Scholars agree that some books are literal translations and others are paraphrases, much like the Living Bible. Given that Greek manuscripts are the earliest witnesses to the Hebrew Old Testament, a more literal manuscript can be helpful for textual criticism. The non-literal translations, however, may shed light on the theology, philosophy, or religious practices of the Jewish faith in the late Second Temple period.
Ryan Reeves, “What is the Septuagint?.”
The above-quoted statement is common in popular literature on the Septuagint (LXX). The idea is usually undergirded by a presupposition regarding the texts that the LXX translators had to work with, in many cases assuming something like the Masoretic Text (MT) as the exemplar. Take note of Reeves’ comment on the helpfulness of “a more literal manuscript” for textual criticism while the alleged “non-literal translations” shed light on things other than the text.
But what if these non-literal translations weren’t as non-literal as one might initially suspect? Could they then be helpful for textual criticism? In other words, rather than assuming that a supposedly non-literal rendering is an example of the translator assuming a certain freedom in their translation, why not ask if there might be another text serving as the foundation for their translation?
After referencing Rudolf Kittel’s comment that the “LXX is not a real translation but a theological commentary,” Natalio Fernández Marcos assures us that:
Once we get into the actual text, as a general rule the translation of the Pentateuch is faithful to the Hebrew text, more than was thought at the beginning of the century. And in the light of recent discoveries at Qumran, the great divergences in the historical books between the LXX and the Hebrew have to be interpreted more as a witness of the pluralism of the Hebrew text before its consonantal fixation at the synod of Yamnia, c. 100 CE, than as the result of the exegetical preferences of the translators.
Natalio Fernández Marcos, The Septuagint in Context: Introduction to the Greek Versions of the Bible (Leiden: Brill, 2000), 23.
Laying aside the reference to the mythical council of Jamnia, this is an important point to note. We cannot presuppose that the LXX translators had something like the MT before them, nor can we assume that they were all working with the same text. We have to allow for divergent Vorlagen and then assess how strict or free they were in their translations.
James Barr has an instructive essay called “The Typology of Literalism in Ancient Biblical Translations” in which he expounds on what I’d call the limits of freedom. Around the same time that Marcos’ volume appeared in its original Spanish edition, Barr said:
Thus, in general, where new evidence has become available it has on the whole increased our conviction that, at least in many books, the LXX worked fairly literally and elements in their rendering stand for something that was actually there, rather than being free invention or fancy.
James Barr, “The Typology of Literalism in Ancient Biblical Translations,” (MSU XV, Göttingen: Vandenhoeck & Ruprecht, 1979), 286.
Barr follows these comments with a quick reference to Ben Sira, which contains a number of “Hebrew variants semantically grossly diverse from one another” (286). So where one thinks that the LXX translator is offering a free rendering of one variant, it is quite possible, even likely, that they are offering a literal rendering of another. But what of a case where there is no known textual variant? Barr uses Sir 5.1 as an example. The Hebrew reads:
אל תשען על חילך ואל תאמר יש לאל ידי
The LXX translates this to say:
Μὴ ἔπεχε ἐπὶ τοῖς χρήμασίν σου
καὶ μὴ εἴπῃς Αὐτάρκη μοί ἐστιν.
Barr says that this appears to be an “excellent free rendering” before asking whether or not it was possible that the final word in the Hebrew text was read as יד rather than ידי (see also W. Bacher, “The Hebrew Text of Ben Sira,” JQR 12/2 , 283) in which case the LXX translator would have been translating quite literally. It is also noteworthy that the Syriac translation of Ben Sira understood the Hebrew in the same way (see W. Th. Van Peursen, Language and Interpretation in the Syriac Text of Ben Sira: A Comparative Linguistic and Literary Study [SSVBTCC 16; Leiden: Brill, 2007], 29, 42), which could indicate that the Syriac version is dependent on the LXX, or that the translators of the LXX and Syriac version had the same or a similar Hebrew text that has been lost, or finally “polygenesis,” that is, each translator read the Hebrew text in the same way independently of one another as Van Peursen suggests.
The point here is that it’s not always necessarily the case that the translator is offering a paraphrase of the source text even if there is no known variant to explain the difference. In this example, the translation itself could be thought to serve as a variant reading of sorts, which gives the reader pause to think about how this reading could have arisen. In this case it’s not difficult to imagine our translator missing the final י in the line. It is, after all, the smallest letter in the Hebrew alphabet.
So in closing, I want to say two things: First, the LXX translators should be given more credit than they sometimes are for the translations they provide. They’re not like those who produced the Targumim nor are they the Eugene Petersons of their day. There was a limit to the freedom they could exhibit. Second, Barr’s essay is brilliant and says much more than I’ve hinted at in this short post. It should be required reading for anyone working with the LXX.
Happy International Septuagint Day!