Why This Research Matters

In the previous post I presented some preliminary results of my Septuagint study of the divine Name renderings into Greek in the Psalms, noting only around 18% of nominative and genitive occurrences of the hundreds surveyed to include the definite article: the remaining 82% are "anarthrous", lacking the article.

I am now starting to see already why this research matters. Professor Albert Pietersma is the lead translator of the NETS New English Translation of the Septuagint. This is a major scholarly work combining expertise across the Septuagint field, including Larry Perkins, whose paper on Exodus we have already discussed on this blog.

Pietersma includes in the introduction to the NETS translation of the Psalter the following note (p. 546):

Since the Greek Psalter provides no evidence that the translator made any serious attempt at distinguishing between the divine names Yahweh, including the short form "Yah", and Adonai, I have in accordance with NETS policy rendered all occurrences of kyrios, when representing either, by "Lord".

The decision is a difficult one, because even if he is right about this (no serious distinguishing going on), then you still have to work around the Adonai plus Yahweh problem with increased difficulty if both are translated by "Lord" (see Psalm 68:20: Our God is a god to save, and to the Lord Lord belong the escape routes of death).

But I don't think he is quite right about that assumption of no difference, and here're two reasons why.

1. In contrast to the article treatment in Yahweh translations in Psalms (82% are rendered anarthrous in nominative and genitive forms), when Adonai is translated into Greek only half of the translations are anarthrous (53%, that is 9 out of 17 times with respect to genitive and nominative cases - note however that one of these anarthrous instances, Ps 16:2, could almost certainly never have been confused with Yahweh in that it translates a possessive, "my Yahweh" being unheard of). This partial similarity between Yahweh and Adonai translation policy may represent a vague gesture at the sanctity of the Yahweh solution, but nonetheless, a significant difference remains (from an albeit small sampling).

2. Less significantly, but most intriguingly, is a translation of Psalm 130:6, my soul waits for the Lord. Here the Greek reads: ἤλπισεν ἡ ψυχή μου ἐπὶ τὸν κύριον.  Since we are focussing on the more meaningful cases of nominative and genitive, I almost missed this accusative construction, but it rang a bell. As it turns out, "ἐπὶ" before κύριον when translating "to Yahweh" systematically removes the accusative article τὸν: Ps 4:5, Ps 21:7, Ps 22:8, Ps 31:24, Ps 32:10 and 11, Ps 37:3, Ps 40:3, Ps 55:22. Not so in Psalm 130:6 translating Adonai.

So to conclude, I can't help but wonder if Pietersma has considered these two important pieces of evidence when he dismisses the possibility of distinguishing efforts on the part of the Psalter translator. Only a more thorough investigation of the benchmark for anarthrous renderings, the Pentateuch, and the Adonai goldmine of Ezekiel will provide us with more evidence. If it can be shown that the "NETS policy" mi, in fact, be misrepresenting the translation practice of the Greek translators, then that should filter down to less clear-cut cases such as the Psalms.

Kyrios (aka the LORD) in the Psalms: results

Back in August, I focussed closely on the Greek translation of Yahweh (and Yahh), the Israelites' personal name for their God. This was fueled by a steadily growing interest that followers of this blog will have noticed in the notion of Lordship and increased clarity in the usage of those words today among Bible-believing Christians. But why the Psalms? Well, it has also become my favourite book that has been a source of much inner reconnection and spiritual life, to such a point that I am also in parallel preparing a Psalms study of the self that I hope may even make this part of the Bible appealing and useful to non-believers. But the Psalms as a large Old Testament book outside the Pentateuch (Genesis through Deuteronomy) provides a wealth of information about the practices of a subsequent translation period (the Greek Septuagint grew to encompass the whole Hebrew Bible over many decades).

So as a very quick reminder of the grammatical context: the Septuagint translation for Greek-speaking Jewish communities (probably initially in Egypt) had to come up with a solution for God's personal name, YHWH. A complex debate rages about the status of this name by the time of the Greek translation and the extant pre-Christian translations, but by the early centuries of the common era, the standard was to use the Greek word for Lord: Kyrios. This translation has been preserved through the millennia in many modern translations, but with important differences: every since the King James (although maybe earlier, but not Latin, which does not use articles), the article was added, and Lord was capitalised: the LORD (BTW I have an opinion developing within me about this scholarly debate, but let's hold that for another post and a bit more research). Now some scholars had reported that the article-free Kyrios (the technical word for this is "anarthrous") is a more consistent technique in the Pentateuch, and with respect to articles, on a direct par with other personal names, like Moses. In the Pentateuch, occasionally (about 7% of the time in Exodus) you get "the Lord", but absolutely no more frequently than "the Moses". So this "the" is contingent on necessary grammar and by no means requires framing the word as an impersonal title.

Here, then, is what happens in the Psalms:


Some words of explanation are in order!

Since I quickly realised that Greek case was influential (has the Greek translator used a nominative, genitive, accusative, dative or vocative?) to the presence or absence of the article, the main table is a summary of article behaviour for the whole Psalter with increasing "weeding out" of certain cases. Firstly, since the vocative (VMS) κύριε is always anarthrous in LXX Psalms it could hardly be included in research about what I am calling the grammatical "signature" of YHWH translations into Greek: it is totally necessary to exclude them (209 occurrences) altogether. Of the remaining 479 occurrences where a form of Kyrios is used to translate YHWH (a few occurrences of YHWH are not translated that way), 301 are anarthrous. However, since the dative seems to almost always require the article (96% of the time), it seemed necessary to weed that one out too. And once the rather erratic accusative κύριον is also removed, we are left with the only stats that really matter: 18% of the Psalter's κύριος carrying the article and 17% of its κυρίου.

With that task complete, you can see how I was also interested to see if there were notable variations across the Psalter's traditional five volume format, by breaking down these arthrous counts accordingly. This may or may not have been useful. What it shows is when a particular volume strays significantly from the averages above. Since there is variation in both volume length and "YHWH density" (number of YHWH occurrences per verse to translate), it may have been more helpful to simply divide into say 4 or 5 (or more) even sized chunks.

I hope this clarifies the data presented.

I have actually had a little exchange now with Professor Larry Hurtado on this research, which appears to have gained his interest. I am appealing to him for guidance on how I might most usefully develop this research.

One possibility might be to follow Perkins, whose paper I think I previously discussed on the blog. I wonder about doing a similar comparison to that which he performed, to a name like Moses but also a title like theos before proceeding onto some other OT texts.

I am keen to learn from experts also what are the other types of instances that I should be excluding from the data, beyond Greek case? I also have a very rough-and-ready tagging system for some interesting common constructs (like κύριος ὁ θεός: always anarthrous, including a couple of non-nominative cases) that I think could be developed under the guise of "lexical units".

(Updated 25/05/2018)