This summer has been very productive thus far. My attention has been centralized on three broad areas of interest: Family, Work, and Studies. Currently, both in my work and in my studies, I’ve been researching the Septuagint (LXX). Various other subjects have occupied my studies as well, but the LXX has piqued my interest for years, and is one of the reasons why I sought an undergraduate degree in Hellenistic Greek. It’s safe to say that my interest continues to increase in this area of research, and I sense there remains a great deal of research to come in the years ahead as Septuagintalist scholars begin and continue to integrate their research with the discoveries of Judean Desert.
This past week, I’ve been reading through R. Timothy McLay’s monograph The Use of the Septuagint in NT Research. McLay does an extremely good job laying out the various related issues of NT authors using the Septuagint, especially as it relates to an extant Hebrew reading or vorlage. His example par excellence is Acts 15:16–18, which “uses” Amos 9:11–12. His argument and demonstration are excellent for those who are just beginning to wade through the issues of how to juggle comparative philology, Hebrew Syntax, textual transmission, a proto-text or vorlage, Greek Syntax, OG as compared to LXX, just to name a few. There are times, however, when I scratch my head in attempting to understand McLay’s arguments. For example, McLay argues,
The next question is, How did the authors cite Scripture? Did they quote their sources directly or quote from memory? Quoting from memory could be one way to explain the way in which the NT authors sometimes blended several Scriptures together. For example, it might explain the use of ἀνορθώσω in Acts 15:16. The verb is not that common in the LXX (only 16x), but it does occur in 2 Reigns (Samuel) 7:13, 16 where the prophet Nathan tells King David that he is not to build the temple, but where the Lord says that David will οἰκοδομήσει build it through his seed and “I will ἀνορθώσω establish his throne.” The similar themes and vocabulary regarding the kingdom of David are probably responsible for the author’s choice of vocabulary, particularly since 2 Reigns 7:13–14 was a very important messianic text. Memory might also account for the different introduction After these things I will return and, perhaps, for the omission of the second reference to the fallen parts. However, it is noticeable that while Acts 15:16 departs quite significantly from Amos 9:11, the next verse in Acts is almost identical to that of Amos. It would be difficult to believe that one could make mistakes in one verse and then be so accurate in the next. Thus, it seems unlikely that lack of memory could account for the differences in this particular case.
McLay begins this section with a question, “Did they [the NT authors] quote their sources directly or quote from memory?,” to which his response (read: conclusion) is: “. . . it seems unlikely that lack of memory could account for the differences in this particular case.” How does he get to his conclusion; what is his argument? In order to understand we must observe the phenomenon to which he is discussing. Acts 15:16 reads, “μετὰ ταῦτα ἀναστρέψω καὶ ἀνοικοδομήσω τὴν σκηνὴν Δαυὶδ τὴν πεπτωκυῖαν καὶ τὰ κατεσκαμμένα αὐτῆς ἀνοικοδομήσω καὶ ἀνορθώσω αὐτήν . . . “ The issue surfaces in the LXX reading of Amos, the supposed source (καθὼς γέγραπται) of which is Amos, which reads: ἐν τῇ ἡμέρᾳ ἐκείνῃ ἀναστήσω τὴν σκηνὴν Δαυιδ τὴν πεπτωκυῖαν καὶ ἀνοικοδομήσω τὰ πεπτωκότα αὐτῆς καὶ τὰ κατεσκαμμένα αὐτῆς ἀναστήσω καὶ ἀνοικοδομήσω αὐτὴν καθὼς αἱ ἡμέραι τοῦ αἰῶνος . . . ” The LXX text is quoted from Rahlfs and accordingly the two texts show affinities, but do differ—most notably is the NT use which interpolates the verb ἀνορθώσω. McLay draws our attention to this omission—all is good and well. Thus, we need an explanation of how this verb came to be added (assuming we have no valid reason to suggest an emendation to the Masoretic Text). It’s in this context that McLay raises the question regarding memory.
So, let’s track his argument. Greek teachers might get it wrong when they say find the verb first, but Logic teachers are spot on when they tell us to find the conclusion first. What exactly is McLay’s conclusion? In this case, we have a handy keyword, “thus,” signaling the conclusion, which is, “Thus, it seems unlikely that lack of memory could account for the differences in this particular case.” Just what exactly is McLay saying here? We could probably better state his conclusion as, “It is improbable that memory failure adequately explains the additional verb (ἀνορθώσω) of the NT quotation.” Coming to this conclusion is based on the premise that memory failure has explanatory power for textual variants, as well as another premise, namely, “It would be difficult to believe that one could make mistakes in one verse and then be so accurate in the next.” The last premise is itself a conclusion of embedded argument by example. Assuming that word frequency has a direct relationship on memory, McLay overcomes the low frequency of ἀνορθώσω (16x) objection by arguing that 2 Reigns 7:13, 16 are more likely to be memorized due to its messianic interpretation. While this certainly could be the case, McLay doesn’t seek to substantiate it. Rather, he moves to verse 17, and finds a conundrum, insofar as he is perplexed as to how one verse (v. 17) can be so accurate, whereas another immediately contextual verse departs from its source.
Albeit convoluted, McLay’s argument is great for students to encounter. Firstly, why should we be persuaded over his conclusion? Doesn’t memory fail in that sort of manner? That is to say, we could memorize large portions of Scripture, recite this Scripture, and make errors. Will our errors perpetuate throughout our recitation? Will they be sparse? Could we have merged two readings of similar text in our recitation without being cognizant of the matter? Could we intentionally shift or move syntactically significant clauses, phrases, or vocabulary, according to our interpretation of a verse, so as to flush out and make plain our point? All of the above seem rationally plausible for understanding errors or misquotations in memory recital. Yet, there is another large assumption in McLay’s argument: can textual support be found for the Acts reading? In other words, we assume their is a failure in memory for we don’t have an extant text with such a reading. The Göttingen LXX lists 764 Athos, Λαύρα 169, a thirteenth – fifteenth manuscipt as supporting the reading of ἀνορθώσω, but it is most likely a quotation of Acts. Nevertheless, based upon what we can know of memory, I am unsure as to why McLay gives more weight to the premise that memory failure occurs systematically and frequently. It seems to me the converse is the case.