Review of Pitersma, trans., The Psalms (NETS)

This article is also available in text-only format.

Albert Pietersma, translator. A New English Translation of the Septuagint and the Other Greek Translations Traditionally Included under that Title: The Psalms. New York/Oxford, Oxford University Press, 2000. pp. xxvii + 149. ISBN 0-19-529752-0 (cloth), 0-19-529753-9 (paper). US $29.99 (cloth), $12.99 (paper).

1. It is over a hundred and fifty years since the last English translation of the Septuagint was published. Although Brenton's translation is still in print, of itself that time lapse is sufficient reason to tackle the task again.¹ There are precedents. Linguistic change in the receptor language was probably one factor in the production of revisions of the Septuagint around the turn of the era. The interim period since Brenton's work has also seen great developments in the editions of the Greek text used as base for the translation. Now the International Organization for Septuagint and Cognate Studies [IOSCS] and Oxford University Press have come together to produce a new translation, and have produced the first volume in a remarkably brief period. This will catch the attention of those curious about the project as a whole as well as those more particularly interested in the Old Greek Psalter. Albert Pietersma, the translator of this first volume, is himself one of the prime movers and shakers behind the project as a whole.

2. This volume is principally comprised of an address to the Reader of NETS (A New English Translation of the Septuagint), another to the Reader of the Psalms, the translation of the Psalms (including Psalm 151) and the Prayer of Mannaseh. The first address gives a great deal of highly condensed material by which the choices made by the project are made explicit (pp. vii-xviii). These were discussed at length by the participants in the early stages. Perhaps it is unreasonable of this reviewer, but he found the Translation Manual of the project published for the IOSCS by Uncial Books in Michigan in 1996 (Pietersma 1996) much more readable and informative than the version under review.

3. The text here presented represents only the first part of the project, for we are promised a commentary on the translation as well. An adequate assessment of the project can only be made when we have both translation and commentary before us for the complete enterprise. Unlike the two previous English translations of the Septuagint,² different translators will work on each book.³ This should mean that the project will be completed in a shorter time than if it were the labour of one scholar, but will also raise issues of uniformity of approach from one book to another.

4. This could be described in some ways as a marriage between the largely North American-dominated IOSCS and the publications of the Septuaginta-Unternehmen at Göttingen. The figure of J. W. Wevers, who has now completed the edition of the Septuagint Pentateuch (Wevers, ed. 1974, 1977, 1982, 1986, 1991) and produced related textual commentaries (Wevers 1974, 1978, 1982, 1986, 1992), is already a strong link between the two groups. While NETS is based on the eclectic reconstruction of the form of the Greek text as determined by the ongoing Göttingen project (when available), the nineteenth century translations were largely based on the text of Codex Vaticanus. This particular volume is something of a curiosity, in that it is based on the text by Alfred Rahlfs originally published in the Göttingen series in 1931, but without adherence to the punctuation introduced by Rahlfs. Emendations are made to Rahlfs' text on almost every page. The revision of Rahlfs' problematic edition is currently being discussed at Göttingen, and we are told that it will certainly not be completed in our lifetimes. The option taken here was sensible and pragmatic. For the books for which more satisfactory Göttingen texts exist, the situation will be easier, but for others such as Proverbs, for which there is no Göttingen edition, even harder.

5. Specific questions arise in relation to each book and to the work of each ancient translator. For the Psalter, many of these are addressed on pages xix-xxvii. We are told that the texts enclosed in square brackets by Rahlfs as unlikely to be original, despite widespread manuscript support, have been suppressed. Pietersma uses his own long experience with the Greek text to identify further elements of the Greek text which he considers unoriginal (see, for example, Pss 7:12; 67:5; 72:28; 97:1; 148:5) Such suggestions, which belong more to the domain of textual criticism than of translation, are infrequent. They are generally quantitative, that is, indications of addition to or omission from Rahlfs' text, rather than qualitative, suggesting a reading different from Rahlfs' text. For some, however, the single-minded approach to the quest for an original Old Greek text will pose problems, and they may regret the production of a translation of the Septuagint that does not consider the principal accretions to the text (e.g., the Christian additions) in transmission. This is particularly provocative, since it is difficult to ignore the omission of the word "Old" (i.e., "Other Greek Translations" rather than "Other Old Greek Translations") from the title and half title pages of this publication when compared with the Project's statement of principles on its Web site and the 1996 Michigan publication, but it is hardly of great significance, and further volumes may set us straight on the matter.

6. NETS is strongly linked to the New Revised Standard Version (NRSV). The decision was taken to use this base text in order to maximise the synoptic potential (p. x) of the English translation of the Septuagint by comparison with the English translation of the Hebrew text. This decision has its drawbacks, but as a working option it is clear and makes use of the experience underlying the NRSV. There is no indication given in the text of the translation where there are differences from the NRSV. Many readers, working through various volumes to find the relevant material, will find themselves wishing already for a synoptic presentation of the text. It is almost certain that such presentations will be published, and we may even hope that the Greek text will be included. Yet, given the considerations of price and availability of that edition, the possibility of a full synoptic study of the Hebrew, Greek, NRSV, and NETS may be some time away for many. There may be copyright or other reasons why the Göttingen text cannot be included with the translation. At the moment, the translation stands starkly alone. It is rather pleasantly laid out as poetry in a single column per page, following generally the NRSV divisions into stanzas, but with the addition of a blank line wherever "selah" or "interlude on strings" is found. The presentation of numbers is a little confusing: at the beginning of each psalm the Rahlfs Septuagint number is given with the MT numbers in brackets, whereas for verses the numbers in the Rahlfs Septuagint are again given, but this time it is the NRSV numbers that are given in brackets (note that the NRSV does not number the titles), and MT numbers are taken to coincide with Greek (p.xxvii).

7. The synoptic concern is important to the raison d'étre of the translation (p. x). It also sets this project apart from its French counterpart, La Bible d'Alexandrie (Harl and Alexandre 1986-), which seeks to read the Greek as it would have been understood by a Greek speaker who did not know Hebrew, and indeed as it was understood by Hellenistic theological writers and exegetes. By contrast the introduction to NETS speaks of "presupposing a Greek translation which aimed at bringing the reader to the Hebrew original rather than bringing the Hebrew original to the reader" (p.ix). This picks up article six of the NETS statement of principles as published in 1996: "NETS translators will seek to reflect the meaning of the Greek text in accordance with the ancient translator's perceived intent and as occasioned by the ancient translator's linguistic approach, even when this policy may result in an unidiomatic (though grammatical) English rendering" (Pietersma 1996: 9, 49). The relevance of this decision to translations of the books and chapters originally composed in Greek is not yet clear, but doubtless will become so as the project advances. Ecclesiasticus too will be interesting to watch.

8. The addresses to the Reader of NETS and to the Reader of the Psalms should be read carefully. To illustrate this, we quote one passage at length from page xi. Differences between NETS and NRSV may typically be due to one of five reasons:

(1) the lexical choice of the NRSV to represent the Hebrew differs significantly from that of the Greek translator's, even though either rendering, independently, might be regarded as an adequate translation of the same Hebrew; (2) differences in translational approach between the translators of the NRSV and the ancient Greek translators has occasioned noteworthy differences between the two versions (for example, in any given passage, the Greek may be hyper-literalistic, where the NRSV is not, or again it may be very free, which the NRSV is not); (3) an attempt to reflect linguistic features in the Greek, such as word echoes or paratactic style, at times has required that the NRSV wording be revised; (4) the Greek translator has apparently rendered a text at variance with MT, due to textual difference; (5) the NRSV has not translated MT, but opted instead for some other reading. Naturally, where, in such instances, the NRSV has adopted the readings of the Septuagint, NETS and NRSV agree, though not because their parent texts agree! As a rule such cases have been annotated in the NRSV, but the reader should, of course, not take for granted that the precise English word used by the NRSV has necessarily been adopted by NETS.

The third of these factors is an addition to the list of four found in the NETS translators' manual. Not all the cases mentioned in the fifth factor are noted in NRSV, especially where the difference is in vocalisation. At Ps 55:20 MT, the NRSV translates "will hear, and will humble them." The reading adopted here is an emendation of the MT, which reads "will hear and will answer them," but there is no annotation because it is a matter of vowel signs alone and, in the view of the NRSV committee, "the vowel points are less ancient and reliable than the consonants" (Metzger 1995: xiii). Clearly the Reformation controversies of Levita, Buxtorf and Cappel have yet to be resolved (cf. Barthélemy 1982: *10-*15). The unwary reader of NRSV and NETS will imagine that the MT and the NETS have the same text.

9. The factors just mentioned alert the reader to the fact that no conclusions can safely be drawn from a comparison between NRSV and NETS despite the synoptic concern of the translation. It is only in the full development of the publication that the value of the link with NRSV will be seen. It seems that at least until we have the commentary, the reader will have to do a great deal of detective work at the points where NETS and NRSV differ, and even when they do not differ the reader cannot draw solid conclusions about the relationship of the "parent texts." This is particularly evident in the third of the factors just cited. Word echoes and paratactic style in the Greek are represented "at times." How does this compare to representation of the word echoes and paratactic style of the Hebrew in NRSV? Where the Greek differs from the Hebrew, have the translators tried to produce an NRSV-style translation of the Greek which at the same time tries to draw out the particular text of the Greek? This would allow for comparison with the NRSV- influenced translation where Hebrew and Greek coincide. If different translation principles have been adopted, the relationship of the English translation to the Greek base would not be uniform: at times it is influenced by the NRSV, and at times it is a direct translation of the Greek that has an implicit task to convey to the reader the way in which the Greek differs from the Hebrew as well as reflecting the manner in which the Greek translator intended his text to be understood, which is circumscribed or dictated in the view of the NETS editors by the meaning of the Hebrew text.

10. The NRSV tradition has a different purpose and a different target readership than this translation of the Septuagint. Although it takes as its base the (consonantal) Hebrew text, it is still engaged in a text-critical enterprise (unspecified) which seeks to go behind the earliest directly attested Hebrew text on the basis of ancient versions, Hebrew tradition, and scholarly conjecture. This eclecticism of the biblical text rather than any direct relation to the Hebrew tradition is quite a different quest from the translation into English of the eclectic text of a particular Greek translation made at a precise moment in the history of transmission of the Hebrew text. The eclecticism of the Greek text chosen as a base selects from among the Greek version and its daughter translations. The language of the NRSV chooses to continue on the basis of the King James Bible, "but to introduce such changes as are warranted on the basis of accuracy, clarity, euphony, and current English usage" (Metzger 1995: xiv). For NETS by contrast,

When literary beauty occurs it is the exception rather than the rule. Consequently, NETS readers would be remiss in expecting literary elegance in the English. That would have required, from the NETS perspective, a different Greek. Since the Septuagint, with a few exceptions, was not originally composed in Greek, and often used unidiomatic Greek, a fully idiomatic translation into English can scarcely be justified."[p. xv]

Just so, one can imagine conversations about the relative merits of Aquila and Symmachus. Is idiomatic Greek necessarily an index of literary elegance? The complex area of aesthetics in biblical translation has been a recurring issue, and even the significant recent two volume work by David Norton (no relation) does not resolve all the issues (Norton 1993).

11. The treatment of proper names will strike many as unusual. Names which are judged to be translated in the Septuagint are given their standard equivalent in English. The classification as a translation seems to depend on whether the name is known to be in general use in the Hellenistic world apart from the LXX. Otherwise names are treated as transcriptions, de novo productions from the source language. These are consequently given as an English transcription. So we find Dauid and Salomon, Pharao, Beniamin, Ierousalem, Chanaan, rulers of Ioudas, Isaak, Iakob. Is there any benefit to be gained from this idiosyncratic approach to proper names? Can there be any doubt that by Dauid the Septuagintal translator meant the figure the NRSV called David, and so forth?

12. Canonical questions also arise. We are told that Odes is not included in NETS, "since it has dubious integrity as a literary unit, and, in any case, almost all of the individual Septuagint odes have already been included in their native setting in other books" (p. xvi). Only the Prayer of Manasses is translated here, as an appendix to the 151 Psalms of the Septuagint Psalter. Perhaps this belongs to the same world as those who regret the loss of reference to Rahlfs' bracketed texts, but this reviewer at least feels that inclusion of the Odes would have been appropriate.

13. Those involved in this project are familiar with the CATSS (Computer Assisted Tools for Septuagint Studies) project in Philadelphia and Jerusalem (see Kraft and Tov 1986, Tov 1986). This project placed emphasis on formal equivalence as well as dynamic equivalence. This emphasis is also found here in NETS (p. xv), but there also seems to be a more rounded appreciation of semantic correspondence. The introduction sets out a sophisticated approach to the evaluation of the meaning of Greek words when governed by their normal Greek semantic range, with a sliding scale between contextual renderings and stereotypes (or a consistent rendering of a particular Hebrew word). This is distinguished from another scale governed by the Hebrew texts. This scale ranges from "isolate renderings," where the Greek is rendered in a particular way because of the perceived connection with Hebrew morpheme(s) in question, to "calques," where the meaning of the Hebrew has been transferred to the Greek and become part of the living language.

14. The Psalter is a particularly good book with which to begin the publication, for "the NETS paradigm of the Greek as an interlinear translation of the Hebrew fits the Book of Psalms like a glove" (p. xix). A useful translation profile is given of the Greek. The atomistic, word-based interlinear translation of the Greek presents both benefits and difficulties to the translator of the Greek. It will be interesting to see how this is applied in books where the paradigm of interlinear translation is less obvious.

15. The tricky question of tense-aspect in the Greek Psalter is dealt with under the heading of "some problems of grammar" (p. xxv). This is a particular instance where the 1996 publication was clearer than the publication under review. The attentive reader may find this to be one of the more striking points of contrast between the NRSV and the NETS translation. The NRSV by and large translates the aspect of the Hebrew verbs ad sensum. The NETS translates the Greek aspect with reference to the Greek alone, trying to communicate at least some of the quality of "a certain stiltedness and abruptness in the use and sequence of Greek tenses" (p. xxv). This is seen as early as the first verses of the first Psalm, where the Hebrew perfect / qatal is translated "do not follow ... take the path ... sit" in NRSV, but "did not walk ... stand ... sit" in NETS. In the following verse, the Hebrew imperfect (NRSV "they meditate") is translated by Greek and NETS as a future "he will meditate."

16. Psalm superscriptions or titles present a particular difficulty in that they seem to give technical terms without context. Believing that they grew piecemeal, Pietersma treats them atomistically in his translation. Since many are barely intelligible in Hebrew, there is little reason to complain of this. The Greek superscriptions can hardly have brought the ancient readers any closer to the Hebrew text than the NETS translation brings us today to the Greek or Hebrew.

17. Inevitably the English text cannot be identical with the Greek. The items added in translation are clearly indicated by footnotes, but are they always necessary, given the elements just noted? A quick tally would include Pss 22(23):6; 30(31):12(11); 31(32):7; 35(36):3(2); 44(45):9(8); 49(50):2; 51(52):7(5); 71(72):15; 139(140):13(12); 147:4-6(15-17), 9(20).

18. I undertook the review task with great hesitation, for I am hugely enthusiastic about this project, and I hope that it will open up Septuagint studies for all serious students of the Hebrew Bible. The previous comments should be seen in this context. It is a wonderful project. I look forward to further publications in the series.

Gerard J. Norton
University of Birmingham

© TC: A Journal of Biblical Textual Criticism, 2001.

Endnotes

¹Originally published as ,a href="#Brenton1844">Brenton 1844. The 1851 edition of Brenton was initially reprinted by Hendrickson Publishers and has gone through at least three printings.