I purchased Muraoka's Greek-English Lexicon of the Septuagint (2009) in Rome at ISBL. It is superbly bound and printed, as we have come to expect from Peeters of Leuven. We owe Muraoka a huge debt of gratitude for this lexicon. GELS is an indispensable research tool, and is likely to remain so for the foreseeable future. Muraoka thanks one of his students, Max Rogland, who improved the author’s English “in countless places” (XVI).
GELS does not replace LEH, nor LSJ + LSG, nor
GI. It is one starting point among many if the goal is to obtain an overview of
the senses a particular Greek lexeme was given in antiquity, in the Septuagint
and beyond.
I begin this review with a fundamental
question that relates, not to this lexicon of the Septuagint in particular, but
to all such lexica. Does it make sense to compile a lexicon of the Septuagint? The
Septuagint is not a linguistic unity, given that its component parts, from the
Pentateuch to 4 Maccabees and the Aquila-like translation of Qohelet, span at
least three and a half centuries, and given the heterogeneity of its contents.
“The Septuagint” as defined by Rahlfs and taken
over by GELS is not a religious unity either. It does not correspond to a delimited
corpus of authoritative writings in antiquity. The canon of “the Old Testament,”
even “the Old Testament” + the “read” extra-canonical literature as defined by
Athanasius in his 39th Festal Epistle (367 ce) is shorter. Athanasius excluded, though he could not
have been unaware of their existence, 1-4 Maccabees and Psalms of
Solomon.
In point of fact, “The Septuagint” as Rahlfs
delimited it includes, in a diverse arrangement, the known contents of Codex
Alexandrinus (ca. 375-400 ce) in the
sense of the included books beyond those of strictly Christian origin, with Psalms of Solomon, according to an index, appearing in an
appendix, no longer extant, after a New Testament, extant, which concludes with 1
Clement (lacking 57:7-63); 2 Clement (up to 12:5a), Philemon, and Revelation (James,
1-2 Peter, 1-3 John, and Jude appear earlier, placed between Acts and Romans).
Much of Rahlfs’ Septuaginta consists of
translation Greek along an immense continuum in terms of translation technique,
from very stilted to idiomatic. Occasionally, the translators resorted to sheer
transliteration. Quite a bit of LXX gives the impression of being a
word-for-word interlinear – whether or not it was – such that translation
equivalents were chosen one word at a time, with an eye not for their
compatibility with one another, but for mapping the source language in terms of
a goal of strict concordance. Other examples of translation Greek the LXX
contains, such as OG Isaiah, Proverbs, and Ben Sira, give a different
impression, such that a sense-for-sense rather than a word-for-word translation
is the habitual if not the universal result. Still other components, Wisdom
of Solomon and 4 Maccabees, for example, are not translation Greek
at all.
From a linguistic perspective,
the problem with Rahlfs’ Septuagint is not what it includes, but what it
excludes. It would make more sense to compile a lexicon of Jewish literature
and documents in the Greek language, including (for example!) the Hellenistic
Jewish authors found in JSHRZ, Philo, Josephus, and the New Testament, in
brief, from the earliest texts down to an arbitrary cutoff point of some sort,
say, the remains of kaige-Theodotion, Aquila, and Symmachus – “the
Three” form a corpus of translation Greek deriving from Hellenistic Judaism of no
less interest than the Greek version(s) of the particular set of books GELS
limits itself to.
In addition, a work like 1 Enoch in Greek is as
much a part of the corpus of translation Greek from Hebrew and Aramaic sources as
is “the Septuagint” in the sense of the two Greek versions of Daniel and the
Greek version of Judith printed in Rahlfs. While reading along in GELS, the
student of the Letter of Aristeas, 1 Enoch, the Testament of Moses, what have
you, will come across a discussion of a lemma and think, “I know the lemma and
the specific usage noted in GELS as first attested in “the LXX” from my corpus
of specialization.” It is possible that the lemma and usage in question occurs
in “the LXX” as in Psalms of Solomon or 4 Maccabees, but earlier
still, for example, in the works already mentioned, Aristeas the Exegete, Alexander
Polyhistor, or Ezekiel the Tragedian. So far as I can tell, this vast corpus,
parts of which are now available online thanks to Ken Penner and his team (The Online Critical Pseudepigrapha), was not
a systematic term of comparison in the preparation of GELS.
Secondly, since so much of “the Septuagint”
is translation Greek, it makes sense to follow in the footsteps of Schleusner,
the Gesenius, as it were, of Septuagintal lexicography, and point out LXX
translation loci that depend on a vocalization and/or a consonantal text at variance
with MT. This is a feature of LEH, though LEH merely scratches the surface of
the subject matter.
Once in a long while, GELS characterizes an
occurrence of a Greek lemma in terms of its Hebrew Vorlage. But not very
often. Muraoka is a fine Hebraist and Septuagintalist and thus could have (and
already has, elsewhere) provided enlightenment in many instances. We learn from
footnote 35 of the lexicon’s preface that Muraoka hopes to publish a separate
volume that deals with the subject in a comprehensive manner. For now, the
reader must use Muraoka 2002 alongside of Muraoka 2009 if the traditional
preoccupations of the study of the Septuagint - text-criticism, translation
technique – are of interest.
To get a sense of GELS, one may compare the beginning
of LEH with the beginning of GELS.
Coverage. From ἆ
to ἀγαθόω, LEH introduces
24 lemmata; GELS 21. Lemmata discussed by both: 18. In other words, LEH
discusses 6 lemmata GELS excludes from consideration (transliterations); GELS
includes 3 lemmata LEH excludes (adjectives based on PNs). Neither LEH nor GELS
include PNs and GNs.
In the same alphabetical range, HR lists 37
lemmata; separately, 170 PNs and adjectives built from them. The discrepancy
with GELS and LEH is caused by the fact that HR aims to be a comprehensive
reference tool. It gives an account of the Wortschatz of the Septuagint
as attested in Codices A, B, S, the standard critical edition of the day (the
Sistine), and also, in Aquila, Theodotion, Symmachus, “Hebrew,” etc. (listed
separately). Furthermore, HR includes (like LEH) transliterations.
GELS' gaps in coverage are disappointing, at least for
someone who is used to working with a more comprehensive reference tool like HR
or, in a cognate field, BAGD. On the other hand, HR does not cover Psalms of
Solomon (unlike LEH and GELS).
Statistics. LEH offers statistics. GELS does not.
Bibliography. LEH and GELS provide references. Neither, of course,
provides an exhaustive bibliography. The partial lists complement one other.
Accuracy. Both LEH and GELS needed to pass through the hands of a detail-obsessed
editor before going to print. For example: ἆ. According to LEH 1992 [2003 is unavailable to me;
perhaps the error was corrected], it occurs 6 times. But it doesn't. It occurs
4x in Codex B (Judg 6:22 [bis]; 11:35 [bis]), and 2x in Codex A (Judg 6:22 [bis]).
According to GELS, ἆ occurs in Judg 11:35
[Codex] A. But it doesn't. It occurs in Judg 11:35 Codex B (bis). The question
must also be asked: is ἆ a ghost lemma? It’s
possible. HR (alternatively?) lists ἀά, a
transliteration, in effect, of אהה (the Vorlage).
Another example: under GELS ἄβατος, we read: “Cf. ἀβατάω.” But we should read: “Cf. ἀβατόω.”
LEH lists ἀβατόομαι (i.e., ἀβατόω) as a neologism. That is
compatible with LSJ and Rocci, which list ἀβατόομαι/ἀβατόω as occurring only in the
Septuagint. GELS implies that the verb is attested elsewhere prior to the
Septuagint since it does not asterisk it, the symbol it uses to distinguish
lemmata and usages that are unattested prior to “the LXX.” In cases like this
one, unasterisked lemmata need to include one or two examples of
extra-Septuagintal attestations.
Another example: ἀβύσσος. According to GELS, it is a feminine substantive in
the Septuagint, and as such, not attested before the Septuagint.
However, ἀβύσσος has been taken to be employed, in accordance with
usage elsewhere in ancient Greek, as an adjective in ἀπὸ ἀβύσσων πηγῶν κάτωθεν Deut 33:13. It
is so translated in NETS: ‘from unfathomable springs below.’ ἀβύσσων Deut 33:13 is
so glossed in LEH: ‘bottomless, deep.’
GELS implies that in ἀπὸ ἀβύσσων πηγῶν κάτωθεν Deut 33:13, which it cites but does not gloss, ἀβύσσος is to be understood substantivally: “from abysses of
springs below,” rather than “from deep springs below.” Given the parallelisms
in context, it is natural to construe the text in this way. It is so tagged in
GRAMCORD.
Or not: GELS cites an adjectival use of
plural ἀβύσσος from Herodotus
just before citing LXX Deut 33:13. But in that case, it would have been
consistent with practice elsewhere if GELS had distinguished between
established and previously unattested usage of this lemma, and noted the
adjectival use of ἀβύσσος in LXX
Deut 33:13 as such.
LEH classifies lemmata that are sometimes
used as adjectives as adjectives even if they are used as substantives in other
instances. That’s not unusual, though the distinction between noun and
adjective is not as cut-and-dried as some grammarians like to suggest. Consider
the following phrases in English:
He doesn't care about the very poor.
Is *poor* an adjective or a noun? It is an
adjective serving as a component of a noun phrase. The grading adjectives allow
accounts for the *very* poor. Dictionaries often list such a word as an
adjective and then note, in a subheading, that it is also occurs as a noun.
The Eeyore factor explains that.
The noun is used as an adjective.
It is a plus that LEH and GELS seek to sort
these things out. But I’m not sure that either uses a “grid” that is linguistically
and taxonomically consistent.
Features. GELS’ attention to semantic domains is a step forward with
respect to LEH. For example, ἆ
is cross-referenced to οἴμμοι, and vice-versa. At οἴμμοι, one finds a
cross-reference to οὐαί. To be sure, the rabbit trail ends there, which it shouldn’t. ὦ deserves cross-referencing as well.
GELS aims to provide definitions, not
only translation equivalents. This is another step forward with respect
to LEH. The attempt to describe the core meaning of lexical items and establish
a relationship between attested meanings wherever possible is, one would think,
a primary and most important heuristic goal of lexicography. A set of target
language glosses-in-context in no apparent order (a criticism that can be made
of LEH) falls short of lexicography as traditionally understood.
However, the tendency to read the Greek as if
it conformed in all particulars to the Hebrew remains. For example, ἄβατος. Behind this adjective is, of course, the verb βαίνω go (as Rocci notes). The adjective means “not enterable,”
either in the sense of impassable, as Muraoka notes; or in the sense of prohibited
entry, for example, of a temple by unauthorized individuals. Muraoka translates τὸν ἄβατον ἡμῖν ναὸν as ‘the
temple, which is off-limits to us’ (3 Macc 5:43).
As far as I can see, ἄβατος always characterizes a place or thing as impassable,
not passable (off-limits), or never passed over (broached), within
and without the LXX. Against LEH and GELS, I would assert that ἄβατος never characterizes a place as desolate (according
to LEH, in Jer 6:8 μὴ ποιήσω σε ἄβατον γῆν ἥτις οὐ κατοικηθήσεται), waste,
desert (according to LEH, in Jer 33:18 [MT 26:18] Ιερουσαλημ εἰς ἄβατον ἔσται);
or deserted and waste (according to GELS, in Jer 30:2 [= 30:18 in Rahlfs;
MT 49:2] ἔσονται εἰς ἄβατον).
In all these cases, the places in question
are ἄβατος because they are שממה a waste, עי a ruin, or the like. A
wasted location, a ruin, is, by definition, a no-go, impracticable
terrain. Still, the semantic content of the translation equivalent should not
be assimilated to that of the source language.
What is my response to GELS? ὦ, ἆ, and οἴμμοι. Mostly, of course, χάριτας ἔχω.
For another in-depth review, see Mike Aubrey here.
Bibliography
Edwin Hatch and Henry A. Redpath,
A Concordance to the Septuagint and the Other Greek Versions of the Old
Testament (including the Apocryphal Books) (3 vols., Oxford: Clarendon,
1897 = HR); Werner Georg Kümmel, Hermann Lichtenberger, and
Gerbern S. Oegema, eds., Jüdische Schriften aus
hellenistisch-römischer Zeit, Gütersloh: Gütersloher, 1973-2005,
forthcoming = JSHRZ); Henry George Liddell, Robert Scott, Henry
Stuart Jones, and Roderick McKenzie, A Greek-English Lexicon. With
a supplement, 1996 (Oxford: Clarendon, 19409 [= LSJ] + 1996 [ =
LSG]; Johan Lust, Erik Eynikel, and Katrin Hauspie, with
the collaboration of G. Chamberlain, A Greek-English Lexicon of the
Septuagint (2 vols.; Stuttgart: Deutsche Bibelgesellschaft, 1992, 1996;
20032 = LEH); Franco Montanari, GI. Vocabolario della
lingua greca. Con la guida all'uso del vocabolario e lessico di base. Seconda
edizione con CD-Rom (Torino: Loescher, 2006 = GI); Takamitsu Muraoka,
A Greek-English Lexicon of the Septuagint. Chiefly of the Pentateuch and the
Twelve Prophets (Leuven: Peeters, 2002); idem, A Greek-English Lexicon
of the Septuagint (Leuven: Peeters, 2009 = GELS); Alfred Rahlfs and
Robert Hanhart, Septuaginta: Editio altera (Stuttgart: Deutsche
Bibelgesellschaft, 2006 = Rahlfs); Lorenzo Rocci, Vocabolario greco-italiano
(Milano: Alighieri, 1990 = Rocci); Johann Friedrich Schleusner, Johann
Christian Biel, and Esdras Heinrich Mutzenbecher, Novus
Thesaurus Philologico-criticus sive lexicon in LXX et reliquos interpretes
graecos ac scriptores apocryphos veteris testamenti (5 vols.; Leipzig:
Weidmann, 1820-1821)
John,
Although βαίνω is almost universally thought to mean "go," it more essentially means "stride" or "straddle" -- to put one foot apart from or ahead of another. I think that the essential sense of ἀβατος is more precisely, "not to be stepped over" or "not to be stepped into." That doesn't differ from "not to be entered," but it's a bit more precise. This first became clear to me when trying to make sense of the Homeric formula, βῆ δ’ ἰέναι, "he stepped forth to go," but it's also discernible from compounds such as ἀναβαίνω, "straddle and ride (a horse or other beast of burden)" and cognates like βῆμα,a platform to put both feet on.
Carl W. Conrad
Department of Classics, Washington University (Retired)
Posted by: Carl Conrad | July 29, 2009 at 07:44 AM
Hi Carl,
That's an important clarification, so long as one remembers that the verb, compounds of it, and noun and adjective cognates are often used in ways that do not respect the core meaning you identify in all particulars. For example, κατἀβασις is used in reference to the "descent" of snow in Sir 43:17b.
Posted by: JohnFH | July 29, 2009 at 07:52 AM
Thanks, John.
Your response is on target. I guess what bothers me in particular -- lexicographically speaking -- is the one-word gloss (of which I'm as guilty as any); it always suggests that the original word and the target-language word are more than roughly equivalent.
Posted by: Carl Conrad | July 29, 2009 at 02:13 PM