Note to electronic version: for the Hebrew the CCAT code has been used and for the Greek the transliteration given on p. 16 of the Manual.
PRINTED COPIES OF THE MANUAL MAY BE ORDERED FROM UNCIAL PRESS (see below).



TRANSLATION MANUAL


for

"A NEW ENGLISH TRANSLATION OF THE SEPTUAGINT" (NETS)

Albert Pietersma





Published for
The International Organization
for Septuagint and Cognate Studies

by
UNCIAL BOOKS
Ada, Michigan 49301


Copyright © 1996 by Albert Pietersma

Published by Uncial Books
9545 28th Street S.E.
Ada, Michigan 49301-9274

for

The International Organization for Septuagint and Cognate Studies

Printed in the United States of America

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means without the prior written permission of the publisher. The only exception is brief quotations in published reviews.

ISBN 0-9653269-0-X


Contents






Introduction7
1.NETS Translation Procedures9
2.Introduction to Greek Psalms27
Appendix 1 : Statement of Principles49
Appendix 2 : Selected Psalms51







(page 7)

Introduction




If translating is, as some would have it, an act of hybris or presumption, then translating an ancient Greek translation of a Semitic original into modern English must surely border on utter folly. A New English Translation of the Septuagint (NETS) is not being undertaken by the International Organization for Septuagint and Cognate Studies because a way has somehow been found to lessen the folly, but because there appears to be no other way to realize a dream of decades. Among the many that have dreamed that dream, the names of George Howard and David Aiken deserve special mention.

While a translator's manual might ideally provide a solution for every problem that could possibly arise in the course of translating, the Translation Committee is keenly aware that this Manual is not able to answer every question. Instead, it is intended to direct, guide and facilitate the translation of the Greek text into modern English, using the New Revised Standard Version (NRSV) as a translation base. But in spite of the shortcomings that any manual must have, translators are urged to take its direction seriously. Our common goal is to make NETS as perfect a translation as is humanly possible.

For obvious reasons, the focus of the manual is on the Septuagint/Old Greek as a translation of extant Semitic texts, though a variety of directives apply equally to books whose Semitic original has been lost or whose language of composition was Greek.

Since the Manual has grown out of the NETS translation efforts on the book of Psalms, examples have been drawn from that book. This is not to say, of course, that whatever holds true in Psalms ipso facto holds true as well in other translation units of the corpus. It is hoped, however, that the principles and procedures set forth, if not the specific details supplied by way of illustration, will help guide individual translators in their assignments. Moreover, the Translation Committee intends to create an electronic discussion group for the express purpose of NETS translator interaction. It is anticipated that as a result of such interaction additions will be made from time to time to the manual. Translators are, therefore, encouraged to air the problems they encounter to fellow translators and/or seek the advice of the committee.
I want to thank my fellow members on the Translation Committee, Leonard Greenspoon, Bob Kraft, Moises Silva, Bernard Taylor and Ben Wright, for their input, Bob and Moises especially.

Albert Pietersma
Toronto
28 September 1995

(page 9)

Chapter 1

NETS Translation Procedures

1. Relevant Articles from the NETS' Statement of Principles (see Appendix 1)

Article. 5. NETS will normally translate the best available editions. Footnotes will indicate significant semantic departures from the base text.
Comment: Though departures from any text edition, including critical editions, are allowed they are not encouraged. In other words, translators must have solid reasons for setting aside any given reading in the Greek edition on which their English translation is based. Moreover, all departures from the base text which affect the meaning in English must be carefully and fully noted (see 4a below).
Article. 6. 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. Appropriate footnotes will inform the reader.
Comment: NETS focuses on translator's intent rather than on audience perception. That is to say, NETS attempts to reflect the manner in which the Greek translator intended his text to be understood rather than the manner in which a Greek audience might understand that text. While from a purely statistical

(p.10)
point of view "translator's intent" and "audience perception" will most often coincide, such coincidence should be demonstrated rather than presupposed. Not infrequently, however, "translator's intent" (in distinction from possible perceptions by the readers) is circumscribed or dictated by the meaning of the Hebrew text (see the "Introduction to Psalms.)
Article. 7. Since much of the Septuagint derives from the Hebrew Scriptures, it is important to reflect that dependence as consistently and comprehensively as possible for the English reader. For these reasons, NETS consciously attempts to employ the wording and approach of a standard modern English translation of the Hebrew Bible in situations in which the Greek understands the Hebrew text in the same way as the English. The New Revised Standard Version (NRSV) has been chosen to provide this English base. Where the Greek does not correspond to the Hebrew as understood by the NRSV, translators will make every reasonable effort to represent the differences fully and accurately.

Comment: NETS strives to maximize synoptic use of the Greek and Hebrew Scriptures. To that end all modifications of the NRSV should reflect as closely as possible what meaningful differences exist between the Greek and Hebrew texts and, conversely, no departures from the NRSV should be introduced without good reason. As in the case of Article 5, a useful modus operandi might be: "Retain what you can, but change what you must."

2. The English Base of NETS: The New Revised Standard Version.
a. General working principle of the NRSV: "As literal as possible, as free as necessary." NETS is based on the same principle (cf. 3 below)

(p. 11)
b. Spelling and linguistic usage:
(1). General guide: H. W. Fowler, A Dictionary of Modern English Usage. (Oxford, 1965).To run a search on the NRSV is, however, often equally efficient.
(2). Some specifics:
(a). Traditional distinction between shall (should) and will (would), adopted by the NRSV, is continued in NETS. Thus, simple future (plain auxiliary): shall/will/will; emphatic future (colored auxiliary): will/shall/shall (see Fowler, Dictionary of Modern English Usage, pp. 713-14.
(b). Capitalization of LORD when it represents the Hebrew tetragrammaton is not applicable to NETS, unless, in a given translation unit, it can be demonstrated that a distinction between YHWH (Yahweh) and )DNY (Adonai) has been carried over into the Greek.
(c). Omissions, additions, singular/plural number etc for reasons of English style: as a rule of thumb it is suggested that, when the Hebrew and Greek texts are in agreement, no change be made to the NRSV for NETS. Conversely, a change is recommended when the two texts do not agree.
(d). While rigid one-to-one Greek-English renderings are usually unacceptable on linguistic grounds (and eschewed by the NRSV), within pericopes (e.g., individual Psalms) literary echoes such as repeated words may frequently be successfully reproduced. In certain circumstances, this may qualify as a good reason for modifying an NRSV decision even where MT and LXX/OG do not (significantly) disagree.

(p 12)
c. Gender specific/inclusive language: NRSV policy is to avoid "all masculine language referring to human beings apart from texts that clearly" refer "to men." [Footnote 1 B. M. Metzger, R. C. Dentan, and W. Harrelson, The Making of the New Revised Standard Version of the Bible (Grand Rapids, 1991), p. 76).] Thus, the NETS guidelines are as follows:
(1) One should generally retain the NRSV reading (especially its circumlocution in the plural number) . . .
(a) if its reading does not run counter to the option exercised by the Greek translator and
(b) if its reading helps avoid misleading masculine singular pronominal references.
(2) One should revise the NRSV when the Greek translator can be shown to have opted for gender-specific language (or non-gender-specific as the case may be). E.g., since Hebrew )Y$ in Pss is rendered by both ANTHROPOS and ANER it may be concluded that gender-specific ANER constitutes a deliberate choice. The same is true for such Hebrew words as )DM, BN, or GBR. Compound terms such as HUIOS/HUIOI ANTHROPON can, of course, readily be rendered by "people", "person" or the like.


3 Translational Approach.
a. Semantic scale (or spectrum): the semantic scale in figure 1 may be useful in determining the meaning of the Greek text. [Footnote 2 For added detail see chapter 2: "Introduction to Psalms."]
(1). Words or lexemes distributed along the entire continuum of this scale can be found in all translated books of the "LXX." What may be expected to differ from translation unit to translation unit is the relative concentration (clustering) of lexemes along the continuum. Hence each unit may be expected to show its own distinctive profile.




(p. 13)
Contextual
Renderings
--------
Stereotypes
|
Calques
--------
Isolate
Renderings

Figure 1. Semantic Scale




(p. 14)
(2). The vertical line on the scale marks an important line of demarcation between two different sorts of semantic representation. Lexemes ranging from Contextual to Stereotypes are dealt with in accord with normal Greek meanings; those ranging from Calques to Isolates are effectively assigned the meaning of their Hebrew counterparts (= so-called Hebraisms). "Normal Greek meanings" is simply meant to designate one or more meaning component(s) that fall within the usual Greek semantic range of a given word.

b. Practical steps (for FULL lexemes)
Step 1. Sketch the respective semantic range of a given Greek lexeme and of its Hebrew counterpart. By "counterpart" is meant the actual extant Hebrew text. For the translator's convenience, the electronic parallel Hebrew-Greek text of Jewish scriptures is available, which also includes Emanuel Tov's suggestions regarding the Hebrew text understood by the LXX/OG translator.
Step 2. Work out the quantitative interrelationships, to determine whether the Hebrew-Greek equation is closed, standard, occasional or rare. The aim here is simply to ascertain whether all/most/some/few occurrences of Hebrew X are equated with Greek Y. This information should assist in step 3, since stereotypes and calques are typically found in the "all/most" categories.
Step 3. Determine the place of the Greek lexeme on the semantic scale. That is to say: does it fall to the left or to the right of the vertical line.
Step 4. If the Greek lexeme falls in the range from contextual to stereotype, apply the Greek semantic range. [Footnote 3 In the use of LSJ, it may be useful to suspend, though not necessarily reject, all English glosses based solely on LXX usage. That is, if LSJ simply records a supposed meaning based on LXX/OG translational passages, the NETS translator is probably in as good or better position to draw appropriate semantic conclusions.]

(p. 15)
(a). If the semantic ranges of the respective Greek and Hebrew lexemes (see step 1. above) overlap or intersect, select the shared component of meaning, since that component probably best reflects the translator's intent.
(b). If the semantic ranges do not overlap or intersect, check for the following reasons in stated order:
(1) the translator's interpretive predilections, sometimes skewed by etymological rendering
(2) the translator's unfamiliarity or limited familiarity with the Hebrew lexeme
(3) possible differences from the MT in Hebrew text, written or read
Step 5. If the Greek lexeme falls in the range from calque to isolate, apply the semantic range of the Hebrew counterpart
Step 6. Establish whether the NRSV falls within the established semantic range of the Greek lexeme as determined by the aforementioned processes.
(a). If it does not, substitute your unmarked/default English equivalent (if one [or several synonyms]) has been selected.
(b). If it does, let stand any clearly acceptable synonym of the unmarked/default English equivalent.[Footnote 4 "Acceptable" is intended to mean that the synonymity of each word is checked independently against the Greek.]

c. Function words (bleached lexemes): perhaps even more than elsewhere, the treatment of function words must arise from the translation unit concerned.
(1) Articles: ignore if Greek = Hebrew; do not ignore if Greek does not = Hebrew for other than purely structural reasons and if the difference can be represented appropriately in idiomatic English.
(2) Prepositions: take semantically seriously only marked (i.e., non-standard) instances.

(p. 16)
(3) Singular/plural number: normally follow the Greek, but see 2. b. (2) (c) above, when purely stylistic considerations are at stake.
(4) Conjunctions: normally follow the Greek, but see 2. b. (2) (c) above, when purely stylistic considerations are at stake.

d. Transliterations and Names.
(1). Transliterated elements in the Greek should be transliterated into English, using the following system [Footnote 5 Adapted from H. W. Smyth, Greek Grammar (Cambridge, Mass., 1956), pp. 7, 9)]:

a (alpha)= al (lambda)= lf (phi)= ph
b (beta)= bm (mu)= mc (chi)= ch
g (gamma)= g/nn (nu)= ny (psi)= ps
d (delta)= dx (xi)= xw (omega)= o
e (epsilon)= eo (omicron)= oau = au
z (zeta)= zp (pi)= peu = eu
h (eta)= er (rho)= reu = eu
q(theta) = ths/[final]s (sigma)= sou = u
i (iota)= it (tau)= tui =ui
k (kappa)= ku (upsilon)= y

(2). Names should normally be treated in one of two ways:
(a).As a translation when the Greek name replaces the Hebrew name:
HebrewGreekEnglish
MizraimAigyptosEgypt
AramSyriaSyria
KushAithiopiaEthiopia

(p. 17)
(b).As a transliteration when the Greek name is (basically) a transliteration of the Hebrew name;
HebrewGreekEnglish
MosheMoysesMoyses
DawidDauidDauid
AbrahamAbraamAbraam

Of the diacritics, only initial aspiration is of relevance for NETS. When the Goettingen LXX uses diacritics, follow it; where it lacks them, follow Swete; when translating Rahlfs, follow Swete.

4. Editorial Policy
a. Footnotes.
(1). In conformity with the NRSV, annotated items that consist of a single word in the English text should be followed by small raised letter (a). Items of more than a single word should be enclosed by a pair of small raised letters (a...a) and appear in the apparatus as a-a. In the text, the closing letter follows punctuation: .a (not a.). As in the NRSV, only the lemma text (or its suggested alternative) should be italicized in the apparatus.
(2). Footnotes should be one of four kinds: [Footnote 6 It is suggested that all four kinds of notes be used with great discretion. Translators are, of course, encouraged --nay, urged--to keep fulsome notes of all kinds in their private electronic files, for use in electronic publication.]
(a) to mark a departure from the base Greek text : a+ Lord = Ra/Zi
(b) to give an alternative English translation: aOr my son

(p. 18)
(c) to mark the Greek text as being of uncertain meaning:
.a-a Gk uncertain

(d) to clarify:
aGk schoinos = a land-measure equivalent to several miles

(3). Other sigla used in the footnotes:
prpreceded by
omomitted by
+followed by
=equivalent to
i.e.that is
oralternative translation
possiblypossible translation
GkGreek
RaRahlfs
ZiZieler
HaHanhart
WeWevers
MaMargolis


If more sigla need to be introduced in given translation units, translators must clear them with the Translation Committee.

b. Chapter/verse alignments: when chapter and/or verse numbering in the Greek and therefore in NETS differs from the NRSV, the NRSV numbering should be added in parentheses: Ps 29:2(30:1) means 29:2 in NETS and 30:1 in NRSV. Normally NETS will correspond to whatever Greek text is used and NRSV to MT.

c. Introduction to individual books or translation units should include the following information:

(p. 19)
(1). Details on the Greek text edition on which the translation is based.
(2). Translational profile of items specific to the given unit, for instance, words and grammatical constructions treated as stereotypes or calques.
(3). Explanation of other particular facts and procedures that will assist the reader of NETS.


5. Miscellaneous.
a. New Testament Quotations: since NETS seeks to promote the synoptic use of the Bible, it is of some importance to harmonize, if feasible, NETS and passages that are quoted verbatim in NT texts. If, however, the NRSV itself differs in its translation, between the source text and the NT quotation of that source text, NETS should keep in mind its primary goal, namely, optimum fidelity to the source text.

b. Electronic Tools and Resources.
(1) Texts and Data. The following materials exist in electronic form and are available through the NETS project to members of the translation team. Each member will need to provide some basic information about the machine and programs available, since this may vary greatly among the team members. It will be especially difficult in some instances to combine effective use of the materials with multi-font (Greek, Hebrew, English) display; there is much to be said, despite certain inelegance, for conducting searches of the parallel Hebrew-Greek texts in ASCII (transliterated) form. It is important that the computer search for what the translator wants to find, and codes for displaying foreign fonts can be detrimental to that goal unless great care is exercised.
Team members knowledgeable about using the Internet and the World Wide Web (WWW), may obtain or view most of the following materials on
(p. 20)

the gopher at ccat.sas.upenn.edu, which is linked to the home page of Robert Kraft: http://ccat.sas.upenn.edu/rs/rak/kraft.html (a home page for the NETS project will be established on the CCAT machine in the near future). Address queries to kraft@ccat.sas.upenn.edu.
(a)
NRSV: the NRSV is availalable in electronic form, including the deutero canonical books (Apocrypha). Since the National Council of Churches retains the control of the NRSV, the NETS project will need to insure that its members are acting responsibly in using these materials for our purposes.

(b).
LXX/OG (Rahlfs), with morphological analysis: the two-volume Rahlfs handbook edition of LXX/OG, with all known typographical errors corrected, is available either by itself, or with morphological analysis (not completely verified yet). The latter is especially valuable for determining the degree of consistency of translational practice at the grammatical and syntactical levels, and can be used very effectively alongside the parallel Hebrew-Greek text (see below). Currently, the LXX/OG Morph files are available on the gopher at ccat.sas.upenn.edu, under archives of CCAT > religious texts > biblical. Table 1 contains a sample (coding is available with the files).

(c).
BHS, with morphological analysis. Similarly, the morphologically tagged MT/BHS material is also available, as well as the BHS text by itself (with or without vocalization and/or cantillation). Again, this information will prove most effective in conjunction with the Greek morphological patterns, and the basic information contained in the parallel Hebrew // Greek files. Currently, information about the BHS Morph


(p. 21)
Table 1. Habakkuk 1:2
E(/WS P E(/WS
TI/NOSRI GSM TI/S
KU/RIE N2 VSM KU/RIOS
KEKRA/COMAI VFX FMI1S KRA/ZW
KAI\ C KAI/
OU) D OU)
MH\ D MH/
EI)SAKOU/SH|S VA AAS2S A)KOU/W EI)S
BOH/SOMAI VF FMI1S BOA/W
PRO\S P PRO/S
SE\ RP AS SU/
A)DIKOU/MENOS V2 PPPNSM A)DIKE/W
KAI\ C KAI/
OU) D OU)
SW/SEIS VF FAI2S SW/ZW

Table 2. Joshua 1:1
WA\ Pc W
Y:HI^Y vqi3msj HYH
)AX:AR"^Y Pd )XRY
MO^WT ncmsc MWT
MO$E^H np M$H
(E^BED ncms (BD
Y:HWF^H np YHWH
WA\ Pc W
Y.O^)MER vqi3ms )MR
Y:HWFH^ np YHWH
)EL- Pp )L
Y:HOW$U^(A np YHW$(
B.IN- ncmsc BN
N^W.N np NWN

(p. 22)
files is available on the gopher at ccat.sas.upenn.edu, under archives of CCAT > religious texts > biblical. Table 2 contains a sample.


(d).
The Parallel Hebrew-Greek texts: probably most basic for the tasks of translating is the parallel Hebrew-Greek text, which can be searched for patterns, anomalies, etc., on either side of the evidence. Further help with interpreting the evidence can also be gained by observing the annotations on differences between the two texts that Emanuel Tov and his team have supplied in the full form of these files [Footnote 7 See Tov, A Computerized Data Base for Septuagint Studies: the Parallel Aligned Text of the Greek and Hebrew Bible. (Computer Assisted Tools for Septuagint Studies 2 = Journal of Northwest Semitic Languages Supplementary Series 1, Stellenbosch, 1986). The unannotated parallel texts are available on the gopher at ccat.sas.upenn.edu, under archives of CCAT > religious texts > biblical. Table 3 contains a sample.

(e).
Greek textual variants (for some sections): the CATSS project to encode all published Greek variants to the LXX/OG texts continues at the University of Pennsylvania, and some of the results are available or can be made available. Currently, samples may be seen on the gopher at ccat.sas.upenn.edu, under archives of CCAT > religious texts > supplemental. It should be remembered, however, that NETS translators are not encouraged to deviate from the Greek text which forms the basis for their English translation, though such deviation is allowed.

(f).
Greek Lexica: an electronic version of the full LSJones lexicon has recently been prepared, and we will attempt to add it to the available tools. The


(p. 23)
Table 3. Genesis 1:8
W/YQR) KAI\ E)KA/LESEN
)LHYM O( QEO\S
L/RQY( TO\ STERE/WMA
$MYM OU)RANO/N
--+ '' =;W/YR) <1.10> KAI\ EI)=DEN
--+ '' =;)LHYM <1.10> O( QEO\S
--+ '' =;KY <1.10> O(/TI
--+ '' =;+WB <1.10> KALO/N
W/YHY KAI\ E)GE/NETO
(RB E(SPE/RA
W/YHY KAI\ E)GE/NETO
BQR PRWI/
YWM H(ME/RA
$NY DEUTE/RA


Table 4. Jonah 1:6
<1KAI\>1
<1EI)=PEN>1
+<<1PROS>1 A'`-410-764 86{mg} Bas.N.
<1AU)TW=|>1 W 86{c} Clem.] > 86{*} Arm
    : <1AUTON>1 A'`-410-764 86{mg}
    Bas.N.
<1*TI/>1] > 106(~)
<1SU\>1] > Syh
+ <1TI/>1 106(~)
+<<1KAQEUDEIS>1 Syp
+ <1KAI>1 Syp
<1R(E/GXEIS>1]
    : <1REGXEI>1 S{*}
;
<1A)NA/STA>1 W(vid.) Chr. II 310]
    : <1ANASTHQI>1 S{c} Q{c} <2L>2'-86(mg)-613 <2C>2'(-538)-68-239
Clem.Cyr.{F}(2nd): cf. Mich. 2:10
    : <1ANASTAS>1 <2lII>2

(p. 24)

brief Newman Lexicon to the Greek New Testament is already available, and it is possible to find the "middle" Liddell-Scott from some distributors.

(g).
Hebrew lexica: it is possible that some of the information found in BDB can be made available; otherwise, the less-valuable, old Hebrew lexicon by Davidson is in electronic form.

(2) Software and basic procedures: it is difficult to provide instructions or advice that will cover all machines and situations. In general, everyone who makes use of computers will probably know how to conduct simple searches through "word processing" software or through special search programs such as BROWSE or LIST (for DOS), or "Find" on the Mac menu. Ideally, one would like to have sophisticated search software that operates on very large files and is capable of looking for multiple targets in various combinations (this AND that, this OR that, this BUT NOT that, etc.). If we share our experiences and our frustrations, many solutions will doubtless be found for our respective needs. For example, LIST is a shareware program that is relatively easy to use and relatively quick for simple searches, and thus convenient to have available; it will not do more complex tasks, but is a good place to start for DOS based people. A shareware program for Windows that permits multiple searches is also available through the Internet. Similar tools are available for the Mac, and more powerful options exist for people who can use the UNIX capabilities on network sites such as ccat.sas.upenn.edu.
As noted in describing the available data, there will be times when the ability to consult two or more different files on the same screen may be important--for example, parallel text and morph analysis,

(p. 25)
or perhaps a dictionary. This is relatively simple on the Mac or in Windows, so the use of such an interface is encouraged. Some people may even, at times, prefer to use two machines placed next to each other (e.g., a portable and an office machine), to achieve the same effects as Windows performance, or possibly to take advantage of the special capabilities of a particular machine (an IBYCUS SC, for example).
c. Other Tools and Resources.
  1. English translations. since NETS is an English translation of the Greek, other English translations such as Thomson and Brenton are always of interest, often of direct help, but sometimes to be ignored. Hard copy of these, for individual units, can be made available.
  2. Lexica: along with standard Greek and Hebrew lexica, the following work may be used with great profit: E. A. Nida, and J. P. Louw (eds) Greek-English Lexicon of the New Testament Based on Semantic Domains (2 vols.; New York, 1988).

(p. 27)

Chapter 2

Introduction to Greek Psalms

English Translation
Ancient texts, including biblical texts, have been translated since time immemorial, and the need for such work is beyond dispute. What is often less clear is the precise reading-public any given translation ought to target. Because of its widely varied audience, this is perhaps especially true for that body of literature called biblical literature. Writing specifically on the topic of Bible translations, Eugene A. Nida envisaged no fewer than three such audiences:

It is usually necessary to have three types of Scriptures: (1) a translation which will reflect the traditional usage and be used in the churches, largely for liturgical purposes (this may be called an "ecclesiastical translation"), (2) a translation in the present -day literary language, so as to communicate to the well-educated constituency, and (3) a translation in the "common" or "popular" language, which is known to and used by the common people, and which is at the same time acceptable as a standard for published materials. [Footnote 1 E. A. Nida and C. R. Taber, The Theory and Practice of Translation (Leiden, 1982), p. 31]

Though we have sought to aim our translation of the Septuagint (in some circles better known as the Old Testament in Greek) primarily at the reading public identified in

(p. 28)
Nida's second grouping (a biblically well-educated audience) in point of fact its more precise character is inevitably and in large part governed by the nature of the English version of the Hebrew Bible which we have used as our base text, namely, the New Revised Standard Version (1989). That any existing English translation of the Hebrew Bible should have been used in this manner perhaps needs some elucidation and justification. Why not, one might object, simply translate the Greek in the tradition of, say, L. C. L. Brenton (1844) and C. Thomson (1808)? The answer to this question is based, we believe, on both considerations of principle and of practicality.
Notwithstanding the (relative) independence which the Greek corpus enjoyed from its inception, it is undisputed historical fact that the so-called Septuagint is, in the first instance, a Greek translation of a Hebrew original. Consequently, whatever else the LXX (or Old Greek) may be considered to be--a repository of textual variants to the Masoretic Text, the oldest commentary on the Hebrew Bible, Holy Writ for Hellenistic Jewry--the fundamental nature of this collection is one of derivation and dependency. Due recognition of this reality, we believe, entails an important methodological directive, namely, that any translation of the (translated) Greek ought to stand in a dependent relationship to a translation of the Hebrew. Put another way, if we are to reflect the historical relationship between the Hebrew and Greek Bibles, we ought to begin by relating their translations into English in a similar manner. That the Hebrew parent text of the Greek often differed from our so-called Masoretic Text and that not all Greek translators took an identical stance vis-a-vis their source text are realities that will clearly emerge along the way in individual passages, sections and books.
Furthermore, alongside of this dictum of principle, there is an important practical consideration that leads one to the same conclusion. Surely central to the very raison d' être of a translation of the LXX (i.e., a translation of a

(p. 29)
translation) is its synoptic potential. That is to say, the users of such a translation should be able to utilize it to the greatest possible extent in a comparative study of the Hebrew and the Greek Bibles. This aim can best be realized, we believe, if English translations of the two are as closely interrelated as the two texts themselves dictate or allow, both quantitatively and qualitatively. In other words, ideally the user of our translation should be able to determine not only matters of longer or shorter text and major transposition of material, but also questions of textual and interpretational difference. We are quite aware that in the present translation that ideal has not always been reached; yet it remains worth pursuing with all one's scholarly might.
Given the above position, essentially two concrete options are open to a translator of the Septuagint: (1) one can first translate the MT into English and then make it the basis for one's translation of the Greek, or (2) one can use an existing English translation of the MT as a point of departure. Clearly the latter route commends itself as being the more practical and economical one, and it is furthermore difficult to see how one could significantly improve upon the work of the committees of scholars that produced the major English translations of the Hebrew.
The New Revised Standard Version, based as it is on the maxim "as literal as possible, as free as necessary" has been deemed eminently suitable to our purpose, not least because of the predominant nature of the Greek translations within the LXX. Consequently, in our translation of the Greek Psalms we have sought to approximate the NRSV to the extent that the Greek text, in our understanding of it, directs or permits. Two principles have guided us in choosing English version: (1) compatibility of translational approach with that of the LXX and (2) widespread use among readers of the Bible.
When our English translation differs from the New Revised Standard Version, the reason is likely to be one of four. First, the lexical choice of the NRSV to represent the

(p. 30)
Hebrew may differ markedly from that of the Greek translator, even though either independently might be regarded as an adequate translation of its source. Secondly, differences in translational approach between the translators of the NRSV and the Greek translator may have occasioned significant differences between the two versions. Thirdly, the Greek translator may have rendered a text at odds with MT, due to interpretational predilection or textual variation. Fourthly, the NRSV did not translate the Masoretic Text, but opted for some other reading. Naturally, where, in such instances, the NRSV has adopted the reading of the Septuagint, the two English translations agree, though not because their parent texts agree! As a rule such cases are annotated in the NRSV.
As explained in the NRSV's preface to the reader, its Committee has sought to eliminate masculine-oriented language to the extent that this does not violate the textual and cultural integrity of biblical passages. Inclusive third-person plurals have thus often been introduced in the NRSV where the Hebrew is thought to allow such an interpretation. Similarly, at times explicit referents have been added for the purpose of fostering a certain understanding of the original, or singular nouns have been rendered as plurals (and vice versa) for reasons of English style or usage. Our rule of thumb in all such matters has been not to modify the NRSV unless this is, in our judgment, demanded by the sense of the Greek. Consequently, inclusive third-person plurals, as well as stylistic plurals or singulars and explicit referents, have been kept wherever possible, when the Hebrew and Greek texts do not disagree. Conversely, when these two foundational texts disagree, every effort has been made to reflect such disagreement in NETS. Differently put, we have done our best to have every deviation from the NRSV count as either a difference between the NRSV and the Septuagint Psalms or between the latter and the Hebrew (or both), whatever the precise reason for that difference.

(p. 31)
In sum, the operating principle has been: Retain what you can, change what you must.
Our desire to enable the reader to make use of the NETS in synoptic manner with the NRSV, has been second only to our commitment to giving a faithful rendering of the original.

Greek Text

The Greek text which we have translated is based on, but is not identical with, Alfred Rahlfs, Psalmi cum Odis. (Septuaginta: Vetus Testamentum Graecum Auctoritate Academiae Litterarum Gottingensis editum 10; Goettingen, 1967 [1931]). We have revised this edition in the light of additional textual information (chiefly ii-v CE) and more recent study. As a matter of course all textual items enclosed by Rahlfs within square brackets, for the purpose of indicating their patently secondary nature, have been eliminated without comment. All other deviations from Rahlfs that affect an English translation have, however, been noted faithfully in a brief apparatus to our English text.

Translating a Translation

Translating an ancient text, what might be called a "permanently fixed expression of [ancient] life" [Footnote 2 J. G. Droysen apud H.-G. Gadamer, Truth and Method (New York, 1986[1960]) p. 349.] can only be described as a profoundly difficult undertaking, for not only do translators have to contend with the gulf that naturally exists between languages and with the absence of the author who wrote the piece in question, but they also suffer from the lack of native speakers of the ancient languages, who might be cajoled into giving some much-needed help. Consequently, what the modern translator is trying to do is something like starting up a one-way conversation, or a monologue that passes for a dialogue.

(p. 32)
The difficulties of the undertaking are certainly not decreased when one attempts to translate an ancient translation into a modern language. If translating is indeed interpreting and not simply reproducing, as semanticists insist it is, a Greek interpretation of a Hebrew original can be expected to reflect what the translator understood the Hebrew text to mean. The end result is therefore inevitably a biblical commentary of sorts written at a specific historical time and place by an individual person, whose understanding of the Hebrew will often have been at variance with our own, though perhaps equally viable. Clearly, at this particular point in our discussion we can properly speak of the independence of the Greek text vis-à-vis the Hebrew, albeit an independence that is circumscribed and relative. But more troublesome for the English translator of the Greek than gauging the extent and impact of this substitution of meaning are the freshly introduced problems based on partial or complete misunderstanding of the source text and/or the (to us) faulty linguistic method used to resolve perceived difficulties. Furthermore, the IOSCS has properly decided to translate the Septuagint, not in the first instance as its reading public would have read it, but rather as the ancient translators themselves presumably understood and intended it (more on this below). Needless to say, such an approach is possible only where a Hebrew or Aramaic text is actually extant or can be reconstructed with some confidence. In other words, one can only discern translator's intent where such can be contrasted with that of the source that is being translated or can be reconstructed with confidence.
Consequently, at the very outset of the undertaking, we have forced ourselves to ask the following question: What precisely is it that we should aim at translating? For the sake of argument we may begin by suggesting two alternative approaches to the Greek text. That is to say, either one can aim at rendering the meaning of the text as it was understood by the Greek-reading public (therefore, based on audience perception of its meaning), or one can seek to

(p. 33)
uncover the meaning of the Greek text from the point of view of the translator (therefore, in accordance with the translator's intent). For those who hold that one's interpretive approach to (ancient) texts should be informed by the principle of authorial intent, the choice is not difficult to make. Seemingly, all we need to do is substitute "translator's intent" for "author's intent."
Although on a theoretical level it is simple enough to make a distinction between audience perception on the one hand and translator's intent on the other, in practice such differentiation is decidedly much more difficult. Moreover, since neither authors nor translators do their work in isolation, but instead target a more or less specific public, the notion of "author's/translator's intent," to some extent, must be made to include that of "audience perception." In other words, author/translator and audience belong to the same language community. Nevertheless, we believe the distinction to have validity and, furthermore, hold it to constitute an important methodological stance for translators of the Greek, with not infrequent practical consequences. In its concrete application to the Greek text, however, the concept of translator's intent has been construed to include both what was (as best one can tell) overtly intended by the translator and also what resulted (inadvertently, one surmises) from the use of certain translational techniques (cf. isolate or etymological renderings). Obvious examples of translator's intent (in distinction from audience perception) are cases in which, due the ambiguity inherent in the Greek, only the syntactic relationships (e.g. subject/object) of the Hebrew can guide the English translator to the Greek translator's intended meaning.
As a guide in our translation of the Greek Psalms, a variety of steps and procedures have been developed.

Semantic Scale or Spectrum

From the English translator's perspective it may be useful to think of the lexical meaning of words in terms of a spectrum

(p. 34)
or a semantic scalen (see fig. 1 on page 13). Whatever the precise image, it is important to view it as a genuine continuum extending from fully contextual segments of meaning at the one end of the scale to completely isolate segments at the other. That is to say, at the one end of the scale (or spectrum) we find meaning as defined primarily by context and, at the other, meaning that is wholly restricted to the individual word or phrase. In the latter case, not surprisingly, such isolated segments may exist in tension with or even stand in contrast to their context. Indeed, it is obvious, we believe, that the more a given translator retreats into an isolate mode of translation (i.e. a translation which pays little attention to context), the less he or she is able to perform the translator's duty to the reader, namely, to convey the total linguistic meaning of the source text.
For obvious reasons, transliterations, including most names, have not been included in our semantic scale since these simply carry the meaning of the source (Hebrew) text and, as such, are not subject to translation into English anymore than they were translated into Greek. Visually, one may picture them as being to the right of isolate end of the scale.
As will be noted presently, the vertical line on the scale represents a crucial semantic demarcation, since lexemes placed to the left are governed by their normal Greek semantic range while those to the right are effectively governed by the semantic range of their Hebrew counterparts.

The Greek Translator of the Psalter

Though there can be little doubt that the full range of the semantic scale is represented in all translational units of the so-called Septuagint, naturally, not all units would show the same distributional profile. In addition to our earlier emphasis on the LXX or Old Greek as a corpus, one has to recognize not only that particular units originated at different dates and in various settings, but also that these were produced by distinct individuals with at times widely diverse approaches to their task of translating. Two factors that

(p. 35)
have no doubt exercised a direct influence on a book's profile are its degree of literalness and its relative chronological place within the corpus. That is to say, later books may be expected to contain a greater number of calques than earlier ones.The book of Psalms is no exception in this and it too has its own particular profile.
But since lexical items on the isolate end of the scale cause the English translator the most grief, it is these that inevitably (and somewhat unfairly to the Greek translator) receive one's most attention. Nonetheless, it can readily be demonstrated that the Psalter, like other "more literal" books of the Septuagint, features a relatively high number of isolate renderings, and in more general terms, contains many items that tend toward the isolate end of the semantic scale. To that extent the translator often clearly shows interest in the formal detail of the Hebrew at the expense of communicating its meaning. Indeed, it could be argued that the Greek translator's starting point is at times almost directly opposite to that advocated by modern semanticists. Not atypically, he begins with the individual word and its supposed core meaning rather than with meaning as it is conveyed by the interrelationship of words in syntactic units. Differently put, one could say that he at times maximizes the individual lexeme and minimizes the context in which it stands, rather than vice-versa. In illustration, one may point to the titles of individual Psalms. Bolder than modern translators of the Hebrew, the Greek translator evidently felt obligated to render into Greek whatever he found in his Hebrew text. (Consequently, transliterations are few.) When he was thus bent on rendering into Greek what he found in his Hebrew, the result of his efforts can be laconic or may even be gibberish at times . What did he think his text meant, for example, when he translated LMNCX, which is now commonly understood to mean "to the leader/conductor," by the phrase EIS TO TELOS, which is seemingly viewed as an isolate phrase ( NCX BDB: "be pre-eminent, enduring") and has to be glossed something like

(p. 36)
"unto/for the end/completion"? Admittedly, the phrase in question stands in minimal context, as do others in the titles, but the translator's focus seems too exclusively on its supposed isolate meaning. Yet, on the positive side of the ledger, one may note that although M&KYL is rather mechanically translated by SYNESIS, the latter is interpretively put in the genitive. That is to say, in the translator's perception the entire Psalm in question "pertains to understanding."
But if the Greek translator's approach to his Hebrew text is all too often indeed what we have alleged it to be, namely, an approach indicative of a one-word-one-meaning understanding of language, what are the options of the modern translator who seeks to uncover his intent and transmit his understanding? The answer seems disturbingly clear: she or he, in such instances, is basically locked into the same or similar approach. When, for example, the Greek with painful consistency translates HPC by THEL- (thus creating a stereotype), is the English translator not compelled to translate the latter consistently by the same English word, or more properly by a narrow range of English synonyms, even though this violates some central principles of modern semantics? Clearly worse than such stereotypes are instances in which the translator (because he failed to understand his Hebrew text) resorted to an etymological translation or rather to a presumed isolate meaning of a Hebrew root. Typically, such renderings stand in some tension to their immediate linguistic environment. Take, for example, 4:7 (4:6) ESEMEIOTHE EPH EMAS TO PHOS TOU PROSOPOU SOU, KYRIE. For the first word MT has the root NSH an orthographical variant of N&H ("raise") but not recognized as such by the translator. In his perplexity he connects it instead with NS ("sign"), which in 60(59):6 he translates as SEMEIOSIS, and so arrives at SEMEIOO ("signal") in 4:7. But how should one gloss this line in English? Probably, "The light of your face has been marked upon us, O Lord!" is about the most one is permitted to do. That Greek readers may well have contextualized the offending

(p. 37)
item in increasing manner as time progressed is probable but irrelevant for the modern translator.
The modern English translator, we believe, has little option but to follow in the footsteps of the ancient Greek tranlator when it comes to questions of what to translate or what not to translate. Accordingly, we have normally translated into English whatever the Greek translator translated into Greek, even though we cannot always be certain of precise meaning. Thus the obscure Hebrew SLH is translated by Greek DIAPSALMA, which in turn is translated into English even though the meaning of the latter is not certain. An exception is PSALMOS (psalm), seemingly a technical term referring to a song sung to the accompaniment of a stringed instrument (harp or lyre), which was then plucked (PSALLEIN) with the fingers rather than struck with a plectrum. A second exception is Hades, the translation of Hebrew Sheol, since the former cannot be equated with English "hell," irrespective of time period.
The above approach has also meant that we have felt compelled to follow the Greek text both in its intelligibility and (though reluctantly) in its relative unintelligibility. Therefore, we have sought to offer contextualized sense as well as isolate sense, with a full range of variation between these poles. Equally, we have left untranslated what the Greek translator left untranslated. Thus the transliterations that our text contains.

Categories of Renderings

As noted above, the semantic scale we have suggested extends from fully contextualized renderings on the one hand to isolate renderings on the other. At the risk of obscuring its nature as a continuum, we will now attempt a somewhat further delineation, based on what we find in the Greek Psalms (but equally applicable to other translation units):
  1. The choice of the Greek word or lexeme may be based primarily on contextual requirements and, as a result, no standard (a term intended to refer exclusively to frequency

    (p. 38)
    of occurrence, not primacy of meaning) Hebrew-Greek equivalency is established. For obvious reasons, this kind of rendering occurs typically but by no means exclusively on the verbal (i.e. the least concrete) side of the language and less often on the nominal (i.e. the most concrete) side. For example, (LC (BDB: "rejoice/exult") which occurs four times in Psalms, is translated by three quite distinct Greek verbs: AGALLIAOMAI ("rejoice") and KATAGELAO ("deride"), KAUCHAOMAI ("boast"). On the nominal side, one might cite (Z (BDB:"strength/might") which is often translated by words signifying power or strength, such as DYNAMIS (20x), ISCHYS (4x), KRATOS (4x) but also AGIASMA ("sanctity"), AGIOSYNE ("holiness"), AINOS ("praise"), BOETHOS ("helper"), DOXA ("glory") and TIME ("honour"). It is to be borne in mind, of course, that terms such as "verbal" and "nominal" are to be taken more in a semantic than a morphological sense. That is to say, semantic classifications into objects, actions/events, attributes/characteristics and relationals are of greater importance than word classes based on formal characteristics.
    Not unexpectedly, in numerical terms the majority of renderings in the Psalter fall into this contextualized category.
  2. The choice of Greek lexeme may be based less on contextual sense than on perceived Hebrew-Greek verbal equivalency. The result is a combination of standard and non-standard equivalents. That is to say, though a given Greek lexeme is established as an unmarked or default rendering, exceptions do appear. Thus M$P+ (BDB: "judgment") is almost always glossed by KRIMA ("judgment") (41x) or KRISIS ("judgment") (21x) and only once each by DIATAXIS ("arrangement"), DIKE ("right"), and PROSTAGMA ("decree"). Needless to say, the line of demarcation between categories 1 and 2 is a fluid one and can, moreover, best be drawn on a case by case basis.
    We include in this category of renderings so-called closed Greek-Hebrew equations, which, however, are in all

    (p. 39)
    their individual occurrences fully concordant with the context, due to the high degree of semantic overlap between the given Hebrew and Greek counterparts.
  3. The choice of Greek lexeme may be based primarily on the perceived meaning of isolated words and results a stereotyped equivalent. That is to say, not only is the established Hebrew-Greek equivalence rarely if ever departed from, but more importantly in some of its uses the Greek word or phrase stands in tension with its context. In fact, it is precisely the latter characteristic that distinguishes a stereotype from other standard or closed Greek-Hebrew equations which, however, fit smoothly into all their contexts.
    Since stereotypes clearly arise out of a perceived one-to-one Hebrew-Greek equivalence, it comes as no surprise that they share certain features with what we have termed isolate translations.
    In our perception, a true stereotype, in which the semantic range of the Hebrew word is adjusted to fit that of its Greek counterpart, exists only for its creator. That is to say, a given translator may create a stereotype (due to his strained use of lexemes) but if such an item becomes fully contextualized, through continued use within the Greek reading/speaking-community, it turns into a calque, i.e., a Greek word with a Hebrew meaning, so to speak (see further below). As noted above and true to the meaning of the term, a stereotype (unlike a calque) smacks of woodenness and strained use. Items in the Psalter include PIKRAINO ("embitter"), YPOMONE ("endurance") + verbal cognates, THELO ("want") + nominal cognates, EIS TELOS ("completely"), NEPHROS ("kidney") etc.
    To the extent that one can speak of points on a continuum, it is from this point onward on the semantic scale that the Hebrew semantic range of a given lexeme should be regarded as dominant, as has been noted above. That is to say, whereas a stereotype may be said to carry the meaning of the target language (Greek), a calque must effectively be

    (p. 40)
    assigned the meaning of its counterpart in the source language (Hebrew).
  4. Sometimes the selected Greek lexeme is a calque, that is to say, an item moulded by prior linguistic usage rather than one shaped, on an ad hoc basis, by the individual translator. In other words, we perceive a calque to be a stereotype that has been acclimatized to the host language. Consequently, the semantic range of a calque is fully concordant with its context and may thus be viewed effectively as having regained its Hebrew meaning. An obvious example is DIATHEKE, a word which no doubt began its life in Jewish usage as a stereotype, but which in the Psalter is clearly a calque and therefore appropriately glossed or translated by "covenant" (in the sense of Hebrew BRYT rather than extra-biblical Greek "testament"). Others are deictic IDOU ("behold"), recitative LEGON ("saying"), SOTERIA ("salvation") + cognates, ELEOS ("steadfast love"), PAROIK- ("sojourn"), KYRIOS ("Lord"), and other words typically of central religious importance to the translator. On the more structural side we have the ways of dealing with the infinitive absolute etc.
    As is clear from the above delineation, the essential difference between a stereotype and a calque is perceived to be one of degree of integration into the target (Greek) language. In practical terms, however, the distinction is not always easily made, mainly for two reasons: (1) the early stage of development from stereotype to calque may pre-date our written corpus of literature (this is particularly true in the oldest portion of our corpus, namely, the Greek Pentateuch) and (2) the positing of such development entails questions about the relative chronology of the books or translational units within our corpus. For example, one more than suspects that important religious terms such as DIATHEKE ("covenant") and NOMOS ("law") had already received calque status by the time the translation enterprise commenced. For example, the determination that THELO ("want") + cognates is a calque rather than a stereotype in

    (p. 41)
    the Psalter means that the translator of Psalms was the first to make it the standard and rigid equivalent of H.PS. .
  5. The choice of Greek lexeme may be based purely on the perceived meaning of individual words in (virtual) semantic isolation, resulting in isolate translations. Typically, etymology of the Hebrew word plays a central role. By way of illustration, in Ps 7:7 Hebrew (BRH (BDB: "overflow, arrogance, fury") is translated, via (BR = PERAN ("across"), by PERAS ("end")--- a word representing entirely different Hebrew lexemes elsewhere in Psalms----and the Greek line as a whole should be glossed something like: "Be exalted at the ends (deaths) of my enemies." A glance at Ps 38:5 (39:5 MT) confirms that PERAS within the Psalter can indeed refer to the terminus of human life. Or should one translate with Thomson "Exalt Thyself in the borders of mine enemies"? Whatever the case, as a result of the translator's histrionics his text, on the one hand, means something quite different from MT and, on the other, is scarcely intelligible in Greek. Of course, there can be little doubt that contemporary as well as succeeding readers of the Greek Psalms, in all such cases would maximize context and minimize the individual lexemes; but a modern translator who seeks to uncover the Greek translator's intent is more restricted.
    As has been noted repeatedly, the semantic spectrum or scale we have tried to portray is wholly continuous. Consequently, the lines of demarcation we have suggested should not be overdrawn.
    Given the difference in vocabulary stock, between Hebrew and Greek, it is not in and of itself surprising to find a relatively large number of one-to-many equations. Thus, for example, ANTILEMPTOR and BOETHOS translate seven Hebrew words each. It would be incorrect to suggest, however, that one-to-many equations obtain exclusively in the Greek-Hebrew relationship. The reverse also occurs, particularly on the verbal side of the language (including the semantic category of "events"). Hebrew $WB, for example, is

    (p. 42)
    rendered by eleven Greek lexemes. Nevertheless, in view of our task at hand, we are interested in the process solely from a Hebrew to Greek perspective, and in the semantic shifts it entails.
    Clearly, in creating Hebrew-Greek equivalences, at times the translator was motivated not simply by linguistic considerations but by ideological ones. Interestingly, for instance, when (NH and $M( have the deity as subject they are given exclusive Greek equivalents, EISAKOUO/EPAKOUO and EISAKOUO respectively. Since such equations occur in (virtually) identical contexts, we have felt free to adopt a single English gloss, in the cases in question.

Problems of Grammar

What has been argued with respect to words or lexemes in the Greek is also applicable to certain details of grammar. A number of these deserve special attention.
  1. One of the most problematic of these is the tense-aspect equation. Quite clearly, the Greek translator of Psalms worked with a set of unmarked or default equivalences. Thus the Hebrew prefixed form equals future indicative, the affixed form equals aorist indicative and the participle equals present indicative. Fortunately, though these equations constitute his default mode, apparent concern for contextual sense did not allow him to adhere to them rigidly. When all is said and done, however, infelicitous and even awkward use of Greek tenses is not an infrequent occurrence. An arbitrary example must suffice. NRSV renders 6:7(6):

I am weary (= suffixed form) with my moaning;
every night I flood (= prefixed form) my bed with tears;
I drench (= prefixed form) my couch with my weeping

The Greek tenses, however, should be translated (apart from other variations):

(p. 43)
I was/became weary (= aorist) with my moaning;
every night I will wash (= future) my couch;
with tears I will (= future) wet my bed

Then the Greek continues with more aorists. Even though one may readily grant (on the basis on recent scholarship) that aspect is more fundamental to the Greek verb than tense and that Greek tenses do not have completely fixed temporal reference, it scarcely follows, in our view, that temporal reference can be ignored at will. There can therefore be little doubt that the abrupt shifts in the use of tenses, occasioned by the translator's system of defaults, has given rise to a degree of awkwardness in the Greek. Of course, Greek readers other than our translator may well have contextualized and thus have glossed over such infelicities, but as we have suggested above, the English translator of the Greek should try to follow in the footsteps of the Greek translator rather than look through the eyes of his readers. Thus, in our English translation we have sought to reflect the usual temporal sense of the Greek tenses, even to the point of converting present perfects of the NRSV into simple pasts, when the Greek text features an aorist. Understandably, such an approach at times produces less than elegant English style, but in our view it reflects at least in some measure the less than elegant style of the Greek translator.
  1. A similar and perhaps equally thorny problem involves the use of prepositions in a translation. Since their primary linguistic function is that of indicating relationships between constituents of a sentence, semanticists have included them under the heading of relationals (in distinction from objects, actions/events, and characteristics). At the same time, however, prepositions, though clearly semantically bleached, are not lexically empty (as is the case, for example, with case inflections). It would seem, therefore, that because of their dual role as relationals (structural morphemes) and lexemes ([bleached] lexical morphemes)

    (p. 44)
    they cause particularly serious problems when treated by a translator as (virtual) isolates. All readers of the Septuagint will no doubt be able to supply their own lists of bloopers. Clearly, for the English translator faced with such infelicities in the Greek there are no easy and obvious solutions. We have been guided by a number of considerations. Though prepositions do have some lexical content (and thus might be treated like full lexemes), it is equally true that they tend to have an extraordinarily wide range of possible meanings (in comparison with full lexemes or words). Consequently, one assigns unmarked/default meaning to them at some risk. Moreover, since prepositions typically function as complements of verbs, their meaning is integrally bound up with verbs. From that perspective, it is difficult to see how they could be given any kind of independent semantic status, even from the Greek translator's perspective. In many cases, therefore, their misuse has been judged a matter of style rather than one of semantic difference. As a result, we have taken note of only those instances which mark a fairly clear semantic shift in the Greek away from the Hebrew. Differently put, when a translator clearly deviates from his default Hebrew-Greek equations, one may with some confidence conclude that he is interpretively at work.
  2. Another primarily structural item is conjunctive KAI, the standard gloss for Hebrew W. Understandably, the NRSV renders the latter often by "and" but also in a number of other ways. Since one is dealing here with an item that, like prepositions, has minimal lexical content, we have not changed the NRSV when it quantitatively corresponds to the LXX. At times, however, the NRSV has chosen (for reasons of English style) not to represent Hebrew W. Again in such cases we have followed the NRSV wherever the LXX simply uses KAI, lest we suggest that MT and LXX are textually at variance in such cases. When on the other hand, the LXX deviates from its default mode by using a marked equivalent, such as E("or") or DE ("but"), we have

    (p. 45)
    revised the NRSV accordingly, since the item has been counted as an interpretive variation from the NRSV, though it quantitatively equals MT. Naturally, wherever MT and LXX are quantitatively at odds, we have revised the NRSV. In sum then, these have been our two guiding principles: (a) when the MT and LXX quantitatively disagree we have reflected this difference, and (b) when the MT and LXX quantitatively agree against NRSV, we have revised the latter only if LXX uses a nonstandard (i.e. marked) equivalent.

Practical Steps


On the most practical level of translating the Greek text into English, on the basis of the NRSV, a number of procedural steps have been found useful in ferreting out the meaning intended by the ancient translator (listed above on pp. 14-15).
In conclusion, as should be clear from the above discussion, we have not attempted to produce a rigid (or even semi-rigid) one-to-one English translation of the Greek, for the simple reason that such an approach, we believe, would fly in the face of the nature of language. Semantic accuracy and so-called literalness are not to be linked as cause and effect. Thus, our general approach and degree of dynamics in translation on the hand dovetails, as far as we have deemed responsible, with that of the ancient Greek translator, and on the other hand is suitably concordant with the NSRV. In revising the latter to reflect the Greek, we have of course allowed English synonyms to stand wherever warranted by the Greek, though a special effort has been made within individual psalms to reflect the verbal echoes of the original.

(p. 46)

Select Bibliography:

Standard lexica and grammars

Barr, J. The Semantics of Biblical Language. Oxford, 1961.
Biguenet, J., and R. Schulte. The Craft of Translation. Chicago, 1989.
Brenton, L. C. L. The Septuagint Version of the Old Testament, according to the Vatican Text, Translated into English; with the Principal Various Readings of the Alexandrine Copy. 2vols. London, 1844.
Chamberlain, G. A. "Method in Septuagint Lexicography." Pp. 177-91.in Uncovering Ancient Stones. Essays in Memory of H. Neil Richardson. Edited by L. M. Hopfe. Winona Lake, 1994.
Kraft, R. A. (ed) Septuagintal Lexicography. (Septuagint and Cognate Studies 1.) Missoula MO, 1972
Louw, J. P. Semantics of New Testament Greek. (SBL Semeia Studies.) Atlanta GA, 1982.
Louw, J. P., and E. A. Nida (eds.) Greek-English Lexicon of the New Testament Based on Semantic Domains. Second edition. 2 vols. New York, 1989.
Metzger, B. M. , R. C. Dentan, and W. Harrelson. The Making of the New Revised Standard Version of the Bible. Grand Rapids, 1991.
Nida, E. A. and Louw, J. P. Lexical Semantics of the Greek of the New Testament. (SBL Resources for Biblical Study 25.) Atlanta GA, 1992.
Nida, E. A. and C. R. Taber. The Theory and Practice of Translation. Leiden, 1982.
Rahlfs, A. Psalmi cum Odis. Setuaginta: Vetus Testamentum Graecum Auctoritate Academiae Litterum Gottingensis editum 10. Goettingen, 1967 (1931).

(p. 47)
Schulte, R. and J. Biguenet. Theories of Translation: An Anthology of Essays from Dryden to Derrida. Chicago, 1992.
Silva, M. Biblical Words & their Meaning: An Introduction to Lexical Semantics. Revised edition. Grand Rapids, 1994
Smyth, H. W. Greek Grammar. Cambridge, Mass., 1956.
Thomson, C. The Holy Bible, Containing the Old and New Covenant, Commonly Called the Old and New Testament. 4 vols. Philadelphia, 1808.
Tov, E. "Greek Words and Hebrew Meanings." Pp. 83-125 in Melbourne Symposium on Septuagint Lexicography. Edited by T. Muraoka. Society of Biblical Literature Septuagint and Cognate Studies 28. Atlanta, 1990.
Tov, E. "Three Dimensions of LXX Words," Revue Biblique 83 (1976) 529-44.

(p. 48) [Blank]
(p. 49)

Appendix 1

Statement of Principles


1. The title of the projected work is A New English Translation of the Septuagint and the Other Old Greek Translations Traditionally included under that Title (herein abbreviated as NETS).
2. The International Organization for Septuagint and Cognate Studies (IOSCS) is the primary sponsoring organization for this project. Grant applications, preliminary and final publications, etc., will be prepared in the name of or on behalf of the IOSCS.
3. For the purposes of NETS, the term Septuagint is understood to be exemplified by, but not in all respects (e.g. exact text, detailed contents) congruent with, Alfred Rahlfs's Septuaginta (1935).
4. Translators will have native or close-to-native fluency in standard English.
5. NETS will normally translate the best available editions. Footnotes will indicate significant semantic departures from the base text.
6. 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. Appropriate footnotes will inform the reader.
7. Since much of the Septuagint derives from the Hebrew Scriptures, it is important to reflect that dependence as consistently and comprehensively as possible for the English reader. For this reason, NETS consciously attempts to employ the wording and approach of a standard modern English translation of the Hebrew Bible in

(p. 50)
situations in which the Greek understands the Hebrew text in the same way as the English. The New Revised Standard Version (NRSV) has been chosen to provide this English base. Where the Greek does not correspond to the Hebrew as understood by the NRSV, translators will make every reasonable effort to represent the differences fully and accurately.
8. The target audience of NETS closely approximates that of the NRSV. Publication in both electronic and print formats is envisioned.
9. The organizational structure of the project will consist of (1) an administrative committee, and (2) a translation committee.
9.1 The president of IOSCS will chair the administrative committee and the Treasurer of IOSCS will serve as financial officer for NETS. Funds specifically raised or earmarked for NETS will be held and administered separately from general IOSCS revenues. The administrative committee will annually prepare an audited statement of NETS funds and submit it to the annual meeting of the IOSCS. Because NETS is primarily sponsored by the IOSCS, the IOSCS Executive Committee will play a major role in the administration of the NETS project.
9.2. The translation committee will be chaired by a member of IOSCS other than the President. The carrying out of this project will be entrusted to the translation committee. If the position of translation committee chair falls vacant, a new chair will be appointed by the IOSCS on the recommendation of the remaining members of the translation committee.
10. An advisory board of senior scholars not actively engaged in the project, whose advice may be sought when deemed appropriate, will be appointed jointly by the administrative and translation committees.


(p. 51)

Appendix 2

Selected Psalms

(Note to the electronic reader: these psalms have been updated to the most final rendition).

Psalm 4


  1. Unto completion, among psalms. An Ode
    for Dauid.

  2. (1) When I would call, the God of my righteousness listened to
       me!
    You gave me room when I was in distress.
    Have compassion on me, and listen to my prayer.

  3. (2) How long, you people, will you be dull-witted?a
    Why do you love vanity, and seek after falsehood?

    Interlude on strings

  4. (3) But know that the Lord made his devout to be
       marvelous;
    the Lord will listen to me when I cry to him.

  5. (4) Become angry, but do not sin;
    speak in your hearts and on your beds be prickedb.

    Interlude on strings

  6. (5) Offer a sacrifice of righteousness,
    and hope in the Lord.

  7. (6) There are many who say, "Who will show us
       good things!
    The light of your face was cmade a signc upon us, O
       Lord!"
  8. (7) You put gladness in my heart;
    because of the season of their grain and wine and oil
       they multiplied.

  9. (8) I shall both lie down and sleep in peace;
    for you alone, O Lord, settled me in hope.



aOr hard-hearted | bPossibly stunned into silence | c-cPossibly stamped



Psalm 7


  1. A Psalm for Dauid, which he sang to the
     Lord because of the words of Chusi son of Iemeni.

  2. (1) O Lord my God, in you I hoped;
    save me from all my pursuers, and deliver me,
  3. (2) or like a lion they will seize me,
    with no one to redeem or save.

  4. (3) O Lord my God, if I did this,
    if there is injustice in my hands,
  5. (4) if I repaid with evil those who were doing me
       wrong,
    then may I afall away empty froma my
       enemies;
  6. (5) mayb the enemy pursue and overtake me,
    trample my life to the ground,
    and make my glory pitch a tent in the dust.

    Interlude on strings

  7. (6) Rise up, O Lord, in your wrath;
    be exalted cin the bordersc of my enemies;
    andd wake up, O mye God, because of the decree
       which you issued.
  8. (7) The congregation of the peoples shall be gathered
       around you;
    and abovef it return on high.
  9. (8) The Lord will judge the peoples;
    judge me, O Lord, according to my righteousness
    and according to the innocence that is in me.

  10. (9) O let the evil of sinners be brought to an end,
    but you shall direct the righteous person;
    God is one who tests hearts and kidneys.g
  11. (10) Right is my help from God,
    who saves the upright in heart.
  12. (11) God is a righteous judge,h
    not bringing on wrath every day.

  13. (12) If you do not turn back, he will make his sword
       gleam;
    he bent his bow and prepared it;
  14. (13) and with it he prepared his implements of death;
    he forged his arrows ifor those that are being burnt.i
  15. (14) See how they were in travail with injustice;
    they conceived toil,
    and brought forth lawlessness.
  16. (15) They dug a pit and cleaned it out,
    and shall fall into the hole they made.
  17. (16) Their toil shall return upon their own heads,
    and on their own pates their injustice shall descend.

  18. (17) I will give to the Lord the acknowledgement due to
       his righteousness,
    and make music to the name of the Lord, the Most
       High. I am here



a-aProbably be totally ineffective against | bPr then = Ra | c-cPossibly at the death | dOm = Ra | ePr Lord = Ra | fPossibly on behalf of | gI.e. emotions | h+ both strong and patient = Ra | i-i Gk uncertain



Psalm 20 (21)

  1. Unto completion. A Psalm for Dauid.

  2. (1) In your might the king shall rejoice, O Lord,
    and in your salvation he shall exult greatly!
  3. (2) You gave him his heart'sa desire,
    and did not deprive him of the wish of his lips.

    Interlude on strings

  4. (3) For you anticipated him with blessings of kindness;
    you set a crown of precious stone on his head.
  5. (4) He asked you for life, and you gave it to him--
    length of days forever and ever.
  6. (5) His glory is great through your saving act;
    glory and majesty you will bestow on him.
  7. (6) You will give him blessing forever and ever;
    you make him glad with joy through your presence.
  8. (7) For the king hopes in the Lord,
    and through the steadfast love of the Most High he
       shall not be shaken.
  9. (8) bMay your hand be found for all your enemies; b
    may your right hand find all those who hate you.
  10. (9) You will make them like a fiery oven
    at the time of your presence.
    The Lord will confound them in his wrath,
    and fire will devour them.
  11. (10) You will destroy their offspring from the earth,
    and their children from among humankind,
  12. (11) because they turned evils against you,
    they devised a plan, which they could never carry
       out.
  13. (12) For you will cplace them as a back;c
    din your remnants you will prepare their faced.

  14. (13) Be exalted, O Lord, in your might!
    We will sing and make music to your sovereignty.



asoul's = Ra | b-bOr May all your enemies feel your hand | c-cProbably turn them to flight| d-dPossibly you will make them face your survivors.