freely. The preceding section indicates that the translators of P were concerned with readability and clarity as much as with accuracy. The tendencies to add explanatory material, to harmonize, and to produce an aesthetically pleasing translation are evidence of 51 S. R. Driver, Notes on the Books of Samuel, lxxii. 52 Ibid., lxxi-lxxvi. 53 Johann Cook, “Composition of the Peshit@ta, 153-64. 54 Ibid., 153, 167-68. 55 Bar-Efrat, Narrative Art, 200.

179

their concern. They also share certain theological concerns with other translations, especially T. In particular, they often use the preposition in a manner similar to T, and they occasionally display other theological concerns in common with Jewish rabbinic circles. Elimination of Variants Eslinger’s observations about the variations in 3:1 as compared with 2:11 do not apply to P, since the two verses are identical. In fact, as noted above, the translators of P sometimes engaged in harmonization of similar passages. The evaluation of variants 2 and 52 as nonsignificant stands, despite the addition of in LXX. The repeated calls of Samuel verses 4-10 produce some variants in P (variants 10, 11, 49, 71), though not as many as in LXX. P generally agrees with MT in these verse, in particular, in the number of times “Samuel” appears in each call. However, P does show some variation, much of which, unlike the case of LXX, is definitely in the direction of harmonization or consistency of expression. For example, whereas MT uses both and when describing Eli’s orders and Samuel’s reactions, P consistently uses , and while LXX uses both and , P uses only . These findings confirm that variants 10, 11, and 71 are nonsignificant. They say nothing about variant 49, a variation in word order, though, so that variant continues to be significant. The tendency toward consistency among similar passages noted in 3:1 and in the calls of Samuel also affects the evaluation of variant 84. The explanations of Simon and Polzin do not apply to P, since Samuel does say exactly what Eli told him to: “Speak, Lord, for your servant is listening.” This variant, then, is another example of the tendency of the translators toward consistency of expression in parallel passages, and it remains nonsignificant. Both Driver and Cook have pointed out the tendency of P to make explicit what is implicit in the text, particularly in regard to specifying the subject or the object of a sentence. This tendency is evident in variants 108 and 111 (where again the explanations of Simon and Polzin do not apply to P) and in variants 77, 106, 110, and 115. All of these variants remain nonsignificant. Bar-Efrat’s comments about the use of an “envelope” as a stylistic device raises the issue of the use of the plural at the end of 3:17 (variant 31), whereas the singular appears at the beginning of the verse. The only difference between the singular and the plural in the text of most mss (which have diacritical points, but no vowels) is the two dots that indicate the plural, called sey m This and other diacritical marks appear in most mss, but their use is not consistent. In fact, one group of mss, 12a1fam (see above, p. 53), does not have sey m here. On the other hand, the use of a plural at the end of the verse may be related to the preceding it. In either case, the evidence suggests that the

180

variant arose in Syriac, either as a result of textual disruption (misplaced sey m ) or out of a concern for clarity (association with ), so the evaluation of the variant as nonsignificant does not change. A literary analysis of P does not change the evaluation of any of the variants, so the final translation technique given above remains valid. Targum Relevant Literary, Historical, and Theological Data Present in the Translation The difficulty of characterizing T as literal or non-literal has been discussed above, but it will be helpful at this point to remember that T is a mixture of literal and midrashic renderings. In addition to the descriptions of Díez Macho and Le Déaut already given in Chapter 3, those of other scholars may be noted. Metzger lists three characteristics of the various targums: “(a) a tendency to avoid direct reference to the ineffable name of God, frequently by the use of the word M mra (“the Word”); (b) in passages referring to God, anthropomorphisms and anthropopathisms are usually avoided and the narrative is recast; (c) some of the Targums include longer or shorter stories (similar to the Midrashim) that serve to illustrate the scripture text.”56 Sperber believes that the present form of the official targums (Onkelos and Jonathan) is the result of a combination of the work of two schools: one concerned with a basically literal translation (though deviating from a strict rendering for the sake of clarity or to give an explicit rabbinic interpretation) and the other inclined to render the text freely, even paraphrastically, and to insert midrash and haggadic material at will.57 Of importance for the present study is that 1 Samuel 3, as a narrative passage, falls in the first category. Levine also discusses the characteristics of the targums. The translators were concerned with literal interpretations, though not necessarily literal renderings. Thus, both halakic and haggadic material incorporated in the targums are designed to elucidate the literal (or accepted) meaning of the passage (often through application in halakic passages). He lists some of the theological motives that result in deviation from a literal rendering, including reverence for God, incorporation of recent interpretations (e.g., the afterlife), and desire to insert exemplaristic legends to illustrate a point.58 The previous discussion of the variants in Chapter 3 has also identified several aspects of T’s theological concerns. Among the theological concerns of the translators are the following: avoiding the impression that God comes in direct contact (either ocular or aural) with humans, reverence for God (often called anti-anthropomorphisms), preserving 56 Metzger, “Versions, Ancient,” 750. 57 Sperber, Bible in Aramaic, 4b:3. 58 Levine, Aramaic Version, 37-42.

181

the sovereignty of God, and avoiding situations that would violate the halakic traditions (especially those preserved in the Mishna). Some of the literary characteristics of the translation (which may also have some theological motivation) are the consistent rendering of both and by and the tendency to render singular words for sin by plurals. The translators also frequently clarified obscure words or phrases,59 and they often used constructions containing the relative particle, which has no equivalent in biblical Hebrew. Final Translation Technique Generally speaking, T shows a greater degree of consistency in rendering both lexical and grammatical categories than do the other versions. The translators were concerned to render literally Hebrew compounds as well as word order. When the variants related to T’s theological concerns and those that deal with expressions using the relative particle are eliminated from the quantitative variants, T’s rendering of the Hebrew is quite literal. The large number of theological renderings, though, must not be forgotten in a description of the translation technique. Since the text of T is so close to that of its Vorlage (with the exception of the renderings based on theological concerns), little can be said about the literary concerns of the translators. The only purely stylistic difference between T and its Vorlage appears to be its preference for constructions that use the relative particle, though this use is not consistent. As already noted, the preference for the divine name and the tendency to use plural forms of words for sin may have some theological reasoning behind them. The theological concerns of the translators are manifold, but many have already been spelled out, especially those that are relevant to 1 Samuel 3. Most deal in some way with a concern to stress the distance between God and humankind and thus preserve the reverence that people should have for God. One specific aspect that has not yet been mentioned in this context is the tendency to use the preposition instead of before or, less 60 frequently, before references to people. Elimination of Variants Because of the nature of T, theological concerns were dealt with in Chapter 3 for two reasons: (1) in order to avoid a totally erroneous picture of T’s translation technique because theological renderings were not properly accounted for, and (2) because, like some of the stylistic concerns in the other versions, the theological concerns of the translators of T are generally easy to identify. Thus, no theological renderings remain to be 59 The comment by Churgin bears repeating: T is concerned “to render intelligible to the fullest possible degree that which is obscure”; Churgin, Targum Jonathan, 78. 60 M. L. Klein, “The Preposition

,” 502-7.

182

considered, and since T possesses few stylistic characteristics of its own, few variants need to be reconsidered from a literary or theological point of view. In fact, only four significant variants have been identified in T, fewer than in any other version. Of the variants in T mentioned earlier in this chapter as possibly affected by one of the literary approaches to the text, the only significant one is variant 24. The context for the discussion of this variant is the observations of Bar-Efrat and Polzin on verses 11-14. As already noted in relation to LXX, the analysis of Polzin is not particularly convincing. Furthermore, T’s concern with rendering verbs literally and its lack of much of a distinct literary style (apart from theological concerns) would seem to override any such analyses. Thus, the variant remains significant. The other variants mentioned above are accounted for by the translation technique of T. Concerning variants 1 and 2, it is characteristic of T that it uses the preposition before , and this usage is exactly the opposite of the point Eslinger is making for MT. However, the rendering of by may in fact support an understanding of T similar to the way in which Eslinger understands MT. As noted in the discussion of these variants in Chapter 3, it is possible that the translators specifically avoided before (unlike P) in order to emphasize that Samuel was primarily the Lord’s, rather than Eli’s, servant (cf. also 2:11 T). Eli’s failure to mention the name of God in verse 17 in MT is dealt with in different ways by the versions: LXX and V change the active verbs to passives, P inserts the name of God, and T renders by . Because of T’s consistent rendering of , any impression that may have been in the Vorlage that Eli was averse to speaking the name of God is annulled. None of the evaluations of the variants in T is changed by literary analysis or further investigation from a theological viewpoint, so the translation technique described above as final stands. Vulgate Relevant Literary, Historical, and Theological Data Present in the Translation The situation with V is different from that of any of the other translations, for the translator is not only a single individual about whom much is known, but he himself commented on his approach to translation. One such quotation has already been mentioned, namely, his concern to render “with complete fidelity what stands in the Hebrew,” though not slavishly.61 Another quotation, cited by Brock, explains his reason for choosing a literal style of translation, when the common practice of translators of his day was to render freely. He says, “in scripture, even the order of the words is a 61 Jerome, Epistle to Sunnia and Fretela, cited in Pfeiffer, Introduction, 124

183

mystery.”62 In addition to Pfeiffer and Brock, other scholars have also pointed out characteristics of V. Metzger notes that since Jerome consulted from time to time with Jewish rabbis, certain common renderings exist between V and T. Vööbus says that Jerome’s translation was not homogeneous, since he consulted the various Greek versions. The influence of the Old Latin (to a large extent) and of rabbinic exegesis (to a small extent) is also evident. However, he was by and large an original translator in the historical books.63 Plater and White have a somewhat more extensive discussion of the characteristics of V. They describe the version as “at once correct and natural, accurate and idiomatic.”64 This description emphasizes both Jerome’s conservative attitude toward the text and his skillful use of the Latin language. V frequently translates the same Hebrew word by different Latin equivalents, and it often renders place-names etymologically. Jerome is particularly graphic in his renderings of the natural features of the land (rivers, mountains, deserts, etc.). He was occasionally obliged to use nonclassical words or even to coin words (based on Hebrew or Greek) for concepts not usually expressed in Latin.65 Roberts’s evaluation of V is somewhat more tempered than that of Plater and White. He says that when translating the OT, Jerome imitated the Latin of the NT, which was already familiar to many Christians. In addition to using vocabulary common in the NT, Jerome also imitated the style of the NT and some of the secondary Greek versions (especially Symmachus) by introducing subordinate constructions where the Hebrew had its ubiquitous coordinate constructions.66 His dependence at one time on Old Latin, at another on LXX, at still another on Symmachus or Aquila or Theodotion, and even occasionally on the targums leads Roberts to say, “Our conclusion, then, regarding the nature of Jerome’s translation is that when due allowance is made for all external influences, it must be admitted that his method was neither straightforward nor consistent.”67 While this judgment may hold true for V as a whole, it does not adequately describe Jerome’s approach in 1 Samuel 3. Pfeiffer says that in the historical books, Jerome showed his greatest skill and originality as a translator, largely because the passages were 62 Sebastian P. Brock, “The Phenomenon of Biblical Translation in Antiquity,” in Studies in the

Septuagint: Origins, Recensions, and Interpretations, ed. Sidney Jellicoe, Library of Biblical Studies, ed. Harry M. Orlinsky (New York: Ktav, 1974), 556. In particular, this statement demonstrates his concern for following the Hebrew word order, already noticed in the statistical evaluation in Chapter 3. 63 Metzger, “Versions, Ancient,” 753; ISBE, 1988 ed., s.v. “Versions,” by A. Vööbus, 972. 64 Plater and White, Grammar, 7. 65 Ibid., 6-10. 66 B. J. Roberts, OT Text and Versions, 254-58. 67 Ibid., 258.

184

easy to understand and because they contained few passages that might shock his Christian readers.68 A perusal of tables 51-53 indicates that one aspect of Jerome’s style was his free use of a wide range of vocabulary, particularly conjunctions. Part of the reason he used so many different conjunctions was his tendency, mentioned above, to change the structure of the sentence from compound (paratactic) to complex (hypotactic). In the area of sentence structure V deviates far more from the Hebrew than any of the other secondary versions. Following these observations, the final translation technique of V may be described. Final Translation Technique In the previous chapter, analysis of V indicated that Jerome was concerned for accuracy rather than consistency in the strictest sense. He would vary the lexical choice if he felt a different word would convey the meaning better in Latin, and he frequently used words of one class to render words of another, particularly if the Hebrew contained verbals. Jerome has sometimes been criticized for excessive reliance on such a wide variety of other versions, but rather than a detriment, Jerome’s method may have proved a benefit, since he was able to take advantage of the wisdom of earlier translators with regard to the rendering of difficult passages. In fact, V is probably closer to the idiom of the readers than any other secondary version, with the possible exception of P (which, after all, was a Semitic language), and also most of the tertiary versions. In addition to variety in lexical rendering and word class, Jerome was also liberal in the addition or omission of conjunctions (as well as using subordinate for coordinate conjunctions), and he showed little interest in the strict rendering of Hebrew compounds. Despite taking these freedoms in his translation, he was generally quite consistent in rendering those grammatical categories which Latin could reasonably be expected to render accurately. Above all, he was faithful in following the word order of the Hebrew, the one area in which his version differs significantly from the spoken language. The previous section has discussed Jerome’s literary concerns for his translation. The important point to remember is that he was more concerned with rendering the meaning than with rendering the literal words of his text. He does not show the same interest as the translators of P in producing an aesthetically beautiful translation, but he is interested in at least producing an acceptable style, as far as the Hebrew word order will allow. Elimination of Variants The first set of variants to be addressed is those that involve repetition (variants 84, 92, 97, 99, 104, 107, 108, 113, 118). Strictly speaking, only variants 92, 107, and 108 in this list deal with repetition, but their resolution will affect the other variants. The 68 Pfeiffer, Introduction, 124.

185

conclusion reached in Chapter 3 was that all three of these variants were significant, since the translation technique of V showed no concern for eliminating redundancy, but the caveat was added, “as determined to this point.” In order to clarify matters, two variants which are clearly nonsignificant need to be reexamined. Variant 104, the omission of in verse 13, is indeed in the midst of a difficult text, but Jerome could have chosen to render the phrase in some way, perhaps by the intensive pronoun ipsos. In Hebrew (if the tiqqun sopherim is not followed) is necessary to the text, for it identifies the object of Eli’s sons’ blasphemy (or curse), namely, themselves.69 However, when Jerome restructures the clause, becomes redundant and is omitted. Similarly, in variant 99 becomes redundant when Jerome restructures the text and is omitted, even though he could easily have inserted the personal pronoun ego without changing the meaning. The point in both these cases is that Jerome was more concerned with good Latin style than with rendering every single element of his Hebrew text, especially since both items omitted are still clearly implied in V. With this new insight into Jerome’s method, variant 92 can be reexamined. V omits the subject Heli but retains the object ad Samuhel. The comparison with LXX, which omits both and , remains instructive, but from a different perspective. Whereas LXX omits any reference to Samuel, Eli remains the implicit subject, since he was just mentioned as the subject of the previous independent clause. Awareness of Jerome’s concern for avoiding redundancy as long as no content is lost—revealed in the foregoing discussion of variants 104 and 99—sheds new light on the present reading. Contrary to the initial impression, Jerome does not really omit the subject Heli at all, but rather subsumes it in the verbal ending of ait. Thus, V, rather than standing halfway between MT and LXX, agrees with MT completely: the subject of the verb is still crystal clear, since Eli was just mentioned, but the word Heli has been omitted because it seemed redundant to Jerome. Therefore, variant 92 should no longer be considered significant. Variants 107 and 108 can also be reexamined in the light of a better understanding of Jerome’s approach to translation. As stated in Chapter 3, these variants are substitutional variants, that is, they are basically equivalent to one another. It was stated in the initial discussion of these variants that no evidence for a concern to avoid redundancy had yet been discerned. Matters are different now, however, and it is probable that the appearance of the phrase domui Heli earlier in the verse led Jerome to choose another, equivalent, rendering at the end of the verse. Thus, variants 107 and 108 are no longer to be considered significant. In the discussion of variant 97, it was noted that V does not omit pronouns without a good reason. While this statement remains true, the newly identified aversion to 69 Admittedly, the reading does not make much sense, but the omission of

the meaning.

would not clarify

186

redundancy seems likely to have supplied Jerome with a reason. The pronominal suffix in MT refers to the content of the message God gave to Samuel. While is a subordinate conjunction that introduces a relative clause, the fact that it is undeclinable requires further specificity concerning its antecedent, so the pronominal suffix is added to the participle to clarify that the clause is apposite to . 70 In V, however, quod already specifies verbum as its antecedent, since it is a neuter singular relative pronoun. Thus, is omitted as a redundant element already specified in the relative pronoun, and variant 97 is no longer significant. Whereas the deletion or change of redundant elements has been demonstrated to be a concern of Jerome, the question remains whether he felt free to add clarifying elements to the text, and if so, under what circumstances. A survey of the quantitative variants reveals several additional elements present in V, even when conjunctions are omitted. In particular, the adverbs adhuc, deinceps, indigne, and usque (variants 88, 95, 103, and 109), the verb erat (variant 76), the participle respondens (in both variants 78 and 112), the demonstrative pronoun ille (variant 118), and the personal pronouns te, eum, and te (variants 84, 113, and 116) are all present in V without any corresponding word in MT. Taking these words all together, it seems that V does indeed tend to add clarifying words when necessary or desirable. Though most of these have already been dismissed as nonsignificant, variants 84, 88, and 113 have been considered significant. In light of the fact that the additions in V include conjunctions (not listed), adverbs, verbs (and verbals), and pronouns, any excess text in V must now, upon further reflection, be viewed with even more suspicion. Thus, though variant 84 does share a common reading with LXX, the additional te could just as well be a clarifying addition implied by the context, perhaps influenced by Jerome’s knowledge of LXX. Variant 84, then, should no longer be considered significant. Variant 113 could have been added to specify the object of the verb, and Jerome might have felt it particularly appropriate following the verb interrogavit. Since this variant was a borderline case to begin with, the additional data on Jerome’s translation technique is enough to change its evaluation to nonsignificant. Finally, the additional adhuc in variant 88 could have been an explanatory addition; since the construction adiecit vocare was not idiomatic Latin, the word presumably would not have been considered redundant (cf. deinceps in variant 95). It is, of course, possible that one or more of these readings were based on a variant Hebrew Vorlage, but an approach based on determining the translation technique does not give any basis for determining which, if any, they were. Since the whole object of this approach is to use the translation technique as a guide in place of uninformed intuition, all these variants (84, 88, and 113) should be considered nonsignificant. 70 Cf. S. R. Driver, A Treatise on the Use of the Tenses in Hebrew and Some Other Syntactical

Questions, 3d ed. (Oxford: Clarendon Press, 1892), 265-67.

187

The only remaining significant variant is in the category of consistency, specifically, grammatical consistency. The evaluation of variant 28, which reads a past tense, whereas MT implies a future, is affected by the literary approaches described above. As already mentioned, Bar-Efrat’s observations on the importance of flashbacks in the narrative are preferable to the arguments of Polzin, and the support of LXX and (especially) T affirm the recognition of the variant as significant. Similarly, the evaluation of nonsignificant variants 39 and 40 does not change as a result of literary analysis, for it is likely that Jerome understood his Hebrew Vorlage as reading a pual in variant 40, and the plural in variant 39 can be explained as a reference to the collective idea behind the Hebrew singular. It is probable that Jerome was influenced by his acquaintance with LXX in his renderings. The list of significant variants for V has changed as a result of a better understanding of Jerome’s approach to translating, as revealed in the text. Since all the changes occurred in quantitative variants, no changes to the grammatical tables need to be made, but the percentage of agreement in quantitative representation returns to the original figure of 86.6%. The new material discovered about the translation technique may now be summarized. In addition to the description given above in the section dealing with the final translation technique, it should be recognized that V is characterized by a certain type of omission of Hebrew material, namely, omission of redundant words. Furthermore, V does apparently add explanatory material of many types, not just conjunctions and adverbs, as suspected earlier. In spite of its great number of variants, the Vorlage of V seemingly lay closer to MT than that of any other secondary version, even T. Evaluation of Partial Secondary Witnesses Since no significant variants were identified in any of the minor Greek versions, only the two recensions will be examined. The Lucianic Recension The characteristics of the Lucianic recension described by Driver (see above, p. 162), along with those noted in the analysis of the chapter, may stand as the final translation technique of the Lucianic recension, to the extent that one can actually speak of a translation technique. A survey of the variants identified as probably significant reveals one variant that, upon further reflection, should probably be considered nonsignificant. Variant 13 is an attempt to make the third call of Samuel resemble the first two and the fourth, a characteristic of LXXL but not, as discussed above, of MT or LXX. Thus, variant 13 should be eliminated. The other variants listed in Chapter 3 remain significant.

188 The Hexaplaric Recension All of the variants for LXXO listed in Chapter 3 as significant remain so. None of the literary approaches described above seems to have any bearing on the discussion of the text of this partial secondary witness. Other Possible Hebrew Readings All of the significant readings identified above should still be considered significant. Conclusion In conclusion, it should be noted that the purpose of this chapter is not to evaluate the relative merits of the various literary approaches of 1 Samuel 3, though some weighing of conflicting interpretations is inevitable. On the contrary, the examination of the literary nuances present in the text of the different versions can be a helpful tool in making textual judgments, especially in borderline cases, just as it is a helpful tool in exegesis. Of course, for a literary approach to be valid, it must fit the text and not be imposed on it from the outside, unless one just wants to hear one’s own voice in the text. Nevertheless, it is clear that different approaches yield different insights, and apparently conflicting analyses may be the result of ambiguities present in the text.71 Therefore, textual critics need to be aware of and engage in analysis of each of the texts before them from a literary point of view, even if the ultimate goal is the reconstruction of a single text lying behind the extant witnesses. 71 Cf. Fishbane’s term “bivalent image” mentioned above. Similarly, James Sanders speaks of the “mutivalency” of the biblical text; James Sanders, Canon and Community: A Guide to Canonical Criticism, Guides to Biblical Scholarship, Old Testament Series (Philadelphia: Fortress Press, 1984), 2223. David Robertson also discusses the tensions and ambiguities present in the biblical text; D. Robertson, The OT and the Literary Critic, 7.

CHAPTER 5 CONSIDERATION OF VARIANTS FROM PRIMARY WITNESSES After attempting to determine the original text of each of the secondary witnesses and sorting through the hundreds of variants to determine those that are significant, all that remains to do in compiling the significant variants is to list the variants from the primary witnesses. Once these are obtained, they may be set alongside the significant variants from the secondary witnesses for further consideration in determining the oldest form of the Hebrew text possible. The primary witnesses for the text of 1 Samuel 3 fall into two main categories: Masoretic variants and non-Masoretic variants, that is, variants from 4QSama. 1 The Masoretic variants can be further divided into three groups. The first group consists of those mss identified by Goshen-Gottstein as those whose readings diverge enough from the mass of Masoretic mss to warrant further investigation (mss 70, 89, 174, and 187).2 The second group is the qere readings found in the margins of the mss. The third group is the tiqqune sopherim, or scribal corrections, listed in the rabbinic literature.3 The variants listed in this chapter may be divided into two groups: purely orthographic variants, and those that are more substantial, the latter of which are treated as significant. In each section, the orthographic variants, where present, will be listed before the substantial variants.4 All variants are collated against MT (ms L, as reflected in BHS) as the base text. 1 For a complete list of the contents of 4QSama, see Ulrich, Qumran Text of Samuel, 271. In

1 Samuel 3, the ms contains portions of verses 1-4 and 18-20. 2 Goshen-Gottstein, “Hebrew Biblical Manuscripts,” 287. 3 For the entire book of Samuel, other primary witnesses would include the puncta extraordinaria

(special points), sebirin, and itture sopherim (scribal omissions). See Würthwein, Text of OT, 17-20. 4 In his discussion of orthography in the Masoretic mss, Barr concludes that the spelling patterns of MT suggest that one ms (or a few mss) are the basis for the entire Masoretic tradition, so little evidence remains of the orthography of the original form of most books. Another implication of his study would seem to be that few purely orthographic variants are true reflections of the spelling found in a particular ancient Hebrew ms, though scribes involved in the earliest transmission of the standard text after the first century C. E. might have introduced orthographic variants (especially defective spelling) independently of any ms tradition. Cf. James Barr, The Variable Spellings of the Hebrew Bible, Schweich Lectures of the British Academy, 1986 (Oxford: Oxford University Press for the British Academy, 1989), 204-8.

189

190 Variants in Masoretic Manuscripts Important Manuscripts as Identified by Goshen-Gottstein Although Goshen-Gottstein states that no extant medieval mss “may be termed ‘valuable’ or be worthy of our attention more than any other,”5 he does identify four mss which “stand out” and “have to be considered” in Samuel.6 Since some way of identifying these mss is necessary, the phrase “important manuscripts” will be used, though GoshenGottstein’s doubts concerning them are recognized. At this point, no attempt is made to differentiate which variants in these mss, if any, might possibly go back to times before the standardization of the text in the first few centuries C. E. and which arose in medieval times, though of course all of the orthographic variants probably belong to the latter category.7 Before listing the variants, however, a brief description of each of the four important mss is in order.8 Ms 70 contains the former prophets, was written at the beginning of the fifteenth century, and is housed at Oxford. Ms 89 is a full Bible (OT) ms, with a Spanish character. It is often difficult to distinguish the letters and and the letters and . This ms contains about twelve thousand variants from the majority of the Masoretic mss. It was written at the beginning of the thirteenth century and is found at Cambridge. Ms 174 contains the former prophets. In it, the letters and and the letters and are sometimes similar. Frequent qere readings are present in the (kethib) text. Written in 1346, it is housed at Copenhagen. Finally, ms 187 contains the former prophets, Megilloth, Jeremiah, and Isaiah. It has a German character and comes from the end of the thirteenth century. It is kept in Milan. Orthographic Variants 1 2 3 4 5 6 7

3:2

] ] ]

3:3 3:6 3:8

] ] ] ]

89 89 174 187 174 174 174 174

5 Goshen-Gottstein, “Hebrew Biblical Manuscripts,” 277. 6 Ibid., 287, n. 3. 7 Cf. the discussion in ibid., 274-76. It is Goshen-Gottstein’s opinion that none of the variants in

Masoretic mss can be connected with any non-Masoretic tradition; ibid., 277. 8 The descriptions of these mss are taken from Benjamin Kennicott, Dissertatio generalis in Vetus Testamentum Hebraicum; cum variis lectionibus, ex codicibus manuscriptis et impressis (Oxford: Clarendon Press, 1780), 77-87.

191

8 9 10 11 12 13 14 15 16 17

] ] ]

3:9 3:11 3:13 3:17 3:18

]

3:21

]

] ] ] ] ]

70 89 174 187 89 70 174 174 174 89 70 174 187 70

Substantial Variants 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

3:2

3:3 3:5 3:6

3:8 3:9 3:14 3:15

3:16 3:17

] > 89 ] 187 ] 187 ] 70 2°] 89 ] mg + 70 ]+ 187 1°] + 70 ] > 89 187 2°] > 174 ] > 70 ]+ 70 verse 6 fin] + verse 6 (repeated) 70 2° (in repeated occurrence of verse 6)] ]+ 174 1°] + 70 ] 70 ] > 89 ] > 70 2°] > 187 2°] 89 ] 89 174 ] > 70 ] 21 187

70

] > 70 ]

187

192

44 3:18 ]+ 174 45 ] > 174 46 3:20 ] 174 47 3:21 ]+ 70 48 ] > 187 49 ] > 89 50 ] 89 No attempt will be made here to evaluate each of the substantial variants, since they are by definition significant, occurring as they do in Hebrew mss that are deemed important. However, it is clear at a glance that most of these variants are inner-Masoretic developments, either scribal errors (e.g., parablepsis in variant 42), reading qere for kethib (orthographic variants 2 and 13, both in ms 174 [see description of this ms above]), or other common scribal modifications (e.g., interchange or omission of particles and prepositions [variants 37, 38], addition or omission of a conjunction [variant 19], explanatory additions [variant 24]). Two of the most interesting readings are found in ms 70: (1) the twofold inclusion of verse 6, with differences between the two occurrences (variants 25, 28, 29, 30, 31), and (2) the addition of the words “to Manoah” (variant 47), an apparent reference to Judges 13. Kethib-Qere Orthographic Variants 51 52

3:2 3:18

K] K]

Q Q

Tiqqune Sopherim Substantial Variants 53

3:13

]

tiq soph Variants in 4QSama`

Part of 4QSama was published in a preliminary fashion by Cross in 1953,9 but he only discussed that part of the scroll containing 1 Sam 1:22-2:6 and 2:16-25. Both Ulrich and McCarter refer to portions of chapter 3 in the scroll, though neither gives the full 9 Cross, “New Qumran Biblical Fragment," 15-26. See Ulrich, Qumran Text of Samuel, 9-10, for a description of the scroll. David Noel Freedman describes the orthography of 4QSama as proto-Samaritan;

David Noel Freedman, K. A. Mathews, and R. S. Hanson, The Paleo-Hebrew Leviticus Scroll (11QpaleoLev) (Winona Lake, IN: Eisenbrauns for the American Schools of Oriental Research, 1985), 56, cf. 61-62.

193

contents of the chapter.10 Only a small part of chapter 3 has been identified on the fragments of 4QSama, namely, 3:1-4, 18-20.11 A look at a photograph of the scroll indicates that only seventeen words or parts of words have been preserved on two fragments. In order to collate 4QSama with MT, a tentative reconstruction of the two fragments is necessary.12 Fragment 1: ] [ (3:1) ] [ (3:2) ] [ (3:2-3) ] [ (3:4) Fragment 2: ] [ ] (3:18) ] (3:18-19) ] (3:19-20) ] [ ] (3:20) Based on this reading of the fragments, the following variations from MT result. Orthographic Variants 1

3:18

]

4QSama

Substantial Variants 2 3

3:3 3:4

] > 4QSamavid13 ] > 4QSama14

10 Ulrich, Qumran Text of Samuel, 63; McCarter, I Samuel, 95. 11 Ulrich, Qumran Text of Samuel, 271. 12 It must be stressed that this reconstruction is tentative, since it is based only on an examination of a photograph of the ms, as well as reference to Ulrich, Qumran Text of Samuel, and McCarter, I Samuel. The photograph was provided by the Ancient Biblical Manuscript Center and is courtesy of the Israel Antiquities Authority; Israel Antiquities Authority, “PAM 41.766,” photograph (Claremont: Ancient Biblical Manuscript Center, 1992). Official publication of the scroll will be in Frank Moore Cross and Eugene Charles Ulrich, eds., Discoveries in the Judaean Desert of Jordan, vol. 11 (Oxford: Clarendon Press, forthcoming). 13 Ulrich notes that the scribe of 4QSam a wrote about fifty letters per line, at least in this part of

the scroll; Ulrich, Qumran Text of Samuel, 63. Though McCarter does not say so specifically, this letter count seems to be behind his suggestion that the scroll omits five words found in MT; McCarter, I Samuel, 95. If so, it would agree with LXX in omitting “Yahweh,” but it would be alone in omitting the other words. Though space considerations do not force this exact appraisal, the readings of both and are certain. Thus, it is clear that the scroll is different from MT at this point, and it is probable that it is about twenty letters shorter. 14 Both Ulrich, tentatively, and McCarter, without qualification, suggest another variant, namely, that 4QSam a agrees with LXX in adding a second “Samuel”; Ulrich, Qumran Text of Samuel, 63; McCarter, I Samuel, 95. This suggestion, however plausible, is not directly supported by the fragment, since no trace of the second appears, but is based on the omission of (variant 3). In light of the fact that MT also omits in verse 8 (cf. v. 6), a second “Samuel” is not an assured reading and remains a conjecture.

194 Though few variants from MT remain in 4QSama, it is noteworthy that the number of variants per word of MT in 4QSama is much higher than that of any one of the Masoretic mss. Conclusions: Significant Variants in 1 Samuel 3 The preceding analysis has demonstrated the importance of understanding the translation techniques of the various translators of the secondary witnesses. It has also clarified certain aspects of those translation techniques. A description of the translation techniques of each of the versions will not be repeated here, since they were described and developed in an inductive manner as the study progressed. It is important to note, however, that the summaries given in the text, especially in Chapter 3, are just that: summaries. A more detailed understanding of the translation technique will require study of the statistical tables that describe the data. Nevertheless, an overview of the translation technique of each of the versions can be gathered by reading (1) the description of the preliminary partial translation technique in Chapter 3, (2) the modification of that description after the analysis of the data later in the same chapter, (3) the initial description of the final translation technique in Chapter 4, and (4) the modifications to that description, if any are necessary, later in Chapter 4. Out of a total of 527 variants analyzed in the secondary witnesses, only 81 were identified as significant, and 45 of those were in LXX, plus another 16 in other Greek witnesses. A list of the significant variants for each of the secondary and partial secondary witnesses follows. Septuagint Consistency

3 6 12 13 15 18 26 27 30 32 33

Ref 3:1 3:2 3:5 3:6 3:7 3:9 3:12 3:13

Variation

Septuagint active lexeme lexeme verb lexeme lexeme lexeme lexeme perfect plural lexeme

Masoretic Text niphal

infinitive

perfect w/c singular

195

34 40

word division lexeme

3:16

Word Order

67 68

Ref Number of Variations 3:16 3 3:19 1

Greek Variant

Quantitative Representation 70 71 72 74 75 76 78 79 80 81 82 83 84 86 87 89 90 91 92 93 94 95 96 98 100 101 102 103 104

3:1 3:2 3:3 3:4 3:5 3:6

3:7 3:9 3:10 3:11 3:13 3:15 3:16 3:17 3:18 3:21

+2 +1 -1 -1 +1 +1 +1 -1 +1 -3 +1 -2 +1 -1 -3 +1 +1 -2 +1 -1 -2 +2 +4 -2 -2 +3 -2 +1 -5

4°

2° 2°

2°

2°

196

105

+31

Peshitta Consistency

8 27 30 36 37

Ref 3:3 3:15 3:17 3:20 3:21

Variation 1°

Ref 3:1 3:6

Number of Variations 1 2

Peshitta lexeme singular lexeme lexeme plural

Word Order

48 49

Syriac Variant

Quantitative Representation

56 79 81 94 123 124

Ref 3:2 3:9

Number of Variations +1 -1 -1 3:14 -1 3:21 +1 -1

Variant - 4° - 3° - 1° 3°

Masoretic Text plural

singular

197 Targum Consistency

24 30

Ref 3:13 3:16

Variation -

Targum perfect preposition

Masoretic Text perfect w/c DDO

Vulgate perfect

Masoretic Text perfect w/c

Quantitative Representation

46 56

Ref Number of Variations 3:2 +1 3:11 -1

Variant - 4°

Vulgate Consistency

28

Ref 3:13

Variation praedixi

The Lucianic Recension

8 12 20 30 40 46 47 48

Ref 3:6 3:7 3:12 3:15 3:19 3:21

Variation (from LXX) ]+ ] ] 2°] post 3°] > 2°] > ] ]

The Hexaplaric Recension

1 2 3 8 9 10 11

Ref 3:3 3:8 3:10 3:15 3:21

Variation (from LXX) ] ] pr ]+ ] 2°] > ] 3°] >

198 Other Possible Hebrew Readings Ref Variation (from indicated base text) 1 3:2 ] (LXX) 2 3:5 ]+ (P) 3 3:9 ] (T) When the 36 substantial variants found in the primary witnesses are added to the total above, a total of 117 significant variant readings have been identified for 1 Samuel 3. Armed with this list of variants considered to be significant for determining the text of 1 Samuel 3, a few other works may be consulted as an example of how variants from the versions, particularly LXX, are often used in evaluating the text. Schulz implies that the translators of LXX read in 3:19,15 but the fact that LXX frequently uses the active voice in a causative manner makes Schulz’s suggestion questionable. A corollary of the rule that translation technique must be considered before using the evidence of a version is that the text of the version itself must first be determined. So, for example, many commentators assume that ms B is the text of LXX, or at least they use its text as if it were. In fact, it would better to use Rahlfs’s text of LXX, since he examines other mss besides B, in lieu of a volume on Samuel in the Göttingen edition. McCarter occasionally uses B as though it were equivalent to the OG, for example, in verse 9 where he accepts the reading “Speak” (without “Lord”) on the basis of B, or when he omits “and thus may he add” in verse 17, following only mss Ba2 (plus cop it), but not the vast majority of the Greek mss or Rahlfs. 16 Similarly, Schulz says of verse 17 that “G hat den zweiten Teil ( ) nicht übersetzt.”17 The other versions are not used nearly as often as LXX in textual discussions, but a few examples will show that they are often used in a manner similar to LXX, that is, without considering the translation technique. Caspari, in his discussion of verse 2, says that the text of P implies a Vorlage . 18 While it is true that forms of are the regular way of rendering forms of , it does not follow that the occurrence of necessarily implies a Vorlage of . In fact, a look at 4:15 shows otherwise: again Eli’s eyes are the subject, and this time they are described as “set,” in MT. Again P reads a form of (this time ), so it appears that is a typical way of describing the blindness of eyes, and Caspari’s assessment of the text of P does not hold.19 Caspari says 15 Schulz, Bücher Samuel, 66; cf. Conybeare and Stock, Grammar, 76-77. 16 McCarter, I Samuel, 95-96. 17 Schulz, Bücher Samuel, 65. Cf. also R. Fruin, “Oudtestamentische Studiën, Nieuw theologisch tijdschrift 20 (1931): 108, who says that the Hebrew text of Judges 18:30 was changed from “Moses” to “Manasseh” before LXX was translated. However, only B reads “Manasseh”; most mss read “Moses.” 18 Caspari, Samuelbücher, 53. 19 Ironically, Caspari fails to mention the reading of LXX in 3:2, which is in fact imply a form of in the Vorlage (see above, pp. 86-87; LXX reads

and may in 4:15,

199

of verse 8 that P supports an additional , but the discussion of variant 71 demonstrates that no such assumption can be made on the basis of P, since the translators were free with their explanatory additions. In verse 4, he says of the word “Samuel”: “Aeth schützt Voc durch .” 20 Though it is a minor point, since eth (and cop) are witnesses to LXX rather than a Hebrew text, he should say that it points to a reading of ; as it stands, his comment gives more importance to the reading of eth than it deserves. Schwartz says that in verse 1 is equivalent to , 21 but the discussion of variant 2 clearly demonstrates that the text of P implies no other reading than of MT. Thenius says that V adds after in verse 10.22 As with the examples of equating ms B with LXX mentioned above, the problem here is that Thenius gives the reading of the Clementine Vulgate, which does not contain the best text of V at this point. In verse 19, he claims that V (joined by LXX and Josephus) supports the reading instead of MT’s . 23 Since neither Latin nor Greek can express causative ideas without circumlocution, V does not imply a qal in the Vorlage any more than LXX does. Such citations of the versions could be multiplied greatly, even for so small a unit as 1 Samuel 3. However, enough have been listed to conclude that the translation technique of the versions must be studied in detail before any evidence from them can be cited.24 While it is easy to identify certain treatments of the secondary witnesses as invalid, it is not as easy to decide one particular set of cases. When a significant variant has been identified in one of the witnesses, should variants in other witnesses that seem to agree with it, but have been deemed nonsignificant, be cited as supporting the reading? For example, BHS cites V as a witness to be consulted for the reading (MT reads ) in verse 21. However, the discussion of variant 47 above (pp. 152-53) concludes that V shows too much freedom in its use of prepositions to consider this variant significant. Since the variant is significant, though (it is read by many Masoretic mss, including the important ms 89), is it permissible to cite V in support of the reading, despite Jerome’s somewhat cavalier approach to rendering prepositions? The answer to this question is a qualified yes. The adage that there is strength in numbers applies to this situation: borderline readings that have been classified as nonsignificant might warrant supporting MT [but cf. LXXL,

]).

20 Caspari, Samuelbücher, 54. 21 Schwartz, Syrische Uebersetzung, 11. 22 Thenius, Bücher Samuels, 17. 23 Ibid. 24 Lest the singling out of these sources as negative examples be seen as an overall negative

evaluation of their work, it should be noted that at least they studied the text and tried to make intelligent decisions. They were not content to rely without question on the traditional text (i.e., MT).

200

reevaluation if a similar reading elsewhere turns out to be significant. Even if, upon further consideration, the reading remains nonsignificant, it can and should be cited alongside the evidence of the significant variant, though due caution concerning its relevance should be noted.25 For example, in 1 Sam 3:12 McCarter advocates reading , which is supported by all the secondary versions. However, the only variant that is significant is in LXX, since the semantic ranges of and are larger than that of , and since the adversum of V may be another instance of Jerome’s freedom in rendering prepositions. Since LXX does have a significant variant here, the other versions may be cited in support, but it should be noted that their testimony is not as weighty as that of LXX. Thus, a comment such as “reading al lî with LXX ( pi l i); cf. P, T, V” would be preferable to “reading al lî with LXX ( pi l i) and Syr., Targ., Vulg.”26 One option would be always to cite the readings of the versions in their own languages, rather than (or before) giving the Hebrew equivalent, since giving the Hebrew leaves the impression that their Vorlage did indeed read what is listed. This suggestion applies as much to critical apparatuses as to commentaries and studies, if not more so. The first two steps of the proposed text-critical methodology, namely, (1) determining the lexical, grammatical, and stylistic characteristics of the secondary witnesses, and (2) determining the literary, theological, and historical characteristics of those witnesses, are now complete, and the significant variants from the secondary witnesses have joined the readings of the primary witnesses as claimants to the title of earliest recoverable text. It should perhaps be noted once again that identifying variants as significant does not imply that they should be preferred to the text of MT in any or all cases. That determination is the goal of the remainder of this study. 25 Some nonsignificant readings that support significant readings in other witnesses should not be cited, for example, when the reading is a common one that probably arose independently (such as the addition or omission of a conjunction) or when the reading probably arose as a result of intra-witness contamination (e.g., an inner-Greek error). When nonsignificant variants are cited as supporting a particular reading, they should be distinguished in some way from those readings considered significant. See, below, Chapter 8, for a suggested way of citing nonsignificant variants alongside significant variants. 26 McCarter, I Samuel, 96. Driver also cites the support of all the versions, but it is unclear

whether his “Pesh. Targ. ” is supposed to be an Aramaic/Syriac reading given only in Aramaic characters or an indication of the Hebrew text underlying the two versions; S. R. Driver, Notes on the Books of Samuel, 43.

CHAPTER 6 RETROVERSIONS OF THE SECONDARY WITNESSES It is commonplace in critical apparatuses to cite the testimony of secondary and even tertiary versions, and ample justification for presenting the readings in their original language exists. However, before any comparison between the Hebrew and non-Hebrew witnesses is possible, the latter must be retroverted into Hebrew, at least implicitly. Retroversion into Hebrew will be done explicitly in this chapter, not only to demonstrate the methodology, but also because retroversion of a versional reading forces the scholar to deal more directly with the purported development of the text. Many versional readings that appear to presuppose a different Hebrew Vorlage turn out to be inconclusive, or even support MT, when one is forced to propose a Hebrew text underlying the version. Developing a Methodology for Retroverting Translations Though many commentators and editors of both critical and diplomatic texts present readings that are retroversions from translations, few have proposed a methodology for doing retroversions. As a result, many retroverted readings remain questionable, and even retroversions that are probably correct suffer from an insufficient theoretical foundation. A few scholars, however, beginning with Max Margolis, have proposed guidelines for those who would attempt to recover the Vorlage behind the present, translated reading. Proposed Methodologies for Retroverting Translations Max Margolis In a 1910 article, Max Margolis discusses the possibility of retroversion from Greek to Hebrew by means of a process which he calls “complete induction.”1 He begins with a discussion of Paul de Lagarde’s first canon, which states that in order to arrive at the Hebrew reading lying behind the Greek, one must first have a knowledge of the style of the 1 Max Margolis, “Complete Induction for the Identification of the Vocabulary in the Greek

Versions of the Old Testament with Its Semitic Equivalents: Its Necessity and the Means of Obtaining It,” Journal of the American Oriental Society 30 (1910): 301-12. Although he acknowledges the importance of retroverted readings, he is not particularly optimistic that convincing retroversions are obtainable, except in those cases in which “the translator has misread or misinterpreted the original” (p. 303). Cf. his comment on pp. 302-3: “As a matter of fact, in passages wanting in the Hebrew, all attempts at retroversion are unscientific.”

201

202

individual translators.2 As a corollary of this principle, one must also determine the limits of the unit of translation, that is, the material rendered by the same translator.3 Not only must one know the style of the translators, Margolis says, but one must also be familiar with the style of the individual Hebrew writers.4 As a preface to discussing several examples, Margolis lays down one more proposed principle: “In order, however, to discover the total sum of criteria, the student must obviously collect his data from the whole of the Greek Old Testament, whereupon he may proceed to distribute them among the various groups of translators thus brought to light.”5 He proceeds to list several examples in which a Hebrew coordinate clause is rendered by a Greek subordinate clause (e.g., Gen 3:6; 4:1; Num 21:16; Deut 23:13; 1 Kgs 14:18), where plural and generic singulars are interchanged (Gen 4:20; Neh 12:44; Prov 11:10; Sir 4:12), and where finite verbs and participles are interchanged (Exod 20:2; Ruth 4:15; 1 Esdr 5:69).6 Margolis’s goal is to replace uninformed conjectures with substantiated lexical and grammatical equivalents, some of which defy intuition. Searching the biblical text for data to inform textual decisions in unquestionably important. However, some doubt must attach itself to Margolis’s contention that his examples are “certain.” For example, he cites Isa 40:6, where is formally equivalent to , and uses this equivalence to support the legitimacy of as a rendering of in Sir 44:1. However, since corresponds to only once out of more than 250 occurrences of , Winton Thomas’s suggestion in BHS that in Isa 40:6 might reflect a form of must be accorded at least equal consideration, especially in light of the fact that renders eleven times in the OT, including four times in Isaiah. If is original in Isa 40:6, then the supposed equivalent in Sir 44:1 has no support. In addition to this line of argument, one must also recall that LXX Isaiah is a free translation, so one formal equivalent implies very little. Despite this questionable example, most of Margolis’s examples of his method do seem probable (and they usually have a wider basis of support). Nevertheless, his contention that the student must collect his data from the whole Greek OT, though it may seem obvious, is not always valid. Though data from other parts of the OT is frequently 2 Ibid., 301. Cf. also p. 302: “After an elimination of the irrational element of chance corruptions

or of the disfiguring element of conscious alteration . . . , there remains the stupendous task of retroversion for which indeed a knowledge of the style of each individual translator is an all-important prerequisite.” 3 Ibid., 304. 4 Ibid., 303. Thus, a knowledge of literary criticism is essential. Margolis gives an example based

on different Pentateuchal sources. 5 Ibid., 304; italics his. 6 Ibid., 305.

203

helpful, more weight should be placed on his previous suggestion that the translation technique of that particular unit of translation be determined. After demonstrating his method of complete induction for recovering the Hebrew Vorlage behind a LXX reading, Margolis turns to show how the method can also be used to recognize inner-Greek corruptions. He suggests that the reading in Isa 28:20 is a corruption of (= ´), which in turn reflects the Hebrew in place of of MT. The retroversion is supported by the equivalent ( ) + infinitive = + infinitive (cf. Isa 54:9).7 Margolis summarizes his approach by saying that the method of complete induction requires two separate procedures. The first requires the use of a concordance to establish all lexical equivalents of a given word. Related words should generally be dealt with together (e.g., and ). The second procedure is the establishment of a list of grammatical equivalents, for example, the correspondence of the Greek active voice with the qal stem, the aorist with the Hebrew perfect, and so forth. These equivalences must be derived from the text. Margolis stresses the importance of determining grammatical equivalents alongside lexical equivalents if retroversion is to be attempted. He says, “Complete induction, at all events, can be had only by means of the two lines of investigation, the lexical and the grammatical. It is a stupendous work, but it must be done. . . .” 8 It is his emphasis on the necessity of a thorough investigation of the biblical texts, and in particular his recognition of the importance of grammatical equivalents, that most clearly separates Margolis from his predecessors, and from many of his successors. Emanuel Tov In The Text-Critical Use of the Septuagint in Biblical Research, Emanuel Tov consciously builds on the foundation laid by Margolis for reconstructing the Vorlage of the LXX. Tov identifies three steps one must follow when retroverting a reading from Greek to Hebrew: (1) identify those elements in LXX which presumably reflect elements in MT; (2) isolate those readings of LXX which could reflect Hebrew readings different from MT; (3) attempt to identify which Hebrew words the translator had in front of him or had in mind.9 The first two of these steps are prior to retroversion proper, which occurs in the 7 Ibid., 308. He is perhaps somewhat overzealous in some of his attempts to see Hebrew readings

behind corrupt Greek texts. For example, on p. 309, n. 3, he suggests that in Jer 44 (37):12 (ms 239) might reflect the Masoretic , whereas most LXX mss read = . However, a more plausible explanation would seem to be that is simply an inner-Greek corruption of (a possibility he does note). Similarly, there is no need to search for a Hebrew equivalent for in ms 26, for this reading, too, probably arose from the graphically similar (perhaps with the aid of a lacuna in the ms tradition). 8 Ibid., 311-12. 9 Tov, Text-Critical Use, 99. Those readings in LXX and the other translations that are isolated in

204

third step. This model is helpful in that it specifies for which words retroversion is to be attempted. The next question is, how exactly does one proceed to identify those words which the translator had in front of him (or in mind)? Tov answers this question by offering a number of guidelines for determining the presumed Vorlage. (1) Retroversions are based either on vocabulary equivalences between LXX and MT or (2) on the scholar’s intuition. (3) Correct retroversions should be probable from a textual point of view (i.e., “should have developed by known procedures of textual change from the reading of MT or vice versa, or else its place in the textual history of MT should be easily definable”), and (4) “they should be plausible from the point of view of the grammar, vocabulary and style of the Hebrew Bible, and in particular of the book in which the reading is found.” (5) Finally, some retroversions are supported by identical readings elsewhere.10 He gives numerous examples of each of these guidelines. Vocabulary equivalences between LXX and MT can be gleaned from a concordance, particularly that of Hatch and Redpath, or from various computerized databases, such as the CATSS database.11 However, one must use the data with caution, for some of the apparent equivalents are formal equivalents only, that is, the Hebrew and Greek words occupy the corresponding space in MT and LXX, respectively, but the Greek reading may not actually render the word found in MT.12 In 2 Kgs 17:20, in MT corresponds to in LXX. Though HR lists as an equivalent of , the present verse is the only example of this correspondence, and also renders the verb . Thus, the presumed Vorlage of the LXX reading is ; one of the two Hebrew readings probably developed from the other by metathesis.13 The search for vocabulary equivalences need not be limited to the exact Greek word in question, for some presumed equivalents are based on related Greek words (e.g., compounds) or on words with similar meanings.14 The textual critic’s intuition cannot be emphasized too strongly, according to Tov, step two are the same as the “significant variants” discussed in this study. 10 Ibid., 101. 11 For a description of this database, see Robert A. Kraft and Emanuel Tov, eds., Computer

Assisted Tools for Septuagint Studies, vol. 1: Ruth, Septuagint and Cognate Studies, no. 20 (Atlanta: Scholars Press, 1986); Emanuel Tov, A Computerized Data Base for Septuagint Studies: The Parallel Aligned Text of the Greek and Hebrew Bible, Computer Assisted Tools for Septuagint Studies (CATSS), vol. 2, Journal of Northwest Semitic Languages Supplementary Series, no. 1 (Stellenbosch: Journal of Northwest Semitic Languages, 1986). 12 See Tov’s discussion in Text-Critical Use, 101-2; idem, “Background of the Greek-Hebrew

Alignment,” in Kraft and Tov, CATSS, vol. 1, 37-38. 13 Tov, Text-Critical Use, 103. 14 Ibid., 106-7.

205

for the word on which a Greek reading is based may not ever be translated with that particular Greek word (or a related word), and it may not even occur elsewhere in LXX. For example, in Jer 2:16 MT is formally equivalent to in LXX, but the meaning is different. Despite the lack of attestation in LXX, Tov suggests that the LXX reading goes back to a Hebrew reading , which is semantically similar to and graphically similar to . 15 Intuition is an invaluable tool, but it remains subjective, and it will sometimes happen that the Vorlage behind a Greek reading is indeterminate. This problem will be discussed below. Tov’s third guideline states that retroversions should be probable from a textual point of view. Tov lists the interchange of graphically similar letters and metathesis as common scribal phenomena.16 Elsewhere, he lists factors such as parablepsis, dittography, phonetic similarity, differences in word division, and various intentional changes.17 For example, the suggested retroversion of a Greek reading should follow the orthography used at the time that the translation was made (see below, pp. 212-16). Furthermore, the textual critic must remember that a retroversion is based not only on the meaning of the Greek text, but also on the graphic form of the Hebrew text. Thus, in Jer 5:6 should not be retroverted as , but rather as , which is graphically closer to of MT.18 The plausibility of the retroverted reading in terms of grammar, vocabulary, and style reminds the textual critic that most convincing retroversions will usually fit the context of the passage and the stylistic characteristics of the book (or section) in question. However, it is entirely possible that the Hebrew Vorlage behind some Greek readings was anomalous. An unusual grammatical form, for example, may reflect a legitimate archaic or dialectical survival in the text, which was changed at some point in the proto-Masoretic tradition. On the other hand, an anomalous reading may simply be an error that crept into the text. However, Tov stresses that “the correctness of a given retroversion should never be confused with its originality within the history of the biblical text.”19 In other words, retroversion is a different step entirely from evaluation. In a further comment on linguistic plausibility, Tov reminds the textual critic that “retroversions should follow the grammar and lexical understanding of the translator rather than the modern scholar’s understanding 15 Ibid., 116-17. 16 Ibid., 120. 17 Idem, Textual Criticism of the Hebrew Bible, 236-84. See also P. Kyle McCarter, Jr., Textual

Criticism, 26-61, and Michael Fishbane, Biblical Interpretation in Ancient Israel (Oxford: Clarendon Press, 1985), 23-88. 18 Tov, Text-Critical Use, 121-22. 19 Ibid., 124 (italics his).

206

of Hebrew philology.”20 For example, the LXX reading for MT’s in Ps 23(22):4 probably reflects the translator’s understanding of as equivalent to the Aramaic , “in the midst” (cf. also Jer 49:19[29:20]).21 Finally, some retroversions of LXX have external support from other ancient versions, Qumran mss, and scriptural citations or allusions in other ancient sources (e.g., Josephus, extracanonical books, the Talmud, etc.). Caution must be exercised, however, when one claims that a reading in one version supports that in another, for some parallel readings developed independently of one another. For example, the agreement between LXX and the Samaritan Pentateuch in saying that God completed his work of creation on the sixth day, rather than the seventh in MT, may be the result of independent contextual harmonization.22 The textual critic must be especially careful when using data from medieval Hebrew mss to support a reading in LXX, for the history of the development of the Masoretic mss makes it unlikely that such agreements are genetically related. Only in the cases of a few specific mss may there be exceptions to this generalization.23 Tov’s guidelines for retroverting the text go well beyond those of Margolis, yet Tov reminds the textual critic of the subjective nature of most retroversions: “No retroversion— with the exception of some personal names—is beyond doubt, but some retroversions are more reliable than others.”24 Some reliable types of retroversions include those supported by scribal errors in Hebrew, those supported by Hebraisms in LXX, and retroversions of variants which are preferable to the readings of MT.25 Types of retroversions which are less reliable, or even doubtful, include retroversions of words or phrases in non-literal translation units, retroversions of difficult words (especially hapax legomena), additions and omissions of personal names for the sake of clarity, mechanical disturbances of the text (e.g., haplography, dittography, or parablepsis) which could have occurred in either Greek or Hebrew, and harmonizations.26 Tov’s discussion of retroversions is helpful and illuminating. Nevertheless, some questions remain. For instance, Tov’s numerous examples deal almost exclusively with what he calls “content words.” He says that the reconstruction of “grammatical words,” that is, prepositions, particles, and conjunctions, and all grammatical categories must be 20 Ibid., 125. 21 Ibid. 22 Ibid., 128. 23 Ibid., 130-31. See also Goshen-Gottstein, “Hebrew Biblical Manuscripts,” 276-84. 24 Tov, Text-Critical Use, 131. 25 Ibid., 131-35. Of course, retroversions which produce readings preferable to MT cannot be

identified until the evaluation stage. 26 Ibid., 137-39.

207

considered less reliable than that of content words.27 However, though this statement may be accurate to a greater or lesser extent, he makes no effort here to test it. An examination of “grammatical words” and grammatical categories in 1 Samuel 3 lends some support to his position, yet contradicts it in other ways. For example, the deviation factor of Greek verbs, nouns, and adjectives (i.e., “content words”) in the chapter, 0.33, is less than the deviation factor of Greek adverbs, prepositions, and particles (included in Tov’s “grammatical words”), 1.27, thus indicating a greater degree of freedom in rendering adverbs, prepositions, and particles, and, by implication, a lesser likelihood of determining the exact Hebrew reading of the Vorlage. However, the deviation factor of the Greek conjunctions (also included among “grammatical words”) is 0.00, so it appears that the translators were in fact concerned to render conjunctions exactly, and therefore the Hebrew conjunctions lying behind the Greek conjunctions can be determined with precision, assuming that the pattern holds up in other chapters. Concerning grammatical categories, the tables in Appendix 3 that reflect the final translation technique indicate that in many cases the deviation factors for the renderings of certain grammatical categories are comparable to, or even less than, that for the rendering of “content words” (i.e., rendering of Hebrew verbs by Greek tense and mood [excluding verbals]: 0.43; rendering of Hebrew stem by Greek voice: 0.30; rendering of person in verbs: 0.00; rendering of number in verbs: 0.12; rendering of the use of Hebrew substantives by Greek case: 0.37 [when grouped as described on pp. 96-97]; rendering of number in pronouns: 0.00; rendering of the use of Hebrew pronouns by Greek case: 0.06 [when grouped as described on pp. 9697]). It seems, then, that “grammatical words” and grammatical categories may at times be rendered with precision. When they are, they are candidates for retroversion alongside “content words.” In his discussion of the support of variants among external sources, Tov, speaking of independently arising parallel elements in various witnesses, suggests that the correction of grammatical inconsistencies in the witnesses is relatively common, and such corrections should not be viewed as genetically related.28 Because he does not elaborate extensively on this suggestion, various questions arise: How exactly is grammatical inconsistency to be defined? Does it only refer to lack of agreement between subject and verb, as is sometimes the case with collective nouns? How widespread is grammatical inconsistency in the OT or in particular books? Under what circumstances should clearly anomalous forms be corrected? The answers to these questions are not straightforward, but more complete answers would lead to more accurate retroversions. Finally, a couple of other points raised by Tov’s discussion may be mentioned. First, Tov says that retroversions which are supported by Hebraisms in LXX are a type of 27 Ibid., 111. 28 Tov, Text-Critical Use, 129.

208

reliable retroversion. Although this statement seems reasonable, can it be supported that translators never introduced syntactical Hebraisms which were not called for by their Vorlage?29 Second, he mentions both the addition and the omission of proper names for the sake of clarity as types of retroversions that are not reliable. While many examples of the addition of names, pronouns, and other elements to a text to enhance its specificity exist, can the same be said for the omission of names?30 These are questions which warrant investigation. John R. Miles John R. Miles’s revised doctoral dissertation is devoted to the study of retroversion from one language to another.31 Though he uses retroversion from Ethiopic into Greek as his example, his methodology is equally applicable to other versions and their daughter translations. He seeks a more scientific approach to retroversion, one based less on the textual critic’s intuition and more on reliable data. Our concern is rather with “retroversion” as a more general problem in critical method. Briefly, we submit that if the modern critic can get from translation to original, it can only be because he knows how a given ancient translator got from original to translation; and that since this knowledge can only be acquired when both the original and the translation are available for study, his ability to “retrovert” results less from insight into the translation he is considering, presumably one for which the original is lost, than it results from insight transferred from his work on other translations for which the originals were available.32 To obtain data about a particular version’s translation technique, he suggests building a “syntacticon,” which compares syntactic structures in the source language to that in the target language.33 In his first chapter, he builds this “syntacticon” inductively, constructing it from a comparison of the Greek and the Ethiopic in Esther 1-8. He acknowledges some of the difficulties inherent in his method, for example, the assumption that the particular Greek and Ethiopic texts he analyzes are actually related to one another as source and translation, when in fact they are not. Nevertheless, he believes that these problems do not significantly alter the outcome of his study. He also notes that his choice 29 The analogy of the gospel of Luke may be instructive here. After the prologue (Lk 1:1-4), which is written in a Greek that approaches classical style, the remainder of the gospel is written in a kind of “Semitic Greek,” perhaps based on the style of LXX, though no (complete) Hebrew or Aramaic composition lies behind it. Cf. H. F. D. Sparks, “The Semitisms of St. Luke’s Gospel,” Journal of Theological Studies, o.s., 44 (1943): 129-38. 30 Though it requires further substantiation in a wider context, it seems probable that Jerome

omitted proper names, pronouns, and other elements that he felt were redundant in the context (e.g., in 3:9 and the pronominal suffix in 3:11; cf. also the substitution of eius for in 3:14, apparently to avoid redundancy); see above, pp. 184-86. No tendency to omit elements for stylistic reasons was discerned in any of the other versions in 1 Samuel 3. 31 John Russiano Miles, Retroversion and Text Criticism. 32 Ibid., 5. 33 Ibid., 1.

209

of a fairly literal unit of translation enhances his probability of successful retroversion.34 His “syntacticon” is based on the translation patterns not of individual words or grammatical structures but of groups of words that are related syntactically in the sentence. Specifically, he groups his examples according to what he sees as patterns of subordination: verb-to-verb subordination, verb-to-substantive subordination, substantiveto-substantive subordination, and so forth. He mentions briefly coordination of sentences and, more extensively, formulaic language. He describes formulaic language as follows: “Within the translation language, given forms of expression can be bound to given language situations in such a way that when the translator comes upon the situation in the original language, the form of his translation will be determined immediately and without reference to the form of the original.”35 For example, Ethiopic dates are often rendered according to a set form, regardless of the exact reading of LXX. Once his “syntacticon” is developed, he first tests its accuracy and usefulness by applying it to the Greek text of Esther 9 and comparing the results with the Ethiopic text. He begins with an exercise in translation rather than retroversion because the initial form of his “syntacticon” is a function that operates on Greek readings and produces Ethiopic readings. In addition, he says, it is more logical to go from Greek to Ethiopic first because that is the historical direction of the translation process. “Retroversion is not a matter of relating the translation to the original, it is a matter of recovering and reversing the relationship which the original once had to the translation.”36 For example, in his analysis of Esth 9:17, he finds five syntactical structures in the Greek text that are reflected in his “syntacticon.” Comparing the translational guidelines he has gleaned from chapters 1-8 with the extant Ethiopic text, he finds that the guidelines have been followed three times out of the five possibilities.37 Overall, he discovers that the Ethiopic text conforms to his predicted results in 70% of the cases.38 He suggests a number of improvements to his initial “syntacticon,” and the result is a complex system of logic tables and decision paths that he admits seem to require a computer to analyze.39 Having demonstrated the use of his “syntacticon” as a tool for predicting the Ethiopic text when confronted with the Greek, he next takes the Ethiopic text of Esther 10 and attempts to determine the Greek text behind it. In order to accomplish this retroversion, it is necessary for him first to invert his tables so that one can begin with Ethiopic rather than Greek. After doing so, he uses the transformed tables to predict the 34 Ibid., 6. 35 Ibid., 69. 36 Ibid., 130-31. 37 Ibid., 140. 38 Ibid., 149. 39 Ibid., 155.

210

retroverted Greek text. He finds that his tables produce the actual Greek text in 88% of the cases for which a syntactic entry exists. After applying his method to 1 Esdras 3 (Greek to Ethiopic) and 4 Baruch 1 (Ethiopic to Greek), he concludes, “the mechanical prediction of translation syntax and the recovery from translation of original syntax would seem on the basis of our study to be possible.”40 Miles’s study of the predictability of retroversion is interesting and informative. His work demonstrates the necessity of using hard data, rather than pure intuition, when discussing translation technique and when attempting retroversion. Nevertheless, a number of questions arise from his study. The first concerns his assumption that determining the rules of translation from source to target language guarantees that the process may be inverted, going from target to source language.41 In fact, the possibility of mechanical retroversion does not logically follow from a determination of rules for translation, since more than one syntactical structure in the source language may be resolved into a single structure in the target language.42 More significant is the question of the structure of his “syntacticon.” Not all syntactic relationships in a sentence can be described in terms of subordination, and the mapping of syntactical structures rather than the grammatical characteristics of individual words (i.e., case, gender, number, tense, etc.) begs the question of translation technique. It may be that the Ethiopic translators (or any other translators) looked beyond individual words at groups of words when translating, but it must be shown that they did so.43 If it can be demonstrated that the translators of a particular unit of translation primarily translated word by word, rather than phrase by phrase, the need for a “syntacticon” such as Miles describes disappears. On the other hand, if the translators show a propensity for translating certain syntactic constructions in a way that violates their normal word by word technique, a more limited “syntacticon” may be helpful.44 Noticeably lacking in Miles’s discussion of retroversion is any discussion of lexical choice. It is true that his main focus is on the predictability of syntax; nevertheless, since 40 Ibid., 200. 41 Ibid., 2. 42 In mathematical terms, if his “syntacticon” describes a function whose domain is the source

language and whose range is the target language, that function may not be invertible because a one-to-one correspondence between discrete structures in the two languages may not exist. He discusses the difficulties that arise from multiple references in the reversed (better: inverted) tables (pp. 158-59), but he does not adequately address what is a potentially complex problem. 43 Cf. the criticism of Anneli Aejmelaeus, review of Retroversion and Text Criticism: The

Predictability of Syntax in an Ancient Translation from Greek to Ethiopic, by John Russiano Miles, Theologische Literaturzeitung 111 (1986), col. 343: “Est ist kaum sinnvoll, syntaktische Erscheinungen nach den Wortklassen zu klassifizieren, die je miteinander verbunden und einander subordiniert erscheinen, nämlich Verb zu Verb, Verb zu Substantiv, Substantiv zu Substantiv usw.” 44 Thus, Aejmelaeus suggests a map of the translations of various prepositional phrases; ibid.

211

retroversion does require that lexical choices be made, some brief discussion of the subject would be helpful (cf. his discussion on word order, pp. 153-54). Also, his discussion of “omission, mistranslation, paraphrase, and unpredictable translation” (pp. 81-84) does not adequately address the problem of what the textual critic should do when confronted with such material. He says that no one can predict “what does not happen when an accurate mechanical translation is not made. . . .”45 However, if one can demonstrate that omission is caused by parablepsis, for example, one can predict at least certain aspects of the Vorlage. Moreover, both Margolis and Tov point out that certain mistranslations do point to specific readings in the Vorlage, especially if the word apparently read is graphically similar to the presumed original. In conclusion, Miles’s methodology for retroverting a translation by using tables that map the correspondence of syntactic structures in the source and target languages supports the need to have hard data before deciding upon a particular retroverted reading. However, it may be doubted whether retroversion can really be as mechanical as he claims, especially in the light of varied renderings of identical or similar syntactic structures.46 Furthermore, he has not demonstrated that such a complex approach to translation, operating on the level of syntactic structures rather than on the level of individual words, is reflected in the Ethiopic translation, much less in the translations used in this analysis. Thus, his method has limited application in the present study. Other Suggestions In addition to these more extensive studies of retroversion, shorter observations on various aspects of retroversion have been made. Isac Leo Seeligmann, in an article discussing contemporary Septuagint research, discusses the relationship between the Hebrew and Greek texts of the OT. He stresses the importance of first establishing the text of LXX itself and of determining the translation technique of the particular book in question.47 In agreement with Margolis and Tov, he observes that Hebraisms and errors in translation often allow the reconstruction of the original reading.48 Finally, Seeligmann 45 Miles, Retroversion, 83. 46 Cf. the comment of Aejmelaeus, review, col. 344: “Davon ist die Rez. [i.e., Aejmelaeus]

jedoch immer stärker überzeugt, daß übersetzungstechnische Studien und Rückübersetzung nicht mechanisch nach gegebenen Regeln zu betreiben sind.” In another context, F. E. Deist says, “[these characteristics of P] sound a warning to the retroverter not to go about his work in a mechanical way, but to take the character of the whole into account so as to make wise decisions on whether a particular reading does in fact constitute a variant reading or not”; Ferdinand E. Deist, Witnesses to the Old Testament, The Literature of the Old Testament, vol. 5 (Pretoria: NG Kerkboekhandel, 1988), 179. 47 Isac Leo Seeligmann, “Problemen en perspectieven in het moderne Septuaginta-onderzoek,”

Jaarbericht ex oriente lux 7 (1940), 382. 48 Ibid., 377; cf. also p. 382. This assertion has been noted and discussed briefly above (pp. 207-

8).

212

says that the correspondence of a LXX variant with another unrelated witness suggests a variant Hebrew Vorlage. 49 If true, this observation is applicable not only insofar as it identifies a variant to be retroverted, but it also needs to be considered at the stage of evaluation and when creating the critical apparatus (but see above, pp. 199-200). F. E. Deist does not deal with retroversion in a systematic way, but he does offer several tips on retroversions for the various versions. He bases his remarks on retroverting LXX on Tov’s Text-Critical Use of the Septuagint, so that information will not be repeated. For T, Deist says, one must bear in mind that the translators produced targums for liturgical purposes, sometimes adapting their text to fit the liturgical circumstance;50 different targums utilize different translation techniques; and targums frequently make use of various types of interpretation imbedded in the text: peshat, halakah, midrash, and haggadah. 51 Speaking of P, Deist points out that the translators frequently translate Hebrew synonyms with a single Syriac word, though occasionally the reverse occurs; they often concretize Hebrew metaphors; they sometimes vocalize or divide words differently than MT; they render difficult Hebrew words inconsistently; and they sometimes translate proper names.52 When attempting to retrovert a reading in V, one must be aware of Jerome’s stylistic tendencies (e.g., preference for “eloquent” or vivid readings and a somewhat negative attitude toward women and childbearing) and his occasional use of different vocalization, word division, or sentence division.53 Before detailing the conclusions about retroversion that have been garnered from the preceding discussion and from work with the text itself, the question of the orthography of the reconstructed texts and of Hebrew mss no longer extant requires discussion. Orthography Though the mss of MT were produced in the Middle Ages, they reflect a much older text, both in content and in orthography. However, probably no scholar would claim that the spelling now found in MT (with all its variety in individual mss) is an accurate reflection in all its particulars of the orthographic practices in vogue at the time when the various biblical books were composed, in the case of Samuel, probably sometime in the sixth century B. C. E. At most, MT reflects an orthography current in perhaps the third or fourth centuries B. C. E. 54 If MT reflects an orthography later than that employed when 49 Ibid., 383. 50 Cf. the discussion of Christian lectionaries, which were also used for liturgical purposes; Aland

and Aland, Text of NT, 166. 51 Deist, Witnesses, 133. 52 Ibid., 178-79. 53 Ibid., 189-91. 54 Frank Moore Cross and David Noel Freedman, Early Hebrew Orthography: A Study of the

213

Samuel was written, should it be modified to reflect earlier orthographic practices? Furthermore, what kind of orthography should reconstructions based on other versions reflect? These questions must be answered before a reconstruction of an earlier form of the text is attempted. In 1952, Cross and Freedman attempted to trace Hebrew orthographic practices as reflected in extant inscriptions and other extrabiblical material, most of which were preexilic. Basing their work on earlier studies by Albright, they discerned a development in Hebrew orthographic practices, as follows: (1) prior to the tenth century, Hebrew writing was purely consonantal (phonetic consonantism); (2) beginning in the ninth century, Hebrew scribes began using yod, waw, and he as matres lectionis to represent word-final vowels, probably borrowing the practice of their Aramaean neighbors; (3) by the sixth century, vowel letters were used sporadically as internal matres lectionis in Hebrew texts; (4) after the sixth century, the use of internal matres lectionis increased greatly.55 In the light of subsequent epigraphic discoveries, they later modified their portrayal, allowing that rare instances of internal matres lectionis began as early as the eighth century.56 Cross and Freedman’s characterization of the development of Hebrew orthographic practices serves as a starting point, or at least a backdrop, for many recent discussions on orthography. In addition to those studies that accept Cross and Freedman’s characterization with little or no significant modification, but only further refinement,57 some have criticized their conclusions to a greater or lesser extent.58 Probably the most important of these critiques has been that of Ziony Zevit, who concludes on the basis of Epigraphic Evidence, American Oriental Series, no. 36 (New Haven: American Oriental Society, 1952), 69-70. Cf. also Francis I. Andersen and David Noel Freedman, “Another Look at 4QSam b,” Revue de Qumran 14 (1989): 22: “. . . we can infer that the Massoretic system and set of spelling rules were firmly in place in all principles and particulars by the third century BCE.” 55 Cross and Freedman, Early Hebrew Orthography, 45-60. 56 Idem, Studies in Ancient Yahwistic Poetry, Society of Biblical Literature Dissertation Series,

no. 21 (Missoula, MT: Scholars Press for the Society of Biblical Literature, 1975), 182. 57 E.g., Francis I. Andersen and A. Dean Forbes, Spelling in the Hebrew Bible: Dahood Memorial

Lecture, Biblica et Orientalia, no. 41 (Rome: Biblical Institute Press, 1986), 31-65, esp. 65: “Between the extremes of Bange and Zevit, the Cross-Freedman schema remains the best working hypothesis”; Freedman, Mathews, and Hanson, The Paleo-Hebrew Leviticus Scroll , 51-53. Cf. now also David Noel Freedman, A. Dean Forbes, and Francis I. Andersen, eds., Studies in Hebrew and Aramaic Orthography, Biblical and Judaic Studies from the University of California, San Diego, vol. 2 (Winona Lake, IN: Eisenbrauns, 1992). 58 E.g., Donald Watson Goodwin, Text-Restoration Methods in Contemporary U.S.A. Biblical

Scholarship, Pubblicazioni del Seminario di Semitistica, Ricerche, no. 5 (Naples: Istituto Orientale di Napoli, 1969); L. A. Bange, A Study in the Use of Vowel-Letters in Alphabetic Consonantal Writing (Munich: UNI-DRUCK, 1971); Ziony Zevit, Matres Lectionis in Ancient Hebrew Epigraphs, American Schools of Oriental Research Monograph Series, no. 2 (Cambridge, MA: American Schools of Oriental Research, 1980).

214

extensive epigraphic evidence that the use of internal matres lectionis was much more prevalent in preexilic times than Cross and Freedman would allow.59 Though all of his conclusions have not been unanimously accepted,60 his reviewers have consistently praised his work as an important study, and some have agreed with Zevit that the system of Cross and Freedman is in need of reevaluation in the light of new discoveries.61 The issue of the development of Hebrew orthography applies most significantly to the task of reconstructing earlier forms of the text of a biblical book when one considers whether or not it is possible to reconstruct accurately the orthography of the period in question. Cross and Freedman quite confidently reconstruct the texts of early Israelite poems, using purely consonantal orthography to do so.62 However, many scholars view the reconstruction of the original orthography of a passage in a biblical book as problematic,63 and the relatively late date of the final editing of Samuel64 (sixth century B. C . E., a period of transition in the orthographic practices even according to the calculations of Cross and Freedman) makes certainty in the reconstruction of the orthography of this book impossible. Another factor that complicates the search for the original orthography is the use of variable spellings of the same word in MT,65 a trait now 59 Cf. Zevit, Matres Lectionis, 35: “From the 7th century on, Judean scribes had available to them

a system of matres lectionis which they could use, if they wished, to indicate long vowels both in word final and medial positions. Judging from scribal practice as exemplified in the inscriptions and letters of this period, composition with m.l. was the norm rather than the exception. It can therefore be assumed that m.l. were employed in compositions originating during this period: Deuteronomy, the Deuteronomic history, Isaiah 1-39, Micah, Jeremiah, Habakkuk, etc.” 60 See especially Joseph Naveh, review of Matres Lectionis in Ancient Hebrew Epigraphs, by

Ziony Zevit, in Israel Exploration Journal 33 (1983): 139-40; Dennis Pardee, review of Matres Lectionis in Ancient Hebrew Epigraphs, by Ziony Zevit, in Catholic Biblical Quarterly 44 (1982): 503-4. 61 See especially James Barr, review of Matres Lectionis in Ancient Hebrew Epigraphs, by Ziony

Zevit, in Journal of the American Oriental Society 104 (1984): 374: “It does indeed deeply damage the position taken by Cross and Freedman.” 62 Cross and Freedman, Ancient Yahwistic Poetry. Cf. also William F. Albright, “The Oracles of

Balaam,” Journal of Biblical Literature 63 (1944): 207-33; idem, “The Psalm of Habakkuk,” in Studies in Old Testament Prophecy Presented to Professor Theodore H. Robinson, ed. H. H. Rowley (Edinburgh: T. & T. Clark, 1950), 1-18. 63 Cf. Tov, Textual Criticism of the Hebrew Bible, 223: “The biblical books that were composed

in an early period, and in particular the ancient poetry, were probably written in a very defective orthography, but this assumption does not provide a solid basis for the reconstruction of that orthography.” 64 Further editorial modifications of Samuel apparently continued even after the text reached its

final form in most respects in the sixth century. See Tov, “The Composition of 1 Samuel 16-18,” 97130; idem, Textual Criticism of the Hebrew Bible, 334-36, 346-47; and discussion below, pp. 237-40. 65 Francis I. Andersen and A. Dean Forbes, “Orthography and Text Transmission: Computer-

Assisted Investigation of Textual Transmission through the Study of Orthography in the Hebrew Bible,” Text: Transactions of the Society for Textual Scholarship 2 (1985): 25-53; eidem, Spelling in the Hebrew Bible; Barr, Variable Spellings. Andersen and Forbes purport to find patterns of defective and plene spelling in the various books, with patterns of defective spelling more prominent in the books whose text was standardized earlier, notably the Primary History and the three major prophets (Spelling in the Hebrew Bible, 317). Barr denies that such patterns prove the earlier composition or standardization of the text (cf.

215

observed in other ancient Near Eastern literature as well.66 Perhaps the most pertinent orthographic issue, at least in 1 Samuel 3, involves the spelling of the sound at the time the book was written, and especially as it relates to the third masculine singular pronominal suffix attached to masculine singular and plural nouns. Though Zevit believes that a few instances of waw as final mater lectionis for exist (examples 41, 53, 99[?], 109), Anson Rainey finds Zevit’s discussion weakest at this point.67 Andersen and Freedman offer an extended discussion of the orthographic representation of the third masculine singular pronominal suffix. They say that the replacement of the preexilic suffixes -h and -w on masculine singular and plural nouns by -w and -yw, respectively, probably occurred simultaneously sometime during the fifth century, as evidenced by the rather frequent appearance of the older suffixes in Samuel and Ezekiel but their rarity in Chronicles.68 More than one hundred instances of the older suffix -w with a plural noun are preserved in the kethib-qere notations. Since many of them occur in Samuel, including two in 1 Samuel 3 (vv. 2, 18), the older orthography of this suffix will be preferred throughout the reconstruction. Similarly, the shorter reading of forms like and will be preferred to the more common Masoretic forms and , and the yod before pronominal suffixes in masculine plural nouns will be omitted as a later form.69 In light of the previous discussion, the following procedure regarding orthography will be adopted in the present study. Whenever the reading of MT (not its orthography) is determined to be original, its orthography will not be modified, with the exception of the third masculine singular suffix and certain preposition-pronoun combinations, as discussed in the preceding paragraph. When a reading reconstructed from one or more of the versions is preferred to MT, the orthography of the reconstruction will agree with the p. 38), claiming instead that a single Masoretic orthography exists for the entire Bible (p. 204). The position of Andersen and Forbes is strengthened to some extent by a comparison of 4QSamb with MT: though sharing many of the plene readings of MT, 4QSamb is consistently defective in its renderings of - , thus indicating the preservation of one aspect of an older orthographic system (Andersen and Freedman, “Another Look at 4QSam b ,” 28). 66 A. R. Millard, “Variable Spelling in Hebrew and Other Ancient Texts,” Journal of Theological

Studies, n.s., 42 (1991): 106-15. 67 Anson F. Rainey, review of Matres Lectionis in Ancient Hebrew Epigraphs, by Ziony Zevit, in

Journal of Biblical Literature 102 (1983): 632. 68 Andersen and Freedman, “Another Look at 4QSam b ,” 23-27. However, the singular form with

-w may have occurred by the sixth century in Hebrew, at least occasionally, as in lw in an inscription from Khirbet Beit Lei; see Zevit, Matres Lectionis, 30-31. G. I. Davies, though mentioning lw as a possible reading, prefers an alternative reading of the inscription: G. I. Davies, Ancient Hebrew Inscriptions: Corpus and Concordance (Cambridge: Cambridge University Press, 1991), 89. 69 Cf. Zevit, Matres Lectionis, 27-28, where he notes that

appears in a sixth century ostracon from Arad. See also Freedman, Mathews, and Hanson, The Paleo-Hebrew Leviticus Scroll, 80, where two occurrences of are noted in 11QpaleoLev (15:2; 17:2). Several of these shorter forms occur in MT of Samuel.

216

orthography of other instances of the word in MT, if they exist. An exception to this rule will be made if it can be determined that a particular spelling of the Hebrew Vorlage led to readings found elsewhere in the text-traditions. It is true that the procedure adopted may result in a somewhat mixed orthography, but that is also the nature of all the extant Hebrew witnesses, including MT.70 Finally, the role of orthography in textual evaluation requires some mention. If older portions of the Bible were composed using a greater concentration of preexilic orthographic practices (especially defective spelling), can the presence of an older orthographic form be used as evidence of the antiquity of a given reading? Andersen and Freedman suggest that it can, noting that the reading in 1 Sam 16:4 (4QSamb, in agreement with LXX), were it added by postexilic scribes, would probably be spelled . 71 However, even if earlier spelling practices, such as defective spelling, did predominate in portions of the Bible that were composed earlier, scribal copying has tended to replace most of these older readings, albeit somewhat inconsistently. The existence of defective readings even in patently late books (e.g., Dan 11:38; Esth 8:16; Neh 11:1) suggests the doubtfulness, if not impossibility, of equating archaic orthography with early readings. Such arguments, if they are advanced, must be made with due reservation. Conclusions concerning Methodology It must be admitted that retroverting a translation is a subjective venture in most cases (with the general exception of proper nouns). However, as Tov points out, certain categories of retroversions are reasonably reliable, namely, those supported by identifiable scribal errors in Hebrew, those supported by Hebraisms, and those that result in readings that are preferable to MT. Concordances and lexicons will serve as useful tools in the process of retroversion, as will the lists of lexical equivalents between the target languages and the Hebrew in 1 Samuel 3 for each of the secondary versions, found in Appendix 3.72 When the reconstruction requires grammatical structures not present in MT, the grammatical tables in Appendix 3 will be used for reference.73 It will sometimes occur that the Vorlage of a version is uncertain, either because no equivalent that can be easily explained as a deviation from MT or one of the other versions exists, or because more than one possible reading exists. In the latter case (e.g., the 70 This statement is true regardless of the position one adopts concerning the possibility of

recovering an earlier orthography of portions of MT. Thus, Cross and Freedman say, “The Hebrew Bible which tradition has delivered to us is in reality a palimpsest; underlying the visible text, the varied spelling customs of older ages have been recorded” (Cross and Freedman, Early Hebrew Orthography, 1); cf. Barr’s “zone of variable spellings” (Barr, Variable Spellings, 204-5). 71 Andersen and Freedman, “Another Look at 4QSam b ,” 8. 72 See below, ***-*** (LXX), ***-*** (P), ***-*** (T), ***-*** (V). 73 See below, ***-*** (LXX), ***-*** (P), ***-*** (T), ***-*** (V).

217

decision of whether to render with , , or -), the use in the chapter (as indicated in the lexical and grammatical tables) and the rest of the book is an important guideline, but it must be acknowledged that at times a subjective, almost arbitrary, decision must be made when the data does not favor one reading over the others. In the case where no good Hebrew equivalent seems to exist for a versional reading, one that reflects as much as possible both the versional evidence and the possibility of a scribal error in Hebrew will be attempted. In some cases, it may be preferable simply to admit that no single reconstruction is compelling, and to offer more than one. In other cases, it may be best to retain the reading of MT and simply note the versional variant. Retroversions of the Secondary and Partial Secondary Witnesses In the retroversions that follow, MT is taken as the starting point for reconstructing whole verses, and deviations from MT based on the version in question are indicated by text in a larger type. If the reconstructed Hebrew text omits one or more items found in MT, the symbol «» (European quotation marks) will appear in place of the omitted item(s). The texts of the versions on which the following reconstructions are based can be found for the most part in the standard editions of those versions, though modifications of T and the fully reconstructed texts of LXX, LXXL, and LXXO appear in Appendix 1.74 Individual comments and references to earlier discussions are given for each of the readings. The symbol ‘<’ in the following paragraphs is used to denote the Hebrew reading retroverted from a versional reading. Septuagint (3:1)

The deviations from MT in this verse are LXX variants 70 ( < , see above, pp. 100-101) and 3 ( < , see above, pp. 92-93). Variant 70 is a quantitative variant, not represented in MT. Although does not occur in 1 Samuel 3 MT, in the 700+ cases of that also appear in MT, all of them render , so this retroversion is certain. The other variant, variant 3, is a variant in consistency. It appears that the translators read (which could conceivably be a qal passive participle, but which they took as a qal active participle) rather than . Since their rendering reflects an active rather than a passive meaning, it is doubtful that was intended to render a niphal verb with a passive meaning. 74 See above, pp. 36-37, for complete details.

218

(3:2) This verse contains LXX variants 6 ( < , see above, pp. 86-87) and 71 (- < , see above, p. 101). and related Greek words are usually used in LXX to render Hebrew words related to ; only in the present case does a form of correspond to in MT. Though the use of in Gen 48:10 (rendering ) could have suggested the unique rendering, the possibility of graphic confusion makes it probable that the translators were seeking to render a form of . 75 Although conjunctions are rather easily added or omitted in all witnesses, the concern of the translators of LXX for fidelity to their text in most instances and the fact that the additional conjunction appears in many Masoretic mss and in T makes it likely that the Vorlage of LXX contained a conjunction. The absolute consistency with which the translators of LXX render conjunctions, as well as the context, makes it certain that that conjunction was - . «»

(3:3)

This verse contains LXX variant 72 («» < omission of , see above, pp. 100101). The translators apparently did not find in their Vorlage. «»

(3:4)

Verse 4 contains two LXX variants, 74 («» < omission of , see above, p. 102) and 75 ( < , see above, p. 102). Though the omission of would not normally be considered significant, in the present verse it is possible that in MT is actually the remnant of an original , as LXX seems to imply. 76 (3:5)

This verse exhibits two deviations from MT, LXX variants 76 ( < , see above, pp. 100-101) and 12 ( < , see above, p. 87). In the first case, the retroversion could probably just as easily be (cf. in 3:5, 6, 8), but the occurrence of in the immediate context (3:9) suggests that the longer form may also have appeared in the Vorlage. 77 75 Cf. also the reading of P,

. Since is a common rendering of , it is possible that the translators read or something similar here. However, the occurrence of in 4:15 as a rendering of (MT) precludes considering the variant in 3:2 significant, since seems to be a common Syriac idiom for speaking of blindness. 76 Thenius, Bücher Samuels, 16; Wellhausen, Bücher Samuelis, 52; S. R. Driver, Notes on the

Books of Samuel, 42. 77 In Samuel and Kings, cf. also 1 Kgs 8:43, 52. The only occurrence of

with in these books is 1 Sam 28:15, but it is possible that the preceding seghol influenced the choice of prepositions

219

Concerning variant 12, is the regular rendering of in Kingdoms and throughout the OT, rendering it some forty times in Kingdoms alone. In addition, if the Vorlage of LXX indeed differs from MT, the occurrences of and together in 3:5, 6 makes the present retroversion certain. «»

«»

(3:6)

«» Several differences from MT occur in the reconstructed Vorlage of LXX in verse 6: LXX variant 78 (- 2° < , see above, pp. 101-102), variant 13 ( [ ]< , see above, pp. 90-91), variant 79 («» 1° < omission of , see above, pp. 100-101), variant 80 ( 2° < , see above, p. 102), variant 81 («» 2° < omission of , see above, p. 103), variant 82 ( < , see above, pp. 100101), variant 84 ( < , see above, pp. 102-3), and variant 83 («» 3° < omission of , see above, pp. 102-3). Variants 78 and 13 may be considered together. As mentioned above, the consistency of the translators in rendering conjunctions, as well as the context and the paucity of conjunctions in Hebrew, makes the retroversion of the conjunction certain in variant 78. The idiom (reading right to left) “imperfect + ” is a Hebraism found elsewhere in the OT, including 1 Sam 19:21,78 so its retroversion is also reasonably certain. The three omissions in the verse require little comment in the area of retroversion, since all of them are “content words” or phrases. They are not generally omitted by literal translators if present in the Vorlage, unless by mistake (e.g., parablepsis), as may be the case in variant 81.79 Similarly, the retroversion of the proper name in variant 80 is obvious. For a brief discussion of the retroversion of in variant 84, see the discussion of variant 76 in verse 5. The only variant in this verse that provides any difficulty at all in retroverting is variant 82, where renders . 80 It is true that renders a number of different Hebrew words, but the constructions and always render some form of the Hebrew , so this retroversion is also fairly certain. here. 78 Cf. BDB, s.v. “

.”

79 Since it cannot be determined in many cases whether parablepsis occurred in the Vorlage, at the

point of translation, or early in the transmission of the translated text, it seems best to operate as though the error were in the Vorlage. If parablepsis is confirmed in the evaluation phase as a contributing factor to the disturbance, the reading will be regarded as secondary at that point. 80 For a discussion of the reading

vs.

of B, see above, pp. 44-45.

220

«»

(3:7)

The variants in this verse are LXX variants 15 ( < , see above, pp. 8788) and 86 («» < omission of , see above, p. 104). Concerning variant 15, although and are equivalent in the sense that both refer to the God of the Israelites, the translators of LXX generally make a distinction in their rendering of the two names, using for and for . The tendency in at least some scribal circles to replace with on a more or less regular basis (cf. Pss 14 and 53) suggests that the exchange may have occurred in Hebrew rather than at the point of translation, especially since the translators render all fifteen other occurrences of in the chapter by . The retroversion of variant 86 is straightforward since it involves an omission. «»

(3:9)

The following LXX variants appear in verse 9: variant 87 («» < omission of , see above, pp. 100-101), variant 18 ( < , see above, p. 87), and variant 89 ( < , see above, pp. 100-101). Like previous omissions, that in variant 87 requires no comment as far as retroversion is concerned. Variant 18, like variant 12 in verse 5, can only be a rendering of the root , if indeed the variation from MT was present in the Vorlage. A wooden retroversion of (variant 89) would be , but the appearance of the equivalent < in 3:16, the reading in 3:6 MT, and many other similar readings involving the vocative (1 Sam 2:24; 4:16; 24:17; 26:17, 21, 25, and numerous instances throughout the OT) demonstrate the certainty of the retroversion. «»

(3:10)

The deviations from MT in verse 10 are LXX variants 90 ( < , see above, pp. 100-101) and 91 («» < omission of , see above, p. 102). As is the case with variants 76 and 84, either or are possible retroversions, and both forms appear in the chapter. However, a survey of the usage of and with the verb in Samuel and Kings shows that appears in 2 Sam 9:2; 11:13; 1 Kgs 12:3; 2 Kgs 18:4; only in 2 Sam 15:2 does follow , and in this case an intervening word occurs between the verb and the prepositional phrase. Thus, the shorter form has been chosen for the retroversion. The retroversion of the omission of in variant 91 requires no comment.

221

«»

(3:11)

Two variants from MT occur in verse 11: LXX variants 92 ( - < , see above, pp. 100-101) and 93 («» < omission of , see above, p. 104). In variant 92, LXX has an additional personal pronoun in the genitive case, which can only be retroverted with a pronominal suffix. The omission in variant 93, though involving a “grammatical word” rather than a “content word,” seems to have been based on the absence of any conjunction in the Vorlage, since the translators regularly render with either a relative pronoun or an article (substituting for the relative pronoun). (3:12) The retroverted version of verse 12 has two variants from MT, LXX variant 26 ( < , see above, p. 88) and variant 27 (- < , see above, p. 88). Since the translators normally render in one of two ways, either by or with only a case ending (usually dative), the renderings in both cases are suspicious. In variant 26, the reading of LXX suggests that the translators rendered rather than , two words that are frequently interchanged in Samuel.81 Although variation between and in Hebrew mss and their similar usage in Samuel are valid text-critical issues, the question at the point of retroversion is not whether or not the words were interchangeable in the mind of the author, but rather how the translators rendered them. The distribution in 1 Samuel 3 suggests that is not used to render , whereas it frequently renders , and the graphic and aural similarity between the two prepositions strengthens the retroversion. On the surface, in variant 27 seems to be a perfectly good translation of ; however, the translators of Samuel do not seem to have equated the two. On the other hand, frequently renders - , - , and - (he locale); either of these first two is a possible retroversion in the present case. In 1 Sam 19:3, 4; and 25:39, is used with - when describing a conversation about someone else, but in each case LXX renders the preposition with . Every instance of with - is rendered with or some other preposition, or simply the dative case; nowhere in the OT does render after . Furthermore, - is used primarily to refer to direct address, not in reference to a third party. The only other example of with in LXX is in 2 Kgdms 13:20, where Absalom tells his sister, “ .” 82 Unfortunately, no Hebrew equivalent for occurs in MT. 81 Sperber, Bible in Aramaic, 4b:111, gives several examples of the indiscriminate use of

and

in MT. 82 Another instance of

with occurs in Esth 4:8, but is used in a way semantically dissimilar to the one in 1 Sam 3:12 (in a terminal accusative phrase), so it is irrelevant.

222

However, the third party connotation associated with the preposition confirms the same connotation in 1 Sam 3:12, and this connotation in turn suggests that a may have slipped out of the Vorlage of LXX by haplography. Thus, seems to be the best possible retroversion, though one cannot rule out other possibilities, including corruption in the Greek text. «»

(3:13)

Verse 13 contains several sets of variants from MT: LXX variant 30 ( [ ]< [ ] , see above, p. 92), variant 32 ( < , see above, pp. 9495), variant 33 ( < , see above, pp. 88-89), variant 94 («» < omission of , see above, p. 103), and variant 95 ( < , see above, p. 103). The first of these variants, variant 30, involves a difference in time: MT reads (apparently a waw consecutive with a perfect verb, and thus a reference to the future), whereas LXX has (a perfect, and clearly a reference to past time). The two other occurrences of waw consecutive with perfect in the chapter are rendered by Greek future tenses. Although three cases are not sufficient to establish a definite trend, the connotation of future time usually associated with the waw consecutive plus perfect and the support of other versions (T and V) makes it likely that the Vorlage of LXX contained some other construction, one referring to a past revelation to the house of Eli (1 Sam 2:2736). The most obvious construction is a waw consecutive with an imperfect verb, the most common verbal form in Hebrew narrative. Although the graphic similarity of the two forms is not great, no other solution seems more likely.83 Variant 32 has a plural in LXX where MT has a singular (cf. also T, but see above, p. 139), so retroversion is straightforward. Variants 94 and 95 may be considered together, since they are substitutional variants of one another. Though the origins of these two readings may be related, it is impossible at this point to determine with confidence what that relationship is, since there is little if any graphic or aural similarity. Despite the difficulties surrounding the question of the origin of the readings, their retroversion is a simple matter. The omission in variant 94 requires no comment, and the retroversion of is obvious. The final variant in verse 13 involves the reading for in variant 33. The Masoretic notes indicate that is a scribal correction, an attempt to avoid the combination . The retroversion is supported by both the tiqqun sopherim and the graphic similarity of and . 83 Some graphic similarity between

and does exist in certain forms of the Egyptian cursive script, and this similarity may have contributed to the confusion; cf. Kautzsch, ed., Gesenius’ Hebrew Grammar, Table of Alphabets following p. xvi, col. 13.

223

(3:14)

LXX variant 34 ( < , see above, p. 89), is the only significant variant in verse 14, and it involves a difference in word division and the presence or absence of a mater lectionis. is the most common rendering of LXX, and the graphic similarity of and makes the retroversion certain.

in

(3:15)

The only LXX variant in 3:15 is variant 96 ( < , see above, p. 103). is the regular rendering of in LXX, and is the most common (and in 1 Kingdoms the exclusive) rendering of the independent . The latter reading is supported by the probability that the shorter text is the result of parablepsis from to . The reading is also graphically similar to earlier in the verse, and this similarity may also have contributed to the parablepsis. «» (3:16) This verse contains three sets of variants: LXX variants 40 ( < , see above, p. 88), 67 (position of < position of , see above, pp. 99-100), and 98 («» < omission of , see above, pp. 103-4). The first variant may be disposed of quickly. Whereas is never rendered by any Greek equivalent in 1 Samuel 3 (the present case excluded), is one of the two standard renderings of (the other being no Greek equivalent except case endings). Furthermore, the verb is frequently followed by before the object. Thus, it is probable that here represents rather than . The other two sets of variants need to be considered together. MT reads , whereas LXX reads . Apparently some sort of disturbance has occurred in the text, since (which is regularly rendered by ) is not represented, and (normally equivalent to ) comes at the beginning of the verse, the place occupied in MT by . Since is the normal rendering of , occurring later in the same verse, and since rarely renders elsewhere (only in Gen 45:1; 1 Kgs 13:21; Zech 7:13; Isa 32:5; 44:5 [ms A], out of almost 4000 occurrences of ), it is almost certain that was present in the Vorlage of LXX. Thus, even though the precise mechanism by which the two diverse readings emerged from an earlier text are unclear, the retroversion given above is sound.

224

(3:17) «» Verse 17 contains two variants, LXX variant 100 («» < omission of , see above, p. 103) and variant 101 ( < , see above, p. 103). Since these variants are substitutional variants of one another, they may be considered together. It is likely that the rather cumbersome Greek phrase is related to the of MT, so the retroversion is plausible from that perspective. Once this relationship is observed, retroversion is straightforward. «»

(3:18)

The two variants in the present verse are LXX variants 102 («» < omission of , see above, pp. 100-101) and 103 ( < , see above, pp. 100-101). Since these two variants are of the sort that are certain retroversions (i.e., an omission and a proper noun), no more need be said. (3:19) The variant in verse 19, LXX variant 68 ( [ ]<[ ] , see above, p. 100) involves a difference in word order, for MT reads . It is easy to see the graphic similarity between the readings found in MT and in the retroversion, and the confusion would have been heightened if the Vorlage of LXX used an abbreviation for such as or . The switch from perfect in MT to imperfect in the retroversion is the result of the waw consecutive. «»

(3:21)

Verse 21 has only two variants, but both involve multiple words: LXX variant 104 («» < omission of , see above, p. 105) and variant 105 ( <

, see above, p. 105). The first of these variants may be dealt with summarily, since it involves the omission of several words, all of which are “content words.” Variant 105, on the other hand, involves the retroversion of many words, but

225

approximately the first half of the addition seems to be a variation of verse 20, so the wording there can help in the retroversion process. Verse 20 reads . Starting with this sentence as a base, the following differences may be noted. First, the structure of the sentence is changed, so that the subject ( ) and its adverbial modifier (... ... ) in verse 20 are now the object of the preposition < , while the subordinate noun clause beginning with in verse 20 becomes the main clause (after and are dropped). The change in order and in conjunctions shifts the participle of verse 20 to an imperfect with a waw consecutive in the reconstruction. The absence of the preposition before in verse 21 suggests that should not be preceded by , as in verse 20. The presence of , which is as superfluous in Greek as is its equivalent in Hebrew, suggests that the infinitive was in the Vorlage (cf. the similar constructions with in 2 Sam 7:8 and 1 Kgs 1:35 [where, however, is used in LXX]; cf also 1 Sam 24:16, and ). Finally, the adverbial phrase that appears as in verse 20 is replaced by another, approximately equivalent, phrase: (cf. Deut 13:8; 28:64). The second half of the retroversion can be divided into two phrases. The first phrase is based on . The word renders in twenty-eight of its thirty-two occurrences in LXX, and renders in at least ninety-five percent of its 350+ cases. The second phrase, , begins with a Hebraism, which surely reflects a construction with the infinitive absolute followed by the perfect, carrying the idea of continuous behavior (cf. Judg 14:9; 2 Sam 5:10).84 In the second part of phrase, and are by far the most common translations of and , respectively, and regularly renders (though is also possible), so the retroversion is relatively certain.85 Peshitta (3:1)

The only deviation from MT in this verse is P variant 48 ( < see above, p. 121), a variation in word order. It is probable that the variation arose in Hebrew rather than Syriac because of the graphic similarity of and . Since no 84 Waltke and O’Connor, Introduction, 589-90. 85 Cf. the retroversions offered in the commentaries: Thenius, Bücher Samuels, 18; Dhorme,

Livres de Samuel, 45; Budde, Bücher Samuel, 29; Klostermann, Bücher Samuelis, 13.

,

226

change in vocabulary or grammatical form is required, the retroversion is a simple one. (3:2) Verse 2 also contains one significant variant in P, variant 56 (- < - , see above, p. 123). The translators of P were consistent in using w to render the Hebrew (cf. the nonsignificant P variant 35, above, pp. 116-17), so this retroversion can be made with certainty. (3:3) The variant in verse 3 is P variant 8 ( < , see above, pp. 114-15). in P corresponds to in MT only here in chapter 3, compared to fifteen occasions where it corresponds to . is used twice in the chapter to render . Data from the rest of 1 Samuel 1-16 suggests that the translators were reasonably consistent in distinguishing between and in their renderings, though they do show more consistency in rendering with . 86 Nevertheless, it is likely that the translators read in their Vorlage. (3:6)

One significant variant appears in 3:6, P variant 49 ( < , see above, pp. 121-22). MT here reads , so this variant involves a change in word order. Since the translators were meticulous in following the word order in the rest of the chapter, it is likely that the deviation from MT occurred in their Vorlage, which has been reconstructed above. «»

«»

(3:9)

Two variants in quantitative representation occur in verse 9, P variants 79 («» < omission of -, see above, p. 127) and 81 («» < omission of - , see above, p. 127). Since both variants involve omissions from the Hebrew text of elements normally represented in translation, the retroversions of the omissions are straightforward. In the case of variant 81, however, the verb itself must also be changed, either to an imperative ( ) or to an imperfect ( ). Table 21 (see Appendix 3, p. ***) indicates that all eight imperatives in MT are rendered by imperatives in P, but none of the others is part of a conditional sentence, as in the present case. Insufficient data exists in 1 Samuel 3 to accurately predict which form might have appeared in the Vorlage. However, Hebrew idiom suggests that the imperfect would have been preferred, though the waw is usually 86 Cf. de Boer, I Samuel i-xvi, 23-24.

227

attached as well. Perhaps the waw was smeared and read as a taw at some point in the transmission of the text, leading to the anomalous form , from which a scribe dropped the final taw. It is possible that this variant, and possibly also variant 79, would need to be reconsidered if a larger amount of material, which included more conditional sentences, were analyzed. «» (3:14)

Verse 13 also involves the omission of an element present in MT, P variant 94 («» < omission of - , see above, pp. 123-24). As in variant 81, the conjunction appears to have been missing in the Vorlage of P. (3:15)

One significant variant occurs in verse 15, P variant 27 ( < , see above, p. 119). P has a singular noun, whereas MT is plural, and the present case is the only instance in the chapter in which P has a singular noun that corresponds to a plural noun in MT. It seems, then, that the Vorlage of P might have omitted a final from by haplography, especially if it were written defectively, . (3:17)

Verse 17 contains what is probably the most interesting variant in P, variant 30 ( < , see above, p. 116). Instead of a word meaning “to hide,” reflecting of MT, P has a word meaning “to fear.” Since the very same Hebrew word occurs later in the same verse and is translated by , “to hide,” it is probable that the Vorlage of P contained something other than . The Hebrew word closely resembles and means “to fear,” so it is likely that a form of this word appeared in the Vorlage rather than a form of . (3:20) This verse contains one variant, P variant 36 ( < , see above, pp. 114-15). This variant is similar to variant 8 in 3:3, except that here appears parallel to in MT. Of sixteen occurrences of in chapter 3, only here does P have instead of the expected . In fact, in chapters 1-16, according to de Boer’s calculations, stands for only six out of 222 times.87 If P shared certain exegetical traditions with T, as some scholars have claimed, the rendering of 87 Ibid.

228

by (or the replacement of by in the transmission of P) might be expected, but not the opposite, as in the present case. It is probable, then, that the Vorlage of P read . «»

(3:21)

The last verse contains three variations from MT: P variant 37 (< , see above, p. 119), variant 123 ( - < y -, see above, pp. 127-28), and variant 124 («» < omission of , see above, pp. 127-28). These three variants are interrelated, so they may be discussed together. The plural indicates that the translators probably read a plural in their Vorlage, a reading which might well have arisen as a result of graphic confusion between (or, more likely, its abbreviation, perhaps ) and the third masculine plural ending -. 88 If so, this confusion also accounts for the extra pronominal suffix of variant 123 that is not reflected in MT, as well as for the omission of , variant 124. Targum (3:2) One variation from MT occurs in 3:2, T variant 46 (- < - , see above, p. 142). The data indicates that the translators of T were concerned to render conjunctions as accurately as possible, and they show no tendency either to add or to delete conjunctions present in their Vorlage. Furthermore, this extra conjunction is supported by several Masoretic mss (including ms 187) and LXX, and its presence is consistent with common Hebrew idiom. It is therefore probable that the conjunction was present in the Vorlage, as given above. «»

(3:11)

This verse also contains a single deviation from MT, T variant 56 («» < omission of , see above, pp. 142-43). This omission is supported by LXX (see above on LXX variant 93), and its retroversion is obvious.

(3:13) Verse 13 contains T variant 24 ( [ ]< [ ], see above, p. 138), in which T has a perfect where MT has a perfect with a waw consecutive. One would expect T to 88 The longer ending is preserved here in the retroversion since the graphic similarity is greater

than with the shorter ending -.

229

render this form, had it appeared in his Vorlage, with an imperfect, as in the rest of the chapter and generally throughout the book. The imperfect of the retroversion is supported by both LXX and V. (3:16) The final significant variant in T is variant 30 ( < - , see above, p. 137). Although can be used in Aramaic to indicate the direct object of a verb, the translators of T do not seem to have taken advantage of this equivalent when appeared in their Vorlage, preferring instead the particle . In fact, in every other occurrence of in the former prophets, T uses to render . It is probable, then, that the Vorlage of T contained the preposition (cf. LXX, Masoretic mss 89, 174). Vulgate

(3:13) After all the nonsignificant variants are eliminated from consideration, only one significant variant remains in V, variant 28 ( [ ] < praedixi, see above, p. 155). In contrast to the anticipatory tone of God’s statement in MT, V emphasizes that the warning has already been given (probably a reference to 2:27-36) by using both the perfect tense and the prefix prae-. Though Jerome exhibits freedom in several aspects of translation technique, he shows quite a bit of consistency in rendering Hebrew inflection with Latin tense and voice.89 It is likely, then, that Jerome’s Vorlage contained an imperfect with a waw consecutive. The Lucianic Recension The retroverted Hebrew texts of the Lucianic recension and of the other partial secondary witnesses should be viewed somewhat differently than those of the previous witnesses. Whereas LXX, P, T, and V were translated directly from Hebrew, the partial secondary witnesses are revisions of a secondary witness in the direction of one or more Hebrew mss, often, but not always, sharing readings with MT. The extent of the revision varies from witness to witness, but this issue is of little consequence in the present discussion. The Hebrew texts given below are attempts to retrovert not the Vorlage of the secondary witness, of which the partial secondary witness is a revision, but rather the readings of the Hebrew ms or mss which were used to correct the secondary witness. Thus, only those variants that differ both from MT and the secondary witness on which the partial secondary witness is based are retroverted. Since most of the revisions of 89 See below, table 55, pp. ***-***.

230

secondary witnesses toward another Hebrew text draw the secondary witnesses closer to MT, the retroversions of the significant variants in the partial secondary witnesses are embedded in a text that mirrors MT as closely as possible; in the case of the major addition in verse 21, the retroverted text of that part of LXX is used. However, it should be clear that, although the resulting retroversion is probably close to the text found in the mss used to revise the base text, that part of the text copied from MT merely provides a context for the retroverted variant. It is impossible to know for certain that other deviations from MT did not appear in the mss used to revise the secondary witnesses. No extensive discussion of the variants in the partial secondary witnesses has been given above, so references to previous mentions of the variants are omitted when describing each of them. (3:6) One significant variant occurs in verse 6, LXXL variant 8 ( < ). This variant involves the placement of the adverb , as does P variant 49 (cf. LXX variant 79). Since no more than a change in word order is involved in the variation from MT, retroversion is a simple matter. (3:7) Verse 7 contains LXXL variant 12 ( < ). Like the previous significant variant, this one revolves around a difference in the placement of a word (i.e., word order). MT reads , and LXX reflects the same word order with . It seems likely that the revisers of LXX changed the word order in their base text to reflect that found in a Hebrew ms used to correct the text. (3:12) Verse 12 contains one deviation from MT, LXXL variant 20 ( < ). The additional conjunction in LXXL restructures the sentence, making the last four words an independent clause. As noted above, interchange of and is fairly common in Hebrew mss of Samuel, so it would not be surprising for the Hebrew exemplar used by the revisers to read . Furthermore, is a common rendering of , while it rarely if ever renders . 90 90 For a similar construction (with

), cf. 1 Kgs 2:27.

231

«»

(3:15)

One Lucianic variant occurs in this verse, namely, variant 30 (transposition of from after - to the end of the verse). This variant is represented in the retroversion by the marks «» where the word appears in MT and by the larger at the end of the verse. The retroversion of a transposition requires little comment, except perhaps the reminder that all the translators examined in this study show a propensity to follow the word order of their Vorlagen whenever the grammar of the target language permits. It should also be noted that the transposition changes the conjunction to a waw consecutive, and the verb that follows becomes an imperfect rather than a perfect. However, since the verbal root begins with , the consonants of the two forms are identical (if the imperfect is spelled defectively). «»

(3:19)

In verse 19, LXXL has one significant variant, variant 40 («» < omission of - ). As an omission, the retroversion of this variant requires no comment. «»

(3:21) «»

Verse 21 contains three variants from the base text (LXX in the long addition): LXXL variant 46 («» < omission of ), variant 47 ( < ), and variant 48 («» [omission of - ] < for ). The retroversion of the omission in variant 46 needs no explanation. Variant 47 involves both the reversal of the two words present in the retroversion of LXX ( ) and the addition of the preposition before . As noted above, often represents the preposition , especially in the context of the second accusative of a double accusative or similar construction. Finally, in variant 48, LXXL differs from LXX in interpreting the final as a genitive rather than a dative, suggesting that the preposition was not present in the Hebrew exemplar used to correct the text. The Hexaplaric Recension (3:3) Verse 3 contains one significant variant in LXXO, variant 1 ( < ). Here LXX reflects MT (though LXX has nothing that corresponds to ), so seems to be a correction in the direction of a different Hebrew reading. Such a reading is present in one Masoretic ms (cf. also 1:9), so the retroversion is justified.

232

(3:8) This verse contains one hexaplaric variant, LXXO variant 2 ( < ). Other variants involving the presence, absence, or placement of have been noted in other witnesses. Since the retroverted text is identical to that at the beginning of verse 6, the accuracy of the retroversion is secure. (3:10) One variant also appears in verse 10, LXX O variant 3 ( < Retroversion of is straightforward, since it reflects the proper name Furthermore, is also found in 3:9. «»

). in Hebrew.

(3:15)

Two hexaplaric variants occur in 3:15: LXXO variants 8 ( < ) and 9 («» < omission of ). That LXXO alters in LXX to suggests that a Hebrew reading instead of precipitated the change. As is the case with LXXL variant 30, the omission of from its position in MT changes the following verb to an imperfect with a waw consecutive, but it does not alter the consonantal text of the verb.

«»

(3:21) «»

The last verse has two significant variants from the base text (the long addition of LXX): LXXO variant 10 («» [omission of - ] < for ) and variant 11 («» < omission of ). The retroversion of the omission of in variant 11 requires no comment, but variant 10 is more complex. The substitution of the genitive in LXXO for the dative of LXX implies that the preposition was not present in the Hebrew exemplar used to correct the text, as in LXXL variant 48. However, whereas LXXL reads , LXXO reads , with intervening between and . Though such a construction is perfectly good Greek, a literal retroversion into Hebrew is not possible, for , as the word order in LXXO suggests, separates the nomen regens ( ) from the nomen rectum ( ) with the infinitive . Two options present themselves: either preceded the phrase , or else it was omitted entirely. The solution to this conundrum is evident when one recalls that, although LXXO often corrects the Greek text

233

of LXX by adding words, changing forms, and changing word order that does not match the Hebrew exemplar, it does not as consistently omit words. Origen preferred rather to mark words not found in his Hebrew mss with an obelus and a metobelus, signs which were frequently omitted when the fifth column of his Hexapla was copied. It is probable, then, that the reading of LXXO reflects a Hebrew text in which the infinitive was absent and in which formed a construct chain, as presented above. Other Possible Hebrew Readings Other possible Hebrew readings (abbreviated OPH below) are readings present in individual secondary witnesses that might reflect an attempt to revise the witness toward a particular Hebrew ms. Only those readings which differ from both the versional base text and from MT are considered. In 1 Samuel 3, one other possible Hebrew reading was discovered in some of the mss of each of the following witnesses: LXX, P, and T. The secondary version whose ms contains each reading is listed before the retroversion. (3:2) (LXX) One significant variant occurs in some mss of LXX in verse 2, OPH variant 1 ( < ). This reading has a plural rather than a singular verb, shifting the subject from Eli to his eyes. Well attested in Greek mss, this reading also occurs in mss of V and MT, including Masoretic ms 187, whose reading of the verb is identical with the retroversion given above. (3:5) (P) Verse 5 contains one significant variant found in various mss of P, OPH variant 2 ( < ). It is possible that this reading reflects that of various Hebrew mss in existence at least as early as the fifth or sixth century C. E. The same reading seems to be reflected also in two Masoretic mss (including the margin of ms 70) and in two it mss. Since the retroversion matches the rendering in the Masoretic mss, it can be assumed to be accurate. (3:9) (T) Verse 9 has one significant variant, OPH variant 3 ( < ). Whereas the majority of T mss read in agreement with MT, at least one reads , perhaps reflecting the Hebrew . The latter reading is found in two Masoretic mss listed by

234

Kennicott, so the retroversion is valid.91 Conclusions It is important to note that no claim has been made that the retroversions given above represent the precise Vorlagen used for the various translations. Three factors preclude such a claim. First, no attempt was made to reconstruct the Vorlagen away from the places where significant variants were found. Other differences in orthography and content undoubtedly existed; however, no reliable means of determining what they were and where they occurred has yet been devised. Second, the retroverted Hebrew texts produced in this study are not an end in themselves. Rather, they are tools that will be used in the following chapter to attempt to determine the oldest form of the Hebrew text of 1 Samuel 3 that can be determined text-critically. Thus, some variants that have been deemed significant are probably the result of choices and errors at the point of translation or during the transmission of the versional witness.92 However, since many or most of these misevaluations lead to clearly secondary readings, they will be eliminated at the stage of evaluation in the next chapter and so will not affect the final outcome. It seems better to include too much than to omit what might be a reflection of an early reading. Third, the retroversions given for the partial secondary witnesses obviously cannot claim to reflect any single Hebrew ms formerly extant, since Hebrew mss were only consulted or used for corrections sporadically. Thus, the Hebrew text surrounding the retroverted significant variants in the partial secondary witnesses merely provides a context for the readings to be examined in the following chapter. Before proceeding to the evaluation stage, it is important to remember Tov’s dictum that the accuracy of a retroversion says nothing about the originality of the reading in question.93 If a retroversion is accurate, that the reading occurred in at least one Hebrew ms is all that is claimed. The evaluation of the Hebrew readings, both original and retroverted, is reserved for Chapter 7. 91 Cf. the Masoretic list

(list 2752 in Weil), which gives four instances in which is written, but is not to be pronounced. These may be other instances of the substitution of and ; Gérard E. Weil, ed., Massorah Gedolah iuxta Codicem Leningradensem B 19 a, vol. 1: Catalogi (Rome: Pontifical Biblical Institute, 1971), 307. 92 See again the definition of “significant variant”, above, p. 6: “those variants which have some

probability of representing a Hebrew Vorlage different from the base text, MT.” How much probability “some probability” is is not defined. However, even variants with a fifty percent probability of reflecting a deviation in the Vorlage also have a fifty percent chance of having originated with the translators or transmitters of the version. 93 See above, p. 205, n. 19.

CHAPTER 7 EVALUATION OF THE SIGNIFICANT VARIANTS As noted in the Introduction, Paul Maas has said that the goal of textual criticism is “to produce a text as close as possible to the original.”1 This assessment is probably valid for the textual criticism of most works, both ancient and modern, but many scholars question its validity for the text of the OT. The original text is not the only possible goal of the textual critic, who may be more interested in determining—or believe that all that is possible is to determine—the form of the text at a particular stage of its development. One factor that complicates the discussion of the “original text” of the OT is the nature of the growth of the biblical text from its earliest oral and written forms to its final form, which was accepted as authoritative. An overlap may exist between textual criticism and literary criticism if, after one form of a text began to be transmitted, the text was subjected to further revision, and this form was also transmitted. This problem, the problem of multiple editions of a text, will be discussed in the next section. Given the complexity of historical development of the text, the textual criticism of the OT is anything but straightforward. Once the problem of multiple editions is solved (if possible), the goal towards which the textual critic strives can still vary according to theological, philosophical, and pragmatic factors. The confessional stance of the textual critic may play a role in determining which text is sought. For example, if one particular form of the text is considered authoritative (e.g., MT or V),2 then more effort may be expended in attempting to reconstruct that form than the presumed original. If, on the other hand, one believes that the most authoritative form of the text is the original form, then one has more incentive to search for that original.3 Yet another theological position, dogmatic majoritarianism, holds that that form of scripture that exists in the greatest numbers is the form that is authoritative.4 One’s philosophical approach to the various extant texts can 1 Maas, Textual Criticism, 1. 2 MT, of course, is the official Bible of rabbinic Judaism, and the Council of Trent attributed a special status to V. 3 In an extreme form of this view, the nineteenth century Princeton school advocated the “inerrancy

of original autographs.” This outlook has been revived by modern fundamentalists. 4 For the OT, that form would be MT, though not necessarily exactly in the form preserved in BHS. This view has its greatest impact on the textual criticism of the NT, where dogmatic majoritarianism holds to the authority of the so-called Majority Text, which is similar in type (but not identical) to the Textus Receptus. See Arthur L. Farstad and Zane C. Hodges, The Greek New Testament According to the

235

236

also affect one’s text-critical preferences. Those textual critics who would probably consider themselves pluralistic in outlook tend to have a greater appreciation, and interest, in those texts used in all of the various faith communities. Some might even suggest that reconstruction of other, hypothetical texts is irrelevant. Others whose interests are more particularistic might find only those forms of the text used by certain communities (e.g., rabbinic Judaism or “orthodox” Christianity) to be relevant. Finally, pragmatists might insist that since forms of the text not directly reflected in extant witnesses cannot be reconstructed with any scientific certainty, conjectural emendation should be studiously avoided. Other, more idealistic, textual critics, while recognizing the abuses of the past, might continue to stress the need for well-reasoned conjectures that clarify difficult passages (cruces interpretum) or explain the origin of anomalous readings. These factors—and others besides—all play a role in determining the textual critic’s goals. Even those textual critics who want to go beyond extant text-traditions and reconstruct some earlier form of the text do not all seek the same end. Some scholars see textual criticism as a means of restoring the original text. Though this goal is probably tacit in the minds of many people who are not experts in the field and explicit in the minds of most of those who hold to some doctrine of inerrancy, the lack of early Hebrew witnesses to much of the OT, the uncertainties involved in the methodological use of the versions, and the problem of determining which literary form should be considered original (especially when multiple editions exist) make the search for the original text problematic at best and dubious at worst.5 Other scholars believe that textual criticism should produce the Majority Text (Nashville: Thomas Nelson, 1982). 5 Cf. the discussion of the “original” text by Eugene Ulrich, “Double Literary Editions of Biblical Narratives and Reflections on Determining the Form to Be Translated,” in Perspectives on the Hebrew Bible: Essays in Honor of Walter J. Harrelson, ed. James L. Crenshaw (Macon: Mercer University Press, 1988), 113-14. See also Emanuel Tov, “The Original Shape of the Biblical Text,” in Congress Volume: Leuven 1989, ed. J. A. Emerton, Supplements to Vetus Testamentum, no. 43 (Leiden: E. J. Brill, 1991), 355-56, who offers what he calls a “moderate formulation” of the idea of the original text. His definition deserves to be quoted at length:

At the end of the process of composition of the biblical books stood at least one entity (a tradition or single copy) which was completed at the literary level. Possibly at one point parallel compositions were created as well, but they are not evidenced, and in any event, textual criticism takes into consideration only the literary composition that has been accepted as authoritative in Judaism. Even if we assume a very complicated literary development, at some time that process was ended. At the end of that process stood a finished literary product which at the same time stood at the beginning of a process of copying and textual transmission. . . . This entity forms the textual source aimed at by textual criticism, even if that aim can be accomplished in some details only. Reference to the originality of details in the texts pertains to this entity and not to an earlier or later literary stage. Its date differs from book to book and usually cannot be determined. For textual criticism this entity thus forms the “original” text, though in a moderate formulation, since it was preceded by oral and written stages. Tov’s placement of the term “original” in quotes shows the dubiety of the term if taken literally. More significant is his choice of the “literary composition that has been accepted as authoritative in Judaism.” This choice, of course, is a theological one, and all textual critics may not agree with it (cf. Ulrich,

237

best possible text. Of course, those who are looking for the original text consider that the best text, but even those who despair of the search for the original text may speak of the best reading in a certain context. However, “best” is a subjective term; what seems best to a modern textual critic may not have seemed best to earlier generations. For example, scribes who “corrected” their copies of LXX to make them more closely resemble Attic Greek may have produced better texts in some sense, but probably no modern textual critic would agree with the qualitative assessment of those scribes. Furthermore, some readings that are patently ungrammatical or obscure may be the very readings sought by the textual critic, but “best” hardly seems the way to describe these readings. It seems preferable, then, for textual critics to seek the earliest possible reading, and to use this terminology in place of references to the “original” or “best” text or readings. “Earliest” is a more inclusive term than either of the other two, yet it surely encompasses the goals of those would seek both the “original” reading (what could be earlier than the original reading?) and the “best” reading (the earliest possible reading should usually explain the existence of the other readings). The various readings to be analyzed in this chapter will be evaluated from the standpoint of finding the earliest possible set of readings that is possible to reconstruct on the basis of the evidence. Even the search for the earliest reading has its problems, however, particularly when one deals with the possibility of multiple literary editions. It is to this possibility that the discussion will now turn. The Problem of Multiple Editions Eugene Ulrich defines “multiple literary editions” as “a literary unit—a story, pericope, narrative, poem, book, etc.—appearing in two or more parallel forms (whether by chance extant or no longer extant in the textual witnesses), which one author, major redactor, or major editor completed and which a subsequent redactor or editor intentionally changed to a sufficient extent that the resultant form should be called a revised edition of that text.”6 That multiple editions of some biblical books exist is proved by a simple comparison of the books of Daniel and Esther in Catholic and Protestant Bibles. The versions translated in Catholic Bibles are significantly longer and have additional material not found in the Protestant Bibles. The reason for these differences lies in the fact that the Catholic versions of these books are basically translations of LXX, whereas Protestant Bibles rely on the overall form of the text preserved in MT. Other examples of multiple literary editions of OT books which are preserved in the extant witnesses include Jeremiah “Double Literary Editions,” 114-15). 6 Eugene Ulrich, “The Canonical Process and Textual Criticism,” in “Sha arei Talmon”: Studies in the Bible, Qumran, and the Ancient Near East Presented to Shemaryahu Talmon, ed. Michael Fishbane, Emanuel Tov, and Weston W. Fields (Winona Lake, IN: Eisenbrauns, 1992), 278. For an early discussion of the relationship of textual criticism and literary criticism, see Talmon, “Textual Study of the Bible,” 327-32.

238

and Ezekiel.7 Ulrich identifies four different types of intentional variant editions of sections of the OT. In the book of Exodus, MT and LXX preserve an earlier form of the text, while 4QpaleoExodm has an edition based on this earlier form, but expanded by multiple harmonizations throughout the book. In Jeremiah, the earlier form testified to by LXX and 4QJerb is rearranged and systematically expanded by numerous small additions in MT and 4QJera,c. Daniel 4-6 in MT and LXX (OG) expand an earlier, now lost, common ancestor in different directions. Finally, he notes that in the story of David and Goliath in 1 Samuel 16-18, MT supplements the earlier form of the text preserved in LXX with diverse traditions about David. Furthermore, he suggests, more hesitantly, that LXX may have in 1 Samuel 1-2 an intentionally altered portrait of Hannah, Elkanah, and the events surrounding Samuel’s birth.8 It is his observations concerning Samuel that are most relevant to the current study. That the story of David and Goliath in MT and LXX represent different literary editions is indisputable, but is this pericope an isolated example of literary activity subsequent to the completion of the book in substantially its final form,9 or is there evidence of editorial activity in other passages? Answers to this query come from two different directions. The first is the study of Stanley D. Walters on 1 Samuel 1 in MT and LXX (ms B).10 Walters contends that MT and LXX are “discrete narratives, each with its own Tendenz.” 11 Moreover, he says, I doubt that there ever was an original text which has given rise—by known processes of transmission—to the two stories M[T] and B [LXX]. The present MS evidence attests alternate traditions—perhaps prophetic and priestly—rather than a series of successive variations on a single tradition.12 Specifically, whereas MT stresses the joint activity of Hannah and Elkanah in making the sacrifice after Samuel’s birth and in presenting him to Eli, LXX makes Hannah dependent on her husband for all her actions.13 Reactions to Walters’s analysis have been mixed. Though he disagrees with some 7 See, e.g., Emanuel Tov, “The Literary History of the Book of Jeremiah in the Light of Its Textual History,” in Empirical Models for Biblical Criticism, ed. Jeffrey H. Tigay, pp. 211-37 (Philadelphia: University of Pennsylvania Press, 1985); Johan Lust, ed., Ezekiel and His Book: Textual and Literary Criticism and Their Interrelation, Bibliotheca Ephemeridum Theologicarum Lovaniensium, no. 74 (Leuven: University Press, 1986). 8 Ulrich, “Canonical Process and Textual Criticism,” 278-86, esp. 285-86. See also idem,

“Double Literary Editions,” 103-8. 9 The insertion of John 7:53-8:11 into its present place in the Gospel of John in many mss is an example of an isolated insertion unrelated to further literary activity. 10 Walters, “Hannah and Anna,” 385-412. 11 Ibid., 409. 12 Ibid., 410. 13 Ibid., 408-9.

239

specifics of Walters’s arguments, Ulrich agrees with “his general conclusion that, as I would rephrase it, in 1 Samuel 1 the MT and the LXX (in basic fidelity to its Hebrew Vorlage) may well present two different editions of the text, one intentionally different from the other, each internally consistent.”14 Tov explicitly rejects the notion of different pristine texts of Samuel, believing instead that the readings in the various extant witnesses are genetically related.15 However, he leaves open the possibility that 1 Samuel 16-18 is part of a larger revision of the whole text of Samuel, a possibility that must be considered especially when LXX has a significantly shorter text.16 Johann Cook also disputes the notion of independent texts in 1 Samuel 1-2. In a study of 1 Sam 1:28 and 2:11, Cook concludes that the variant versions of MT and LXX are based on an earlier (Hebrew) version that excluded the Song of Hannah.17 The possibility of separate literary editions of 1 Samuel 1-2 cannot be said to have been ruled out, but neither has it been satisfactorily demonstrated. In addition to the proposals of Walters concerning 1 Samuel 1, many scholars posit two or more separate editions of the entire Deuteronomistic History.18 For example, Richard D. Nelson, who sees two distinct editions, says that the first edition was composed by a true historian during the reign of Josiah, and the second was revised by an editor early in the exile.19 Even if the analyses of Nelson and others are accurate, their relevance for the text-critical study of Samuel is problematic. In the first place, scholars find few Deuteronomistic intrusions in the books of Samuel, particularly after 1 Samuel 12. 20 Secondly, no correlation has been shown to exist between the earlier edition of the Deuteronomistic History and any textual witness; all the witnesses testify to the final, exilic edition. There are certainly substantial differences between MT and LXX, 14 Ulrich, “Canonical Process and Textual Criticism,” 281. 15 Tov, Textual Criticism of the Hebrew Bible, 173-76. In particular, he denies that MT and LXX

offer independent traditions in 1 Sam 1:23, as Walters proposes (ibid., 176; S. D. Walters, “Hannah and Anna,” 410-12). 16 Tov, Textual Criticism of the Hebrew Bible, 346-47. 17 Johann Cook, “Hannah and/or Elkanah on Their Way Home (1 Samuel 2:11)? A Witness to the

Complexity of the Tradition History of the Samuel Texts,” Old Testament Essays 3 (1990): 253-54. 18 See, e.g., Helga Weippert, “Die ‘deuteronomistischen’ Beurteilungen der Könige von Israel und Juda und das Problem der Redaktion der Königsbücher,” Biblica 53 (1972): 301-39; Frank Moore Cross, Canaanite Myth and Hebrew Epic: Essays in the History of the Religion of Israel (Cambridge and London: Harvard University Press, 1973), 274-89; Richard D. Nelson, The Double Redaction of the Deuteronomistic History, Journal for the Study of the Old Testament Supplement Series, no. 18 (Sheffield: JSOT Press, 1981). 19 Nelson, Double Redaction, 42 and passim. 20 Ibid., 14. Cf. also Gerhard von Rad, Old Testament Theology, 2 vols., trans. D. M. G. Stalker (New York: Harper & Row, 1962-65), 346: “For a long stretch after the end of the Deuteronomistic Book of Judges in I Sam. XII the Deuteronomist’s interpreting hand abandons us, and only again comes into action with the story of Solomon (I Kings III).”

240

for example, in the books of Kings, including differences in the chronology of the kings and significant differences in content and order of the narratives.21 Differences exist between MT and LXX in Joshua and Judges, too.22 It remains to be demonstrated convincingly that such a pattern of differences exists between the witnesses of Samuel. The evaluation of 1 Samuel 3 shows no conclusive evidence of differences on the literary level, though it has been suggested that the long addition in 3:21 and 4:1 in LXX may be the result of literary and not just textual differences.23 In view of the lack of evidence at this point of different editions in chapter 3, the variants there will be treated as purely textual variants. However, the matter of separate editions will have to be addressed once again when verse 21 is evaluated. Conjectural Emendations The term “emendation” is used in at least four different ways by scholars. First, some scholars refer to any reading retroverted from one of the secondary versions as an emendation. However, since evidence of the reading does occur in an extant text-tradition, the term “retroverted reading” is more appropriate. Second, any change to the Masoretic vocalization or accents can be called an emendation, especially if it is not supported by one of the versions (and so is not a retroverted reading). Since the goal of this thesis is to reconstruct a purely consonantal text, such emendations are irrelevant to the task at hand. Similarly, philological emendations that involve no change in the consonantal text but only the recognition of a new root related to a cognate language or of a newly discovered grammatical structure are largely irrelevant to the present discussion, since they do not affect the consonantal text. In this study, the term “emendation” will refer only to proposed or accepted readings that (1) require a change in the consonantal text (including changes in word division, since it is likely that the earliest forms of the text used some means of separating words), and (2) are not documented in the extant witnesses.24 Scholars of earlier generations resorted to conjectural emendation of their text quite 21 See, e.g., Shenkel, Chronology and Recensional Development; Ralph W. Klein, “Archaic

Chronologies and the Textual History of the Old Testament,” Harvard Theological Review 67 (1974): 25563; Julio C. Trebolle Barrera, Jehú y Joás: Texto y composición literaria de 2 Reyes 9-11, Institución San Gerónimo, no. 17 (Valencia: Edilva, 1984); Baruch Halpern and David S. Vanderhooft, “The Editions of Kings in the 7th-6th Centuries B.C.E.,” Hebrew Union College Annual 62 (1991): 179-244. 22 See Tov, Textual Criticism of the Hebrew Bible, 327-32, 344-45, and the bibliographies there. 23 So Barthélemy, Critique textuelle, 1:152; but contrast the explanations in McCarter, I Samuel,

97; R. W. Klein, 1 Samuel, 30. See also below, 268-70, where further possible evidence of literary differences is discussed. 24 This definition of emendation is substantially the same as that of Tov, Textual Criticism of the

Hebrew Bible, 351-53. Tov identifies three different types of emendations: contextual emendations, linguistic emendations, and emendations for metrical reasons (ibid., 357-69). These types of emendations are not distinguished in this study.

241

frequently, with what many modern scholars would call reckless abandon. In reaction to their excesses, some moderns have eschewed the practice altogether.25 Others have for the most part tried to maintain the consonantal text of MT (with the exception of matres lectionis) and to solve textual difficulties on the basis of comparative philology.26 Still others advocate the continued judicious use of conjectural emendations.27 It is this last approach which is followed in the present thesis, for at least three reasons. First, emendation is a recognized part of the text-critical process, whether one is examining biblical, classical, medieval, or modern texts.28 Second, the MT of Samuel is patently poor in comparison with other books.29 Third, it seems methodologically improper to exclude or limit the use of conjectures in advance. To quote Albrektson, “[when one encounters a difficult reading,] two possible explanations must be compared: is a particular difficulty due to an error in the textual transmission or to a linguistic anomaly, puzzling but explicable? The answer cannot be given in advance, and the possibilities must be considered on equal terms.”30 Guidelines for Evaluating Variants Once the various original and reconstructed Hebrew variants are assembled, how are they be evaluated? The text-critical value of a certain variant may be measured according to two different sets of criteria, external and internal. External criteria include the evaluation of a variant on the basis of the age or presumed worth of the witnesses 25 For example, the Committee for the Textual Analysis of the Hebrew Old Testament expresses

extreme reservations about making conjectures because of the danger of corrupting the text still further. In addition, they are concerned that some conjectures may restore a precanonical form of the text (e.g., the text of J in the Pentateuch) rather than the text of the final redactor. See Barthélemy, Critique textuelle, 1:7477. 26 The most famous proponent of the this method was surely Mitchell Dahood, though many others have also used the method. See, e.g., Mitchell Dahood, “The Value of Ugaritic for Textual Criticism,” Biblica 40 (1959): 160-70; idem, “Ebla, Ugarit, and the Bible,” Afterword to The Archives of Ebla: An Empire Inscribed in Clay, by Giovanni Pettinato (Garden City, NY: Doubleday & Co., 1981), 271-321. For a critique of abuses of this method, see Barr, Comparative Philology. 27 Bertil Albrektson, “Difficilior Lectio Probabilior: A Rule of Textual Criticism and Its Use in

Old Testament Studies,” in Remembering All the Way . . .: A Collection of Old Testament Studies Published on the Occasion of the Fortieth Anniversary of the Oudtestamentisch Werkgezelschap in Nederland, ed. A. S. van der Woude, Oudtestamentische Studiën, no. 21 (Leiden: E. J. Brill, 1981), 14-17; Tov, Textual Criticism of the Hebrew Bible, 351-69. 28 Maas, Textual Criticism, 11-17. Although noting the difficulties involved with many emendations, he comments, “It is far more dangerous for a corruption to pass unrecognized than for a sound text to be unjustifiably attacked” (p. 17). 29 Numerous scholars and commentators could be cited who hold similar opinions, including

S. R. Driver, Notes on the Books of Samuel, xxxv-xxxvi; McCarter, I Samuel, 5; and Harry Meyer Orlinsky, “The Textual Criticism of the Old Testament,” in The Bible and the Ancient Near East: Essays in Honor of William Foxwell Albright, ed. George Ernest Wright, 113-32 (Garden City, NY: Doubleday & Co., 1961), 150. 30 Albrektson, “Difficilior Lectio Probabilior,” 16.

242

containing it, the number of witnesses containing the variant, the geographical distribution of the variant, and the distribution of the variant among different text-types or local texts. A comparison of the methods of textual critics of the OT and the NT reveals an interesting phenomenon: whereas most NT textual critics put a fairly heavy emphasis on external criteria, most OT textual critics do not.31 One group of OT scholars that sees value in some types of external evidence might be called the American, or perhaps the Albright, school. These scholars are influenced by Albright’s proposal that divergent local texts of the Hebrew Bible emerged in various locations (Albright suggested Babylonia, Palestine, and Egypt). The clearest expression of this position is Frank M. Cross’s article on the theory of local texts.32 Cross’s theory is an amplification not only of Albright’s work, but also of that of Barthélemy.33 According to the theory of local texts, three distinct textual families can be discerned in the witnesses of Samuel. A reading that appears in two of the local texts has a greater possibility of being original than one that appears in only one local text.34 A related external phenomenon is the occurrence of related readings in witnesses from different text-traditions, such as the agreements of LXX with P, LXX with a reading from the apocrypha or rabbinic literature, or LXX with Masoretic mss that sometimes preserve significant readings.35 Connected with this phenomenon is the question of whether variants in one witness that have been judged nonsignificant should play a role in the evaluation of a parallel significant reading in another witness, a question that has already been addressed above (pp. 199-200), where it was decided that, though the nonsignificant readings should be considered, they should in no way be accorded the same value as significant variants. The issue of how such agreements between significant and nonsignificant variants should be represented in the critical apparatus is addressed below. External considerations do play some role in evaluating variant readings, but internal factors are more important.36 Numerous rules and guidelines have been developed 31 For an overview of the two major approaches to NT textual criticism, rigorous (thoroughgoing) eclecticism and rational (modified) eclecticism, see above, 13-15, and, in greater detail, Brooks, “The Text of the New Testament and Biblical Authority,” 19-20. For a more extensive comparison of textual criticism as practiced by OT and NT textual critics, see James R. Adair, “Old and New in Textual Criticism: Similarities, Differences, and Prospects for Cooperation,” TC: A Journal of Biblical Textual Criticism 1 (1996). 32 Cross, “Theory of Local Texts,” 306-20. 33 Barthélemy, “Redécouverte d’un chaînon manquant,” 18-29; idem, devanciers d’Aquila. See also R. W. Klein, Textual Criticism, 69-73; Ulrich, Qumran Text of Samuel, 4-9. 34 Cf. Cross, “Theory of Local Texts,” 317, n. 11. Of course, a reading that appears in two local

texts is not necessarily original, particularly if those two are the Palestinian and the Egyptian texts, which share a common ancestor, according to the theory. The point here is only that scholars that hold to this theory put greater emphasis on external factors than do other scholars. 35 See the discussion of this “zeer gecompliceerde vraagstukken” in Seeligmann, “Problemen en perspektieven,” 382-84. 36 McCarter, Textual Criticism, 71-72, outlines the hazards of using external criteria when

243

to assist the textual critic in his or her decision-making process. Examples of such guidelines include preference for the shorter readings, preference for the more difficult reading, preference for the reading more consistent with the author’s vocabulary and style, and consideration given to possible mechanical errors (e.g., parablepsis [homoioteleuton and homoioarkton], dittography, haplography).37 As long as these suggestions are seen as guidelines or helps, they can be of benefit, particularly to the beginning student. However, the idea that they are fixed rules should be avoided, since every case must be considered individually, and many factors often come into play.38 As Tov notes, “the quintessence of textual criticism is to select from the different transmitted readings the one reading which is the most appropriate in the context.”39 The emphasis on one reading is especially important when one is attempting to produce a critical text, as will be done in the following chapter. The following section may be considered a commentary on that critical text. Evaluations of the Variants The reading of MT is given as a collating base for each verse. The variant readings from the secondary witnesses are then grouped into units that will be considered together. Instead of repeating the entire retroversion of each verse (these retroversions are taken from the previous chapter), only those parts of the verse necessary for the collation will be listed, following the reading of MT. To this point, only variants in the secondary witnesses have been discussed to any extent, since variants in the primary witnesses are by definition evaluating readings. Particularly important is his observation that a reading from a ms with a preponderance of better readings should not automatically be preferred to a reading from a generally inferior ms. If one had some assurance that one ms was accurate ninety percent of the time and another only seventy percent of the time, then one could justifiably rely on the more accurate ms except in the case of obvious errors. However, since such an assurance would of necessity come from outside the ms itself, and since the quality of a ms in the first place is determined by internal considerations, the critic has no reason to believe that the reading from the one ms has a greater probability of being correct than that from the other, just because previous readings of the first ms have been better. One possible exception to this characterization of the value of external evidence exists, however. McCarter correctly notes that “the stemma of the biblical text is extremely intricate, and its various lines of transmission are not distinct and independent” (ibid., 71). Nevertheless, if a partial stemma can be reconstructed, as Cross’s local text theory attempts to do, parallel non-trivial variants in unrelated sources should be considered—at least as evidence that the reading originated in Hebrew—alongside internal evidence. In addition, if the internal evidence provides no clues whatsoever to the older reading, some external factor will have to be used to decide which variant to print in the critical text. 37 Cf. R. W. Klein, Textual Criticism, 73-83; McCarter, Textual Criticism, 26-61; Deist, Text of the OT, 38-50; Würthwein, Text of OT, 106-10; Tov, Textual Criticism of the Hebrew Bible, 236-84. Alongside these lists of guidelines, discussions of scribal habits are also informative. See especially Fishbane, Biblical Interpretation in Ancient Israel, passim; Shemaryahu Talmon, “DSIa As a Witness to Ancient Exegesis of the Book of Isaiah,” Annual of the Swedish Theological Institute 1 (1962): 62-72; idem, “Textual Transmission of the Bible,” 95-132. 38 See the discussion in Emanuel Tov, “Criteria for Evaluating Textual Readings: The Limitations

of Textual Rules,” Harvard Theological Review 75 (1982): 429-48. 39 Ibid., 444-45 (italics his).

244

significant and require no retroversion; all Hebrew variants, whether from primary witnesses or retroverted from secondary witnesses, will be discussed below. The sigla in the collations are consistent with those used in Chapter 2 above. The reading of the base text (MT) will be given first, followed by a large right bracket ‘]’. If more than one such reading occurs in the verse, the one in question will be identified by a numeral followed by a small superscript o: ‘1°’, ‘2°’, and so forth. Next, the variant reading(s) will each be given (retroverted, if necessary), along with the witnesses that support the reading. Each of the variants following the base reading will be separated from the previous one by a vertical bar ‘|’. Witnesses will be listed in the following order: primary (Hebrew) witnesses,40 secondary witnesses, partial secondary witnesses, conjectures (abbreviated ‘cj’, followed by the names of scholars or Bible versions that propose or accept the conjecture). Witnesses that support a reading in most respects (or the most important respects) but differ in small details will be enclosed in parentheses ‘()’. Next, the symbols ‘+’ and ‘>’ represent an addition and an omission with respect to the base text, respectively. The abbreviations ‘pr’ and ‘post’ mean that the variant precedes or follows the word or phrase given as the base text, respectively. A superscripted vid means that one may infer the reading from the witness, though it does not explicitly contain it (e.g., as a result of a lacuna), and a superscripted mss following the abbreviation of a witness refers to a reading that is not the main rendering within the text-tradition. Finally, nonsignificant variants cited in support of a significant variant in one of the secondary witnesses will be printed in smaller type following the significant reading (e.g., if the reading of LXX is significant and the readings of P and V, while agreeing with LXX, are nonsignificant, that part of the collation would read: LXX P V ). No attempt will be made to cite every nonsignificant variant that agrees with a significant variant. An asterisk ‘*’ following a reading indicates that it is the one that will be printed as the base text in the critical edition of 1 Samuel 3 given in the next chapter. If no asterisk appears next to any reading, the discussion of that unit of variation should be read for an explanation. (3:1)

*] +

LXX P

is probably an explanatory addition to the text, perhaps based on 2:11 (cf. 1:9). The reading of P might reflect the influence of LXX; if so, the reading might be 40 Especially Masoretic mss 70, 89, 174, 187; 4QSama, where extant; kethib or qere, cited as K

and Q, respectively; and tiqqune sopherim, cited as tiq soph. Other Masoretic mss may occasionally be cited in support of significant readings, though their readings are not considered significant, as indicated by the smaller font size used in the references.

245

secondary in P, though it is preserved in all extant mss. However, it is also possible that the translators added the word independently. It is almost certainly secondary. *]

P

Although it is possible that the verb is secondary in all the traditions, it is probable that this variant arose as a result of graphic similarity between and , causing the verb to be omitted accidentally, only to be replaced later in the wrong place (perhaps as a result of being written in the margin). In addition to the argument from graphic confusion (which would not have occurred had the order supposed in P been original), normal Hebrew idiom seems to favor the medial position of the verb.

]

* LXX

The in could have arisen as a result of dittography from the previous in . On the other hand, one could argue that the was omitted as a result of haplography. It is probable that the translators of LXX had before them the reading , which they took as a qal active participle, but which in fact was probably a qal passive participle. Though the niphal and the qal passive have identical meanings here, the ambiguity of the form without and the relative infrequence of qal passives in the OT might have led a scribe to insert the of the niphal, either to clarify the meaning, or perhaps under the false assumption that it had accidentally fallen out of the text. The reading reflected in LXX, then, is to be preferred, though only by a small amount. (3:2) *] > 89 The omission of in ms 89 is surely secondary, resulting either from accidental haplography (perhaps aided by the common ending - in and [kethib in ms 89]) or from the difficulty involved with pointing as an adjective. *]

LXX P

The renderings of LXX ( ) and P ( ) suggest the possibility that their respective Vorlagen read . This possibility is far from certain, even in LXX, where the variant was considered significant. Thus, to replace the rendering of MT as the probable earlier reading, the case for should be strong. However, the evidence is at best a toss-up. Although both readings make good sense in the context and are idiomatic, is the more commonly used word, and a scribe might have wanted to replace the less

246

common with the familiar . When this observation is coupled with the uncertainty concerning the retroversion, the reading of MT remains preferable. *]

187 LXX P T

The additional conjunction in many witnesses is natural in the context and smooths out what might have been felt to be a rough spot in the flow of the narrative. On the other hand, the absence of a conjunction serves to stress Eli’s blindness (perhaps more than mere physical blindness41 ). Scribes who were not sensitive to the nuances of the text might have inserted a conjunction that they felt belonged there. It is less likely that a scribe would have purposely deleted an existing conjunction, though the possibility of accidental omission is certainly not remote. All in all, the reading of MT seems more likely to have been the earlier reading. *]

187 LXXmss

By reading a plural instead of a singular, the variant exhibited in ms 187 and in several mss of LXX shifts the subject of the verb from Eli himself to his eyes. While such a reading does fit the context, it seems more likely that Eli is the intended subject, rather than his eyes. If overtones of spiritual blindness are present in the verse, it is surely Eli who would be criticized and not his eyes. Thus, the reading of MT is preferable. *]

70

The reading of ms 70 at this point is a misplaced dittography; it is clearly secondary. (3:3) * 1°]

P

The phrase does not appear again in the OT, and the phrase occurs only at Prov 20:27. The shift from to could have occurred in Hebrew as easily as the shift from to in Syriac. It is true, however, that the Syriac translators, perhaps under the influence of T, do occasionally show some tendency to have when would be expected on the basis of MT. Of greater significance is the parallel between at the beginning of the verse and at the end. Anticipating a textual evaluation later in this same verse, the absence of 41 Gnuse, Dream Theophany, 152; for a different interpretation of Eli’s blindness, see Polzin,

Samuel and the Deuteronomist, 52-54.

247

in the verse (in the critical text) makes the reading of MT somewhat more likely from a stylistic perspective. *]

LXXO 96

The reference to the temple in Shiloh apparently caused some tradents of the text difficulty. Since Solomon’s temple was not yet constructed, some scribes apparently thought the term was inappropriate, preferring instead a term that did not connote the Jerusalem temple (so also mss of MT and P in 1 Sam 1:9). It is probable, however, that is in fact the more ancient reading.

]

* LXX

LXX omits the word after , and one must immediately consider whether the word might be an addition to the text in MT. Though the term does add specificity, there is no doubt that the temple in question was dedicated to Yahweh. However, it is possible that a scribe might have felt the need to emphasize the fact; on the other hand, the word might have been added inadvertently because of the frequency of the phrase in the OT (cf. 1 Sam 1:9; 2 Kgs 18:16; 23:4; 24:13; Jer 7:4 [ter], etc.). Yet another possibility is that and in LXXO and ms 96 were substitutional variants and that is the result of conflation. In any case, it is likely that is secondary here.

] > 4QSamavid 4QSama has a lacuna at this point, but based on letter counts, it probably had a text that was about twenty characters shorter than that of MT. Ulrich and McCarter have postulated the omission not only of but also of the rest of the verse. Although this supposition is as likely as any, it is impossible to be certain about it. It seems best, then, to omit only , with LXX (see previous unit of variation). * 2°]

89

Ms 89 has an article attached to , a reading which is equivalent, though less common. Nevertheless, the reading with the article is not compelling, so the reading of the majority of Masoretic mss will be retained.

248 (3:4)

] [?

]

4QSama |

LXX

A discussion of these variants is reserved for later (see below, pp. 270-72). (3:5)

*] +

70mg Pmss it mss | +

187 | +

LXX | +

or +

P mss

The additional word is probably the result of harmonization with verse 6, though it may have arisen independently. Either way, it is probably secondary. As for the prepositional phrase, the context of verse 5 certainly supports its presence, but it does not require it. The addressee in the more concise statement of MT is equally clear. It is likely, then, that the prepositional phrase is a contextual addition.

]

LXX A discussion of this unit of variation is reserved for later (see below, 270-72). (3:6)

]

P|

* LXX |

LXXL

All these variants revolve around the placement—or existence—of the temporal adverb (the additional conjunction and different verb form in LXX will be considered below). The order of the words remains constant in every witness, but appears in every possible place: before, between, and after and , and it is also absent in one tradition. The accidental addition, omission, or transposition of the adverb does not change the meaning of the sentence, since specifies repetitive action. Furthermore, the present verse is not the only one in which or its equivalent moves around in the witnesses (cf. 3:8 P V LXXO; 3:9 LXXmss ; 3:21 P). The omission of the word in LXX (combined with a change in the sentence structure in LXX, to be discussed in the next section) and the varied placement of in the other witnesses lead one to suspect that the form now found in LXX is the earliest form.

]

* LXX Both readings are acceptable Hebrew constructions, though the reading of MT is

249

both more common and present in the immediate context (3:8 all witnesses). That readings of the type found in LXX are not foreign to the idiom of Samuel can be seen from 1 Sam 19:21; 2 Sam 18:22. It is probable that the original form found in the Vorlage of LXX was changed during the transmission of MT to reflect the more common idiom, which was also present in the context. However, the possibility that LXX here reflects a different literary stage than MT cannot be ruled out, though the evidence is inconclusive at this point. This issue will be taken up further below. * 1°] +

70

The phrase in ms 70 is almost certainly an imitation of Samuel’s actions in 3:4, so the reading of MT is to be preferred. 1°] +

LXX

A discussion of this unit of variation is reserved for later (see below, 270-72). *] > 89 187 LXX One could argue that the longer reading of MT is a later insertion designed to make the present verse more closely resemble verse 8. However, the immediately preceding verse lacks any reference to Samuel arising, and no attempts to correct that verse exist. It is more likely that the omission of is due to parablepsis in either Hebrew or Greek from “Samuel” to “Samuel.” The same error could have occurred independently in the Hebrew witnesses on the one hand and LXX on the other, though some sort of genetic connection cannot be ruled out. In either case, the longer reading of MT is to be preferred. * 2°] > 174 The missing after in ms 174 could have developed from either the longer or the shorter text of the previous unit of variation. If the reading was based on the majority MT, might have dropped out as a result of haplography. If the reading was based on the shorter text, might have been inserted either from verse 8 or from memory of other mss. If the evaluation of the preceding unit of variation is correct, then the reading of ms 174 is probably not original, since the opportunity for parablepsis would no longer be present. The reading of the majority of Masoretic mss should be retained. *] +

LXX

The word functions in LXX similarly to the way in which functions earlier in the verse in most other witnesses: it distinguishes the second call of Samuel from

250

the first. Even without these additions, the two descriptions are not identical. Nevertheless, scribes apparently felt some need to differentiate the incidents further. has a stronger rhetorical impact than , and its addition may be seen as an attempt by the 42 tradents of the Vorlage of LXX more sharply to distinguish the second call from the first. may be compared with in verse 8, which may have inspired the addition in verse 6. Since appears to be an attempted improvement of the text, and since no reason for deleting the word were it original is apparent, the reading of MT here seems preferable. * 2°]

70

The reading of ms 70 is a clear error caused by haplography.

]

* 70 |

LXX

This unit of variation is similar to one of the units of variation in verse 5 (see above, p. 148). The witnesses supporting one reading or another have changed, but the reasons for accepting the shortest reading remain the same, notwithstanding the fact that the shortest reading appears only in ms 70. It is uncertain whether ms 70 is genetically related to what is probably an older reading, or whether it is simply a correction to the preceding verse. Regardless of which possibility is true, the variation between and in the witnesses suggests that neither is original. *] +

+ whole verse (repeated) 70

The repetition of from verse 5 and the entirety of verse 6 in ms 70 is apparently the result of a form of parablepsis in which the scribes eye skipped from at the end of verse 6 up to the same words at the end of verse 5. Whether the deviations from the majority MT present in the first rendition of the verse are also present in the second is not indicated in Kennicott’s apparatus. If not, preservation of variant readings may also have been involved in this long dittography. Clearly, however, the repetition itself is secondary. 42 Rather than the translators themselves, probably, since analysis has shown LXX to exhibit a

fairly literal translation technique. It is possible, of course, that —or , the reading of many mss (see above, pp. 44-45)—is an early inner-Greek addition, but the fact that all extant mss read one or the other of these readings suggests the presence of in the Vorlage.

251 (3:7)

]

* LXX

It is difficult to determine which of these words predates the other, and no compelling criteria exist for deciding the matter. On the one hand, the tendency of scribes would probably have been to change to , as the authors of Chronicles often did with material borrowed from Samuel (so, e.g., 1 Chr 14:13-17; 2 Sam 5:22-25). To argue that a scribe would also have changed other occurrences of to in the same chapter is invalid, since an examination shows that neither Chronicles nor the Elohistic Psalter (Pss 42-83) replaces every single occurrence of with . 43 On the other hand, the word is present numerous times in the context, including once in the same verse, so a scribe might have inadvertently written instead of . LXX shows no particular tendency to replace with (or vice versa) in Samuel. However, a closer examination of some passages in Chronicles and in the Elohistic Psalter reveals that scribes tended to replace groups of occurrences of rather than isolated cases, so the replacement of a lone case in the middle of a passage densely populated with instances of seems somewhat less likely than the accidental replacement of , so the reading of LXX is preferred by a small amount.

]

* LXX

M. O’Connor discusses the phenomenon of “prepositional override” in both his examination of Hebrew poetry and his grammar.44 He cites a number of instances in the poetic sections of the Hebrew Bible where prepositional override occurs. The example he lists in his grammar, 1 Sam 15:22, is also a poetic fragment. The question that arises is whether or not the same phenomenon can occur in Hebrew prose. The reading of LXX would seem to suggest that the Hebrew Vorlage used by the Greek translators omitted the second , though if this instance of prepositional override is unique in Hebrew prose, one would suspect some sort of error. The only example of the phenomenon in MT seems to be Ezek 39:4, and the editor of Ezekiel in BHS, K. Elliger, suggests that the preposition has dropped out and should be restored.45 The possibility exists that Ezek 39:4 and 43 Cf. The Anchor Bible Dictionary, s.v. “Names of God in the OT,” by Martin Rose, 1006. 44 O’Connor, Hebrew Verse Structure, 310-11; Waltke and O’Connor, Introduction, 222-23.

Mitchell Dahood describes the same phenomenon under the rubric “double-duty prepositions” in Psalms III: 101-150, The Anchor Bible, ed. William F. Albright and David Noel Freedman, vol. 17A (Garden City, NY: Doubleday & Company, 1970), 435-37. Cf. also Kautzsch, ed., Gesenius’ Hebrew Grammar, §119hh. 45 Hebrew mss and other versions in Ezek 39:4; 1 Sam 15:22; and other similar passages record

variants that contain the missing preposition, but, though the reading of certain passages may be questionable, the phenomenon itself is well established in poetry. Elliger’s suggestion in the case of Ezekiel is probably unfounded.

252

1 Sam 3:7 LXX preserve an archaic prose construction that has elsewhere been replaced by multiple prepositions, but without further documentation, the evidence is meagre. Another possibility, and one that deserves more consideration, is that both of these prose passages are either remnants of older poetic material or are themselves to be considered poetic in some sense. Walther Zimmerli, for example, considers this section of Ezekiel to be “rhythmically elevated prose.”46 Is it possible that behind the story of Samuel’s call to prophecy preserved as a prose literary work lies an earlier poetic oral work? It is easy to see both parallelism and meter in 1 Sam 3:7, especially if the second is omitted as in LXX and the two-word phrase is taken as compensation for the missing 47 preposition. Finally, since a scribe would be much more apt to add the preposition than to delete it, the text of LXX should be preferred. *]

LXXL

The variant reading in LXXL involves the placement of the prepositional phrase . The most common word order for the second half of the verse would be passive verb, indirect object (prepositional phrase), direct object, as in MT, but the word order reflected in LXXL is also used in the OT. If one accepts the argument in the previous section that verse 7 might reflect the remnants of an earlier, poetic form of the story, the word order of MT would seem to preserve the parallelism better, since the direct objects of the verbs come at the ends of the two half-verses. If not, then the word order of MT can still be maintained as the most likely, the reading of LXXL being the result of an accidental alteration of the text. (3:8)

* 1°] +

70 LXXO

The presence or absence of has been noted in other witnesses in other verses (see above, p. 248). Since the presence of the word here conforms verse 8 to verse 6, its 46 Walther Zimmerli, Ezekiel, Hermeneia—A Critical and Historical Commentary on the Bible,

2 vols., trans. James D. Martin, ed. Paul D. Hanson and Leonard J. Greenspoon (Philadelphia: Fortress Press, 1979-83), 2:299. The whole passage runs as follows: [The language of Ezek 38:1-9*; 39:1-5, 17-20,] as has been observed again and again in Ezekiel, can be described neither as prose nor as tightly controlled speech. Rather, it reveals the character of a rhythmically elevated prose, in which there appear two-stress and three-stress lines which are occasionally connected in clear parallelism (see, e.g., 38:9; 39:17f) without being linked by fixed laws into a metrically self-contained whole. 47 Cf. The Interpreter’s Dictionary of the Bible, s.v., “Poetry, Hebrew,” by Norman K. Gottwald,

832.

253

originality is doubtful, especially since the following *] +

makes it superfluous.

174

The extra further specifies the subject of the verb, though who the subject is is clear from the context. The interesting aspect of this reading is that it is found in the same Hebrew ms that omitted in a parallel context in verse 6. The shorter reading of MT should be preferred. (3:9)

]

70 |

* LXX |

V

MT (with a slight variant in ms 70) specifies both the subject and the object of the verb, whereas LXX specifies neither. One can readily see that scribes might want to identify the speaker and the addressee were they missing, since the last subject mentioned in the previous verse was Yahweh. No apparent reason for deleting the words exists, so it is probable that they represent scribal additions to the text. (V’s Vorlage was probably identical to MT, but Jerome omitted “Eli” for stylistic reasons; see above, p. 185.)

]

LXX A discussion of these variants is reserved for later (see below, 270-72). *] +

LXX

The presence or absence of a vocative in the witnesses is similar to the situation with mentioned above: various witnesses include it in some place, while others exclude it, only to include it elsewhere ( found in 3:5 Pmss ; 3:6 MT P T). The only verse in which is present in all the witnesses is verse 16. Since it is more likely that was added than that it dropped out, the reading of MT will be retained in the present verse. *]

P

The presence or absence of makes no difference to the meaning of the verse, and one could argue that P’s text is actually better Syriac than a strict rendering of MT would have been. Nevertheless, the translators tend to render almost every item in their Hebrew Vorlage, including a similar expression in 3:2. It is likely, then, that at some point fell out of the stream of tradition that resulted in P, either as an attempt at stylistic

254

improvement or, more likely, by simple haplography. Thus, the text of MT is preferable. *]

Tms 145 182

As mentioned above, the Masora preserves several apparent instances of the interchange of and . It is possible that at least one ms of T reflects such a substitution here. The citation of the two Masoretic mss (neither of which is considered important by Goshen-Gottstein, hence the smaller type) does not suggest any sort of genetic connection with T. However, they illustrate the possibility of interchange between the two words. The evidence is admittedly slim, from the viewpoint of external evidence,48 and since the two words have the same meaning in the context, the reading of MT will be retained. *]

P

Most Hebrew conditional sentences have the apodosis begin with a waw, and it is likely that the alteration of the text, if it indeed occurred in Hebrew, was inadvertent, as explained in the previous chapter. Thus, the reading of MT should be preferred. (3:10)

*]

LXX

The reading in LXX is a fairly obvious addition, specifying the object of the verb. It is therefore secondary and should be rejected.

] > LXX A discussion of these variants is reserved for later (see below, 270-72). *] +

LXXO LXXmss P Vmss arm

This unit of variation could be seen as part of the larger set of variants which deal with repetition and variation among the witnesses, which will be considered below. The question that will be asked of those variants is whether they indicate different literary editions of 1 Samuel 3. Here, however, the distribution of the variants among the various witnesses indicates that the longer reading probably arose independently in many of the witnesses in an effort to conform Samuel’s action in verse 10 with Eli’s instruction in 48 See above, pp. 241-43.

255

verse 9. 49 Thus, the reading of MT, LXX, and others should be retained. (3:11)

*]

LXX

The addition of the pronominal suffix - does more than simply add specificity; it also changes the connotation of the word from “thing” in MT (“I am doing something”) to “word” in LXX (“I am accomplishing my word”), a more prophetic idea. The context seems to support the claim of MT, since “my word” seems forced and stilted; furthermore, prophecy is not the focus of the passage. Thus, MT’s reading should be retained.

] >* LXX T Many mss of both LXX and T support the reading of MT, but textual analysis of each of the versions suggests that the original texts of the translations omitted the subordinating conjunction. Waltke and O’Connor give several examples of asyndetic relative clauses, though they note that such clauses are more common in poetry.50 No purely mechanical reason presents itself as a reason for the omission of , but one can readily suppose that a scribe might have added the conjunction, either accidentally or with the motivation of improving the style, so that the phrase corresponded with more typical prose usage. In light of these considerations, the reading of LXX and T will be preferred to that of MT. (3:12) 1°]

* LXX P T V

Many commentators have noted that each of the prepositions and is often 51 used in contexts in which one would normally expect the other. This phenomenon could indicate either that the semantic fields of the two prepositions overlapped to some extent at 49 Several mss of LXX apparently reversed this procedure, removing in verse 9 in an attempt to make it conform to verse 10. Because one of the mss to do so is B, the base text in the Cambridge edition of LXX, scholars often cite the reading of B as though it were the reading of “the” LXX. Cf. the discussion of these LXX variants above, p. 47. 50 Waltke and O’Connor, Introduction, 338. Cf. also Kautzsch, ed., Gesenius’ Hebrew Grammar,

§155f-m, where prose examples are given. 51 E.g., S. R. Driver, Notes on the Books of Samuel, 12: “There is a tendency, however, in these two books to use and interchangeably.”

256

the time of composition or during transmission, or that scribal errors have frequently crept into the text, perhaps because of aural confusion. One would primarily suspect semantic overlap in areas or times of Aramaic influence (i.e., either in northern Israel or during the Persian period or later).52 BDB notes the interchange of and in the books of Samuel, Kings, Jeremiah, and Ezekiel, suggesting that the interchange in many cases is due to transcribers.53 The seemingly better semantic fit of during the preexilic period, combined with the possibility of scribal change during the period of transmission, suggests that this unit of variation may be related to the other textual problems associated with MT in Samuel. It seems preferable at this point, then, to adopt the text of LXX.

]

LXXL (P T V omit conjunction)

* LXX |

The different prepositions in these three variants suggest that whichever one was original, it was accidentally replaced by one of the others, perhaps after having dropped out altogether. Any of the prepositions could have been omitted by simple haplography, but the chances are slightly better that one of the two beths in the second variant might have been accidentally omitted when beth was written only once. Assuming that the original preposition dropped out, which preposition would a scribe have inserted in its place? Based on the context, either or seems a likely candidate. Thus, is less likely to be a secondary insertion. Based on this evidence, then, seems to be the most probable reading.54 (3:13)

]

* LXX T V |

cj S. R. Driver

Though some commentators have argued that the construction in MT is a waw conjunctive with the perfect, referring to the past,55 it seems better to take it as a waw consecutive construction, referring to the future. The translators of P certainly understood their Vorlage, presumably the same as MT here, to refer to the future (or perhaps the present). The other secondary witnesses, however, all read the waw consecutive imperfect form, referring to past time. In fact, the variant in V is the only significant variant that 52 BDB notes the use of

postexilic period; BDB, s.v. “ 53 BDB, s.v. “

with the force of a dative by writers of the “silver age,” i.e., the .”

,” note 2.

54 The conjunction in the third variant restructures the sentence, so that

the passage that follows, rather than that which precedes. 55 Hertzberg, Die Samuelbücher, 29.

belongs with

257

remains after all the other potential variants have been eliminated. If the reference is to past time, the most obvious point of reference would be 1 Sam 2:27-36, a passage that speaks of a prophet (literally, a man of God) delivering a message of judgment to Eli. Since verse 12 refers to a prior message of judgment of which Eli is apparently already aware, one would also expect the present verse to do the same. If the reading of MT were accepted, then the message given to Samuel would be new and unknown, but such is not the case. Of course, one could argue that the reading reflects an earlier stratum in the literary process, particularly since many scholars take 2:27-36 to be a late addition to the book. However, though an editor might not strive to make a verse inserted at one point match another some distance away, he would be more likely to work to make consecutive verses correspond. Thus, the reading of LXX, T, and V seems preferable to that of MT at this point.

]

LXX |

* cj BHK REB NJB |

cj

Wellhausen Although with some imagination it is possible to look at the two extant variants and find some similarly shaped letters in the Old Hebrew alphabet, it is doubtful if one variant arose from the other because of graphic confusion. It is possible that this unit of variation should be considered below along with several others as a possible instance of differences on a literary level. However, the differences here do not relate to repetition of similar elements in parallel constructions (or the lack thereof), as most of the others do. The variants seem unrelated to both the preceding unit of variation and to the next one, the tiqqun sopherim, which is also reflected in LXX. The two readings are not substitutional variants, for, though they correspond in position, they are not semantically equivalent. In fact, one could easily envision a conflate reading arising that read . Another possibility is that both variants are later, independent additions to a text that originally read simply . The problem with the text of MT as it stands is that, while not ungrammatical, it is certainly awkward, and the phrase is unusual in that it follows the expression , an expression that usually ends a phrase.56 By rearranging the accents of MT, one could translate “I am judging his house forever because of sin, which he knew about, because his sons . . .” However, this rendering is still somewhat cumbersome. LXX avoids the awkwardness of MT, but its reading is redundant, mentioning Eli’s sons twice in a span of only a few words. BHK suggests replacing with , a suggestion followed by the translators of the Revised English Bible (REB) and the New Jerusalem Bible (NJB). Wellhausen suggests omitting and taking as a conjunction, thus joining the phrase with the following 56 MT does have an athna under

connection of this phrase with what precedes.

, but it joins

with

, indicating the close

258

clause.57 A corruption in the text is apparent, and it seems to have infected both MT and LXX. Although Wellhausen’s solution cannot be ruled out, the graphic similarity between and , combined with the fact that often follows to form a kind of compound conjunction, suggest that the emendation of BHK may be preferable.58

]

* tiq soph LXX

The tiqqune sopherim, or scribal corrections, were primarily designed to preserve the dignity of God, especially by avoiding certain combinations of words that could be taken as disrespectful or blasphemous. Masoretic tradition records that in this verse was originally , a reading both graphically similar and better suited to the context. Of the ancient versions, only LXX preserves the pre-corrected reading, but that reading is undoubtedly the one to be preferred. (3:14) *]

LXX |

PV

LXX has a different word division and an extra . The effect of this difference is to throw the phrase back to the previous verse, which in LXX ends, “and he did not rebuke them; and not only thus (or, and that is not all).” and are occasionally confused (e.g., 1 Sam 2:16; 20:2), and the present confusion probably also arose out of graphic or, more likely, aural confusion.59 Of these two variants, the reading of MT flows more smoothly and corresponds more closely to typical Hebrew idiom, so should be preferred to . As for the conjunction, which P lacks, the addition and omission of conjunctions is not uncommon in the mss of any of the witnesses, but the presence of the conjunction in both MT and LXX strengthens the case for the presence of the conjunction in the text that lay behind both of them, especially since they differ in other respects. Thus, the reading of MT should be preferred over that of LXX or P. *] > 89 The failure of ms 89 to include is probably the result of haplography caused by the similar ending on the preceding word ( ). Since the resulting sentence is grammatically anomalous, because of the loss of the nomen regens, the reading of the base 57 Wellhausen, Bücher Samuelis, 53. 58 Cf. BDB, s.v., “

.”

59 Though most cognate languages also contain a negative in the form l , the corresponding negative in Ugaritic is spelled with a simple l; see Cyrus H. Gordon, Ugaritic Textbook, Analecta Orientalia, no. 38 (Rome: Pontifical Biblical Institute, 1965), 425 (Glossary, s.v. “l II”).

259

text should be retained. (3:15)

*] > 70 The omission of after the preposition leaves an obviously ungrammatical sentence, so the text of the majority of Masoretic mss should be preferred.

]+

* LXX

According to MT, Samuel was sleeping one moment and opening the doors of the temple the next. Though not impossible, this sequence seems to lack a transitional element. That missing element is present in LXX, which includes the phrase after . This phrase was probably omitted inadvertently as a result of parablepsis, specifically homoioteleuton, when the scribes eye skipped from to . It is possible that homoioarkton also played a role, since is similar to . Therefore, the reading of LXX has a greater likelihood of being the older text. *]

P

The difference between MT and P is probably the result of graphic confusion between the singular and the plural in Hebrew. If the Hebrew text at some stage used defective spelling more widely that MT does, the plural of the word in question would have been , and the only difference between this word and the singular form is the added at the end of the word. An extra could have arisen as a result of dittography; on the other hand, a could have disappeared as a result of haplography. It is possible that a scribe might have seen the plural as an anachronism associated with Solomon’s temple, as opposed to the single curtain of the sanctuary, but if so, he ignored the far greater problem of Samuel’s presence in the sanctuary near the ark. Though absolute certainty is impossible, it seems best to retain the plural reading of MT. *]

LXXO

The phrase is used consistently throughout Samuel and Kings, whether in reference to the sanctuary at Shiloh or to Solomon’s temple. However, though continues to predominate, ( ) occurs with some frequency in 2 Chronicles.60 This shift in the direction of substituting for later in Israel’s 60 See 2 Chr 3:3; 5:14; 7:5; 15:18; 22:12; 23:9; 24:7, 13, 27; 25:24; 28:24 bis; 31:13, 21; 33:7;

260

history has already been touched upon above (p. 251). It is probable that crept into the text here as a result of this postexilic tendency, so the reading of MT should be retained as the earlier one.

] LXXL |

* LXXO

These variants deal with the presence and placement of the subject of the main verb, . MT places the subject at the beginning of the clause, a perfectly acceptable position, although one might have expected a position immediately following the verb . LXXL puts the subject at the end of the clause, a position that is good Greek style but that stretches the limits of normal Hebrew style. LXXO omits the word altogether. Which of these, if any, predates the others? If one follows the rule that MT should be followed whenever no problem is evident, then MT should be followed. However, that approach has been considered and rejected. If one considers only the main reading of LXX, or perhaps occasionally one of the other secondary witnesses, the readings of LXXL and LXXO will also both be immediately rejected, since they represent revisions of LXX and thus, it could be argued, tertiary witnesses. But this approach has also been dismissed in favor of an approach that considers the historical development of each of the witnesses. In those places in which the Lucianic and hexaplaric recensions differ from both MT and LXX, they often apparently deviate in the direction of Hebrew texts that were extant at the time. The varied positions, or absence, of in these witnesses indicates the probability that the word is a later addition to the text-tradition, perhaps inserted marginally at first (whence LXXL’s anomalous placement), then given a permanent place in the tradition at the beginning of the clause. If was not originally part of the text, it is possible that a scribe might have wanted to insert it so that the last noun in the previous phrase, , was not taken as the subject of the present clause, especially since the verb was , a verb not suitable for God. The reading of LXXO, then, will be accepted. * 2°] > 187 |

89

The sign of the definite direct object is what one would expect in the present context, appearing as it does before , unless good reason exists for its absence. No such reason appears forthcoming in this verse. Ms 187 omits it altogether, probably by simple haplography. Ms 89 has the preposition instead, but this reading is surely an error, perhaps based both on graphic similarity with and on the occurrence of just two words later. The reading of the majority of mss will be retained here. 34:9; 35:8; 36:18, 19. In most cases, the form used is

, with the article.

261 (3:16)

]

* 89 174 T |

LXX All of the variants in verse 16 revolve around the introductory clause. Mss 89, 174, and T (cf. LXX) all read the preposition instead of after . A survey of the verb in Samuel and Kings reveals that when the object of the verb is a person (as opposed to phrases such as . . . or ), the prepositions and are used almost exclusively at the expense of , though the latter does appear on rare occasion. Moreover, every other occurrence of the verb in the present chapter is followed by one of the two former prepositions. It is likely, then, that is to be preferred to . The question of the omission of and the transposition of in LXX is somewhat more difficult. On the one hand, the double verb in MT is consistent with Hebrew style. On the other hand, the reason for the omission of and the transposition of is not immediately evident. One possibility is that was omitted by simple haplography, leaving an awkward sentence, which a later scribe corrected by moving . A similar scenario is that a scribe’s eye skipped from in the preceding verse to in this verse, omitting . When the resulting sentence was found to be nonsense, the verb was moved to the beginning of the sentence and the subject was reintroduced. It is always possible that and are substitutional variants, only one of which was in the earliest form of the text, but the argument based on Hebrew idiom seems stronger. Thus, the reading of 89 174 T, which differs only slightly from MT, is to be preferred. (3:17)

*] > 70 The omission of the phrase in ms 70 is the result of parablepsis, as the scribe’s eye skipped from to , omitting the words in between. The reading of the base text is preferable. *]

187 |

P

The delayed occurrence of the particle in ms 187 is abnormal and almost certainly secondary, since is what one would expect. The reading of P is more interesting. Instead of warning Samuel not to hide anything from him, in P Eli comforts him by telling him not to be afraid of him. Graphic similarity was clearly a factor in the

262

confusion of and , but which reading is the basis for the other? If Eli had wanted to tell Samuel not to be afraid, the more common way would have been to say, . In addition, Eli’s words of comfort in P are quickly contradicted by his threat of divine judgment if Samuel does not tell him the contents of the vision. Furthermore, at the end of the verse parallels this portion of the verse, implying that stood here. Thus, the reading of MT is to be preferred. *] > 70 The shorter text in ms 70 is the result of parablepsis from 1° to 2°. The resulting reading could easily have gone unnoticed, since it makes good sense as it stands. Nevertheless, it is the reading of the majority of Masoretic mss that is preferable. *]

187 LXX P V

The reading of ms 187—which is supported by LXX, P, and V—has a plural rather than a singular noun, since God spoke several words to Samuel. The singular in the majority of Hebrew mss should be taken as a collective noun and translated “the message” rather than “the word.” The scribe of ms 187 (or his predecessor) altered the singular to the plural, perhaps inadvertently, in order to ensure that the reader did not think that God had spoken only a single word to Samuel. The plurals in the versions arose out of the need the translators felt to create idiomatic readings, and so they were not considered significant. The reading of MT should be retained.

]

* LXX

and are substitutional variants, since they play corresponding roles in their respective sentences. No apparent graphic similarity exists between the two variants, so some other reason for textual alteration must be sought. The phrase is somewhat cumbersome, but thoroughly Hebraic (cf. 1 Sam 15:14; 25:24; 2 Sam 7:22; 18:12; 22:7, etc.). It is more likely that a scribe would have substituted the colorless, though more literal, phrase for than vice versa.61 It is also possible that arose as a gloss on , later to be inserted into the text itself. The reading of LXX probably reflects the earlier reading. 61 Cf. 1 Sam 1:23, where LXX also has the “more Hebraic” text, and Walters’s discussion of this

passage; S. D. Walters, “Hannah and Anna,” 400.

263 (3:18)

] >* LXX The excess text in MT could be an explanatory expansion, or the absence of the word in LXX could be the result of haplography. does usually occur with an indirect object introduced by (or occasionally ), but it also sometimes occurs without any indication of the object. It seems more likely that a scribe would have added in keeping with the prevailing custom than that one would have omitted it for some reason. Thus, the reading of LXX should be preferred. *] +

174

The demonstrative in ms 174 qualifies the noun and is almost certainly secondary. A scribe, perhaps subconsciously, apparently considered the sentence as it stood to be in need of clarification. The reading of MT should be retained. *] +

LXX P

Earlier in this verse it was MT that had an explanatory addition; here it is LXX. specifies the subject of the verb , which might have been considered ambiguous since was the last subject mentioned. Furthermore, since immediately follows the verb, a scribe might have wanted to avoid the implication that it was God who was speaking. As is often the case with explanatory additions, this one appears to be secondary, and the text of MT should be followed. *] > 174 The omission of the pronoun after is probably the result of haplography caused by the graphic similarity of the two words. The fact that the next word also begins with may have been a contributing factor as well. In any case, the reading of MT is to be preferred. (3:19)

]

* LXX

When Hebrew writers wanted to state the fact or express the wish that God would be with someone, they would usually do so without recourse to a form of ; for example, (1 Sam 16:18; 18:14), (Jdg 1:22), and so forth.

264

When they did choose to use a form of , they almost invariably placed it before the subject: (1 Sam 18:12); (2 Kgs 18:7); (1 Kgs 8:57); (Gen 26:3). The only exceptions to this latter rule are the present passage, 1 Sam 17:37, and 2 Sam 14:17. However, the LXX readings in both 1 Samuel passages put the verb first, and the context shows a great likelihood of graphic confusion. Only in 2 Sam 14:17 is the subject before the verb in both MT and LXX (and here LXX is kaige, not OG; the word order in LXXL is slightly different). What this evidence means for the current unit of variation is that, barring some compelling reason for putting the subject first, the verb should probably come first, followed by the subject. Thus, the reading of LXX will be accepted here. *]

LXXL

The lack of a conjunction in LXXL could be the result of haplography triggered by the preceding on the end of . On the other hand, the extra conjunction in MT could be the result of dittography. The lack of a conjunction between clauses is unusual in Hebrew, and it serves to stress the following clause (cf. 3:2). However, verse 19 appears to have an intentional three-fold structure, which would be disturbed by the omission of the conjunction. In light of this observation, the reading of MT will be retained. (3:20) *]

174 |

P

The reading in ms 174, , probably derived from the reading preserved in MT by means of dittography and graphic confusion: became , and the before was dropped. The reading of the majority of Masoretic mss is preferable to that of ms 174. The reading preserved in P substitutes for MT’s . The phrase occurs in several other passages in the OT, while is apparently not represented in MT. However, the relative scarcity of even the phrase (between ten and twenty occurrences) urges caution. Perhaps the change from to can be explained as another example of the tendency of later scribes to change to . In any case, the reading of MT should be preserved. (3:21) *] +

70

One of the most interesting readings in the Masoretic mss in 1 Samuel 3 is this apparent gloss in ms 70. Since the verse notes that the Lord continued to appear, a scribe

265

probably confirmed this statement by writing in the margin “to Manoah,” a reference to Judges 13. Somehow the marginal note found its way into the text. Though clearly not original, it is a good demonstration of the phenomenon of glossing. *] > 187 Ms 187’s failure to render this phrase is probably the result of parablepsis, the scribe’s eye skipping from to . The following and preceding the phrase under consideration somehow survived the parablepsis, or perhaps it was restored at some point in the process of transmission. The reading of MT is preferable. *] > 89 The omission of by ms 89 could be the result of parablepsis (homoioarkton) with the following word , or it might have dropped out accidentally for some other reason. The text of the majority of Masoretic mss should be retained. *]

89

The reason for the duplication in ms 89 must be simple dittography, but the reason for the change in prepositions is less obvious. Graphic similarity may have played a role in the change, but it is just as likely that a scribe felt that was a better fit for the context than . In any case, the reading of ms 89 is inferior. *]

P

The reading of P is probably derived from a Hebrew ms in which the divine name was abbreviated (or something similar). The abbreviation was mistakenly read as a pronominal suffix, changing the noun from singular to plural in the process. The reading of P makes even less sense than the reading of MT, so MT’s rendering should be preferred.

] > LXX |

mss T edd V

This unit of variation is discussed with the following one.

266 3°] + LXX | LXXL | LXXO It is clear that a substantial disturbance has occurred in the witnesses in verse 21. The variants can be broken into two units, the first of which involves the omission in LXX of several words that are found in MT, and the second of which concerns the addition of many words in LXX. Also related to these variants are the differences between MT and LXX in 4:1a. These units of variation are clearly the most significant in the whole chapter in terms of the number of words involved, and the reason for this great difference must be considered. Scholarly opinion is divided over whether these differences are purely textual or whether they should be considered differences on the literary level. The former position is held by such people as Thenius, McCarter, and Klein.62 They attribute the differences in the witnesses largely to mechanical errors and later attempts to remedy those errors. Others, such as de Boer and the Committee for the Textual Analysis of the Hebrew Old Testament, take the latter position.63 They believe the additions in LXX to be drawn from other passages and to reflect a stage of the text prior to the settling of its final form. Thenius, Klein, and McCarter all assert that the shorter text of MT is the result of parablepsis, though the circumstances they envision are slightly different. All three would delete as a variant of 3:20. Thenius would then add the first five words of 4:1 MT (missing in LXX), , followed by the rest of the long addition in LXX. He believes that a scribe’s eye skipped from (4:1 MT) to (4:1 LXX), resulting in the reading of MT. He posits a second parablepsis for another scribe, this time from (3:21) to (4:1 MT), resulting in the reading of LXX.64 Klein believes that the words in 3:21 and 4:1 MT are secondary, added to make sense of the text only after the loss of the original words. A scribe in the tradition of MT skipped from (3:21 65 MT) to (4:1 LXX). McCarter agrees with Klein in omitting from 3:21 MT, but he keeps in 4:1. The 62 Thenius, Bücher Samuels, 17; McCarter, I Samuel, 97, 103; R. W. Klein, 1 Samuel, 30. 63 De Boer, I Samuel i-xvi, 56; Barthélemy, ed., Critique textuelle, 1:152. 64 Thenius, Bücher Samuels, 17. 65 R. W. Klein, 1 Samuel, 30.

267

rest of his proposal is the same as that of Thenius.66 De Boer and the members of the Committee believe that the differences between the two witnesses are literary rather than textual. De Boer says, “III 21b and IV 1a introduce the coming incidents and strengthen the bond between the youth history of Samuel and the stories to come.”67 Barthélemy and the other members of the Committee state that none of the three additional clauses in Gk shows originality, but all are based on other passages (3:20; 4:15; 4:1ff.). They say, “the fact that MT does not offer any joint between the two narratives that critics could recognize as literarily heterogeneous is a remarkable indication of its great antiquity.”68 Which of these two disparate positions is more probable? If it were demonstrated that 1 Samuel LXX offers a text that differs from that of MT on a literary level (see below, pp. 268-70), the case for a literary origin would be strengthened. Even so, however, the likelihood of mechanical error (parablepsis) tips the evidence in favor of a textual solution for 3:21 and 4:1a. The additional material in LXX does not seem to reflect any trend that has been noted elsewhere in LXX.69 In particular, the excess material present in the story of David and Goliath is preserved in MT, not LXX. The Committee’s observation that the additional Greek clauses show no innovation is valid enough, but that fact in itself is insufficient reason for denying their place in the text, since threads of connection run back and forth in the narrative in both MT and LXX and, for that matter, in all narrative. In fact, one could argue just the opposite, that similar style and vocabulary supports the originality of the LXX additions (with the exception, of course, of the material repeated from verse 20). If the differences between LXX and MT in verse 21 are textual rather than literary, one must decide which version contains a witness to the earlier form of the text or whether neither completely preserves it. It is immediately obvious that the phrase is problematic. The different spelling of “Shiloh” in the same verse is unusual, though not unprecedented, and both and appear frequently enough in Samuel. Moreover, the repetition of “Shiloh,” though perhaps redundant to modern ears, seems to accord well enough with Hebrew idiom, so this word, at least, should probably be retained. More difficult is the phrase , whose exact meaning is unclear. Some witnesses read , but this reading cannot be considered a significant variant in any of the witnesses; furthermore, it helps very little, since the phrase remains awkward 66 McCarter, I Samuel, 97, 103. 67 De Boer, I Samuel i-xvi, 56. 68 Barthélemy, ed., Critique textuelle, 1:152. 69 Contra De Boer, I Samuel i-xvi, 56. It must be remembered that the translation technique of LXX has been found to be a literal one, and any differences from MT on a literary level would have arisen in the Hebrew Vorlage.

268

and unnecessary. It is possible that the circumstances that disrupted the text have left a fragment of a lost sentence that would have made sense, but it seems best in light of the extant evidence to dismiss this phrase as secondary, though its origin remains obscure. Although Klein omits the first five words of 4:1, they seem natural enough in the context, and provide a fitting conclusion to the story in chapter 3: Samuel has moved from boy ministering to Eli in the sanctuary to man bring the word of the Lord to all Israel. That this phrase is more fitting as a conclusion to chapter 3 than as an introduction to chapter 4 is made clearer when the long addition in LXX is analyzed. In MT, chapter 4 begins with the note that Israel went out to fight the Philistines. The story as it concerns Eli comes to its primary conclusion in 4:18 with the death of Eli and the statement that he had judged Israel for forty years. Verses 19-22 form a sort of appendix to the story, tying it in with the further adventures of the ark in the following chapters. Commentators see the notice in verse 18 that Eli had judged Israel for forty years (LXX: twenty years) as a Deuteronomistic attempt to fit Eli into the pattern of the judges in the book of Judges.70 This conclusion to the story accords well with the introduction to this pericope in LXX, detailing once again his age (cf. 4:15) and his sons’ sin (cf. 4:11, 17). Furthermore, LXX’s introduction to the circumstances of the battle (“And it came about in those days that the Philistines assembled themselves to fight against Israel”) meshes well with the following notice (“and Israel went out to fight against them”), in contrast to the abrupt start of the narrative in MT (“and Israel went out to fight against the Philistines”). Thus, it is likely that the long addition in LXX, with the exception of the repetition of the material from verse 20, represents the older text.71 Variants That Pertain to the Question of Multiple Literary Editions The time has arrived to discuss the variants that bear upon the question of different literary versions in 1 Samuel 3. It has already been concluded that the variants in verse 21 are of a textual rather than literary nature, but this decision does not prejudice the case against the variants to be considered here. Differences in the story of David and Goliath that probably depend on editorial activity have already been noted. Walters has argued that 1 Samuel 1 also contains evidence of different literary editions in MT and LXX, though not all concur. Cook has decided against a literary solution to the differences between MT and LXX over the differences in 2:11 and the placement of the Song of 70 Cf. H. P. Smith, Books of Samuel, 36; McCarter, I Samuel, 114-15. 71 A comment on the material that parallels verse 20 is in order at this point. It is likely that these words are a textual alternative to verse 20 that was misplaced at some point in the tradition. Which of the versions is earlier? The key to deciding lies in the phrases in MT and in LXX. Both phrases occur in MT, but only the former appears in Samuel and Kings (2 Sam 3:10; 17:11; 24:2, 15; 1 Kgs 5:5), while the latter is limited to Deuteronomy and Jeremiah (Deut 13:8; 28:64; Jer 12:12; 25:33). On this basis, it is probable that the version found in MT is preferable to that in LXX, LXXL, and LXXO.

269

Hannah in 1 Samuel 2.72 Before examining the variants in chapter 3, a couple of methodological question must be answered. First, what are the criteria for identifying a different edition of the text? Is a large number of variants sufficient grounds for concluding that different editions exist, or must identifiable patterns of variation exist? In the first place, a distinction must be made between significant and nonsignificant variants. If all apparent variants are considered, V has more deviations from MT in 1 Samuel 3 than any of the other secondary witnesses. However, when nonsignificant variants are eliminated, V contains only one significant variant (3:13), and that supported by LXX and T. Even when only significant variants are considered, the answer to the question is still not obvious. To try to reach a conclusion by comparing two witnesses to a text, one of which had many deviations from the arbitrarily chosen base text and the other of which did not, would be begging the question. Instead, two analogies may be considered. The first one concerns the development of two daughter languages from a single parent language, such as Spanish and Portuguese from Latin. As the development of the two languages is traced historically, their divergence grows as the temporal distance from the parent language increases. Thus, a large number of variants in a particular witness from a base text may indicate the passage of a great deal of time in separate text-traditions rather than the existence of separate editions. The second analogy involves a comparison of texts which have a common origin but are admittedly different literary editions. If the parallel passages of Kings and Chronicles are compared, one immediately notes a number of differences in the texts, most of which are due to the authors of Chronicles, who had particular historical interests, theological stances, and pastoral concerns (to use an anachronism), all of which are reflected to a large extent in the differences between the texts. Therefore, one should only claim to find a different literary edition when certain patterns of variation exist in a witness.73 The second methodological question is this: if different editions do exist, is one to be preferred as the older set of readings (and how can the older edition be identified?), or should each reading or group of readings be examined independently?74 The answer to this question may seem paradoxical, but it is nevertheless the correct one. One edition may be preferred as containing the older readings, but each reading or group of readings should 72 See above, pp. 237-40. 73 Cf. Tov’s comment in Textual Criticism of the Hebrew Bible, 314: “It is assumed that largescale differences displaying a certain coherence were created at the level of the literary growth of the books by persons who considered themselves actively involved in the literary process of composition.” 74 It should be rememebered that the target selected for this text-critical investigation was the oldest

possible text. Given a demonstrated multiple literary edition, other textual critics might prefer the edition that contains the latest set of variants, arguing that the latest form represents the completion of the canonical process as accepted by one or more faith communities. Thus, Ulrich says, “The texts were authoritative texts, and through the traditioning process they were being made more authoritative”; Ulrich, “Canonical Process and Textual Criticism,” 289.

270

still be examined independently. The reason for examining readings in the later edition is that it may on occasion preserve older readings that have disappeared from the older edition. Thus, the Lucianic edition of LXX, though overall later than OG, sometimes preserves older readings, derived either from the Hebrew mss toward which it was corrected or preserved directly from OG when all typical OG witnesses are corrupt. It is also possible for late readings to creep into a generally older text, so evaluating each set of variants is imperative. Once a pattern of revision and a relative chronological order has been established, the older edition should be preferred whenever revision is apparent. The older edition should also be preferred when literary causes for variation are suspected but neither reading is demonstrably older. Having determined guidelines for detecting the existence of multiple literary editions and choosing readings among them, it is time to turn to the set of readings in 1 Samuel 3 that concern Yahweh’s repeated call of Samuel. The readings are as follows. (3:4) ] [? ] 4QSama | LXX (3:5) ] LXX (3:6) 1°] + LXX (3:9) ] LXX (3:10) ] > LXX A comparison of these variants may perhaps be appreciated better by graphic means. They may be divided into two groups: those that deal with the number of times God calls Samuel and those that treat Eli’s command and Samuel’s response. Call 1 is found in 3:4-5, call 2 in 3:6-7, call 3 in 3:8-9, call 4 in 3:10 (in part). The last call, of course, differs from the others in that God continues speaking to Samuel, so only the first chart, dealing with God’s call of Samuel, contains data from all four calls. The textual decisions decided upon above are integrated into the texts of MT and LXX (which is presented in retroverted form) in order to highlight the possible literary differences.75 CHART 1

God’s Call of Samuel MT (corrected)

LXX (retroverted and corrected)

Call 1 Call 2 Call 3 Call 4

75 For a discussion of these variants as a group from the standpoint of whether or not they are

significant on the basis of literary analysis, see above, p. 174.

271 CHART 2

Eli’s Command and Samuel’s Response MT (corrected)

LXX (retroverted and corrected)

Call 1 Call 2 Call 3

. . .

. . .

These charts reveal two things about LXX in comparison with MT. First, neither is more prone than the other to schematize. Second, the variations between MT and LXX are not theological, nor do they affect one’s perception of any of the characters in the narrative.76 On the contrary, although the variants are somewhat denser in these verses than elsewhere in the chapter, they appear rather trivial. Does some other explanation exist to explain the number of variations in these verses? A comparison with the texts of the gospels reveals that in parallel passages, scribes of one gospel often alter their texts in the direction of another of the Gospels (cf., e.g., Matt 17:1-9 and parallels in Mark and Luke, with eighteen variants attributed to contamination from other gospels in a space of nine verses; and Matt 13:1-9 and parallels from Mark and Luke, with fifteen such variants).77 Thus, the sheer repetition in 1 Sam 3:4-10 may account for the variants (note also the variants in these verses that were not deemed preferable, many of which are based on similar verses in the section; see especially the Masoretic mss). Therefore, the data in 1 Samuel 3 does not support the existence of separate literary editions in MT and LXX, though it must be pointed out again that the data speaks only for the present chapter and not for any other chapter or for the book as a whole. Which of the variants listed above are to be preferred then as the oldest variants, and on what basis? As noted previously (p. 218), the in the phrase in 3:4 MT may be the remnant of a missing , which is in fact present in LXX. In the second call, one could argue either that the missing in MT resulted from haplography or that the extra one in LXX resulted from dittography. Of these two choices, haplography is probably somewhat more likely, though no certainty can attach to such an evaluation. In the fourth call, a perusal of other occurrences of in the OT reveals that the phrase is generally used in order to avoid repetition. However, MT contains precisely such repetition. It is probable, then, that the omission of in LXX is the older reading. In fact, it is possible that the two-fold repetition of the name in verse 10 MT supports the two-fold repetition in verses 4 and 6, though not 76 Thus, Walters’s contention that LXX presents some characters differently from MT finds no

support in chapter 3, though, of course, his evaluation of chapter 1 is not thereby negated. 77 The statistics are taken from Kurt Aland, ed., Synopsis Quattuor Evangeliorum, 13th ed. (Stuttgart: Deutsche Bibelgesellschaft, 1985), 236-39, 174-75.

272

much weight can be attributed to this argument. In the variants related to Eli’s command and Samuel’s response, it is LXX (or rather its Vorlage) that seems to be schematizing. It has been determined already that schematization is not characteristic of LXX in the chapter, but some harmonization between verses is not ruled out. The three-fold in LXX contrasts with the pattern , , in MT. MT also reads twice in the three calls, whereas LXX does so only once, in the third call, which is separated from the command by several words. In these units of variation, the readings of MT are preferable. Conclusions Before the textual decisions are presented in the form of a critical text in the next chapter, a few general conclusions may be drawn from the data. First, no single witness contains an overwhelming majority of older readings, though, in general terms, it is clear that MT and LXX are the most important witnesses to the text in this chapter. In fact, LXX is preferred to MT more often than vice versa when they are in conflict. A comparison of MT with each of the other witnesses yields interesting results. In Chart 3, the significant variants in the other witnesses are compared one by one with MT on the basis of which reading was taken to be preferable. The first column contains the number of times that the witness other than MT was preferred when the two conflicted, and the second contains the number of times MT was preferred. In addition to abbreviations discussed earlier, cj will be used for conjectures (the figure in the cj row is the number of conjectures accepted; no comparative data from MT is given).

CHART 3 Comparison of MT with Other Witnesses Other Witness

MT

20

13

P

0

10

T

3

1

V

1

0

LXXL

0

2

LXXO

1

4

OPH

0

3

mss

2

27

tiq soph

1

0

4QSama

1

1

cj

1

NA

LXX

273

A second conclusion arrived at on the basis of an analysis of the data is that witnesses generally regarded as of little importance for textual criticism do sometimes have superior readings. Nowhere is this fact more obvious that in verse 15, where the reading of LXXO, supported in part by LXXL, was chosen over that of MT. Both T and V occasionally have superior readings, though, perhaps surprisingly, P does not. 4QSama is too fragmentary in the chapter to be of much use, but one reading supported by this ms was preferred over MT. The twenty-seven times that Ms B19a (L, reflected in BHS) prevails over the other Masoretic mss evaluated in this study shows the quality of B19a, but the two times that the other mss have the better reading are reminders that BHS is not equivalent to MT. Finally, the importance of considering conjectures is stressed by the preference for one of them in verse 13. One last conclusion should be noted. The usefulness of external evidence in evaluating readings was discussed above, where it was concluded that internal evidence should receive primary consideration. However, external evidence continues to play some role in textual decisions, particularly when internal evidence fails or is ambiguous. Thus, if two equally acceptable readings are found in MT, LXX, P, and T on the one hand, and another in LXXL on the other, the reading of those witnesses that generally bear witness to an older text will take priority. This contention is particularly true when the divergent witness is one of the Masoretic mss.

CHAPTER 8 A CRITICAL EDITION OF 1 SAMUEL 3 This critical edition of 1 Samuel 3 is arranged on the page in the following manner. At the top is the critical text itself, as reconstructed on the basis of the evaluations in the previous chapter. It should be stressed once again that the evaluation of the evidence of the various witnesses, particularly the secondary witnesses, requires a study of the translation technique of each witness. Thus, this critical text is not really useful apart from the commentary in Chapter 7, which is based in turn on an investigation of the significant variants, particularly Chapters 3 and 4. Each verse begins on a separate line in this edition, except 3:21 and 4:1a, which are joined. The presentation ends with the pericope in 4:1a, before the record of the armies going out to battle begins. Immediately below the text is the first apparatus, which contains the readings of the Hebrew witnesses and retroverted secondary witnesses. Only those units of variation that are supported by significant variants in one or more of the witnesses are cited, though nonsignificant witnesses will sometimes be listed as well, in smaller print. In every unit of variation recorded in the apparatus, the readings of MT, 4QSama (when extant), LXX, P, T, and V will be cited whenever they are significant or agree with either the base text or a significant variant. They will not be cited if their evidence is ambiguous. It should be remembered that witnesses that agree with MT have not been deemed significant or nonsignificant, so the value of their support may be minimal. Other witnesses are cited as needed. Below the first apparatus is the second apparatus, which contains the readings of the secondary witnesses in their original languages. The reference numbers in the first apparatus correspond to those in the second. Secondary witnesses are cited in the second apparatus in the order presented in the first apparatus (except for omissions, for which no indication is given in the second apparatus), separated by the double dagger ‘‡’. Tertiary witnesses will not be cited in their original languages, since these witnesses are primarily witnesses to the texts of the secondary witnesses. Sometimes it is necessary to cite the same passage from a secondary witness in two different units of variation; in order to avoid needless repetition in these cases, an arrow ‘↑’ will be used in the second reference to refer the reader to the previous unit of variation. 274 The arrow ‘←’, followed by a Hebrew phrase (all in parentheses), means that the evidence cited from one of the versions represents a slightly different group of words than that given as the Hebrew equivalent. This sign will

275

only be used in cases where the translation makes it impossible to avoid recording extra data without mutilating the words (cf. v. 13, P). In the two apparatuses, the chapter and verse are given, followed by the variant reference number, which begins again at one for every verse (with the exception of 3:21 and 4:1a, which are considered together). Variants in the same verse are separated by a bullet ‘•’. 1 SAMUEL 3 2

1

(3:1) 3

5

4 5,4

3

2

1 3,2

(3:2) (3:3) (3:4)

1 1

MT T V] + LXX P • 2 MT LXX T V] P • 3 LXX] MT (3:2) 1 MT LXX P T] > 89 • 2 MT T] LXX P • 3 MT mss ms T ] 187 LXX P T mss V it • 4 MT LXX P T V] 187 LXXmss • 5 MT LXX P T V] 70 (3:3) 1 MT LXX V] P • 2 MT LXX P T V] LXXO 96 • 3 LXX 4QSamavid] + MT P T V • 4 MT vid a LXX P T V | > 4QSam • 5 MT] 89 (3:4) 1 LXX] MT P T V | [? ] 4QSama (3:1)

1

(3:1)

1

T:

‡ V: Heli ‡ LXX: ‡ T:

‡ P: ‡ V: Domini erat pretiosus ‡ P:

• 2 LXX:

• 3 LXX: (3:2) 1 LXX: ‡ P: ‡ T: • 2 T: ‡ LXX: ‡ P: • 3 T mss : ‡ LXX: ‡ P: ‡ T: ‡ V: nec • 4 LXX: mss ‡ P: ‡ T: ‡ V: poterat ‡ LXX : • 5 LXX: ‡ P: ‡ T: ‡ V: videre (3:3) 1 LXX: ‡ V: Dei ‡ P: • 2 LXX: ‡ P: ‡ T: ‡ V: in templo ‡ LXXO: • 3 LXX: ↑ ‡ P: ↑ + ‡ T: ↑ + ‡ V: ↑ + Domini • 4 LXX: ‡ P: ‡ T: ‡ V: ubi erat arca Dei (3:4) 1 LXX: ‡ P: ‡ T: ‡ V: Samuhel

276 2

1

6

(3:5)

5

4

3 8

2

1

(3:6)

1

(3:7) (3:8)

7

3

2 2

1

(3:5) 1 MT P T V] + 70mg Pmss it mss | + 187 | + LXX | + or mss + P • 2 MT P T V] LXX (3:6) 1 LXX] MT T V | P| L LXX • 2 MT LXX P T V] + 70 • 3 LXX] MT P T V • 4 MT P T V] > 89 187 LXX | 174 • 5 MT P T V] + LXX • 6 MT LXX P T V] 70 • 7 70] MT P T | LXX • 8 MT LXX P T V] + + whole verse (repeated) 70 (3:7) 1 LXX] MT P V • 2 LXX] MT P V • L 3 MT LXX P T V] LXX (3:8) 1 MT LXX P T V] + 70 LXXO • 2 MT LXX P T V] + 174 ‡ V: vocavi ‡ Pmss : ‡ LXX: ‡ ‡ T: ‡ V: et abiit ‡ LXX: ‡ T: ‡ V: Dominus vocare rursum ‡ ‡ LXXL: • 2 LXX: ‡ P: ‡ T: ‡ V: Samuhel • 3 LXX: ‡ P: ‡ T: ‡ V: Samuhel • 4 P: ‡ T: ‡ V: consurgensque Samuhel • 5 P: ‡ T: ‡ V: Heli ‡ LXX: • 6 LXX: ‡ P: ‡ T: ‡ V: et dixit • 7 P: ‡ T: ‡ LXX: • 8 LXX: ‡ P: ‡ T: ‡ V: et dormi (3:7) 1 LXX: ‡ P: ‡ V: Dominum • 2 LXX: ‡ P: ‡ V: neque revelatus fuerat • 3 LXX: ‡ P: ‡ T: ‡ V: ei sermo Domini ‡ L LXX : (3:8) 1 LXX: ‡ P: ‡ T: ‡ V: et vocavit (3:5) Pmss (3:6) P:

‡ T: • 2 P: 1 LXX: 1

P:

277 6

5,4

3

2

2

1

70 • 2 MT LXX T]

2

1

1

(3:9) (3:10)

1

2

(3:9)

3

(3:11) 1

(3:12)

LXX] MT P T (V omit )| MT P T V] LXX • 3 MT P T V] + LXX • 4 P • 5 MT T] Tms 145 182 • 6 MT LXX T]

P (3:10) 1 MT P T V] LXX • 2 LXX] + O mss MT P T V • 3 MT LXX T V] + LXX LXX P Vmss arm (3:11) 1 MT P T V] LXX • 2 LXX T] MT P V (3:12) 1 LXX P T V] MT • 2 LXX] MT | LXXL (P T V omit conjunction) (3:9) 1 LXX: ‡ P: ‡ T: ‡ V: et ait ad Samuhel • 2 P: ‡ T: ‡ V: vade ‡ LXX: • 3 P: ‡ T: ‡ V: et dormi ‡ LXX: • 4 LXX: ‡ T: ms ‡ P: • 5 T: ‡T : • 6 LXX: ‡ T: ‡ P: (3:10) 1 P: ‡ T: ‡ V: et vocavit ‡ LXX: • 2 LXX: ‡ P: + ‡ T: + ‡ V: + Samuhel Samuhel • 3 LXX: ‡ T: ‡ V: loquere ‡ LXXO: ‡ LXXmss : ‡ P: ‡ Vmss : loquere Domini (3:11) 1 P: ‡ T: ‡ V: verbum ‡ LXX: • 2 LXX: ‡ T: ‡ P: ‡ V: quod quicumque (3:12) 1 LXX: ‡ P: ‡ T: ‡ V: adversum • 2 LXX: ‡ L LXX : ‡ P: ‡ T: ‡ V: super domum eius

278 2

1

(3:13) 3

2

4

3

2

1

(3:14)

1 6

(3:15) 5

1

(3:16)

(3:13) 1 LXX T V] MT P | cj S. R. Driver • 2 cj BHK REB NJB] MT P T V | LXX | cj Wellhausen • 3 tiq soph LXX] MT P T (V) (3:14) 1 MT T] LXX | PV • 2 MT LXX P T V] > 89 (3:15) 1 MT LXX P T V] > 70 • 2 LXX] > MT P T V • 3 MT LXX T V] P • 4 MT LXX P] LXXO • 5 MT T] > 187 | 89 • 6 LXXO] MT LXX P T V | LXXL (3:16) 1 89 174 T] MT Tmss | LXX (3:13)

1

LXX: ‡ T: ‡ V: praedixi enim ‡ P: (← ) • 2 P: ‡ T: ‡ V: propter iniquitatem eo quod noverat ‡ LXX: • 3 LXX: ‡ P: ‡ T: ‡ V: indigne agere (← ) (3:14) 1 T: ‡ LXX: ‡ P: ‡ V: idcirco • 2 LXX: ‡ P: ‡ T: ‡ V: domui (3:15) 1 LXX: ‡ P: ‡ T: ‡ V: mane • 2 LXX: • 3 LXX: ‡ T: ‡ V: ostia ‡ P: • 4 LXX: ‡ P: ‡ LXXO: • 5 T: • 6 LXX O: ‡ LXX: ‡ P: ‡ T: ‡ V: et Samuhel timebat indicare visionem Heli ‡ LXXL: (3:16) LXX:

1

T:

‡ Tmss :

‡

279 2

1

4

(3:17)

3 5

4

3

2

2

1

1

1

(3:18) (3:19) (3:20)

MT LXX P T V] > 70 • 2 MT LXX T V] 187 | P • 3 MT LXX P T V] > 70 • 4 MT T] 187 LXX P V • 5 LXX] MT P T V (3:18) 1 LXX] MT P T V • 2 MT LXX P T V] + ms 174 LXX • 3 MT T V] + LXX P • 4 MT LXX P T V] > 174 (3:19) 1 LXX] MT P T V • 2 MT LXX P T V] L LXX (3:20) 1 MT LXX V] 174 | P (3:17)

1

(3:17)

1

LXX: ‡ T:

‡ P:

‡ T: ‡ V: quem locutus est • 2 LXX: ‡ V: oro te ne celaveris ‡ P: • 3 LXX: ‡ P: ‡ T: ‡ V: haec faciat tibi Deus et haec addat • 4 T: ‡ LXX: ‡ P: ‡ V: verbis • 5 LXX: ‡ P: ‡ T: ‡ V: tibi (3:18) 1 LXX: ‡ P: ‡ T: ‡ V: indicavit itaque ei • 2 LXX: ‡ P: ‡ T: ‡ V: sermones ‡ LXXms: + • 3 T: ‡ V: et ille respondit ‡ LXX: ‡ P: • 4 LXX: ‡ P: ‡ T: ‡ V: est (3:19) 1 LXX: ‡ P: ‡ T: ‡ V: et L Dominus erat • 2 LXX: ‡ P: ‡ T: ‡ V: et non ‡ LXX : (3:20) 1 LXX: ‡ V: Domini ‡ P:

280 3,2

1

(3:21) (4:1a)

4 5

(3:21-4:1a) 1 MT LXX P T V] + MT LXX T V] > 187 P • 3 cj] > LXX | + 89 | 187 • 4 T V] LXX |

70 • 2

5

(3:21-4:1a) 1 LXX: ‡ P: ‡ T: 2 LXX: ‡ T: quoniam revelatus fuerat Dominus Samuheli • 3 T: iuxta verbum Domini ‡ P: • 4 P: ‡ T: evenit sermo Samuhelis universo Israheli ‡ LXX:

MT T V | +

P| MT P

LXXL | LXXO • LXX] > MT P T V ‡ V: ut appareret • ‡ V: ‡ V: in Silo ‡ V: et ‡ LXXL:

‡ LXXO: • 5 LXX:

CONCLUSION This study has proposed an inductive method for reconstructing the critical text of a biblical book, especially one for which versional evidence is important. The text chosen to illustrate the method, 1 Samuel 3, is a relatively brief passage, too brief, in fact, for the conclusions regarding specific readings to be treated as anything other than provisional. The same applies to the descriptions of the translation techniques characteristic of each of the secondary witnesses: specific conclusions must be regarded as tentative, pending the application of the method to a larger section of Samuel. Nevertheless, it should be clear that the method promises to provide a wealth of detailed information about translation techniques, which can then be used in a systematic way to retrovert versional evidence, so that data from secondary witnesses can be used with confidence alongside readings from primary witnesses in an effort to reconstruct a critical text. Of paramount importance to the inductive method is the identification in the secondary witnesses of significant variants, which can be identified only after a detailed study of the translation that includes an analysis of consistency in lexical choices and grammatical correlations, segmentation, word order, and quantitative representation. Much of the data from the foregoing analysis is presented in data tables for each translation, and the data in these tables allows the textual critic to describe the translation technique in great detail on the basis of solid data rather than vague impressions. General characterizations of a translation as “literal” or “free” may be set aside in favor of the detailed descriptions offered by the inductive method.1 Summaries at the bottom of each data table offer an evaluation of the translation technique of the version with respect to the specific aspect of the translation being analyzed. These summaries do not tell the whole story regarding the literalness of the translations, however. For example, the tables dealing with the quantitative representation of elements in the Hebrew text for P and T (see below, Appendix 3) indicate that P deviates from its Vorlage about 20% of the time, while T deviates about 10% of the time. One can say, then, that P varies from its Hebrew Vorlage about twice as often as T does, at least in 1 Samuel 3. However, a closer examination of the data in the tables indicates that the deviations from the Vorlage characteristic of T for the most part fall into two broad

1 For a summary of the arguments presented in this study and a suggestion for the use of descriptive language to describe the translation technique, see James R. Adair, “‘Literal’ and ‘Free’ Translations: A Proposal for a More Descriptive Terminology,” Journal of Northwest Semitic Languages 23 (1997): 181-209.

281

282

categories: theological “improvements” (about two-thirds of the total word count) and stylistic accomodations to Aramaic (most of the rest of the variants). In contrast, there are no obvious theological “improvements” in P. Rather, about two-thirds of P’s changes are stylistic accomodations Syriac, and the other third may be seen as other “improvements” to the text not dictated by Syriac idiom (e.g., specifying subjects or objects of verbs, explicatory additions). Similarly, if one were to calculate the individual deviation factor reflected in the Greek renderings of Hebrew participles (table 5, grouping the data as described there), the low deviation factor of 0.73 might be deceiving. The Greek translators are in fact quite consistent (in the chapter) in rendering Hebrew participles, but the way in which they are consistent is surprising: they rather consistently use full Greek verbs to render Hebrew participles. It is clear, then, that all the data in the tables, not just the summaries, must be considered when attempting to understand the way in which the translators worked. Constructing critical texts of the secondary and partial secondary witnesses, analyzing the translation techniques of the secondary witnesses in great detail, identifying the significant variants, retroverting the secondary witnesses, evaluating the various readings (original and retroverted), and constructing a critical text in the source language: the inductive method—built on the foundation laid by Margolis and elaborated upon by Tov, Barr, and others—demands a great deal of effort (it is a “stupendous work,” according to Margolis), but it offers the clearest possible description of the work of the translators and thus enables critics to use versional evidence with a high degree of confidence. It is hoped that scholars will apply the inductive method to other portions of scripture, perhaps refining and improving the method in various ways. Above all, the inductive method is designed to be used to construct critical texts of Old Testament books that may be used by translators, exegetes, and theologians as they perform their various tasks in service of scholarship, the synagogue, and the church.

APPENDIX 1 RECONSTRUCTED ARCHETYPICAL TEXTS OF SECONDARY AND PARTIAL SECONDARY WITNESSES The text of P accepted in this thesis as archetypical is identical to that of the Leiden edition. Similarly, the text of the Roman edition of V is accepted as archetypical. The texts of the other versions are given, in whole or in part, below. Septuagint In light of the number of readings from the apparatus of the Cambridge LXX that were preferred over the base text (twelve), the entirety of the reconstructed archetype of 1 Samuel 3 is given below. (Readings different from the base text occur in verses 2, 3, 5, 6, 7, 9, 15, and 17). (3:1) (2) (3) (4) (5) (6)

(7) (8) (9) (10) (11) (12)

283

284

(13) (14) (15) (16) (17) (18) (19) (20) (21)

Targum The reconstructed archetypical text of T is identical to that of Sperber in all but verses 2 and 10, which are given below. (3:2) (3:10)

The Lucianic Recension LXXL differs from LXX in every verse except verse 2, a total of forty-nine deviations. Furthermore, Paul de Lagarde’s edition of LXXL is not totally acceptable, since he fails to include a critical apparatus of variant readings and because he sometimes prefers the reading of one (and on one occasion none) of the primary Greek witnesses to LXXL (boc2 e2 ) to that of the other three.1 Inasmuch as the text accepted as Lucianic in this thesis differs from Lagarde’s in eight places, and since his edition is somewhat difficult to find, the entire text of 1 Samuel 3 is given below. 1 Lagarde, Librorum VT Canonicorum. As noted above (p. 69, n. 63), the Lucianic text reconstructed here is basically the same as that found in Taylor, Lucianic Manuscripts, 9-12, with the exception that Taylor reads [ ] in v. 13, whereas the text reconstructed for this study reads simply .

285

(3:1) (2) (3) (4) (5) (6)

(7) (8) (9)

(10) (11) (12) (13) (14) (15) (16) (17)

(18) (19) (20) (21)

286

The Hexaplaric Recension (3:1) (2) (3) (4) (5) (6)

(7) (8) (9)

(10) (11) (12) (13) (14) (15) (16) (17) (18) (19)

287

(20) (21)

APPENDIX 2 ADD-OMS AND PRELIMINARY TRANSLATION TECHNIQUE TABLES Most of the abbreviations used in the tables are self-explanatory. However, the following should be noted: Ch or chain (second or higher item in a construct chain), OR (object of relative particle), SC or subj comp (subjective complement), OP (object of preposition), CPast (continuous past). These tables describe aspects of the translation techniques of the complete secondary witnesses before the significant variants have been eliminated from the data. Tables that are different from the Final Translation Technique Tables in Appendix 3 are marked with an asterisk. Septuagint Add-oms 3:1 3:2 3:3 3:4 3:5 3:6

4°

4° 2° 2°

3:7 3:9

6° 2° 2°

3:10

288

289

Add-oms—Continued. 3:11 3:13 3:15 3:16 - 1° 3:17 3:18 3:21

Consistency *Table 1.—Verbs, Nouns, and Adjectives: Lexemes Heb Lexeme

Gk Lexeme

Frequency 3 12 1 5 1 1 2 5 5 3 3 2 3 2 2 16

Common Root

1 1 1

290

Table 1—Continued. Heb Lexeme

Heb words (>1x): 31 deviation factor: 0.38 total Heb (>1x): 132 Heb roots (>1x): 29

Gk Lexeme

Frequency 1 3 4 3 6 2 1 1 2 2 2 2 2 1 1 3 2 1 1 1 11 1 1 2 6 2 3

2 3

2 2

3 3

Gk words: 42

1.35 Gk words/Heb word

primary Gk: 117 Gk roots: 42

88.6% of Heb words by primary rendering 1.45 Gk roots/Heb root

*Table 2.—Adverbs, Prepositions, and Particles: Lexemes Heb Lexeme

Common Root

Gk Lexeme —

Frequency 4 6

291

Table 2—Continued. Heb Lexeme

Gk Lexeme

rel pronoun — art (=rel pronoun) —

—

—

( , ) , , ,

Heb advs (>1x): 13 total Heb (>1x): 85 deviation factor: 2.78

—

Gk advs: 26 primary Gk: 67

Frequency 1 1 2 1 2 1 2 7 1 12 1 2 6 2 2 11 1 1 6 1 5 1 4 2

2.00 Gk advs/Heb adv 78.8% of Heb advs by primary rendering

Table 3.—Conjunctions: Lexemes Heb Lexeme

Heb conjs (>1x): 2 total Heb (>1x): 66 deviation factor: 0.00

Gk Lexeme

Gk conjs: 2 primary Gk: 66

Frequency 56 10 1.00 Gk conjs/Heb conj 100.0% of Heb conjs by primary rendering

292 *Table 4.—Word Classes percentage of verbs/verbals represented by verbs/verbals: 94/95 = 98.9% percentage of verbs represented by verbs: 71/76 = 93.4% percentage of participles represented by participles: 4/12 = 33.3% percentage of infinitives represented by infinitives: 4/7 = 57.1% percentage of nouns/adjs represented by nouns/adjs: 52/55 = 94.5% percentage of pronouns represented by pronouns: 37/38 = 97.4% *Table 5.—Verbs: Hebrew Inflection, Greek Tense and Mood Heb Infl Ptc pf pf pf pf w/c pf w/c impf impf impf impf w/c impf w/c impv inf/abs inf/cons inf/cons ptc ptc ptc ptc

Gk Tense impf 3 aor 10 pf 5 fut 2 pf 1 fut 4 impf 1 aor 7 pres 1 aor 34 pres 8 fut 2 pres 1 aor 4 pres 6 impf 3 aor 1 pf 1

Gk mood:

Ind

Subj

Opt

Impv Inf

3 7 5 2 1 4 1

1

3

2

2

2

1 34 8 2 1 4 3 1 1

1 3 2

293

Table 5—Continued. deviation factor (discrete tense/mood combinations)1 : 6.68 deviation factor (grouped)2 : 0.97 deviation factor (grouped, without Hebrew verbals)3 : 0.85 *Table 6.—Verbs: Hebrew Stem, Greek Voice Heb Stem qal ni pi hith hi

Gk Voice:

Act 55 2 11 1 5

Mid 2

1st 6

2d

Pass 7 3 2 1

Other4

3d

Other

51

5

1

5

deviation factor: 3.57 Table 7.—Verbs: Person Heb Person 1st 2d 3d

Gk Person:

14

deviation factor: 0.00 Table 8.—Verbs: Number Heb Number Gk Number: Sing sg 79 pl 1

Plur 2 2

Other 4

deviation factor: 0.12 1 Perfect, perfect w/c, imperfect, imperfect w/c are counted as four separate inflections in all calculations in this table. 2 For full verbs, Greek past tenses (imperfect, aorist, perfect) are grouped together, and subjunctive

and optative moods are ignored, since they have nothing to do with time. For verbals, Hebrew infinitives are grouped, as are Greek infinitives, so that the regularily of rendering Hebrew infinitives with Greek infinitives versus rendering Hebrew infinitives with Greek verbs is what is being compared. Participles are similarly grouped, but it should be noted that the “normal” rendering of the Hebrew participle in this chapter is with a full Greek verb. 3 Since Hebrew verbals are often rendered less strictly than full verbs, they are excluded from this calculation for comparison with the previous deviation factor. 4 Items in the “other” columns in the following tables indicate that a word of one class in Hebrew

(verb, noun/adjective, pronoun) was rendered with a word in a different class in Greek. Since these deviations are accounted for in table 5, they are ignored in calculating the deviation factors.

294 Table 9.—Nouns and Adjectives: Gender Heb Gender masc fem

Gk Gender:

Masc Fem 31 8 1 4

Neut 12 1

Other 8 2

Dual

Other 9 1

Dat

Acc

deviation factor: 21.54 *Table 10.—Nouns and Adjectives: Number Heb Number Gk Number: Sing sing 42 plur dual

Plur 8 6 1

deviation factor: 2.29 Table 11.—Nouns and Adjectives: Use vs. Case Heb Use subj subj comp DO chain OP voc

Gk Case:

Nom 36 3

Gen 1

12 8

14

deviation factor (discrete): 6.45 deviation factor (grouped): 0.37 Table 12.—Nouns and Adjectives: Articles Heb Article yes implied5 no

Gk Article:

Yes 6 4 18

No

Other 1

29

9

deviation factor: 11.37 deviation factor (without implied articles): 12.23 5 Implied by Masoretic pointing.

Other 1

1 2

Voc

11 1 7 2

3 4

295

Table 13.—Nouns and Adjectives: Hebrew Definiteness, Greek Articles Heb Definiteness def not def

Gk Article:

Yes 26 2

No 15 14

Other 2 8

Neut 1

Other 1

Plur 1 2

Dual

Other

Gen

Dat

deviation factor: 8.04 Table 14.—Pronouns: Gender Heb Gender masc com

Gk Gender:

Masc Fem 20 3 12

deviation factor: 1.44 Table 15.—Pronouns: Number Heb Number Gk Number: Sing sing 33 plur

1

deviation factor: 0.06 Table 16.—Pronouns: Use vs. Case Heb Use subj subj comp chain OP adj

Gk Case:

Nom 8 1

11 4

3 3

Acc

Voc

1 6

deviation factor (discrete): 3.76 deviation factor (grouped): 0.06 Segmentation

51 52 53 54 55 56

Ref 3:1 3:2 3:5 3:6 3:7 3:8

Hebrew Compound

Greek Rendering

Other

1

296

Segmentation—Continued. 57 3:8 58 3:9 59 3:13 60 61 3:14 62 3:15 63 3:17 64 3:18 65 3:20 66 3:21 Heb compounds: 43 Gk equivalents: 27

62.8% of Heb compounds rendered exactly

*Word Order Ref Number of Variations Greek Variant 67 3:16 3 68 3:19 1 Heb semantic units: 373 variations: 4 Heb word order followed 98.9% of the time *Quantitative Representation

69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85

Ref 3:1 3:2 3:3 3:4 3:5 3:6

Number of Variations +1 +2 +1 -1 -1 -1 +1 +1 +1 +1 -1 +1 -3 +1 -2 +1 +1

Variant 1° 4°

4° 2° 2°

6°

297

Quantitative Representation—Continued. 86 3:7 -1 87 3:9 -3 88 +1 89 +1 90 3:10 +1 91 -2 92 3:11 +1 93 -1 94 3:13 -2 95 +2 96 3:15 +4 97 -1 98 3:16 -2 99 -1 100 3:17 -2 101 +3 102 3:18 -2 103 +1 104 3:21 -5 105 +31

2° 2°

1° 2°

Heb semantic units: 438 Gk deviations: 88 Peshitta Add-oms 3:1 3:5

- 2° - 5°

79.9% agreement

298

Add-oms—Continued. 3:6 - 1° - 2° 3:8 - 1° - 2°

3:9

- 1° - 3°

3:10 3:11 3:12

- 2° - 1°

3:13

- 2°

3:14

- 1° - - 1° 1° - 3° 2°

3:15

2°

3:17 1° 3° 2° 3:18

3:20 3:21

-

- 1°

-

299

Add-oms—Continued. 3° Consistency *Table 17.—Verbs, Nouns, and Adjectives: Lexemes Heb Lexeme

Syr Lexeme

Frequency 2 1 14 5 2 3 2 3 6 3 5 2 3 16 1 3 4 1 2 6 2 1 3 2 2 2 2 2 3 3 12

Common Root

1 1 1 1

2 3

2

300

Table 17—Continued. Heb Lexeme

Heb words (>1x): 29 deviation factor: 0.26 total Heb (>1x): 132 Heb roots (>1x): 27

Syr Lexeme

Frequency 1 1 2 8 3

Syr words: 35

1.21 Syr words/Heb word

primary Syr: 123 Syr roots: 34

93.2% of Heb words by primary rendering 1.26 Syr roots/Heb root

Table 18.—Adverbs, Prepositions, and Particles: Lexemes Heb Lexeme

Syr Lexeme

—

—

—

, ,

Common Root 3 3

—

Frequency 5 3 3 1 2 1 6 4 3 1 13 1 6 3 2 7 7 6 1 6 1 3

301

Table 18—Continued. Heb advs (>1x): 12 Syr advs: 22 total Heb (>1x): 85 primary Syr: 63 deviation factor: 3.41

1.83 Syr advs/Heb adv 74.1% of Heb advs by primary rendering

Table 19.—Conjunctions: Lexemes Heb Lexeme

Syr Lexeme

Frequency 54 1 7 2

Heb conjs (>1x): 2 Syr conjs: 4 total Heb (>1x): 64 primary Syr: 61 deviation factor: 0.16

2.00 Syr conjs/Heb conj 95.3% of Heb conjs by primary rendering

Table 20.—Word Classes percentage of verbs/verbals represented by verbs/verbals: 94/96 = 97.9% percentage of verbs represented by verbs: 74/75 = 98.7% percentage of participles represented by participles: 7/12 = 58.3% percentage of infinitives represented by infinitives: 4/7 = 57.1% percentage of nouns/adjs represented by nouns/adjs: 65/70 = 92.9% percentage of pronouns represented by pronouns: 42/42= 100.0% Table 21.—Verbs: Inflection Heb Infl pf pf w/c impf impf w/c impv inf/abs inf/cons ptc

Syr Infl: Other

Pf

Impf

Impv

Inf

Ptc

17 2 37

Pres CPast Plpf 1

1 9

1 1 8 2

4

1 1

deviation factor: 1.83 deviation factor (without verbals): 0.42

4 5

2

302 Table 22.—Verbs: Stem Heb Stem Syr Stem: Pe qal 65 ni 1 pi 6 hith 1 hi 1

Ethpe Pa

Ethpa Aph Ettaph Other 1 1 1 1 1 6 1

2 6 2

deviation factor: 1.33 Table 23.—Verbs: Person Heb Person 1st 2d 3d

Syr Person:

1st 6

2d

3d

Other

56

1

14

deviation factor: 0.00 Table 24.—Verbs: Gender Heb Gender masc fem com

Syr Gender:

Masc Fem 80 1 1

Com

6

deviation factor: 0.02 Table 25.—Verbs: Number Heb Number Syr Number: Sing sing 85 plur

Plur

Other 1

3

deviation factor: 0.00 Table 26.—Nouns and Adjectives: Gender Heb Gender masc fem deviation factor: 1.38

Syr Gender:

Masc Fem 50 6 6 3

Other 4 1

Other 1

303 *Table 27.—Nouns and Adjectives: Number Heb Number Syr Number: Sing sing 49 plur 1 dual

Plur 8 6 1

Dual

Other 5

deviation factor: 2.03 Table 28.—Nouns and Adjectives: Use Heb Use Syr Use: Subj Ch DO subj 38 subj comp DO 1 9 chain 11 OP 3 1 voc

SC OP Voc Adj Adv OR 3

1 2 4 30 2

6

deviation factor: 1.09 Table 29.—Nouns and Adjectives: Hebrew Articles, Syriac Emphatic State Heb Article yes implied no

Syr Emphatic: Yes 6 4 14

No 1

Other

40

5

deviation factor: 6.06 deviation factor (without implied articles): 6.46 Table 30.—Nouns and Adjectives: Definiteness Heb Definiteness def not def

Syr Definiteness:

Yes 41 8

No 2 14

Com 1 14

Other

deviation factor: 2.09 Table 31.—Pronouns: Gender Heb Gender masc com deviation factor: 0.05

Syr Gender:

Other

Masc Fem 26

Other 1 4

1 1

304 Table 32.—Pronouns: Number Heb Number Syr Number: Sing sing 38 plur

Plur

Dual

Other

3

deviation factor: 0.00 Table 33.—Pronouns: Use Heb Use Syr Use: subj subj comp chain OP adj

Subj 8

Ch

DO IO SC OP Voc Adj

Other

1 1

11 1

1 6

2 8 3

deviation factor: 8.15 Segmentation Ref 3:1 3:5 3:6 3:8

Hebrew Compound

38 39 40 41 42 43 3:9 44 3:10 45 3:13 46 3:15 47 3:20 Heb compounds: 47 Syr equivalents: 36

Syriac Rendering -

76.6% of Heb compounds rendered exactly

*Word Order Ref Number of Variations Syriac Variant 48 3:1 1 49 3:6 2 50 3:11 1 Heb semantic units: 396 variations: 4 Heb word order followed 99.0% of the time

305 *Quantitative Representation

51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86

Ref 3:1

Number of Variations +1 +1 +1 +1 +1 3:2 +1 +1 3:3 +1 +1 +1 +1 +1 +1 3:5 -1 -1 3:6 +1 +1 -1 3:7 +1 +1 3:8 +1 +1 +1 -1 +1 +1 +1 -1 3:9 -1 -1 -1 3:10 -3 +1 +1 3:11 +1 +1

Variant 1° - 1° - 1° - 2° - 4° - 1° - 1° - 2° - 2° - 3° - 4° - 5° - 1° - 2° - 1° - 2° - 1°

- 1° - 3° 2° - 2° -

306

Quantitative Representation—Continued. 87 3:12 +1 88 -1 89 3:13 -1 90 +1 91 +1 92 -1 93 +2 94 3:14 -1 95 +1 96 +1 97 +1 98 +1 99 3:15 -1 100 +1 101 +1 102 +1 103 -1 104 +1 105 -1 106 3:17 +2 107 +1 108 +1 109 -1 110 +2 111 +1 112 3:18 -1 113 -1 114 +1 115 +1 116 +1 117 +1 118 3:19 +1 119 3:20 +1 120 -1 121 3:21 +1 122 -7 123 +1

- 1° - 1°

- 2° - 1° - - 1° 1° - 1° - 1° - 2° - 3° 2°

1° 3° 2° -

- 1°

-

307

Quantitative Representation—Continued. 124 -1 Heb semantic units: 425 Syr deviations: 85

3° 80.0% agreement

Targum Add-oms 3:2

- 1° - 4°

3:3 3:7

3:8 3:11 3:12 3:13 3:14 3:15 3:19

-

3:20 Consistency Table 34.—Verbs, Nouns, and Adjectives: Lexemes Heb Lexeme

Aram Lexeme

Frequency 3 14 5 2 2 5 9

Common Root

1 1

308

Table 34—Continued. Heb Lexeme

Aram Lexeme

6

Heb words (>1x): 31 deviation factor: 0.04 total Heb (>1x): 139 Heb roots (>1x): 29

Aram words: 34

Frequency 4 5 1 1 3 19 3 4 3 6 2 1 3 2 2 2 2 1 1 3 2 3 12 1 1 2 8 3

Common Root

2 3

2 3 3

1.10 Aram words/Heb word

primary Aram: 136 97.8% of Heb words by primary rendering Aram roots: 34 1.17 Aram roots/Heb root

6 The full form is listed, since that is what is rendered in T, although the basic Hebrew word is

.

309 *Table 35.—Adverbs, Prepositions, and Particles: Lexemes Heb Lexeme

Aram Lexeme

—

, Table 35—Continued. Heb advs (>1x): 12 total Heb (>1x): 85 deviation factor: 2.22

Aram advs: 22 primary Aram: 71

Frequency 5 2 2 2 3 5 2 3 1 1 1 15 1 5 3 2 13 1 1 6 7 4 1.83 Aram advs/Heb adv 83.5% of Heb advs by primary rendering

Table 36.—Conjunctions: Lexemes Heb Lexeme

Heb conjs (>1x): 2 total Heb (>1x): 68 deviation factor: 0.00

Aram Lexeme

Aram conjs: 2 primary Aram: 68

Frequency 58 10 1.00 Gk conjs/Heb conj 100.0% of Heb conjs by primary rendering

310 Table 37.—Word Classes percentage of verbs/verbals represented by verbs/verbals: 98/98= 100.0% percentage of verbs represented by verbs: 76/77 = 98.7% percentage of participles represented by participles: 11/12 = 91.7% percentage of infinitives represented by infinitives: 5/7 = 71.4% percentage of nouns/adjs represented by nouns/adjs: 65/68 = 95.6% percentage of pronouns represented by pronouns: 42/42=100.0% *Table 38.—Verbs: Inflection Heb Infl

Aram Infl: Pf Other 18 w/c 1 3 w/c 37

pf pf impf impf impv inf/abs inf/cons ptc

Impf

Impv

Inf

Ptc

Pres CPast Plpf

1 2 9 8 2 5 1

11

deviation factor: 0.29 deviation factor (without verbals): 0.28 Table 39.—Verbs: Stem Heb Stem Aram Stem: Pe Ithpe Pa qal 64 3 ni 1p7 3 1 pi 1 11 hith 1 hi 4

Ithpa Aph/Shaph Ittaph

Other

1 1 1 6

deviation factor: 1.02 7 The “p” in the peal column in the niphal row indicates a peal passive form (participle). Absence

of the letter “p” in other positions in the columns of the active stems (peal, pael, and aphel/shaphel) indicates that all participial forms in those stems are active.

311 Table 40.—Verbs: Person Heb Person 1st 2d 3d

Aram Person: 1st 6

2d

3d

Other

58

1

14

deviation factor: 0.00 Table 41.—Verbs: Gender Heb Gender masc fem com

Aram Gender: Masc Fem 81 1 1 1

Com

Other 1

6

deviation factor: 0.04 Table 42.—Verbs: Number Heb Number Aram Number: Sing Plur sing 86 1 plur 3

Other 1

deviation factor: 0.02 Table 43.—Nouns and Adjectives: Gender Heb Gender masc fem

Aram Gender: Masc Fem 52 5 5 3

Other 2 1

deviation factor: 1.05 Table 44.—Nouns and Adjectives: Number Heb Number Aram Number: Sing Plur sing 54 4 plur 6 dual 1 deviation factor: 0.50

Dual

Other 2 1

312

Table 45.—Nouns and Adjectives: Use Heb Use Aram Use: Subj Ch DO subj 37 subj comp DO 1 8 chain 11 OP 3 voc

SC

OP Voc Adj Adv OR 1 1

3

1 3 5 1

31 8

deviation factor: 1.47 Table 46.—Nouns and Adjectives: Hebrew Articles, Aramaic Emphatic State Heb Article

Aram Emphatic:

Yes

No

Other

yes

5

1

1

implied

4

no

6

49

2

deviation factor: 1.14 deviation factor (without implied articles): 1.21 Table 47.—Nouns and Adjectives: Definiteness Heb Definiteness def not def

Aram Definiteness:

Yes 40 4

No 1 20

Com

Other

deviation factor: 0.52 Table 48.—Pronouns: Gender Heb Gender masc com deviation factor: 0.00

Other

Aram Gender: Masc Fem 27

14

Other 2 1

1 1

313

Table 49.—Pronouns: Number Heb Number Aram Number: Sing Plur sing 38 plur 3

Dual

Other

deviation factor: 0.00 Table 50.—Pronouns: Use Heb Use Aram Use: Subj Ch subj 8 subj comp chain 12 OP 1 adj

DO IO

SC OP Voc Adj Adv Other 1

1

1 15 3

deviation factor: 0.34 Segmentation Ref Hebrew Compound Aramaic Rendering 42 3:18 43 3:20 Heb compounds: 50 Aram equivalents: 48 96.0% of Heb compounds rendered exactly Word Order There are no deviations from the word order of MT found in T, as the following summary indicates. Heb semantic units: 416 variations: 0 Heb word order followed 100.0% of the time *Quantitative Representation

44 45 46 47 48 49 50 51 52

Ref 3:1 3:2 3:3

3:7

Number of Variations +1 +1 +1 +2 +1 +6 +1 +1 +3

Variant - 1° - 4° - 1° - 4°

314

Quantitative Representation—Continued. 53 3:7 +1 54 +1 55 3:8 +2 56 3:11 -1 57 +1 58 3:12 +1 59 3:13 +1 60 3:14 +1 61 3:15 +1 62 +1 63 +1 64 3:18 +1 65 3:19 +1 66 +1 67 +1 68 -2 69 +1 70 3:20 +2 71 3:21 +1 Heb semantic units: 428 Aram deviations: 39 Vulgate Add-oms 3:2 3:3 3:4 3:5 3:6

3:8

- 2° - 1° erat - 2° qui respondens - 3° qui - 3° - 5° qui te et 3° et 2°

-

-

-

90.9% agreement

315

Add-oms—Continued. 3:8 adhuc - 2° qui - 3° 3:9 et 3° deinceps - 3° 3:10 3:11 3:13

vocaverat secundo - 1° 2° indigne

3:14

3:15 3:16 3:17

3:18

3:20

- 1° quod eius in - 3° qui respondens eum est 1° ille quod est 2° - 2° esset

316

Consistency Table 51.—Verbs, Nouns, and Adjectives: Lexemes Heb Lexeme

Lat Lexeme Deus aio dico interrogo respondeo domus filius revelo dico loquor sermo verbum sum abeo vado cognosco nosco scio Dominus dies adiicio addo abscondo celo omnis quicumque universus loco visio praedico indico puer servus

Frequency 3 5 6 1 2 5 2 2 1 4 5 4 3 4 1 1 1 1 19 3 2 2 2 1 3 1 2 2 1 1 2 2 2

Common Root

1 1 1 1

2 3

317

Table 51—Continued. Heb Lexeme

Heb words (>1x): 29 deviation factor: 2.13 total Heb (>1x): 134 Heb roots (>1x): 27

Lat Lexeme aeternum iniquitas oculus facio consurgo suscito voco video apareo revertor dormio iaceo audio

Frequency 2 2 2 3 2 1 12 1 1 2 7 1 3

2 2 3 3

Lat words: 45

1.55 Lat words/Heb word

primary Lat: 108 Lat roots: 43

80.6% of Heb words by primary rendering 1.59 Lat roots/Heb root

Table 52.—Adverbs, Prepositions, and Particles: Lexemes Heb Lexeme

Common Root

Lat Lexeme — ad adversum super si non ubi qui eo — — in propter iuxta ecce praesto

Frequency 6 5 1 1 2 1 1 4 1 8 5 6 1 1 5 1

318

Table 52—Continued. Heb Lexeme

, , ,

Heb advs (>1x): 12 total Heb (>1x): 82 deviation factor: 5.41

Lat Lexeme antequam neque necdum hic — ad ut nec non — a ex in usque Lat advs: 30 primary Lat: 55

Frequency 1 1 1 2 10 1 1 1 5 1 3 2 1 3 2.50 Lat advs/Heb adv 67.1% of Heb advs by primary rendering

Table 53.—Conjunctions: Lexemes Heb Lexeme

Heb conjs (>1x): 2 total Heb (>1x): 55 deviation factor: 30.98

Lat Lexeme ergo -que autem enim et igitur itaque porro enim qui quia quoniam Lat conjs: 12 primary Lat: 37

Frequency 3 3 4 1 32 1 1 1 1 2 5 1 6.00 Lat conjs/Heb conj 67.3% of Heb conjs by primary rendering

319 Table 54.—Word Classes percentage of verbs/verbals represented by verbs/verbals: 95/97 = 97.9% percentage of verbs represented by verbs: 74/76 = 97.4% percentage of participles represented by participles: 0/12 = 0.0% percentage of infinitives represented by infinitives: 3/7 = 42.9% percentage of nouns/adjs represented by nouns/adjs: 57/68 = 83.8% percentage of pronouns represented by pronouns: 36/40 = 90.0% *Table 55.—Verbs: Hebrew Inflection, Latin Tense and Mood Heb Infl pf pf pf pf w/c pf w/c impf impf impf impf impf impf impf w/c impf w/c impf w/c impv inf/abs inf/cons inf/cons inf/cons ptc ptc ptc ptc

Lat Tense Lat Mood: impf 4 pf 12 plpf 3 fut 1 pf 1 fut 3 fut pf 1 impf 2 pf 2 plpf 1 pres 3 impf 5 pf 30 pres 2 pres 8 fut 2 impf 1 pf 1 pres 3 fut 1 fut pf 1 impf 4 pres 4

Ind 4 12 3 1 1 3 1 1

Subj

Opt

Impv Inf

Ptc

1 2

1 3 5 30 2 8 2 1 1 3 1 1 3 3

1

deviation factor (discrete tense/mood combinations): 6.15 deviation factor (grouped): 1.53 deviation factor (grouped, without verbals): 0.25

1

320 Table 56.—Verbs: Hebrew Stem, Latin Voice Heb Stem qal ni pi hith hi

Lat Voice:

Act 62 2 10 1 10

Pass 2 2 1 1

Dep 2

Other

1st 6

2d

3d

Other

56

2

2 2

deviation factor: 0.63 Table 57.—Verbs: Person Heb Person 1st 2d 3d

Lat Person:

14

deviation factor: 0.00 Table 58.—Verbs: Number Heb Number Lat Number: Sing sing 85 plur

Plur 1 2

Other 1 1

Masc Fem 25 14 2 4

Neut 10 1

deviation factor: 0.02 Table 59.—Nouns and Adjectives: Gender Heb Gender masc fem

Lat Gender:

deviation factor: 31.64 Table 60.—Nouns and Adjectives: Number Heb Number Lat Number: Sing sing 41 plur dual deviation factor: 2.95

Plur 9 5 1

Other 10 2

Other 11 1

321 Table 61.—Nouns and Adjectives: Use vs. Case Heb Use subj subj comp DO chain OP voc ptc (adj) verb (ptc) adv

Lat Case:

Nom 37 3

Gen

Dat

Abl

Voc

Other 1

1 2 1

Acc 1

10 4

11 1 14 2

2 16

1 1 6

1 9 1

deviation factor (discrete): 13.58 deviation factor (grouped): 1.02 Table 62.—Pronouns: Gender Heb Gender masc com

Lat Gender:

Masc Fem 22 2 12

Neut

Other 1 2

Dual

Other 3

Dat

Acc

5

1 6

deviation factor: 0.22 Table 63.—Pronouns: Number Heb Number Lat Number: Sing sing 34 plur

Plur 2

deviation factor: 0.00 Table 64.—Pronouns: Use vs. Case Heb Use subj subj comp chain OP adj

Lat Case:

deviation factor (discrete): 9.91 deviation factor (grouped): 3.44

Nom 5 1 2

Gen

5

Abl

Voc

3 4 3

2

Other 3

322

Segmentation Ref 3:1 3:2 3:5 3:6 3:7 3:8

Hebrew Compound

48 49 50 51 52 53 54 55 56 3:9 57 3:10 58 3:13 59 60 3:14 61 62 63 3:15 64 3:17 65 66 67 3:18 68 3:20 69 Heb compounds: 49 Lat equivalents: 27

Latin Rendering coram videre me me ei tertio me puerum te sicut vocaverat secundo ei eos domui victimis et muneribus indicare me tibi tibi ei propheta Domini 55.1% of Heb compounds rendered exactly

Word Order Ref Number of Variations Latin Variant 70 3:17 1 oro te ne Heb semantic units: 372 variations: 1 Heb word order followed 99.7% of the time Quantitative Representation

71 72 73 74

Ref 3:2

3:3

Number of Variations +1 -1 -1 -1

Variant factum - 2° - 1°

323

Quantitative Representation—Continued. 75 3:3 -1 76 +1 77 3:4 -1 78 +2 79 3:5 -1 80 +1 81 3:6 -1 82 -1 83 +1 84 +1 85 +1 86 3:7 +1 87 3:8 +1 88 +1 89 -1 90 +1 91 -1 92 3:9 -1 93 +1 94 -1 95 +1 96 -1 97 3:11 -1 98 3:12 +1 99 3:13 -1 100 +1 101 +1 102 -1 103 +1 104 -2 105 3:14 -1 106 +1 107 +1 108 -1 109 +1 110 3:15 -1 111 3:16 -1

erat - 2° qui respondens - 3° qui - 3° - 5° qui te et 3° fuerat et 2° adhuc - 2° qui - 3° et 3° deinceps - 3° - 1° sum essem quod 2° indigne - 1° quod eius usque - 3°

324

Quantitative Representation—Continued. 112 3:16 +2 qui respondens 113 3:17 +1 eum 114 +1 est 1° 115 +1 est 2° 116 +1 te 2° 117 +1 sunt 118 3:18 +1 ille 119 +1 quod 120 +1 est 2° 121 3:20 -1 - 2° 122 +1 esset 123 3:21 +1 fuerat 124 -1 Heb semantic units: 424 Lat deviations: 57 86.6% agreement

APPENDIX 3 FINAL TRANSLATION TECHNIQUE TABLES See Appendix 2 for a brief description of the abbreviations used in the tables. These tables describe aspects of the translation techniques of the complete secondary witnesses after the significant variants have been eliminated from the data. Tables that are different from the Preliminary Translation Technique Tables in Appendix 2 are marked with an asterisk. Septuagint Consistency *Table 1.—Verbs, Nouns, and Adjectives: Lexemes Heb Lexeme

Gk Lexeme

325

Frequency 3 12 1 5 1 1 2 5 5 3 3 3 2 2 16 3 4 3

Common Root

1 1 1

326

Table 1—Continued. Heb Lexeme

Heb words (>1x): 31 deviation factor: 0.31 total Heb (>1x): 129 Heb roots (>1x): 29

Gk Lexeme

Frequency 6 2 1 1 2 2 2 2 2 1 1 3 2 1 1 1 11 1 1 2 6 2 3

2 3

2 2

3 3

Gk words: 40

1.29 Gk words/Heb word

primary Gk: 116 Gk roots: 40

89.9% of Heb words by primary rendering 1.38 Gk roots/Heb root

*Table 2.—Adverbs, Prepositions, and Particles: Lexemes Heb Lexeme

Common Root

Gk Lexeme —

rel pronoun —

Frequency 4 6 2 1 2 1

327

Table 2—Continued. Heb Lexeme

Gk Lexeme art (=rel pronoun) —

—

—

( , ) , , ,

Heb advs (>1x): 13 total Heb (>1x): 82 deviation factor: 1.27

—

Gk advs: 23 primary Gk: 67

Frequency 2 7 12 1 2 6 2 2 11 1 1 6 1 5 1 4 2

1.77 Gk advs/Heb adv 81.7% of Heb advs by primary rendering

Table 3.—Conjunctions: Lexemes Heb Lexeme

Heb conjs (>1x): 2 total Heb (>1x): 66 deviation factor: 0.00

Gk Lexeme

Gk conjs: 2 primary Gk: 66

Frequency 56 10 1.00 Gk conjs/Heb conj 100.0% of Heb conjs by primary rendering

328

*Table 4.—Word Classes percentage of verbs/verbals represented by verbs/verbals: 93/94 = 98.9% percentage of verbs represented by verbs: 71/76 = 93.4% percentage of participles represented by participles: 4/12 = 33.3% percentage of infinitives represented by infinitives: 4/6 = 66.7% percentage of nouns/adjs represented by nouns/adjs: 52/55 = 94.5% percentage of pronouns represented by pronouns: 37/38 = 97.4% *Table 5.—Verbs: Hebrew Inflection, Greek Tense and Mood Heb Infl pf pf pf pf w/c impf impf impf impf w/c impf w/c impv inf/abs inf/cons inf/cons ptc ptc ptc ptc

Gk Tense Gk mood: Ind impf 3 3 aor 10 7 pf 5 5 fut 2 2 fut 4 4 impf 1 1 aor 7 pres 1 1 aor 34 34 pres 8 fut 2 2 pres 1 aor 3 pres 6 4 impf 3 3 aor 1 1 pf 1 1

Subj

3

Opt

Impv Inf

Ptc

1

2

2

2

8 1 3 2

329

deviation factor (discrete tense/mood combinations)1 : 6.70 deviation factor (grouped)2 : 0.85 deviation factor (grouped, without Hebrew verbals)3 : 0.83 *Table 6.—Verbs: Hebrew Stem, Greek Voice Heb Stem qal ni pi hith hi

Gk Voice:

Act 55 1 11 1 5

Mid 2

1st 6

2d

Pass 7 3 2 1

Other4

3d

Other

51

5

1

5

deviation factor: 3.55 Table 7.—Verbs: Person Heb Person 1st 2d 3d

Gk Person:

14

deviation factor: 0.00 Table 8.—Verbs: Number Heb Number Gk Number: Sing sg 79 pl 1

Plur 2 2

Other 4

deviation factor: 0.12 1 Perfect, perfect w/c, imperfect, imperfect w/c are counted as four separate inflections in all

calculations in this table. 2 For full verbs, Greek past tenses (imperfect, aorist, perfect) are grouped together, and subjunctive and optative moods are ignored, since they have nothing to do with time. For verbals, Hebrew infinitives are grouped, as are Greek infinitives, so that the regularily of rendering Hebrew infinitives with Greek infinitives versus rendering Hebrew infinitives with Greek verbs is what is being compared. Participles are similarly grouped, but it should be noted that the “normal” rendering of the Hebrew participle in this chapter is with a full Greek verb. 3 Since Hebrew verbals are often rendered less strictly than full verbs, they are excluded from this

calculation for comparison with the previous deviation factor. 4 Items in the “other” columns in the following tables indicate that a word of one class in Hebrew (verb, noun/adjective, pronoun) was rendered with a word in a different class in Greek. Since these deviations are accounted for in table 5, they are ignored in calculating the deviation factors.

330 Table 9.—Nouns and Adjectives: Gender Heb Gender masc fem

Gk Gender:

Masc Fem 31 8 1 4

Neut 12 1

Other 8 2

Dual

Other 9 1

Dat

Acc

deviation factor: 21.54 *Table 10.—Nouns and Adjectives: Number Heb Number Gk Number: Sing sing 42 plur dual

Plur 7 6 1

deviation factor: 1.78 Table 11.—Nouns and Adjectives: Use vs. Case Heb Use subj subj comp DO chain OP voc

Gk Case:

Nom 36 3

Gen 1

12 8

14

deviation factor (discrete): 6.45 deviation factor (grouped): 0.37 Table 12.—Nouns and Adjectives: Articles Heb Article yes implied5 no

Gk Article:

Yes 6 4 18

No

Other 1

29

9

deviation factor: 11.37 deviation factor (without implied articles): 12.23 5 Implied by Masoretic pointing.

Other 1

1 2

Voc

11 1 7 2

3 4

331

Table 13.—Nouns and Adjectives: Hebrew Definiteness, Greek Articles Heb Definiteness def not def

Gk Article:

Yes 26 2

No 15 14

Other 2 8

Neut 1

Other 1

Plur 1 2

Dual

Other

Gen

Dat

deviation factor: 8.04 Table 14.—Pronouns: Gender Heb Gender masc com

Gk Gender:

Masc Fem 20 3 12

deviation factor: 1.44 Table 15.—Pronouns: Number Heb Number Gk Number: Sing sing 33 plur

1

deviation factor: 0.06 Table 16.—Pronouns: Use vs. Case Heb Use subj subj comp chain OP adj

Gk Case:

Nom 8 1

11 4

3 3

Acc

Voc

1 6

deviation factor (discrete): 3.76 deviation factor (grouped): 0.06 Segmentation Variant 51 52 53 54 55 56

Ref 3:1 3:2 3:5 3:6 3:7 3:8

Hebrew Compound

Greek Rendering

Other

1

332

Segmentation—Continued. 57 3:8 58 3:9 59 3:13 60 61 3:14 62 3:15 63 3:17 64 3:18 65 3:20 66 3:21 Heb compounds: 43 Gk equivalents: 27

62.8% of Heb compounds rendered exactly

*Word Order All variants in this category have been judged to be significant. Heb semantic units: 369 variations: 0 Heb word order followed 100.0% of the time *Quantitative Representation Ref Number of Variations 69 3:1 +1 73 3:3 -1 77 3:5 +1 85 3:6 +1 88 3:9 +1 97 3:15 -1 99 3:16 -1 Heb semantic units: 357 Gk deviations: 7

Variant 1° 4° 6° 2° 1° 98.0% agreement

Peshitta Consistency *Table 17.—Verbs, Nouns, and Adjectives: Lexemes Heb Lexeme

Syr Lexeme

Frequency 2 14 5 2

Common Root

333

Table 17—Continued. Heb Lexeme

Heb words (>1x): 29 deviation factor: 0.22 total Heb (>1x): 129 Heb roots (>1x): 27

Syr Lexeme

Frequency 3 2 3 6 3 5 2 3 16 3 4 2 6 2 1 3 2 2 2 2 2 3 3 12 1 1 2 8 3

Common Root 1 1 1 1

2 3

2 3 3

Syr words: 32

1.10 Syr words/Heb word

primary Syr: 122 Syr roots: 31

94.6% of Heb words by primary rendering 1.15 Syr roots/Heb root

334

Table 18.—Adverbs, Prepositions, and Particles: Lexemes Heb Lexeme

Syr Lexeme

—

—

—

, ,

—

Heb advs (>1x): 12 Syr advs: 22 total Heb (>1x): 85 primary Syr: 63 deviation factor: 3.41

Frequency 5 3 3 1 2 1 6 4 3 1 13 1 6 3 2 7 7 6 1 6 1 3 1.83 Syr advs/Heb adv 74.1% of Heb advs by primary rendering

Table 19.—Conjunctions: Lexemes Heb Lexeme

Syr Lexeme

Heb conjs (>1x): 2 Syr conjs: 4 total Heb (>1x): 64 primary Syr: 61 deviation factor: 0.16

Frequency 54 1 7 2 2.00 Syr conjs/Heb conj 95.3% of Heb conjs by primary rendering

335 Table 20.—Word Classes percentage of verbs/verbals represented by verbs/verbals: 94/96 = 97.9% percentage of verbs represented by verbs: 74/75 = 98.7% percentage of participles represented by participles: 7/12 = 58.3% percentage of infinitives represented by infinitives: 4/7 = 57.1% percentage of nouns/adjs represented by nouns/adjs: 65/70 = 92.9% percentage of pronouns represented by pronouns: 42/42= 100.0% Table 21.—Verbs: Inflection Heb Infl

Syr Infl: Other

pf pf w/c impf impf w/c impv inf/abs inf/cons ptc

Pf

Impf

Impv

Inf

Ptc

Pres CPast Plpf

17 2 37

1 1 9

1 1 8 2

4

1 1

4 5

deviation factor: 1.83 deviation factor (without verbals): 0.42 Table 22.—Verbs: Stem Heb Stem Syr Stem: Pe qal 65 ni 1 pi 6 hith 1 hi 1

Ethpe Pa 2 6 2

Ethpa Aph Ettaph Other 1 1 1 1 1 6 1

deviation factor: 1.33 Table 23.—Verbs: Person Heb Person 1st 2d 3d deviation factor: 0.00

Syr Person:

1st 6

2d

3d

Other

56

1

14

2

336 Table 24.—Verbs: Gender Heb Gender masc fem com

Syr Gender:

Masc Fem 80 1 1

Com

Other 1

6

deviation factor: 0.02 Table 25.—Verbs: Number Heb Number Syr Number: Sing sing 85 plur

Plur

Other 1

3

deviation factor: 0.00 Table 26.—Nouns and Adjectives: Gender Heb Gender masc fem

Syr Gender:

Masc Fem 50 6 6 3

Other 4 1

deviation factor: 1.38 *Table 27.—Nouns and Adjectives: Number Heb Number Syr Number: Sing sing 49 plur dual

Plur 7 6 1

Dual

Other 5

deviation factor: 1.58 Table 28.—Nouns and Adjectives: Use Heb Use Syr Use: Subj Ch DO subj 38 subj comp DO 1 9 chain 11 OP 3 1 voc deviation factor: 1.09

SC OP Voc Adj Adv OR 3

Other 1

2 4 30 2

6

1 1

337 Table 29.—Nouns and Adjectives: Hebrew Articles, Syriac Emphatic State Heb Article yes implied no

Syr Emphatic: Yes 6 4 14

No 1

Other

40

5

deviation factor: 6.06 deviation factor (without implied articles): 6.46 Table 30.—Nouns and Adjectives: Definiteness Heb Definiteness def not def

Syr Definiteness:

Yes 41 8

No 2 14

Other 1 4

Com 1 14

Other

Dual

Other

deviation factor: 2.09 Table 31.—Pronouns: Gender Heb Gender masc com

Syr Gender:

Masc Fem 26

deviation factor: 0.05 Table 32.—Pronouns: Number Heb Number Syr Number: Sing sing 38 plur

Plur 3

deviation factor: 0.00 Table 33.—Pronouns: Use Heb Use Syr Use: subj subj comp chain OP adj deviation factor: 8.15

Subj 8

Ch

DO IO SC OP Voc Adj 1

1

11 1

1 6

2 8 3

Other

338

Segmentation Ref 3:1 3:5 3:6 3:8

Hebrew Compound

38 39 40 41 42 43 3:9 44 3:10 45 3:13 46 3:15 47 3:20 Heb compounds: 47 Syr equivalents: 36

Syriac Rendering -

76.6% of Heb compounds rendered exactly

*Word Order Ref Number of Variations Syriac Variant 50 3:11 1 Heb semantic units: 393 variations: 1 Heb word order followed 99.7% of the time *Quantitative Representation

51 52 53 54 55 57 58 59 60 61 62 63 64 65 66 67

Ref 3:1

3:2 3:3

3:5 3:6

Number of Variations +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 -1 -1 +1 +1

Variant 1° - 1° - 1° - 2° - 1° - 1° - 2° - 2° - 3° - 4° - 5° - 1° - 2°

339

Quantitative Representation—Continued. 68 3:6 -1 69 3:7 +1 70 +1 71 3:8 +1 72 +1 73 +1 74 -1 75 +1 76 +1 77 +1 78 -1 80 3:9 -1 82 3:10 -3 83 +1 84 +1 85 3:11 +1 86 +1 87 3:12 +1 88 -1 89 3:13 -1 90 +1 91 +1 92 -1 93 +2 95 3:14 +1 96 +1 97 +1 98 +1 99 3:15 -1 100 +1 101 +1 102 +1 103 -1 104 +1 105 -1 106 3:17 +2 107 +1

- 1° - 2° - 1°

- 1° 2° - 2° - 1° - 1°

- 2°

- - 1° 1° - 1° - 1° - 2° - 3° 2°

340

Quantitative Representation—Continued. 108 3:17 +1 109 -1 110 +2 111 +1 112 3:18 -1 113 -1 114 +1 115 +1 116 +1 117 +1 118 3:19 +1 119 3:20 +1 120 -1 121 3:21 +1 122 -7 Heb semantic units: 419 Syr deviations: 79

1° 3° 2° -

- 1°

81.1% agreement

Targum Consistency Table 34.—Verbs, Nouns, and Adjectives: Lexemes Heb Lexeme

Aram Lexeme

Frequency 3 14 5 2 2 5 9 4 5 1 1 3 19 3

Common Root

1 1

341

Table 34—Continued. Heb Lexeme

Aram Lexeme

6

Heb words (>1x): 31 deviation factor: 0.04 total Heb (>1x): 139 Heb roots (>1x): 29

Aram words: 34

Frequency 4 3 6 2 1 3 2 2 2 2 1 1 3 2 3 12 1 1 2 8 3

Common Root

2 3

2 3 3

1.10 Aram words/Heb word

primary Aram: 136 97.8% of Heb words by primary rendering Aram roots: 34 1.17 Aram roots/Heb root

6 The full form is listed, since that is what is rendered in T, although the basic Hebrew word is

.

342 *Table 35.—Adverbs, Prepositions, and Particles: Lexemes Heb Lexeme

Aram Lexeme

—

, Heb advs (>1x): 12 total Heb (>1x): 84 deviation factor: 2.02

Aram advs: 21 primary Aram: 71

Frequency 5 2 2 2 3 5 2 3 1 1 15 1 5 3 2 13 1 1 6 7 4 1.75 Aram advs/Heb adv 84.5% of Heb advs by primary rendering

Table 36.—Conjunctions: Lexemes Heb Lexeme

Heb conjs (>1x): 2 total Heb (>1x): 68 deviation factor: 0.00

Aram Lexeme

Aram conjs: 2 primary Aram: 68

Frequency 58 10 1.00 Gk conjs/Heb conj 100.0% of Heb conjs by primary rendering

343

Table 37.—Word Classes percentage of verbs/verbals represented by verbs/verbals: 98/98= 100.0% percentage of verbs represented by verbs: 76/77 = 98.7% percentage of participles represented by participles: 11/12 = 91.7% percentage of infinitives represented by infinitives: 5/7 = 71.4% percentage of nouns/adjs represented by nouns/adjs: 65/68 = 95.6% percentage of pronouns represented by pronouns: 42/42=100.0% *Table 38.—Verbs: Inflection Heb Infl Aram Infl: Pf pf 18 pf w/c impf 3 impf w/c 37 impv inf/abs inf/cons ptc

Impf

Impv

Inf Ptc 1

Other

2 9 8 2 5 1

11

deviation factor: 0.27 deviation factor (without verbals): 0.26 Table 39.—Verbs: Stem Heb Stem Aram Stem: Pe Ithpe Pa qal 64 3 ni 1p7 3 1 pi 1 11 hith 1 hi 4

Ithpa Aph/Shaph Ittaph

Other

1 1 1 6

deviation factor: 1.02 7 The “p” in the peal column in the niphal row indicates a peal passive form (participle). Absence

of the letter “p” in other positions in the columns of the active stems (peal, pael, and aphel/shaphel) indicates that all participial forms in those stems are active.

344 Table 40.—Verbs: Person Heb Person 1st 2d 3d

Aram Person: 1st 6

2d

3d

Other

58

1

14

deviation factor: 0.00 Table 41.—Verbs: Gender Heb Gender masc fem com

Aram Gender: Masc Fem 81 1 1 1

Com

Other 1

6

deviation factor: 0.04 Table 42.—Verbs: Number Heb Number Aram Number: Sing Plur sing 86 1 plur 3

Other 1

deviation factor: 0.02 Table 43.—Nouns and Adjectives: Gender Heb Gender masc fem

Aram Gender: Masc Fem 52 5 5 3

Other 2 1

deviation factor: 1.05 Table 44.—Nouns and Adjectives: Number Heb Number Aram Number: Sing Plur sing 54 4 plur 6 dual 1 deviation factor: 0.50

Dual

Other 2 1

345

Table 45.—Nouns and Adjectives: Use Heb Use Aram Use: Subj Ch DO subj 37 subj comp DO 1 8 chain 11 OP 3 voc

SC

OP Voc Adj Adv OR 1 1

3

1 3 5 1

31 8

deviation factor: 1.47 Table 46.—Nouns and Adjectives: Hebrew Articles, Aramaic Emphatic State Heb Article

Aram Emphatic:

Yes

No

Other

yes

5

1

1

implied

4

no

6

49

2

deviation factor: 1.14 deviation factor (without implied articles): 1.21 Table 47.—Nouns and Adjectives: Definiteness Heb Definiteness def not def

Aram Definiteness:

Yes 40 4

No 1 20

Com

Other

deviation factor: 0.52 Table 48.—Pronouns: Gender Heb Gender masc com deviation factor: 0.00

Other

Aram Gender: Masc Fem 27

14

Other 2 1

1 1

346

Table 49.—Pronouns: Number Heb Number Aram Number: Sing Plur sing 38 plur 3

Dual

Other

deviation factor: 0.00 Table 50.—Pronouns: Use Heb Use Aram Use: Subj Ch subj 8 subj comp chain 12 OP 1 adj

DO IO

SC OP Voc Adj Adv Other 1

1

1 15 3

deviation factor: 0.34 Segmentation Ref Hebrew Compound Aramaic Rendering 42 3:18 43 3:20 Heb compounds: 50 Aram equivalents: 48 96.0% of Heb compounds rendered exactly Word Order There are no deviations from the word order of MT found in T. Heb semantic units: 416 variations: 0 Heb word order followed 100.0% of the time *Quantitative Representation

44 45 47 48 49 50 51 52 53 54

Ref 3:1 3:2 3:3

3:7

Number of Variations +1 +1 +2 +1 +6 +1 +1 +3 +1 +1

Variant - 1° - 1° - 4°

-

347

Quantitative Representation—Continued. 55 3:8 +2 57 3:11 +1 58 3:12 +1 59 3:13 +1 60 3:14 +1 61 3:15 +1 62 +1 63 +1 64 3:18 +1 65 3:19 +1 66 +1 67 +1 68 -2 69 +1 70 3:20 +2 71 3:21 +1 Heb semantic units: 426 Aram deviations: 37

-

-

91.3% agreement

Vulgate Consistency Table 51.—Verbs, Nouns, and Adjectives: Lexemes Heb Lexeme

Lat Lexeme Deus aio dico interrogo respondeo domus filius revelo dico loquor sermo verbum sum

Frequency 3 5 6 1 2 5 2 2 1 4 5 4 3

Common Root

1 1 1 1

348

Table 51—Continued. Heb Lexeme

Lat Lexeme abeo vado cognosco nosco scio Dominus dies adiicio addo abscondo celo omnis quicumque universus loco visio praedico indico puer servus aeternum iniquitas oculus facio consurgo suscito voco video apareo revertor dormio iaceo audio

Frequency 4 1 1 1 1 19 3 2 2 2 1 3 1 2 2 1 1 2 2 2 2 2 2 3 2 1 12 1 1 2 7 1 3

Common Root

2 3

2 2 3 3

349

Table 51—Continued. Heb words (>1x): 29 deviation factor: 2.13 total Heb (>1x): 134 Heb roots (>1x): 27

Lat words: 45

1.55 Lat words/Heb word

primary Lat: 108 Lat roots: 43

80.6% of Heb words by primary rendering 1.59 Lat roots/Heb root

Table 52.—Adverbs, Prepositions, and Particles: Lexemes Heb Lexeme

, , ,

Lat Lexeme — ad adversum super si non ubi qui eo — — in propter iuxta ecce praesto antequam neque necdum hic — ad ut nec non — a ex in usque

Frequency 6 5 1 1 2 1 1 4 1 8 5 6 1 1 5 1 1 1 1 2 10 1 1 1 5 1 3 2 1 3

350

Table 52—Continued. Heb advs (>1x): 12 total Heb (>1x): 82 deviation factor: 5.41

Lat advs: 30 primary Lat: 55

2.50 Lat advs/Heb adv 67.1% of Heb advs by primary rendering

Table 53.—Conjunctions: Lexemes Heb Lexeme

Lat Lexeme ergo -que autem enim et igitur itaque porro enim qui quia quoniam

Heb conjs (>1x): 2 total Heb (>1x): 55 deviation factor: 30.98

Lat conjs: 12 primary Lat: 37

Frequency 3 3 4 1 32 1 1 1 1 2 5 1 6.00 Lat conjs/Heb conj 67.3% of Heb conjs by primary rendering

Table 54.—Word Classes percentage of verbs/verbals represented by verbs/verbals: 95/97 = 97.9% percentage of verbs represented by verbs: 74/76 = 97.4% percentage of participles represented by participles: 0/12 = 0.0% percentage of infinitives represented by infinitives: 3/7 = 42.9% percentage of nouns/adjs represented by nouns/adjs: 57/68 = 83.8% percentage of pronouns represented by pronouns: 36/40 = 90.0% *Table 55.—Verbs: Hebrew Inflection, Latin Tense and Mood Heb Infl pf pf pf impf

Lat Tense Lat Mood: impf 4 pf 12 plpf 3 fut 3

Ind 4 12 3 3

Subj

Opt

Impv Inf

Ptc

351

Table 55—Continued. impf fut pf 1 impf impf 2 impf pf 2 impf plpf 1 impf pres 3 impf w/c impf 5 impf w/c pf 30 impf w/c pres 2 impv pres 8 inf/abs fut 2 inf/cons impf 1 inf/cons pf 1 inf/cons pres 3 ptc fut 1 ptc fut pf 1 ptc impf 4 ptc pres 4

1 1

1 2

1 3 5 30 2 8 2 1 1 3 1 1 3 3

1 1

deviation factor (discrete tense/mood combinations): 6.26 deviation factor (grouped): 1.54 deviation factor (grouped, without verbals): 0.23 Table 56.—Verbs: Hebrew Stem, Latin Voice Heb Stem qal ni pi hith hi

Lat Voice:

Act 62 2 10 1 10

Pass 2 2 1 1

Dep 2

Other

1st 6

2d

3d

Other

56

2

2 2

deviation factor: 0.63 Table 57.—Verbs: Person Heb Person 1st 2d 3d deviation factor: 0.00

Lat Person:

14

352 Table 58.—Verbs: Number Heb Number Lat Number: Sing sing 85 plur

Plur 1 2

Other 1 1

Masc Fem 25 14 2 4

Neut 10 1

deviation factor: 0.02 Table 59.—Nouns and Adjectives: Gender Heb Gender masc fem

Lat Gender:

Other 11 1

deviation factor: 31.64 Table 60.—Nouns and Adjectives: Number Heb Number Lat Number: Sing sing 41 plur dual

Plur 9 5 1

Other 10 2

deviation factor: 2.95 Table 61.—Nouns and Adjectives: Use vs. Case Heb Use subj subj comp DO chain OP voc ptc (adj) verb (ptc) adv

Lat Case:

deviation factor (discrete): 13.58 deviation factor (grouped): 1.02

Nom 37 3

Gen

Dat

Abl

Voc

Other 1

1 2 1

Acc 1

10 4

11 1 14 2

2 16

1 1 6

1 9 1

353

Table 62.—Pronouns: Gender Heb Gender masc com

Lat Gender:

Masc Fem 22 2 12

Neut

Other 1 2

Dual

Other 3

Dat

Acc

5

1 6

deviation factor: 0.22 Table 63.—Pronouns: Number Heb Number Lat Number: Sing sing 34 plur

Plur 2

deviation factor: 0.00 Table 64.—Pronouns: Use vs. Case Heb Use subj subj comp chain OP adj

Lat Case:

Nom 5 1 2

Gen

5

Abl

Voc

3 4 3

2

deviation factor (discrete): 9.91 discrete deviation factor (grouped): 3.44 Segmentation

48 49 50 51 52 53 54 55 56 57 58 59

Ref 3:1 3:2 3:5 3:6 3:7 3:8

3:9 3:10 3:13

Hebrew Compound

Latin Rendering coram videre me me ei tertio me puerum te sicut vocaverat secundo ei eos

Other 3

354

Segmentation—Continued. 60 3:14 61 62 63 3:15 64 3:17 65 66 67 3:18 68 3:20 69 Heb compounds: 49 Lat equivalents: 27

domui victimis et muneribus indicare me tibi tibi ei propheta Domini 55.1% of Heb compounds rendered exactly

Word Order Ref Number of Variations Latin Variant 70 3:17 1 oro te ne Heb semantic units: 372 variations: 1 Heb word order followed 99.7% of the time Quantitative Representation

71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88

Ref 3:2

3:3

3:4 3:5 3:6

3:7 3:8

Number of Variations +1 -1 -1 -1 -1 +1 -1 +2 -1 +1 -1 -1 +1 +1 +1 +1 +1 +1

Variant factum - 2° - 1° erat - 2° qui respondens - 3° qui - 3° - 5° qui te et 3° fuerat et 2° adhuc

355

Quantitative Representation—Continued. 89 3:8 -1 - 2° 90 +1 qui 91 -1 - 3° 92 3:9 -1 93 +1 et 3° 94 -1 95 +1 deinceps 96 -1 - 3° 97 3:11 -1 - 1° 98 3:12 +1 sum 99 3:13 -1 100 +1 essem 101 +1 quod 102 -1 2° 103 +1 indigne 104 -2 105 3:14 -1 - 1° 106 +1 quod 107 +1 eius 108 -1 109 +1 usque 110 3:15 -1 111 3:16 -1 - 3° 112 +2 qui respondens 113 3:17 +1 eum 114 +1 est 1° 115 +1 est 2° 116 +1 te 2° 117 +1 sunt 118 3:18 +1 ille 119 +1 quod 120 +1 est 2° 121 3:20 -1 - 2° 122 +1 esset 123 3:21 +1 fuerat 124 -1 Heb semantic units: 424 Lat deviations: 57 86.6% agreement

APPENDIX 4 ALGORITHM FOR CALCULATING THE DEVIATION FACTOR Diagram of the Algorithm 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

begin tot ← 0; divisor ← 0; for k ← 1 to entries do begin x ← data; y ← data; sum ← 0; for l ← 1 to y do begin z ← data; select on z > x/2 do sum ← sum + [(y-1)2 ⋅ (x-z)2]; on 0 < z ≤ x/2 do sum ← sum + [(y-1)2 ⋅ z2]; on z = 0 do sum ← sum + x 2 endselect end; tot ← tot + sum; divisor ← divisor + x end; deviation ← tot/divisor end.

Explanation of the Algorithm Each line in the algorithm is numbered for the sake of reference. In the explanations below, the line number of the line to be discussed is given first, followed by comments. The algorithm is explained in terms of an analysis of lexical data, but it works the same way with the grammatical categories. The data that the algorithm reads is laid out as follows. Each line of data represents a distinct Hebrew word that appears more than once in the chapter. The first element in each line of data is the number of times that a particular Hebrew word appears; this number is read into the variable x in line 6 of the algorithm (e.g., is rendered 8 times in LXX, so x = 8 [see table 1 in Appendix 2]). The next element in each line of data is the

356

357

number of distinct equivalents in the target language; this number is read into the variable y in line 7 of the algorithm (e.g., is rendered by 2 different Greek words, and , so y = 2). There are y more elements in the line, each of which corresponds to the number of times the distinct equivalents in the target language occur (e.g., occurs 5 times and occurs 3 times, so the whole line referring to reads 8, 2, 5, 3). One terminological matter needs clarification before the algorithm is explained. The algorithm calculates the “amount of deviation”, not the “deviation factor,” for each line of data. The deviation factor for a line of data is equal to the amount of deviation divided by the total number of renderings for the Hebrew word in question (e.g., for , the amount of deviation determined by the algorithm is 18, which, when divided by the number of renderings [8], yields 2.25 for the deviation factor of the line). It is mandatory that amount of deviation be totaled, then divided by the total number of renderings at the end, rather than that the deviation factors of the lines be totaled, then divided by the number of lines of data at the end. If the second approach were used, a word like , with only 2 entries, would be given the same weight as , which has 17 entries. Now it is time to explain the algorithm. Line 2—tot contains a running total of the amount of deviation in each category. Line 3—divisor contains the total number of items in the category, the sum of all the x’s (e.g., there are 132 Hebrew verbs, nouns, and adjectives that are rendered in LXX and are figured in the calculations in Appendix 2, so divisor = 132 at the end of the calculation [see line 4]). Line 4—entries is the number of Hebrew words that occur more than once (e.g., there are 31 Hebrew verbs, nouns, and adjectives that are rendered more than once in LXX, so entries = 31; these 31 words are rendered a total of 132 times in LXX, so divisor will equal 132 at the end of the algorithm). The block of commands surrounded by “begin” and “end” is repeated entries times (i.e., 31 times in this case). Lines 6 and 7—x and y are read from the line of data on each pass through the block of commands. Line 8—sum will hold the amount of deviation for the current line of data. Line 9—since there are y distinct renderings of the Hebrew word (in the case of , 2), the following block will be repeated y times. Line 11—z is the number of times the item in the target language is used to render the Hebrew word (in the case of , z = 5). Lines 12-19—this calculation is the heart of the algorithm. Within the range 0 to x, the distinct rendering in the target language is considered more consistent if z is closer to either end of the range (i.e., 0 or x) than if it is in the middle. The amount of deviation is the same whether z is 2 units from 0 (i.e., z = 2) or 2 units from x (i.e., z = x-2). The amount of deviation also increases with the number of distinct renderings in the target

358

language (e.g., if a Hebrew word is rendered consistently by a single Greek word, the factor y-1 = 0, and the amount of deviation for that line is 0; in the case of , y-1 = 1). Both y-1 and the second factor (the distance from the closer end point of the range; i.e., the smaller of z and x-z) are squared to reflect larger amounts of deviation further from the end points of the range. z = 0 is a special case, since normally z must be between 1 and x. z is 0 only when the target language fails to render any of the occurrences of a Hebrew form by its natural equivalent (e.g., if 4 Hebrew masculine nouns are rendered in Greek by 3 feminines and 1 neuter, a factor of 0 will be included in the line of data, alongside the 3 and the 1). z cannot equal 0 in lexical data, since lexemes in one language have no natural equivalent in another. Line 21—by the time the algorithm gets here, sum contains the total amount of deviation for the line of data (the individual deviation factor for the line of data can be obtained by dividing x into sum). This figure is added to the running total in tot. Line 22—divisor is increased by x to reflect the total number of words rendered. Line 24—the sum of the amounts of deviation in each line, contained in tot, is divided by the number of elements analyzed, contained in divisor, to yield the deviation factor for the category. A brief example of how the algorithm works is now in order, using the data from table 19, P’s rendering of the conjunctions in MT. The data from table 19 is as follows: Heb Lexeme

Syr Lexeme

Frequency 54 1 7 2

The two lines of data that correspond to this table are: DATA: 55, 2, 54, 1 DATA: 9,2,7,2

{55 ’s, 2 different renderings in P, 54 ’s, 1 } {9 ’s, 2 different renderings in P, 7 ’s, 2

’s}

entries is equal to 2 (for the 2 lines of data). The amount of deviation for the first line of data is 12⋅ 12 + 12⋅ 12 = 2. The amount of deviation for the second line is 12⋅ 22 + 12⋅ 22 = 8. The individual deviation factors are 2/55 = 0.04 (first line) and 8/9 = 0.89 (second line). The overall deviation factor is (2+8) ÷ (55+9) = 10/64 = 0.16.

SELECTED BIBLIOGRAPHY Books Aland, Kurt, and Aland, Barbara. The Text of the New Testament. Translated by Erroll F. Rhodes. Grand Rapids: William B. Eerdmans, 1987; Leiden: E. J. Brill, 1987. Alter, Robert. The Art of Biblical Narrative. New York: Basic Books, 1981. Andersen, Francis I., and Forbes, A. Dean. Spelling in the Hebrew Bible: Dahood Memorial Lecture. Biblica et Orientalia, no. 41. Rome: Biblical Institute Press, 1986. Ap-Thomas, D. R. A Primer of Old Testament Text Criticism. 2d ed. Oxford: Basil Blackwell, 1965. Baer, S. Liber Samuelis: Textum Masoreticum Accuratissime Expressit, e Fontibus Masorae Varie Illustravit, Notis Criticis Confirmavit. Leipzig: Bernhard Tauchnitz, 1892. Bange, L. A. A Study in the Use of Vowel-Letters in Alphabetic Consonantal Writing. Munich: UNI-DRUCK, 1971. Bar-Efrat, Shimon. Narrative Art in the Bible. Translated by Dorothea Shefer-Vanson. Journal for the Study of the Old Testament Supplement Series, no. 70. Bible and Literature Series, no. 17. Sheffield: Sheffield Academic Press, Almond Press, 1989. Barr, James. The Typology of Literalism in Ancient Biblical Translations. Mitteilungen des Septuaginta-Unternehmens, no. 15. Göttingen: Vandenhoeck & Ruprecht, 1979. ________. Comparative Philology and the Text of the Old Testament. Oxford: Oxford University Press, 1968; reprint, with additions and corrections, Winona Lake, IN: Eisenbrauns, 1987. ________. The Variable Spellings of the Hebrew Bible. Schweich Lectures of the British Academy, 1986. Oxford: Oxford University Press for the British Academy, 1989. Barthélemy, Dominique. Les devanciers d’Aquila. Supplements to Vetus Testamentum, no. 10. Leiden: E. J. Brill, 1963. ________. Études d’histoire du texte de l’Ancien Testament. Orbis Biblicus et Orientalis, no. 21. Fribourg: Éditions Universitaires, 1978; Göttingen: Vandenhoeck & Ruprecht, 1978. ________. Critique textuelle de l’Ancien Testament. Vol. 1, Josué, Juges, Ruth, Samuel, Rois, Chroniques, Esdras, Néhémie, Esther. Orbis Biblicus et Orientalis, no. 50/1. Fribourg: Éditions Universitaires, 1982; Göttingen: Vandenhoeck & Ruprecht, 1982. Bauer, Hans; Leander, Pontius; and Kahle, Paul. Historische Grammatik der hebräischen Sprache des Alten Testaments. Vol. 1. Tübingen: Max Niemeyer, 1922; reprint, Hildesheim: Georg Olms, 1962. Bergsträsser, Goffhelf. Hebräische Grammatik. Leipzig: F. C. W. Vogel, 1918. ________. Einführung in die semitischen Sprachen. With an appendix, “Zur Syntax der 359 Sprache von Ugarit,” by Carl Brockelmann. Munich: Max Hueber, 1928. Boer, Pieter Arie Hendrik de. Research into the Text of 1 Samuel i-xvi: A Contribution to the Study of the Books of Samuel. Amsterdam: H. J. Paris, 1938.

Boström, Otto Henry. Alternative Readings360 in the Hebrew of the Books of Samuel. Augustana Library Publications, no. 8. Rock Island, IL: Augustana Book Concern, 1918. Brockelmann, Carl. Syrische Grammatik. 6th ed. Porta Linguarum Orientalium, ed. Richard Hartmann, no. 5. Leipzig: Otto Harrassowitz, 1951. ________. Hebräische Syntax. Neukirchen: Kreis Moers, Verlag der Buchhandlung des Erziehungs-Vereins, 1956. ________. Grundriss der vergleichenden Grammatik der semitischen Sprachen. 2 vols. Berlin: Reuter & Reichard; New York: Lemcke & Buechner, 1908-13; reprint, Hildesheim: Georg Olms, 1961. Brueggemann, Walter. First and Second Samuel. Interpretation: A Bible Commentary for Teaching and Preaching, ed. James Luther Mays, Patrick D. Miller, Jr., and Paul J. Achtemeier. Louisville: John Knox Press, 1990. Buck, Carl Darling. Comparative Grammar of Greek and Latin. Chicago: University of Chicago Press, 1933. Budde, Karl. Die Bücher Samuel. Kurzer Hand-Commentar zum Alten Testament, ed. Karl Marti, vol. 8. Tübingen and Leipzig, J. C. B. Mohr (Paul Siebeck), 1902. Cappel, Louis. Commentarii et notae criticae in Vetus Testamentum. Amsterdam: Jacobus Cappellus, 1689. Caspari, Wilhelm. Die Samuelbücher. Kommentar zum Alten Testament, ed. Ernst Sellin, vol. 7. Leipzig: A. Deichertsche Verlagsbuchhandlung Dr. Werner Scholl, 1926. Churgin, Pinkhos. Targum Jonathan to the Prophets. Yale Oriental Series, vol. 14. New Haven: Yale University Press, 1907. Clark, Albert Curtis. The Descent of Manuscripts. Oxford: Clarendon Press, 1918; reprint, Norwich: Fletcher and Son, 1969. Conybeare, F. C., and Stock, St. George. Grammar of Septuagint Greek. Boston: Ginn & Co., 1905; reprint, Peabody, MA: Hendrickson, 1988. Cox, Claude E. Hexaplaric Materials Preserved in the Armenian Versions. Atlanta: Scholars Press, 1986. Cross, Frank Moore, Jr. The Ancient Library of Qumran and Modern Biblical Studies. 2d ed. Garden City, NY: Doubleday & Co., Anchor Books, 1961. ________. Canaanite Myth and Hebrew Epic: Essays in the History of the Religion of Israel. Cambridge and London: Harvard University Press, 1973. Cross, Frank Moore, Jr., and Freedman, David Noel. Early Hebrew Orthography: A Study of the Epigraphic Evidence. American Oriental Series, no. 36. New Haven: American Oriental Society, 1952. ________. Studies in Ancient Yahwistic Poetry. Society of Biblical Literature Dissertation Series, no. 21. Missoula, MT: Scholars Press for the Society of Biblical Literature, 1975. Dahood, Mitchell J.. Psalms III: 101-150. The Anchor Bible, ed. William F. Albright and David Noel Freedman, vol. 17A. Garden City, NY: Doubleday & Company, 1970. Dalman, Gustaf. Grammatik des jüdisch-palästinischen Aramäsich. 2d ed. Leipzig: J. C. Hinrichs, 1905. Davies, G. I. Ancient Hebrew Inscriptions: Corpus and Concordance. Cambridge: Cambridge University Press, 1991. Dearing, Vinton Adams. Manual of Textual Analysis. Berkeley: University of California Press, 1959.

361 Analysis. Berkeley: University of California ________. Principles and Practice of Textual Press, 1974. Debrunner, Albert. Zur Übersetzungstechnik des Septuaginta. Beihefte zur Zeitschrift für die alttestamentliche Wissenschaft, no. 41. Berlin: Walter de Gruyter, 1925. Deist, Ferdinand E. Towards the Text of the Old Testament. Translated by W. K. Winckler. Pretoria: N. G. Kerkboekhandel Transvaal, 1978. ________. Witnesses to the Old Testament. The Literature of the Old Testament, vol. 5. Pretoria: N. G. Kerkboekhandel, 1988. Dhorme, P. Paul. Les livres de Samuel. Études bibliques. Paris: Librairie Victor Lecoffre, 1910. Dietrich, Walter. David, Saul und die Propheten: Das Verhältnis von Religion und Politik nach den prophetischen Überlieferungen vom frühesten Königtum in Israel. Beiträge zur Wissenschaft vom Alten und Neuen Testament, no. 122. Stuttgart: W. Kohlhammer, 1987. Díez Macho, Alejandro. El Targum: Introducción a las traducciones aramaicas de la Biblia. Textos y estudios “Cardinal Cisneros.” Madrid: Consejo Superior de Investigaciones Científicas, 1982. Dirksen, P. B., and Mulder, M. J., eds. The Peshi a: Its Early Text and History. Papers Read at the Peshi a Symposium Held at Leiden 30-31 August 1985. Monographs of the Peshit@ta Institute, Leiden, no. 4. Leiden: E. J. Brill, 1988. Dorival, Gilles; Harl, Marguerite; and Munnich, Olivier. La Bible grecque des Septante: De judaïsme hellénistique au christianisme ancien. Initiations au christianisme ancien. Paris: Éditions du Cerf and Éditions du C.N.R.S., 1988. Driver, S. R. A Treatise on the Use of the Tenses in Hebrew and Some Other Syntactical Questions. 3d ed. Oxford: Clarendon Press, 1892. ________. Notes on the Hebrew Text and the Topography of the Books of Samuel. 2d ed. Oxford: Clarendon Press, 1913; reprint, Oxford: Oxford University Press, 1960. Eslinger, Lyle M. Kingship of God in Crisis: A Close Reading of 1 Samuel 1-12. Bible and Literature Series, no. 10. Sheffield: JSOT Press, 1985. Fernández Marcos, Natalio. Introducción a las versiones griegas de la biblia. Textos y estudios “Cardinal Cisneros”, no. 23. Madrid: Instituto “Arias Montano” Consejo Superior de Investigaciones Científicas, 1979. ________, ed. La Septuaginta en la investigación contemporánea (V congreso de la IOSCS). Textos y estudios “Cardinal Cisneros” de la Biblia Políglota Matritense, no. 34. Madrid: Instituto “Arias Montano” Consejo Superior de Investigaciones Científicas, 1985. Fernández Marcos, Natalio, and Busto Saiz, José Ramón. Theodoreti Cyrensis Quaestiones in Reges et Paralipomena. Textos y estudios “Cardinal Cisneros”, no. 32. Madrid: Instituto “Arias Montano” Consejo Superior de Investigaciones Científicas, 1984. Fischer, Johann. Das Alphabet der LXX-Vorlage im Pentateuch: Eine textkritische Studie. Alttestamentliche Abhandlungen, vol. 10, no. 2. Münster: Aschendorffschen Verlagsbuchhandlung, 1924. Fishbane, Michael A. Biblical Interpretation in Ancient Israel. Oxford: Clarendon Press, 1985; New York: Oxford University Press, 1985. Fokkelman, J. P. Narrative Art and Poetry in the Books of Samuel: A Full Interpretation Based on Stylistic and Structural Analyses. Assen, The Netherlands: Van Gorcum, 1981.

Frankel, Z. Vorstudien zur der Septuaginta.362 Leipzig: Fr. Chr. Wilh. Vogel, 1841. Freedman, David Noel; Forbes, A. Dean; and Andersen, Francis I., eds. Studies in Hebrew and Aramaic Orthography. Biblical and Judaic Studies from the University of California, San Diego, vol. 2. Winona Lake, IN: Eisenbrauns, 1992. Freedman, David Noel; Mathews, K. A.; and Hanson, R. S. The Paleo-Hebrew Leviticus Scroll (11QpaleoLev). Winona Lake, IN: Eisenbrauns for the American Schools of Oriental Research, 1985. Garsiel, Moshe. The First Book of Samuel: A Literary Study of Comparative Structures, Analogies and Parallels. Jerusalem: Revivim, Rubin Mass, 1990. Gnuse, Robert Karl. The Dream Theophany of Samuel: Its Structure in Relation to Ancient Near Eastern Dreams and Its Theological Significance. Lanham, MD: University Press of America, 1984. Goodwin, Donald Watson. Text-Restoration Methods in Contemporary U.S.A. Biblical Scholarship. Pubblicazioni del Seminario di Semitistica, Ricerche, no. 5. Naples: Istituto Orientale di Napoli, 1969. Gordon, Cyrus H. Ugaritic Textbook. Analecta Orientalia, no. 38. Rome: Pontifical Biblical Institute, 1965. Grandgent, Charles H. An Introduction to Vulgar Latin. Boston: D. C. Heath, 1907. Hertzberg, Hans Wilhelm. Die Samuelbücher. 2d ed. Das Alte Testament Deutsch, ed. Volkmar Herntrich and Artur Weiser, vol. 10. Göttingen: Vandenhoeck & Ruprecht, 1960. Hofmann, J. B., and Szantyr, Anton. Lateinische Syntax und Stilistik. Handbuch der Altertumswissenschaft, division 2, part 2, vol. 2. Munich: C. H. Beck, 1965. Hurtado, Larry D. Text-Critical Methodology and the Pre-Caesarean Text: Codex W in the Gospel of Mark. Studies and Documents, no. 43. Grand Rapids: William B. Eerdmans, 1981. Jellicoe, Sidney. The Septuagint and Modern Study. Oxford: Oxford University Press, 1968; reprint, Winona Lake, IN: Eisenbrauns, 1989. Johnson, Bo. Die hexaplarische Rezension des 1. Samuelbuches der Septuaginta. Studia Theologica Lundensia, no. 22. Lund: CWK Gleerup, 1963. ________. Die armenische Bibelübersetzung als hexaplarischer Zeuge im 1. Samuelbuch. Coniectanea Biblica, Old Testament Series, no. 2. Lund: CWK Gleerup, 1968. Kahle, Paul E. The Cairo Geniza. 2d ed. Oxford: Basil Blackwell, 1959. Kalt, Edmund, and Lauck, Willibald, eds. Herders Bibelkommentar: Die Heilige Schrift für das Leben erklärt. Vol. 3/1, Die Samuelbücher, by Peter Ketter. Fribourg: Herder & Co., 1940. Kautzsch, E., and Cowley, A. E. Gesenius’ Hebrew Grammar. 2d ed. Oxford: Clarendon Press, 1910. Klein, Ralph W. Textual Criticism of the Old Testament: The Septuagint after Qumran. Guides to Biblical Scholarship, Old Testament Series. Philadelphia: Fortress Press, 1974. ________. 1 Samuel. Word Biblical Commentary, vol. 10, gen. eds. David A. Hubbard and Glenn W. Barker, Old Testament editor John D. W. Watts. Waco: Word Books, 1983. Klostermann, Erich. Die Bücher Samuelis und der Könige. Kurzgefaßter Kommentar zu den heiligen Schriften Alten und Neuen Testamentes sowie zu den Apokryphen, ed. Hermann Strack and Otto Böckler, vol. 3. Nördlingen: C. H. Beck, 1887.

Koster, M. D. The Peshi a of Exodus: 363 The Development of Its Text in the Course of Fifteen Centuries. Studia Semitica Neerlandica, no. 19. Assen: Van Gorcum, 1977. Kraft, Robert A., and Tov, Emanuel. A Computerized Data Base for Septuagint Studies: The Parallel Aligned Text of the Greek and Hebrew Bible. Computer Assisted Tools for Septuagint Studies (CATSS), vol. 1, Ruth, by John R. Abercrombie. Journal of Northwest Semitic Languages Supplementary Series, no. 1. Stellenbosch: Journal of Northwest Semitic Languages, 1986. Le Déaut, R. Introduction à la littérature targumique. Part 1. Rome: Pontifical Biblical Institute, 1966. Levine, Etan. The Aramaic Version of the Bible: Contents and Context. Beihefte zur Zeitschrift für die alttestamentliche Wissenschaft, no. 174. Berlin: Walter de Gruyter, 1988. Licht, Jacob. Storytelling in the Bible. Jerusalem: Magnes Press for the Hebrew University, 1978. Lundström, Sven. Übersetzungstechnische Untersuchungen auf dem Gebiete der Christlichen Latinität. Lunds universitets årsskrift, no. 51/3. Lund: C. W. K. Gleerup, 1955. Lust, Johan, ed. Ezekiel and His Book: Textual and Literary Criticism and Their Interrelation. Bibliotheca Ephemeridum Theologicarum Lovaniensium, no. 74. Leuven: University Press, 1986. Maas, Paul. Textual Criticism. Translated by Barbara Flower. Oxford: Clarendon Press, 1958. McCarter, P. Kyle, Jr. I Samuel. The Anchor Bible, ed. William F. Albright and David Noel Freedman, vol. 8. Garden City, NY: Doubleday & Company, 1980. ________. Textual Criticism: Recovering the Text of the Hebrew Bible. Guides to Biblical Scholarship, Old Testament Series. Philadelphia: Fortress Press, 1986. McCarthy, Carmel. The Tiqqune Sopherim and Other Theological Corrections in the Masoretic Text of the Old Testament. Orbis Biblicus et Orientalis, no. 36. Fribourg: Universitätsverlag, 1981; Göttingen: Vandenhoeck & Ruprecht, 1981. Metzger, Bruce M. The Text of the New Testament: Its Transmission, Corruption, and Restoration. 2d ed. New York: Oxford University Press, 1968. ________. A Textual Commentary on the Greek New Testament. Corrected ed. London: United Bible Societies, 1975. ________. The Early Versions of the New Testament: Their Origin, Transmission and Limitations. Oxford: Clarendon Press, 1977. Miles, John Russiano. Retroversion and Text Criticism: The Predictability of Syntax in an Ancient Translation from Greek to Ethiopic. Septuagint and Cognate Studies Series, no. 17. Chico, CA: Scholars Press, 1985. Miscall, Peter D. The Workings of Old Testament Narrative. The Society of Biblical Literature Semeia Studies. Philadelphia: Fortress Press, 1983; Chico, CA: Scholars Press, 1983. ________. 1 Samuel: A Literary Reading. Indiana Studies in Biblical Literature. Bloomington: Indiana University Press, 1986. Mulder, M. J.; Oosterhoff, B. J.; Reiling, R.; Ridderbos, H. R.; and van Unnik, W. C., eds. The World of the Bible. Translated by Sierd Woudstra. Bible Handbook, vol. 1. Edited by A. S. van der Woude. Grand Rapids: William B. Eerdmans, 1986.

364 Muraoka, Takamitsu. Emphatic Words and Structures in Biblical Hebrew. Jerusalem: Magnes Press, 1985; Leiden: E. J. Brill, 1985. Nelson, Richard D. The Double Redaction of the Deuteronomistic History. Journal for the Study of the Old Testament Supplement Series, no. 18. Sheffield: JSOT Press, 1981. Nowack, Wilhelm. Die Bedeutung des Hieronymus für die alttestamentliche Textkritik. Göttingen: Vandenhoeck & Ruprecht, 1875. O’Connor, M. Hebrew Verse Structure. Winona Lake, IN: Eisenbrauns, 1980. Olofsson, Staffan. The LXX Version: A Guide to the Translation Technique of the Septuagint. Coniectanea Biblica, Old Testament Series, no. 30. Stockholm: Almqvist & Wiksell, 1990. Pfeiffer, Robert Henry. Introduction to the Old Testament. New York: Harper & Bros., 1948. Pisano, Stephen. Additions or Omissions in the Books of Samuel: The Significant Pluses and Minuses in the Massoretic, LXX and Qumran Texts. Orbis Biblicus et Orientalis, no. 57. Fribourg: Universitätsverlag, 1984; Göttingen: Vandenhoeck & Ruprecht, 1984. Plater, W. E., and White, H. J. A Grammar of the Vulgate. Oxford: Clarendon Press, 1926. Polzin, Robert. Samuel and the Deuteronomist: A Literary Study of the Deuteronomic History. San Francisco: Harper & Row, 1989. Quentin, Henri. Mémorie sur l’éstablissement du texte de la Vulgate. Vol. 6, pt. 1, Octateuque. Collectanea Biblica Latina, no. 6. Rome: Desclee et Cie., 1922; Paris: J. Gabalda, 1922. Rad, Gerhard von. Old Testament Theology. 2 vols. Translated by D. M. G. Stalker. New York: Harper & Row, 1962-65. Rahlfs, Alfred. Verzeichnis der griechischen Handschriften des Alten Testaments, für das Septuaginta-Unternehmens. Göttingen: Nachrichten von der königlichen Gesellschaft der Wissenschaften zu Göttingen, 1928. ________. Septuaginta-Studien. 2d ed. Vols. 1-3. Göttingen: Vandenhoeck & Ruprecht, 1965. Reider, Joseph, and Turner, Nigel. Index to Aquila: Greek-Hebrew, Hebrew-Greek, Latin-Hebrew: With the Syriac and Armenian Evidence. Supplements to Vetus Testamentum, no. 12. Leiden: E. J. Brill, 1966. Roberts, Bleddyn J. The Old Testament Text and Versions: The Hebrew Text in Transmission and the History of the Ancient Versions. Cardiff: University of Wales, 1951. Robertson, A. T. A Grammar of the Greek New Testament in the Light of Historical Research. Nashville: Broadman Press, 1934. Robertson, David. The Old Testament and the Literary Critic. Guides to Biblical Scholarship, Old Testament Series. Philadelphia: Fortress Press, 1977. Robinson, Theodore H. Paradigms and Exercises in Syriac Grammar. 4th ed. Edited by L. H. Brockington. Oxford: Clarendon Press, 1962. Sanders, James. Canon and Community: A Guide to Canonical Criticism. Guides to Biblical Scholarship, Old Testament Series. Philadelphia: Fortress Press, 1984. ________. From Sacred Story to Sacred Text. Philadelphia: Fortress Press, 1987. Schulz, Alfons. Die Bücher Samuel. Exegetisches Handbuch zum Alten Testament, ed. Johannes Nikel, vol. 8. Münster: Aschendorff, 1919.

365 Schwartz, Emanuel. Die syrische Uebersetzung des ersten Buches Samuelis und ihr Verhältniss zu MT., LXX und Trg. Berlin: H. Itzkowski, 1896. Seeligmann, Isac Leo. The Septuagint Version of Isaiah: A Discussion of Its Problems. Leiden: E. J. Brill, 1948. Segal, J. B. Edessa: The Blessed City. Oxford: Clarendon Press, 1970. Seow, C. L. A Grammar for Biblical Hebrew. Nashville: Abingdon Press, 1987. Shenkel, James Donald. Chronology and Recensional Development in the Greek Text of Kings. Harvard Semitic Monographs, no. 1. Cambridge: Harvard University Press, 1968. Smith, Henry Preserved. A Critical and Exegetical Commentary on the Books of Samuel. Edinburgh: T. & T. Clark, 1899. Smyth, Herbert Weir. Greek Grammar. Edited by Gordon M. Messing. Cambridge: Harvard University Press, 1956. Soisalon-Soininen, Ilmari. Die Infinitive in der Septuaginta. Annales academiæ scientiarum fennicæ, no. 132, 1. Helsinki: Suomalainen Tiedeakatemia, 1965. Sollamo, Raija. Renderings of Hebrew Semiprepositions in the Septuagint. Annales academiæ scientiarum fennicæ, Dissertationes humanarum litterarum, no. 19. Helsinki: Suomalainen Tiedeakatemia, 1979. Sperber, Alexander. A Historical Grammar of Biblical Hebrew: A Presentation of Problems with Suggestions to Their Solution. Leiden: E. J. Brill, 1966. Stoebe, Hans Joachim. Das erste Buch Samuelis. Kommentar zum Alten Testament, vol. 8, pt. 1. Gütersloh: Gütersloher Verlagshaus Gerd Mohn, 1973. Stramare, Tarcisio, ed. La bibbia “Vulgata” dalle origini ai nostri giorni. Collectanea Biblica Latina, no. 16. Rome: Abbazia San Girolamo, Libreria Vaticana, 1987. Streeter, B. H. The Four Gospels: A Study of Origins. London: Macmillan & Co., 1924. Stummer, Friedrich. Einführung in die lateinische Bibel: Ein Handbuch für Vorlesungen und Selbstunterricht. Paderborn, Germany: F. Schoning, 1928. ________. Hauptprobleme der Erforschung der alttestamentlichen Vulgata. Beihefte zur Zeitschrift für die alttestamentliche Wissenschaft, no. 66. Berlin: Walter de Gruyter, 1936. Swete, Henry Barclay. An Introduction to the Old Testament in Greek. With an Appendix by Henry St. John Thackeray. Revised by Richard Rusden Ottley. Cambridge: Cambridge University Press, 1914; reprint, Peabody, MA: Hendrickson, 1989. Thackeray, Henry St. John. A Grammar of the Old Testament in Greek. Vol. 1, Introduction, Orthography and Accidence. Cambridge: Cambridge University Press, 1909; reprint, Hildesheim: Georg Olms, 1978. ________. The Septuagint and Jewish Worship. 2d ed. London: Oxford University Press, 1923. Thenius, Otto. Die Bücher Samuels. 2d ed. Kurzgefasstes Exegetisches Handbuch zum Alten Testament, no. 4. Leipzig: S. Hirzel, 1864. Tigay, Jeffrey H., ed. Empirical Models for Biblical Criticism. Philadelphia: University of Pennsylvania Press, 1985. Tov, Emanuel. The Text-Critical Use of the Septuagint in Biblical Research. Jerusalem Biblical Studies, no. 3. Jerusalem: Simor, 1981.

366 ________. A Computerized Data Base for Septuagint Studies: The Parallel Aligned Text of the Greek and Hebrew Bible. Computer Assisted Tools for Septuagint Studies (CATSS), vol. 2. Journal of Northwest Semitic Languages Supplementary Series, no. 1. Stellenbosch: Journal of Northwest Semitic Languages, 1986. ________. Textual Criticism of the Hebrew Bible. Philadelphia: Fortress Press, 1992. ________, ed. The Hebrew and Greek Texts of Samuel. 1980 Proceedings of the IOSCSVienna. Jerusalem: Acedemon, 1980. Trebolle Barrera, Julio C. Jehú y Joás: Texto y composición literaria de 2 Reyes 9-11. Institución San Gerónimo, no. 17. Valencia: Edilva, 1984. Ulrich, Eugene Charles, Jr. The Qumran Text of Samuel and Josephus. Harvard Semitic Monographs, no. 19. Missoula, MT: Scholars Press, 1978. Verheij, A. J. C. Verbs and Numbers: A Study of the Frequencies of the Hebrew Verbal Tense Forms in the Books of Samuel, Kings, and Chronicles. Studia Semitica Neerlandica. Assen: Van Gorcum, 1990. Walters, Peter [Peter Katz]. The Text of the Septuagint: Its Corruptions and Their Emendation. Edited by David W. Gooding. Cambridge: Cambridge University Press, 1973. Waltke, Bruce K., and O’Connor, M. An Introduction to Biblical Hebrew Syntax. Winona Lake, IN: Eisenbrauns, 1990. Wellhausen, Julius. Der Text der Bücher Samuelis. Göttingen: Vandenhoeck & Ruprecht, 1871. Westcott, Brooke Foss, and Hort, Fenton John Anthony. The New Testament in the Original Greek. New York: Harper & Brothers, 1882. Wilson, Robert Dick. A Scientific Investigation of the Old Testament. Philadelphia: The Sunday School Times Company, 1926. Würthwein, Ernst. The Text of the Old Testament: An Introduction to the Biblia Hebraica. Translated by Erroll F. Rhodes. Grand Rapids: William B. Eerdmans, 1979. Wutz, Franz. Die Transkriptionen von der Septuaginta bis zu Hieronymus. Part 1. Beiträge zur Wissenschaft vom Alten Testament, n.s., no. 9, Texte und Untersuchungen zur vormasoretuschen Grammatik des Hebräischen, no. 2. Berlin: W. Kohlhammer, 1925. Zevit, Ziony. Matres Lectionis in Ancient Hebrew Epigraphs, American Schools of Oriental Research Monograph Series, no. 2. Cambridge, MA: American Schools of Oriental Research, 1980. Zimmerli, Walther. Ezekiel. Hermeneia—A Critical and Historical Commentary on the Bible. 2 vols. Translated by James D. Martin. Edited by Paul D. Hanson and Leonard J. Greenspoon. Philadelphia: Fortress Press, 1979-83. Collections Aejmelaeus, Anneli. “The Significance of Clause Connectors in the Syntactical and Translation-Technical Study of the Septuagint.” In VI Congress of the International Organization for Septuagint and Cognate Studies: Jerusalem 1986, ed. Claude E. Cox, 361-80. Society of Biblical Literature Septuagint and Cognate Studies, no. 23. Atlanta: Scholars Press, 1987.

367 Greek.” In Studien zur Septuaginta—Robert ________. “OTI recitativum in Septuagintal Hanhart zu Ehren aus Anlaß seines 65. Geburtstages, ed. Detlef Fraenkel, Udo Quast, and John William Wevers, 74-82. Abhandlungen der Akademie der Wissenschaften in Göttingen, Philologisch-historische Klasse, 3d series, no. 190, Mitteilungen des Septuaginta-Unternehmens, no. 20. Göttingen: Vandenhoeck & Ruprecht, 1990. Albrektson, Bertil. “Reflections on the Emergence of a Standard Text of the Hebrew Bible.” In Congress Volume: Göttingen 1977, ed. J. A. Emerton, William L. Holladay, A. Lemaire, Roland E. Murphy, E. Nielsen, R. Smend, and J. A. Soggin, 49-65. Supplements to Vetus Testamentum, no. 29. Leiden: E. J. Brill, 1977. ________. “Difficilior Lectio Probabilior: A Rule of Textual Criticism and Its Use in Old Testament Studies.” In Remembering All the Way: A Collection of Old Testament Studies Published on the Occasion of the Fortieth Anniversary of the Oudtestamentisch Werkgezelschap in Nederland, ed. A. S. van der Woude, 5-18. Oudtestamentische Studiën, no. 21. Leiden: E. J. Brill, 1981. Albright, William F. “The Psalm of Habakkuk.” In Studies in Old Testament Prophecy Presented to Professor Theodore H. Robinson, ed. H. H. Rowley, 1-18. Edinburgh: T. & T. Clark, 1950. Barr, James. “Vocalization and the Analysis of Hebrew among the Ancient Translators.” In Hebräische Wortforschung: Festschrift zum 80. Geburtstag von Walter Baumgartner, ed. G. W. Anderson, P. A. H. de Boer, G. R. Castellino, Henri Cazelles, E. Hammershaimb, H. G. May, W. Zimmerli, 1-11. Supplements to Vetus Testamentum, no. 16. Leiden: E. J. Brill, 1967. ________. “Translators’ Handling of Verb Tense in Semantically Ambiguous Contexts.” In VI Congress of the International Organization for Septuagint and Cognate Studies: Jerusalem 1986, ed. Claude E. Cox, 381-403. Society of Biblical Literature Septuagint and Cognate Studies, no. 23. Atlanta: Scholars Press, 1987. ________. “‘Guessing’ in the Septuagint.” In Studien zur Septuaginta—Robert Hanhart zu Ehren aus Anlaß seines 65. Geburtstages, ed. Detlef Fraenkel, Udo Quast, and John William Wevers, 19-34. Abhandlungen der Akademie der Wissenschaften in Göttingen, Philologisch-historische Klasse, 3d series, no. 190, Mitteilungen des Septuaginta-Unternehmens, no. 20. Göttingen: Vandenhoeck & Ruprecht, 1990. Barthélemy, Dominique. “Les problèmes textuels de 2 Sam 11,2 - 1 Rois 2,11 reconsidérés à la lumière de certaines critiques des ‘Devanciers d’Aquila.’” In 1972 Proceedings of the International Organization for Septuagint and Cognate Studies and the Society of Biblical Literature Pseudepigrapha Seminar. Los Angeles: Society of Biblical Literature, 1972. ________. “Les Tiqquné Sopherim et la critique textuelle de l’Ancien Testament.” Chap. in Études d’histoire du texte de l’Ancien Testament. Orbis Biblicus et Orientalis, no. 21. Fribourg: Éditions Universitaires, 1978; Göttingen: Vandenhoeck & Ruprecht, 1978. ________. “La qualité du Texte Massorétique de Samuel.” In The Hebrew and Greek Texts of Samuel, ed. Emanuel Tov, 1-44. 1980 Proceedings of the IOSCS-Vienna. Jerusalem: Acedemon, 1980. ________. “Trois niveaux d’analyse (a propos de David et Goliath).” In The Story of David and Goliath, by Dominique Barthélemy, David W. Gooding, Johan Lust, and Emanuel Tov, 47-54. Orbis Biblicus et Orientalis, no. 73. Fribourg: Éditions Universitaires, 1986; Göttingen: Vandenhoeck & Ruprecht, 1986. Bickerman, Elias. “Some Notes on the Transmission of the Septuagint.” Chap. in Studies in Jewish and Christian History. Part 1. Arbeiten zur Geschichte des Antiken Judentums und des Urchristentums, vol. 9. Leiden: E. J. Brill, 1976.

________. “The Septuagint as a Translation.” 368Chap. in Studies in Jewish and Christian History. Part 1. Arbeiten zur Geschichte des Antiken Judentums und des Urchristentums, vol. 9. Leiden: E. J. Brill, 1976. Brock, Sebastian P. “The Phenomenon of the Septuagint.” In The Witness of Tradition: Papers Read at the Joint British-Dutch Old Testament Conference Held at Woudschoten, 1970, ed. A. S. van der Woude, 11-36. Oudtestamentische Studiën, no. 17. Leiden: E. J. Brill, 1972. ________. “The Phenomenon of Biblical Translation in Antiquity.” In Studies in the Septuagint: Origins, Recensions, and Interpretations, ed. Sidney Jellicoe, 541-71. Library of Biblical Studies, ed. Harry M. Orlinsky. New York: Ktav, 1974. Cook, Johann. “The Composition of the Peshit@ta Version of the Old Testament (Pentateuch).” In The Peshi a: Its Early Text and History. Papers Read at the Peshi a Symposium Held at Leiden 30-31 August 1985, ed. P. B. Dirksen and M. J. Mulder, 147-68. Monographs of the Peshit@ta Institute, Leiden, no. 4. Leiden: E. J. Brill, 1988. ________. “The Plurality of Old Testament Texts and Exegetical Methodology.” In Paradigms and Progression in Theology, ed. J. Mouton, A. G. van Aarde, and W. S. Vorster, 362-77. Pretoria: The Human Sciences Research Council, 1988. Cox, Claude. “The Use of the Armenian Version for the Textual Criticism of the Septuagint.” In La Septuaginta en la investigación contemporánea (V congreso de la IOSCS), ed. Natalio Fernández Marcos, 25-35. Textos y estudios “Cardinal Cisneros” de la Biblia Políglota Matritense, no. 34, Madrid: Instituto “Arias Montano” Consejo Superior de Investigaciones Científicas, 1985. Cross, Frank Moore. “The Development of the Jewish Scripts.” In The Bible and the Ancient Near East: Essays in Honor of William Foxwell Albright, ed. George Ernest Wright, 170-264. Garden City, NY: Doubleday & Co., 1961. ________. “The Evolution of a Theory of Local Texts.” In Qumran and the History of the Biblical Text, ed. Frank Moore Cross and Shemaryahu Talmon, pp. 306-20. Cambridge: Harvard University Press, 1975. ________. “Problems of Method in the Textual Study of the Hebrew Bible.” In The Critical Study of Sacred Texts, ed. Wendy Doniger O’Flaherty, 31-54. Berkeley: Berkeley Religious Studies Series, 1979. Culley, Robert C. “Exploring New Directions.” In The Hebrew Bible and Its Modern Interpreters, ed. Douglas A. Knight and Gene M. Tucker, 167-200. Philadelphia: Fortress Press, 1985; Decatur, GA: Scholars Press, 1985. Dahood, Mitchell. “Ebla, Ugarit, and the Bible.” Afterword to The Archives of Ebla: An Empire Inscribed in Clay, by Giovanni Pettinato. Garden City, NY: Doubleday & Co., 1981. Debrunner, A. “Zur Übersetzungstechnik der Septuaginta.” In Vom Alten Testament: Festschrift für Karl Marti. Beihefte zur Zeitschrift für die alttestamentliche Wissenschaft, no. 41. Berlin: Walter de Gruyter, 1925. Fernández Marcos, Natalio. “The Lucianic Text in the Book of Kingdoms.” In De Septuaginta: Studies in Honor of John William Wevers on His Sixty-Fifth Birthday, ed. A. Pietersma and C. Cox, 161-74. Mississauga, ON: Benben, 1984.

369 ________. “Some Reflections on the Antiochian Text of the Septuagint.” In Studien zur Septuaginta—Robert Hanhart zu Ehren aus Anlaß seines 65. Geburtstages, ed. Detlef Fraenkel, Udo Quast, and John William Wevers, 219-29. Abhandlungen der Akademie der Wissenschaften in Göttingen, Philologisch-historische Klasse, 3d series, no. 190, Mitteilungen des Septuaginta-Unternehmens, no. 20. Göttingen: Vandenhoeck & Ruprecht, 1990. Fishbane, Michael. “I Samuel 3: Historical Narrative and Narrative Poetics.” In Literary Interpretations of Biblical Narrative, vol. 2, ed. Kenneth R. R. Gros Louis and James S. Ackerman, 191-203. Nashville: Abingdon Press, 1982. Fitzmyer, Joseph A. “The Phases of the Aramaic Language.” Chap. in A Wandering Aramean: Collected Aramaic Essays. Society of Biblical Literature Monograph Series, no. 25. Chico, CA: Scholars Press, 1979. Gooding, David W. “An Approach to the Literary and Textual Problems in the DavidGoliath Story.” In The Story of David and Goliath, by Dominique Barthélemy, David W. Gooding, Johan Lust, and Emanuel Tov, 55-86. Orbis Biblicus et Orientalis, no. 73. Fribourg: Éditions Universitaires, 1986; Göttingen: Vandenhoeck & Ruprecht, 1986. Housman, A. E. “The Application of Thought to Textual Criticism.” Chap. in Selected Prose, ed. J. Carter. Cambridge, Cambridge University Press, 1961. Kilpatrick, G. D. “Atticism and the Text of the Greek New Testament.” Chap. in The Principles and Practice of New Testament Textual Criticism: Collected Essays of G. D. Kilpatrick, ed. J. K. Elliott. Leuven: University Press, 1990. ________. “Eclecticism and Atticism.” Chap. in The Principles and Practice of New Testament Textual Criticism: Collected Essays of G. D. Kilpatrick, ed. J. K. Elliott. Leuven: University Press, 1990. ________. “Literary Fashion and the Transmission of Texts in the Graeco-Roman World.” Chap. in The Principles and Practice of New Testament Textual Criticism: Collected Essays of G. D. Kilpatrick, ed. J. K. Elliott. Leuven: University Press, 1990. Kittel, Rudolf. “Das erste Buch Samuel.” In Die Heilige Schrift des Alten Testaments, ed. E. Kautzsch, 378-420. 3d ed. Vol. 1. Tübingen: J. C. B. Mohr (Paul Siebeck), 1909. Koster, M. D. “Which Came First: The Chicken or the Egg? The Development of the Text of the Peshit@ta of Genesis and Exodus in the Light of Recent Studies.” In The Peshi a: Its Early Text and History. Papers Read at the Peshi a Symposium Held at Leiden 30-31 August 1985, ed. P. B. Dirksen and M. J. Mulder, 99-126. Monographs of the Peshit@ta Institute, Leiden, no. 4. Leiden: E. J. Brill, 1988. Lust, Johan. “Second Thoughts on David and Goliath.” In The Story of David and Goliath, by Dominique Barthélemy, David W. Gooding, Johan Lust, and Emanuel Tov, 87-91. Orbis Biblicus et Orientalis, no. 73. Fribourg: Éditions Universitaires, 1986; Göttingen: Vandenhoeck & Ruprecht, 1986. Marquis, Galen. “Consistency of Lexical Equivalents as a Criterion for the Evaluation of Translation Technique as Exemplified in the LXX of Ezekiel.” In VI Congress of the International Organization for Septuagint and Cognate Studies: Jerusalem 1986, ed. Claude E. Cox, 405-24. Society of Biblical Literature Septuagint and Cognate Studies, no. 23. Atlanta: Scholars Press, 1987. Metzger, Bruce Manning. “The Lucianic Recension of the Greek Bible.” Chap. in Chapters in the History of New Testament Textual Criticism. New Testament Tools and Studies, no. 4. Grand Rapids: William B. Eerdmans, 1963.

370 Criticism.” In La Septuaginta en la Mulder, M. J. “The Use of the Peshit@ta in Textual investigación contemporánea (V congreso de la IOSCS), ed. Natalio Fernández Marcos, 37-53. Textos y estudios “Cardinal Cisneros” de la Biblia Políglota Matritense, no. 34. Madrid: Instituto “Arias Montano” Consejo Superior de Investigaciones Científicas, 1985. Muraoka, Takamitsu. “The Greek Texts of Samuel-Kings: Incomplete Translations or Recensional Activity?” In 1972 Proceedings of the International Organization for Septuagint and Cognate Studies and the Society of Biblical Literature Pseudepigrapha Seminar, ed. Robert A. Kraft, 90-107. Septuagint and Cognate Studies, no. 2. N.p.: Society of Biblical Literature, 1972. Nida, Eugene A. “Textual Criticism and Entropy.” In Tradition of the Text: Studies Offered to Dominique Barthélemy in Celebration of His 70th Birthday, ed. Gerard J. Norton and Stephen Pisano, 124-28. Orbis Biblicus et Orientalis, no. 109. Fribourg: Universitätsverlag, 1991; Göttingen: Vandenhoeck & Ruprecht, 1991. Orlinsky, Harry Meyer. “The Textual Criticism of the Old Testament.” In The Bible and the Ancient Near East: Esays in Honor of William Foxwell Albright, ed. George Ernest Wright, 113-32. Garden City, NY: Doubleday & Co., 1961. Peters, Melvin K. H. “The Use of Coptic for Textual Criticism of the Septuagint.” In La Septuaginta en la investigación contemporánea (V Congreso de la IOSCS), ed. Natalio Fernández Marcos, 55-66. Textos y estudios “Cardinal Cisneros” de la Biblia Políglota Matritense, no. 34, Madrid: Instituto “Arias Montano” Consejo Superior de Investigaciones Científicas, 1985. Roberts, Bleddyn J. “The Old Testament Manuscripts, Text and Versions.” In The Cambridge History of the Bible. Vol. 2, The West from the Fathers to the Reformation, ed. G. W. H. Lampe, 1-26. Cambridge: Cambridge University Press, 1969. _______. “The Textual Transmission of the Old Testament (Including Modern Critical Editions of the Hebrew Bible).” In Tradition and Interpretation, ed. G. W. Anderson, 1-30. Oxford: Clarendon Press, 1979. Robinson, Theodore H. “The Syriac Bible.” In The Bible: Its Ancient and English Versions, ed. H. Wheeler Robinson, 83-99. 2d ed. Oxford: Clarendon Press, 1954. Sanders, James A. “Stability and Fluidity in Text and Canon.” In Tradition of the Text: Studies Offered to Dominique Barthélemy in Celebration of His 70th Birthday, ed. Gerard J. Norton and Stephen Pisano, 203-17. Orbis Biblicus et Orientalis, no. 109. Fribourg: Universitätsverlag, 1991; Göttingen: Vandenhoeck & Ruprecht, 1991. Schulz, Alfons. “Narrative Art in the Books of Samuel.” In Narrative and Novella in Samuel: Studies by Hugo Gressmann and Other Scholars 1906-1923, ed. David M. Gunn, 119-70. Translated by David E. Orton. Journal for the Study of the Old Testament Supplement Series, no. 116. Historic Texts and Interpreters in Biblical Scholarship, no. 9. Sheffield: Sheffield Academic Press, Almond Press, 1991. Silva, M. “Internal Evidence in the Text-Critical Use of the LXX.” In La Septuaginta en la investigación contemporánea (V congreso de la IOSCS), ed. Natalio Fernández Marcos, 151-67. Textos y estudios “Cardinal Cisneros” de la Biblia Políglota Matritense, no. 34, Madrid: Instituto “Arias Montano” Consejo Superior de Investigaciones Científicas, 1985. Skehan, Patrick W. “The Qumrân Manuscripts and Textual Criticism.” In Volume du congrès: Strasbourg 1956, ed. G. W. Anderson, P. A. H. de Boer, Millar Burrows, Henri Cazelles, E. Hammershaimb, and Martin Noth, 148-60. Supplements to Vetus Testamentum, no. 4. Leiden: E. J. Brill, 1957.

371 Soisalon-Soininen, Ilmari. “Zurück zur Hebraismenfrage.” In Studien zur Septuaginta— Robert Hanhart zu Ehren aus Anlaß seines 65. Geburtstages, ed. Detlef Fraenkel, Udo Quast, and John William Wevers, 35-51. Abhandlungen der Akademie der Wissenschaften in Göttingen, Philologisch-historische Klasse, 3d series, no. 190, Mitteilungen des Septuaginta-Unternehmens, no. 20. Göttingen: Vandenhoeck & Ruprecht, 1990. Sollamo, Raija. “The Koine Background for the Repetition and Non-Repetition of the Possessive Pronoun in Co-ordinate Items.” In Studien zur Septuaginta—Robert Hanhart zu Ehren aus Anlaß seines 65. Geburtstages, ed. Detlef Fraenkel, Udo Quast, and John William Wevers, 52-63. Abhandlungen der Akademie der Wissenschaften in Göttingen, Philologisch-historische Klasse, 3d series, no. 190, Mitteilungen des Septuaginta-Unternehmens, no. 20. Göttingen: Vandenhoeck & Ruprecht, 1990. Sparks, H. F. D. “The Latin Bible.” In The Bible: Its Ancient and English Versions, ed. H. Wheeler Robinson, 100-127. 2d ed. Oxford: Clarendon Press, 1954. Sutcliffe, E. F. “Jerome.” In The Cambridge History of the Bible. Vol. 2, The West from the Fathers to the Reformation, ed. G. W. H. Lampe, 80-101. Cambridge: Cambridge University Press, 1969. Talmon, Shemaryahu. “The Old Testament Text.” In The Cambridge History of the Bible, ed. P. R. Ackroyd and C. F. Evans, 1: 159-99. Cambridge: Cambridge University Press, 1970. ________. “The Textual Study of the Bible—A New Outlook.” In Qumran and the History of the Biblical Text, ed. Frank Moore Cross and Shemaryahu Talmon, 321-400. Cambridge: Harvard University Press, 1975. ________. “The Ancient Hebrew Alphabet and Biblical Text Criticism.” In Mélanges Dominique Barthélemy: Études bibliques offertes a l’occasion de son 60e anniversaire, ed. Pierre Cassetti, Othmar Keel, and Adrian Schenker, 504-22. Orbis Biblicus et Orientalis, no. 38. Fribourg: Éditions Universitaires, 1981; Göttingen: Vandenhoeck & Ruprecht, 1981. ________. “The Ancient Hebrew Alphabet and Biblical Text Criticism.” In Mélanges bibliques et orientaux en l’honneur de M. Mathias Delcor, ed. A. Coquot, S. Légasse, and M. Tardieu, 387-402. Alter Orient und Altes Testament, no. 215. Kevelaer: Verlag Butzon & Bercker, 1985; Neukirchen-Vluyn: Neukirchener Verlag, 1985. Tov, Emanuel. “Did the Septuagint Translators Always Understand Their Hebrew Text?” In De Septuaginta: Studies in Honor of John William Wevers on His Sixty-Fifth Birthday, ed. A. Pietersma and C. Cox, 53-70. Mississauga, ON: Benben, 1984. ________. “The Composition of 1 Samuel 16-18 in the Light of the Septuagint Version.” In Empirical Models for Biblical Criticism, ed. Jeffrey H. Tigay, 97-130. Philadelphia: University of Pennsylvania Press, 1985. ________. “The Literary History of the Book of Jeremiah in the Light of Its Textual History.” In Empirical Models for Biblical Criticism, ed. Jeffrey H. Tigay, 211-37. Philadelphia: University of Pennsylvania Press, 1985. ________. “Response by E. Tov.” In The Story of David and Goliath, by Dominique Barthélemy, David W. Gooding, Johan Lust, and Emanuel Tov, 92-94. Orbis Biblicus et Orientalis, no. 73. Fribourg: Éditions Universitaires, 1986; Göttingen: Vandenhoeck & Ruprecht, 1986. ________. “The Story of David and Goliath in the MT and LXX.” In The Story of David and Goliath, by Dominique Barthélemy, David W. Gooding, Johan Lust, and Emanuel Tov, 129-37. Orbis Biblicus et Orientalis, no. 73. Fribourg: Éditions Universitaires, 1986; Göttingen: Vandenhoeck & Ruprecht, 1986.

372 ________. “The Nature and Study of the Translation Technique of the LXX in the Past and Present.” In VI Congress of the International Organization for Septuagint and Cognate Studies: Jerusalem 1986, ed. Claude E. Cox, 337-59. Society of Biblical Literature Septuagint and Cognate Studies, no. 23. Atlanta: Scholars Press, 1987. ________. “The Original Shape of the Biblical Text.” In Congress Volume: Leuven 1989, ed. J. A. Emerton, 345-59. Supplements to Vetus Testamentum, no. 43. Leiden: E. J. Brill, 1991. Tsevat, Matitiahu. “Abzählungen in 1 Samuel 1-4.” In Die Hebräische Bibel und ihre zweifache Nachgeschichte: Festschrift für Rolf Rendtorff zum 65. Geburtstag, ed. Erhard Blum, Christian Macholz, and Ekkehard W. Steggemann, 207-14. Neukirchen-Vluyn: Neukirchener Verlag, 1990. Ulrich, Eugene Charles, Jr. “Characteristics and Limitations of the Old Latin Translation of the Septuagint.” In La Septuaginta en la investigación contemporánea (V congreso de la IOSCS), ed. Natalio Fernández Marcos, 67-80. Textos y estudios “Cardinal Cisneros” de la Biblia Políglota Matritense, no. 34, Madrid: Instituto “Arias Montano” Consejo Superior de Investigaciones Científicas, 1985. ________. “Double Literary Editions of Biblical Narratives and Reflections on Determining the Form to Be Translated.” In Perspectives on the Hebrew Bible: Essays in Honor of Walter J. Harrelson, ed. James L. Crenshaw, 101-16. Macon: Mercer University Press, 1988. ________. “The Canonical Process and Textual Criticism.” In “Sha arei Talmon”: Studies in the Bible, Qumran, and the Ancient Near East Presented to Shemaryahu Talmon, ed. Michael Fishbane, Emanuel Tov, and Weston W. Fields, 267-91. Winona Lake, IN: Eisenbrauns, 1992. Wevers, John William. “The Use of Versions for Text Criticism: The Septuagint.” In La Septuaginta en la investigación contemporánea (V congreso de la IOSCS), ed. Natalio Fernández Marcos, 15-24. Textos y estudios “Cardinal Cisneros” de la Biblia Políglota Matritense, no. 34. Madrid: Instituto “Arias Montano” Consejo Superior de Investigaciones Científicas, 1985. Winton Thomas, David. “The Textual Criticism of the Old Testament.” In The Old Testament and Modern Study, ed. H. H. Rowley, 238-63. Oxford: Clarendon Press, 1951; Oxford: Oxford University Press, Oxford Paperbacks, 1961. Wright, Benjamin G. “The Quantitative Representation of Elements: Evaluating ‘Literalism’ in the LXX.” In VI Congress of the International Organization for Septuagint and Cognate Studies: Jerusalem 1986, ed. Claude E. Cox, 311-35. Society of Biblical Literature Septuagint and Cognate Studies, no. 23. Atlanta: Scholars Press, 1987. Articles Adair, James R. “A Methodology for Using the Versions in the Textual Criticism of the Old Testament.” Journal of Northwest Semitic Languages 20 (1994): 111-42. ________. “‘Literal’ and ‘Free’ Translations: A Proposal for a More Descriptive Terminology.” Journal of Northwest Semitic Languages 23 (1997): 181-209. ________. “Old and New in Textual Criticism: Similarities, Differences, and Prospects for Cooperation.” TC: A Journal of Biblical Textual Criticism 1 (1996). Aejmelaeus, Anneli. “Participium Coniunctum as a Criterion of Translation Technique.” Vetus Testamentum 32 (1982): 385-93. ________. “What Can We Know about the Hebrew Vorlage of the Septuagint?” Zeitschrift für die alttestamentliche Wissenschaft 99 (1987): 58-89.

________. Review of Retroversion and 373 Text Criticism: The Predictability of Syntax in an Ancient Translation from Greek to Ethiopic, by John Russiano Miles. In Theologische Literaturzeitung 111 (1986): col. 342-44. Albrektson, Bertil. “Recension eller tradition? Några synpunkter på den gammaltestamentliga konsonanttextens standardisering.” Svensk Exegetisk Årsbok 40 (1975): 18-35. Albright, William F. “The Oracles of Balaam.” Journal of Biblical Literature 63 (1944): 207-33. Althann, Robert. “Northwest Semitic Notes on Some Texts in 1 Samuel.” Journal of Northwest Semitic Languages 12 (1984): 27-34. Andersen, Francis I., and Forbes, A. Dean. “Orthography and Text Transmission: Computer-Assisted Investigation of Textual Transmission through the Study of Orthography in the Hebrew Bible.” Text: Transactions of the Society for Textual Scholarship 2 (1985): 25-53. Andersen, Francis I., and Freedman, David Noel. “Another Look at 4QSamb.” Revue de Qumran 14 (1989): 7-29. Avinery, Iddo. “An Example of the Influence of Hebrew on the Peshitta Translation—The Status Constructus.” Textus 9 (1981): 36-38. Barbone, Pier Giorgio. “L’uso dell’elaboratore elettronico per lo studio della Pes\it@ta.” Henoch 9 (1987): 55-56. Barnes, W. Emery. “On the Influence of the Septuagint on the Peshitta.” Journal of Theological Studies, o.s., 2 (1901): 186-97. Barr, James. “Doubts about Homoeophony in the Septuagint.” Textus 12 (1985): 1-77. ________. Review of Matres Lectionis in Ancient Hebrew Epigraphs, by Ziony Zevit. In Journal of the American Oriental Society 104 (1984): 374-75. Barthélemy, Dominique. “Redécouverte d’un chaînon manquant de l’histoire de la Septante.” Revue biblique 60 (1953): 18-29. Baumstark, A. “Pes\it@ta und palästinisches Targum.” Biblische Zeitschrift 19 (1931): 25770. Bee, Ronald E. “The Use of Statistical Methods in Old Testament Studies.” Vetus Testamentum 23 (1973): 257-72. Bloch, J. “The Influence of the Greek Bible on the Peshitta.” The American Journal of Semitic Languages and Literatures 36 (1920): 161-66. Borbone, Pier Giorgio. “La critica del testo e l’Antico Testamento ebraico.” Review of Critique textuelle de l’Ancien Testament, vol. 1, Josué, Juges, Ruth, Samuel, Rois, Chroniques, Esdras, Néhémie, Esther, by Dominique Barthélemy. In Rivista di Storia e Letteratura Religiosa 30 (1984): 251-74. Bourke, Joseph. “Samuel and the Ark: A Study in Contrasts.” Dominican Studies 7 (1954): 73-103. Brock, Sebastian P. “Lucian Redivivus: Some Reflections of Barthélemy’s Les devanciers d’Aquila.” Studia Evangelica 5 (1968): 176-81. Brooks, James A. “The Text of the New Testament and Biblical Authority.” Southwestern Journal of Theology 34 (1992): 19-20. Churgin, Pinkhos. “The Targum and the Septuagint.” The American Journal of Semitic Languages and Literatures 50 (1933): 41-65. Claassen, Walter T. “1 Sam. 3:19—A Case of Context and Semantics.” Journal of Northwest Semitic Languages 8 (1980): 1-9.

374 ________. “A Research Unit for Computer Applications to the Language and Text of the Old Testament.” Journal of Northwest Semitic Languages 13 (1986): 11-21. Cook, Johann. “Text and Tradition: A Methodological Problem.” Journal of Northwest Semitic Languages 9 (1981): 3-11. ________. “Genesis 1 in the Septuagint as Example of the Problem: Text and Tradition.” Journal of Northwest Semitic Languages 10 (1982): 25-36. ________. “Hannah and/or Elkanah on Their Way Home (1 Samuel 2:11)? A Witness to the Complexity of the Tradition History of the Samuel Texts.” Old Testament Essays 3 (1990): 247-62. Coppens, J. “La critique du texte hébreu de l’Ancien Testament.” Biblica 25 (1944): 9-49. Cross, Frank Moore. “A New Qumran Biblical Fragment Related to the Original Hebrew Underlying the Septuagint.” Bulletin of the American Schools of Oriental Research 132 (1953): 15-26. Dahood, Mitchell. “The Value of Ugaritic for Textual Criticism.” Biblica 40 (1959): 16070. Dearing, Vinton Adams. “Textual Analysis: A Consideration of Some Questions Rasied by M. P. Weitzman.” Vetus Testamentum 29 (1979): 355-59. Deist, Ferdinand E. “Is die Massoretiese Teks die Ou Testament?” Skrif en kerk 10 (1989): 9-20. Dirksen, P. B. “The Peshit@ta and Textual Criticism of the Old Testament.” Vetus Testamentum 42 (1992): 376-90. Driver, Godfrey Rolles. “Some Hebrew Roots and Their Meanings.” Journal of Theological Studies, o.s., 23 (1922): 69-73. ________. “Studies in the Vocabulary of the Old Testament: III.” Journal of Theological Studies, o.s., 32 (1931): 361-66. ________. “Abbreviations in the Masoretic Text.” Textus 1 (1960): 112-31. ________. “Once again Abbreviations.” Textus 4 (1964): 76-94. Freedman, David Noel. “The Massoretic Text and the Qumran Scrolls: A Study in Orthography.” Textus 2 (1962): 87-102. Fruin, R. “Oudtestamentische Studiën.” Nieuw theologisch tijdschrift 20 (1931): 108-15. Garbini, G. “Osservazioni linguistiche a 1 Sam., cap. 1-3.” Bibbia e oriente 5 (1963): 4752. Gehman, H. S. “Exegetical Methods Employed by the Greek Translator of I Samuel.” Journal of the American Oriental Society 70 (1950): 292-96. ________. “The Hebraic Character of Septuagint Greek.” Vetus Testamentum 1 (1951): 81-90. ________. “Some Types of Errors of Transmission in the LXX.” Vetus Testamentum 3 (1953): 141-48. Goshen-Gottstein, M. H. “Theory and Practice of Textual Criticism: The Text-Critical Use of the Septuagint.” Textus 3 (1963): 130-58.

375Their History and Their Place in the HUBP ________. “Hebrew Biblical Manuscripts: Edition.” Biblica 48 (1967): 243-90. ________. “The Development of the Hebrew Text of the Bible: Theories and Practice of Textual Criticism.” Vetus Testamentum 42 (1992): 204-13. Halpern, Baruch, and Vanderhooft, David S. “The Editions of Kings in the 7th-6th Centuries B.C.E.” Hebrew Union College Annual 62 (1991): 179-244. Haupert, Raymond S. “The Transcription-Theory of the Septuagint.” Journal of Biblical Literature 53 (1934): 251-55. Heller, Jan. “Grenzen sprachlicher Entsprechung der LXX: Ein Beitrag zur Übersetzungstechnik der LXX auf dem Gebiet der Flexionskategorien.” Mitteilungen des Instituts für Orientforschung 15 (1969): 234-48. Hulley, Karl Kelchner. “Principles of Textual Criticism Known to St. Jerome.” Harvard Studies in Classical Philology 55 (1944): 87-109. Janzen, J. Gerald. “‘Samuel Opened the Doors of the House of Yahweh’ (1 Samuel 3.15).” Journal for the Study of the Old Testament 26 (1983): 89-96. Johannessohn, Martin. “Zur Entstehung der Ausdrucksweise der lateinischen Vulgata aus den Jüngeren griechischen alttestamentlichen Übersetzungen.” Zeitschrift für die neutestamentliche Wissenschaft 44 (1952-53): 90-102. Kahle, Paul E. “Untersuchungen zur Gechichte des Pentateuchtextes.” Theologische Studien und Kritiken 88 (1915): 399-439. Katz, Peter. “Zur Übersetzungstechnik der Septuaginta.” Welt des Orients 11 (1956): 26773. Kerber, G. “Syrohexaplarische Fragmente zu den beiden Samuelis-büchern.” Zeitschrift für die alttestamentliche Wissenschaft 18 (1898): 177-96. Klein, Michael L. “The Preposition (‘Before’): A Pseudo-Anti-Anthropomorphism in the Targums.” Journal of Theological Studies, n.s., 30 (1979): 502-7. Klein, Ralph W. “Archaic Chronologies and the Textual History of the Old Testament.” Harvard Theological Review 67 (1974): 255-63. Kraft, Robert A. Review of Les devanciers d’Aquila, by Dominique Barthélemy. In Gnomon 37 (1965): 474-83. Lee, J. A. L. “Equivocal and Stereotyped Renderings in the LXX.” Revue biblique 87 (1980): 104-17. Liebreich, L. J. “Notes on the Greek Version of Symmachus.” Journal of Biblical Literature 63 (1944): 397-403. Margolis, Max Leopold. “Complete Induction for the Identification of the Vocabulary in the Greek Versions of the Old Testament with Its Semitic Equivalents: Its Necessity and the Means of Obtaining It.” Journal of the American Oriental Society 30 (1910): 301-12. ________. “Textual Criticism of the Greek Old Testament.” Proceedings of the American Philosophical Society 67 (1928): 187-97. Martin-Achard, Robert. “Some Syntactical Criteria of Translation Greek.” Vetus Testamentum 10 (1960): 295-310. Millard, A. R. “Variable Spelling in Hebrew and Other Ancient Texts.” Journal of Theological Studies, n.s., 42 (1991): 106-15. Mulder, M. J. “The Old Testament Peshit@ta and Its Various Editions.” Old Testament Exegesis 3 (1985): 34-44.

Naveh, Joseph. Review of Matres Lectionis376 in Ancient Hebrew Epigraphs, by Ziony Zevit. In Israel Exploration Journal 33 (1983): 139-40. Nida, Eugene A. “The ‘Harder Reading’ in Textual Criticism: An Application of the Second Law of Thermodynamics.” The Bible Today 32 (1981): 101-7. Pardee, Dennis. Review of Matres Lectionis in Ancient Hebrew Epigraphs, by Ziony Zevit. In Catholic Biblical Quarterly 44 (1982): 503-4. Peters, C. “Zur Herkunft der Pes\itta des ersten Samuel-buches.” Biblica 22 (1941): 25-34. Rabin, Chaim. “The Ancient Versions and the Indefinite Subject.” Textus 2 (1962): 60-76. ________. “The Translation Process and the Character of the Septuagint.” Textus 6 (1968): 1-26. Rainey, Anson F. Review of Matres Lectionis in Ancient Hebrew Epigraphs, by Ziony Zevit. In Journal of Biblical Literature 102 (1983): 629-34. Rife, J. Merle. “The Mechanics of Translation Greek.” Journal of Biblical Literature 52 (1933): 244-52. Saebø, Magne. “From Pluriformity to Uniformity: Some Remarks on the Emergence of the Massoretic Text, with Special Reference to Its Theological Significance.” Annual of the Swedish Theological Institute 11 (1977-78): 127-37. Sanders, James A. “Text and Canon: Concepts and Method.” Journal of Biblical Literature 98 (1979): 5-29. Seeligmann, Isac Leo. “Problemen en perspectieven in het moderne Septuagintaonderzoek.” Jaarbericht ex oriente lux 7 (1940): 359-90. ________. “Indications of Editorial Alteration and Adaptation in the Massoretic Text and the Septuagint.” Vetus Testamentum 11 (1961): 201-21. Segal, M. H. “The Composition of the Books of Samuel.” The Jewish Quarterly Review 55 (1965): 318-39. Simon, Uriel. “Samuel’s Call to Prophecy: Form Criticism with Close Reading.” Prooftexts: A Journal of Jewish Literary History 1 (1981): 119-32. Smith, Henry Preserved. “The Value of the Vulgate Old Testament for Textual Criticism.” The Presbyterian and Reformed Review 2 (1891): 216-34. Sparks, H. F. D. “The Semitisms of St. Luke’s Gospel.” Journal of Theological Studies, o.s., 44 (1943): 129-38. Sperber, Alexander. “Zur Sprache des Prophetentargums.” Zeitschrift für die alttestamentliche Wissenschaft 45 (1927): 267-87. ________. “The Problems of the Septuagint Recensions.” Journal of Biblical Literature 54 (1935): 73-92. Stahl, Rainer. “Die Überlieferungsgeschichte des hebräischen Bibeltextes als Problem der Textkritik—Ein Beitrag zu gegenwärtig vorliegenden textgeschichtlichen Hypothesen und zur Frage nach dem Verhältnis von Text- und Literarkritik.” Theologische Literaturzeitung 105 (1980): cols. 475-78. Stipp, Hermann-Josef. “Das Verhältnis von Textkritik und Literarkritik in neueren alltestamentlichen Veröffentlichungen.” Biblische Zeitschrift, n.s., 34 (1990): 1637. Stummer, Friedrich. “Einige Beobachtungen über die Arbeitsweise des Hieronymus bei der Übersetzung des Alten Testaments aus der Hebraica Veritas.” Biblica 10 (1929): 130. Sutcliffe, Edmund F. “St. Jerome’s Hebrew Manuscripts.” Biblica 29 (1948): 195-204.

Talmon, Shemaryahu. “Double Readings377 in the Massoretic Text.” Textus 1 (1960): 14485. ________. “Synonymous Readings in the Textual Tradition of the Old Testament.” Scripta Hierosolymitana 8 (1961): 335-83. ________. “DSIa As a Witness to Ancient Exegesis of the Book of Isaiah.” Annual of the Swedish Theological Institute 1 (1962): 62-72. ________. “Aspects of the Textual Transmission of the Bible in the Light of Qumran Manuscripts.” Textus 4 (1964): 95-132. Thackeray, Henry St. John. “The Greek Translators of the Four Books of Kings.” Journal of Theological Studies, o.s., 8 (1907): 262-78. ________. “The Bisection of Books in Primitive Septuagint MSS.” Journal of Theological Studies, o.s., 9 (1907-8): 88-98. Tov, Emanuel. “On ‘Pseudo-Variants’ Reflected in the Septuagint.” Journal of Semitic Studies 20 (1975): 165-77. ________. “Lucian and Proto-Lucian.” Revue biblique 83 (1976): 51-54. ________. “Compound Words in the LXX Representing Two or More Hebrew Words.” Biblica 58 (1977): 189-212. ________. “The Nature of the Hebrew Text Underlying the LXX: A Survey of the Problems.” Journal for the Study of the Old Testament 7 (1978): 56-68. ________. “Loan-Words, Homophony and Transliterations in the Septuagint.” Biblica 60 (1979): 216-36. ________. “Criteria for Evaluating Textual Readings: The Limitations of Textual Rules.” Harvard Theological Review 75 (1982): 429-48. ________. “A Modern Textual Outlook Based on the Qumran Scrolls.” Hebrew Union College Annual 53 (1982): 11-27. ________. “The Representation of the Causative Aspects of the Hiph(il in the LXX: A Study in Translation Technique.” Biblica 63 (1982):417-24. ________. “Some Sequence Differences between the MT and LXX and Their Ramifications for the Literary Criticism of the Bible.” Journal of Northwest Semitic Languages 13 (1987): 151-60. Tov, Emanuel and Wright, Benjamin. “Computer-Assisted Study of the Criteria for Assessing the Literalness of Translation Units in the LXX.” Textus 12 (1985): 14887. Ulrich, Eugene Charles, Jr. “Double Literary Editions to Biblical Narratives and Reflections on Determining the Form to Be Translated.” Perspectives in Religious Studies 15 (Fall 1988): 101-16. Volz, Paul. “Ein Arbeitsplan für die Textkritik des Alten Testaments.” Zeitschrift für die alttestamentliche Wissenschaft 54 (1936): 100-113. Vööbus, A. “Der Einfluss des altpalästinischen Targums in der Textgeschichte der Peschitta des Alten Testament.” Le muséon 68 (1955): 215-18. Walters, Stanley D. “Hannah and Anna: The Greek and Hebrew Texts of 1 Samuel 1.” Journal of Biblical Literature 107 (1988): 385-412. Watson, Wilfred G. E. “The Structure of 1 Sam 3.” Biblische Zeitschrift 29 (1985): 9093. Weippert, Helga. “Die ‘deuteronomistischen’ Beurteilungen der Könige von Israel und Juda und das Problem der Redaktion der Königsbücher.” Biblica 53 (1972): 30139.

378 Weissert, D. “Alexandrian Analogical Word-Analysis and Septuagint Translation Techniques.” Textus 8 (1973): 31-44. Weitzman, M. P. Review of Principles and Practice of Textual Analysis, by Vinton A. Dearing. In Vetus Testamentum 27 (1977): 225-35. Wicke, Donald W. “The Structure of 1 Sam 3: Another View.” Biblische Zeitschrift 30 (1986): 256-58. Willis, John T. “An Anti-Elide Narrative Tradition from a Prophetic Circle at the Ramah Sanctuary.” Journal of Biblical Literature 90 (1971): 288-308. ________. “Cultic Elements in the Story of Samuel’s Birth and Dedication.” Studia Theologica 26 (1972): 33-61. ________. “Samuel Versus Eli, I Sam. 1-7.” Theologische Zeitschrift 35 (1979): 201-12. Wutz, Franz Xaver. “Ist der hebräische Urtext wieder erreichbar?” Zeitschrift für die alttestamentliche Wissenschaft 43 (1925): 115-19. Zipor, M. A. “A Striking Translation Technique of the Peshit@ta.” Journal of Semitic Studies 26 (1981): 11-20. Reference Works Aland, Kurt, ed. Synopsis Quattuor Evangeliorum. 13th ed. Stuttgart: Deutsche Bibelgesellschaft, 1985. The Anchor Bible Dictionary. S.v. “Names of God in the OT,” by Martin Rose. Bauer, Walter. A Greek-English Lexicon of the New Testament and Other Early Christian Literature. 2d ed. Translated and edited by William F. Arndt, F. Wilbur Gingrich, and Frederick W. Danker. Chicago: University of Chicago Press, 1979. Baumgartner, Walter, and Stamm, J. J. Hebräisches und Aramäisches Lexikon zum Alten Testament. 3d ed. 5 vols. Leiden: E. J. Brill, 1967-. Biblia Sacra iuxta Latina Vulgatum versionem ad codicem fidem iussu Pii XII. Edited by Cura et studio monachorum abbatiae pontificiae Sancti Hieronymi in urbe ordinis Sancti Benedicti. Rome: Typis Poliglottis Vaticanus, 1926-. Brockelmann, Carl. Lexicon Syriacum. 2d ed. Göttingen: Max Niemeyer, 1928. Brooke, Alan England; McLean, Norman; and Thackeray, Henry St. John, eds. The Old Testament in Greek. London: Cambridge University Press, 1906-. Brown, Francis; Driver, S. R.; and Briggs, Charles A. A Hebrew and English Lexicon of the Old Testament with an Appendix Containing the Biblical Aramaic. Oxford: Clarendon Press, 1906; reprint, 1951. Elliger, K., and Rudolph, W., eds. Biblica Hebraica Stuttgartensia. 2d ed. Stuttgart: Deutsche Bibelgesellschaft, 1983. Encyclopædia Britannica, 1985 ed. Macropædia. S.v. “Languages of the World.” Field, Frederick, ed. Origenis Hexaplorum. 2 vols. Oxford: Oxford University Press, 1875; reprint, Hildesheim: Georg Olms, 1964. Fischer, Bonifatius, ed. Novae concordantiae bibliorum sacrorum iuxta vulgatam versionem critice editam. 5 vols. Stuttgart: Friedrich Frommann Verlag Günther Holzboog, 1977. Fischer, Bonifatius; Gribomont, Jean; Sparks, H. F. D.; and Thiele, W., eds. Biblia Sacra iuxta Vulgatam versionem. 2d ed. 2 vols. Stuttgart: Württembergische Bibelanstalt, 1975. Glare, P. G. W., ed. Oxford Latin Dictionary. Oxford: Clarendon Press, 1982.

379 J. Targum Jonathan of the Former Prophets. Harrington, Daniel J., and Saldarini, Anthony The Aramaic Bible, vol. 10, ed. Kevin Cathcart, Michael Maher, and Martin McNamara. Wilmington: Michael Glazier, 1987. Hatch, Edwin, and Redpath, Henry A. A Concordance to the Septuagint and the Other Greek Versions of the Old Testament (Including the Apocryphal Books). 3 vols. Oxford: Clarendon Press, 1897-1906; reprint, Grand Rapids: Baker Book House, 1983. Holladay, William L. A Concise Hebrew and Aramaic Lexicon of the Old Testament. Grand Rapids: William B. Eerdmans, 1971. Holmes, Robert, and Parsons, James, eds. Vetus Testamentum Graecum cum variis lectionibus. 5 vols. Oxford: Clarendon Press, 1798-1827. The International Standard Bible Encyclopedia. 1988 ed. S.v. “Semites,” by Mario Liverani. Translated by William Sanford LaSor. ________. S.v. “Versions,” by Arthur Vööbus. The Interpreter’s Dictionary of the Bible. S.v., “Poetry, Hebrew,” by Norman K. Gottwald. ________. S.v. “Samaritan Pentateuch,” by Bleddyn J. Roberts. ________. S.v. “Septuagint,” by John William Wevers. ________. S.v. “Versions, Ancient,” by Bruce M. Metzger. ________. Supplementary Volume. S.v. “Abbreviations, Hebrew Texts,” by Michael Fishbane. ________. Supplementary Volume. S.v. “Septuagint,” by Emanuel Tov and Robert A. Kraft. ________. Supplementary Volume. S.v. “Syriac Versions,” by A. Vööbus. ________. Supplementary Volume. S.v. “Targums,” by M. J. McNamara. Jastrow, Marcus. A Dictionary of the Targumim, the Talmud Babli and Yerushalmi, and the Midrashic Literature. 2 vols. New York: Title, 1943. Kennicott, Benjamin, ed. Vetus Testamentum Hebraicum cum variis lectionibus. 2 vols. Oxford: Clarendon Press, 1776-80. ________. Dissertatio generalis in Vetus Testamentum Hebraicum; cum variis lectionibus, ex codicibus manuscriptis et impressis. Oxford: Clarendon Press, 1780. Kittel, Rudolf, ed. Biblia Hebraica. 3d ed. Stuttgart: Privegierte Württembergische Bibelanstalt, 1937. Lagarde, Paul de. Librorum Veteris Testamenti Canonicorum. Vol. 1. Göttingen: Aedibus Dieterich, Arnold Hoyer, 1883. Levy, J. Chaldäisches Wörterbuch über die Targumim und einen grossen Theil des rabbinischen Schriftthums. 3d ed. Leipzig: G. Engel, 1866; reprint, Cologne: Joseph Melzer Verlag, 1959. Liddell, Henry George, and Scott, Robert. A Greek-English Lexicon. Revised by Henry Stuart Jones and Roderick McKenzie. Oxford: Clarendon Press, 1968. Lisowsky, Gerhard. Konkordanz zum hebräischen Alten Testament. 2d ed. Stuttgart: Deutsche Bibelgesellschaft, 1981. Mandelkern, Solomon. . 2d ed. Berlin: F. Margolin, 1925. Margoliouth, J. P. Supplement to the Thesaurus Syriacus of R. Payne Smith, S.T.P. Oxford: Clarendon Press, 1927.

380 The Old Testament in Syriac According to the Peshi a Version. Edited on behalf of the International Organization for the Study of the Old Testament by the Peshit@ta Institute, Leiden. Leiden: E. J. Brill, 1972-. Payne Smith, J. [J. P. Margoliouth], ed. A Compendious Syriac Dictionary. Oxford: Clarendon Press, 1903. Payne Smith, R., ed. Thesaurus Syriacus. 2 vols. Oxford: Clarendon Press, 1879-1901. Rahlfs, Alfred, ed. Septuaginta: id est Vetus Testamentum graece iuxta LXX interpretes. Editio Minor. Stuttgart: Deutsche Bibelgesellschaft, 1979. Rossi, Giovani Bernardo de. Variae lectiones Veteris Testamenti. 4 vols. Parma: n.p., 1784-88. Sperber, Alexander, ed. The Bible in Aramaic. 4 vols. in 5 parts. Leiden: E. J. Brill, 1959-1973. Svenskt Bibliskt Uppslagsverk. S.v. “Bibeln,” by C.-M. Edsman. ________. S.v. “Gamla Testamentet,” by Ivan Engnell. Swete, Henry Barclay, ed. The Old Testament in Greek According to the Septuagint. 3 vols. Cambridge: Cambridge University Press, 1887-1912. Taylor, Bernard. The Lucianic Manuscripts of 1 Reigns. Volume 1: Majority Text. Harvard Semitic Monographs, no. 50. Atlanta: Scholars Press, 1992. Weil, Gérard E., ed. Massorah Gedolah iuxta Codicem Leningradensem B 19 a. Vol. 1: Catalogi. Rome: Pontifical Biblical Institute, 1971. Ziegler, J., et al., eds. Septuaginta: Vetus Testamentum Graecum auctoritate Academiae Scientarum Gottingensis editum. Göttingen: Vandenhoeck & Ruprecht, 1931-. Unpublished Works Beling, Willard A. “The Hebrew Variants in the First Book of Samuel Compared with the Old Greek Recensions.” Ph.D. diss., Princeton University, 1947. Brock, Sebastian P. “The Recensions of the Septuagint Version of 1 Samuel.” D.Phil. diss., Oxford University, 1966. Cook, Johann. “Die pluraliteit van ou-testamentiese tekste en eksegetiese metodologie.” Paper presented at the annual meeting of Die Ou-Testamentiese Werkgemeenskap van Suid-Afrika, 1988. Dunn, Mark R. “An Examination of the Textual Character of Codex Ephraemi Rescriptus (C, 04) in the Four Gospels.” Ph.D. diss., Southwestern Baptist Theological Seminary, 1990. Israel Antiquities Authority. “PAM 41.766.” Photograph. Claremont: Ancient Biblical Manuscript Center, 1992. Kelly, Balmer H. “The Septuagint Translators of I Samuel and II Samuel i-xi.” Ph.D. diss., Princeton University, 1948. McReynolds, Paul Robert. “The Claremont Profile Method and the Grouping of New Testament Manuscripts.” Ph.D. diss., Claremont Graduate School, 1968. Merrill, Arthur L. “I Sam 1-12: A Traditio-Historical Study.” Ph.D. diss., University of Chicago, 1962. Rehm, Martin. “Textkritische Untersuchung der Samuel-Königsbucher und der Chronik.” Ph.D. diss., University of Munich, 1937. Sanders, James A. Lecture delivered at the University of Stellenbosch, South Africa, 17 August 1989.

Thank you for evaluating AnyBizSoft PDF Merger! To remove this page, please register your program! Go to Purchase Now>>

AnyBizSoft

PDF Merger  Merge multiple PDF files into one  Select page range of PDF to merge

 Select specific page(s) to merge  Extract page(s) from different PDF files and merge into one