A Puzzling Article on “the LXX”

I have been a member of the Institute for Biblical Research (IBR) for a few years now. I have attended the past three or four annual meetings, which get squished between the ETS and SBL conferences, and have made a habit of reading IBR’s journal, Bulletin for Biblical Research (BBR) each quarter. In fact, BBR is where I have contributed most of my book reviews by far. It’s a thriving society, and a very good journal for evangelical scholarship.

ibr_logo_headerSo I was excited to find an article in the latest volume that dealt directly with the Septuagint. The article is by Dr. J. Daniel Hays, and entitled “The Persecuted Prophet and Judgment on Jerusalem: The Use of LXX Jeremiah in the Gospel of Luke” BBR 25.4 (2015): 453-73. As I began reading it, however, it seemed to me to contain some flaws, or at least things that left me puzzled. In this post, I want to respond to what I found to be the more unconvincing aspects of Hays’s argument.

In advance of posting, I sent a draft of this critique to Dr. Hays, and his response is at the bottom of the post.

Summary of the Article

Now, I realize that it is probably possible to nit-pick most journal articles into oblivion, so I want to state at the outset that that is not my goal. Rather, I wish to point out how I think Hays’s article represents broader trends in biblical scholarship – especially New Testament scholarship – when interacting with the Septuagint.

I’ll state it up front: Identifying shared vocabulary between the NT and LXX with an indistinct notion of “dependence.” This assumption shows up to some degree in Dr. Hays’s response below.

The article’s abstract is as follows:

This article argues that within Second Temple Judaism, Jeremiah was well known as the paradigmatic “persecuted prophet” and was likewise closely associated with the consequential fall and destruction of Jerusalem. Thus, when the Gospel of Luke portrays Jesus as the “persecuted prophet” in conflict with the leaders in Jerusalem or recounts Jesus’ warnings of judgment on Jerusalem, allusions and parallels to Jeremiah are numerous, implying that the traditions associated with LXX Jeremiah form a critical background for understanding those texts.

Overall, I think that Hays does a fine job defending about 75% his thesis. In fact, I have basically no qualms with anything he proposes about the themes of Jeremiah or its well-known status in Second Temple Judaism. The problems arise when we get to the Septuagint aspect. Again, as his response states, this is the very aspect that Hays assumes.

1) Questionable Textual Basis

One of the first problems in this article appears in Hays’s third footnote. There he states that he cites translations of LXX Jeremiah from the NETS translation (available here). That is fine as far as it goes. But he then states that he will work from Rahlfs-Hanhart’s text of LXX Jeremiah. This is problematic for two reasons. Firstly, Ziegler’s critical edition of OG Jeremiah is the far superior text and should therefore form the basis of any study like this (1). Secondly, the NETS translation of LXX Jeremiah is itself based on Ziegler, not on Rahlfs-Hanhart, which risks misalignment between what Hays cites in English versus Greek, although presumably he checked for this.

2) Questionable References to LXX Jeremiah

Almost as soon as Hays begins to move through his argument, he starts to talk about LXX Jeremiah in terms that are not actually specific to LXX Jeremiah (i.e. Jeremiah in its Greek version, in distinction from the Hebrew original). This is quite pervasive and leads to two problems, a thematic one and a lexical one.

Thematic: First, it causes Hays to illegitimately attribute features proper to Jeremiah in Hebrew to LXX Jeremiah. But any “themes” or “theology” present in Jeremiah in Hebrew are going to carry over into the Greek version, unless you really do some spadework to show how that is not the case (i.e., how LXX Jeremiah somehow changes the message of Jeremiah in Hebrew, which gets picked up by Luke. This is possible, but takes a lot of work to demonstrate). To this extent, wherever Hays talks about Luke adopting a Jeremianic theme, this proves nothing about Luke’s “use” of LXX Jeremiah (i.e., the Greek version specifically). This is like attributing “shared themes” between Charles Spurgeon’s sermons and Romans to his “use” of the KJV. The fact that there are similarities in themes does not arise from the translation itself, but the ideas of the original that the translation conveys.

The fundamental flaw seems to be Hays’ assumption that shared ideas demonstrate the textual dependence of Luke upon LXX Jeremiah (although he never clearly defines “use”). Or, that a relationship of dependence between Luke and LXX Jeremiah does not need to be demonstrated. For example, Hays cites a range of texts from Jeremiah to illustrate the central role of Jerusalem, it’s judgment, and the persecution of the prophet, but there is nothing specific to the Greek text in these matters that isn’t part of the Hebrew also (again, unless you carefully demonstrate this, which Hays does not systematically attempt). This is evident from the fact that Hays often cites both the LXX and MT versification – there is no meaningful difference between the two at the level of concepts or themes.

This problem continues throughout much of the article. See his discussion of:

  • Jeremiah facing rejection on 463
  • Conflict with leaders on 463-64
  • The negative sense of “scribe” and the suffering prophet concept on 464
  • Rejection and Jerusalem themes on 465
  • Judgment on Jerusalem on 465-66
  • Persecution of prophets on 466
  • Lament over Jerusalem on 467-68
  • Den of robbers on 469
  • Destruction of Jerusalem on 470-71
  • Prophetic trials on 471-72
  • “Daughter of Jerusalem” on 472

Lexical: Secondly, Hays repeatedly appeals to “catchwords” to demonstrate what he sees as Luke’s dependence upon LXX Jeremiah. But in the vast majority – if not all – of his examples, the vocabulary identified is far too conventional to necessarily prove any textual relationship. For example, Hays observes that priests are collectively characterized in negative terms in Jeremiah more than any other prophetic book, and points to the appearance of ἱερεύς 35x (p. 456). But, for one thing, the use of the word ἱερεύς is not a “distinctive” of LXX Jeremiah. And if the portrayal of priests in largely negative terms is distinctive, that is a distinctive theme specifically of the Hebrew text, not the LXX version per se. For another thing, nothing about Luke’s use of the word ἱερεύς means he was “dependent” upon the Greek version of Jeremiah, rather than using the word simply to talk about priests. If Luke, too, portrays priests negatively, this has nothing to do with LXX Jeremiah itself (other than perhaps its existence allowing Luke to read and be familiar with the concepts of the Hebrew original, but this too is debatable).

This problem persists through most of the article also, as Hays repeatedly finds shared vocabulary of the broadest sort, then takes the associated concepts and makes them indicative of “use” of LXX Jeremiah. It continues with:

  • ἐκκόπτω on 463
  • αἷμα on 466
  • πῦρ and ἀνάπτω on 467
  • The use of vocatives on 467
  • κλαίω on 468
  • ἐχθρός and ἀνθ᾽ ὧν on 469
  • ἀποστέλλω and δοῦλος on 470
  • ἐρήμωσις on 471

But the fact that both Luke and LXX Jeremiah are written in Greek and share standard vocabulary should not be confused for any special relationship between the two texts themselves, or any special influence upon the theology of Luke from the Greek version of Jeremiah. Agreement in vocabulary is insufficient to prove textual dependence specifically from the Greek version of Jeremiah if that vocabulary accurately reflects the underlying Hebrew. Hays seems to get close to realizing this a few times, but always stops short. On p. 466, for instance, he notes that Luke and Jeremiah both deal with persecution of the prophets, and, although διώκω occurs in other prophetic books, because Jeremiah is the only one to use it in the sense of persecution against God’s prophets like Luke does, there is a relationship of dependence. But again, the Greek vocabulary does not demonstrate any dependence, only shared themes between Luke and Jeremiah, themes that arise specifically from the Hebrew text of Jeremiah. (See a similar case on p. 469 in his discussion of ἐχθρός and its related themes).

The Connection to Lamentations

Hays also observes how in the Greek translation of Lamentations the figure of Jeremiah is framed as the speaker, a feature not present in the Hebrew text. This is one occasion where Hays has identified something that is actually unique to the Greek version of a book versus the MT. Hays attempts to tie the notion of Jeremiah weeping with Luke’s statement in 19:41-48 that Jesus wept over Jerusalem (p. 468). While it is true that Jeremiah weeps on several occasions (9:1; 13:17; 22:10), this is not a feature unique to LXX Jeremiah (as if Jeremiah does not weep in the Hebrew version). And while the reference to Jeremiah in LXX Lamentations is unique to the Greek version, at that point we are talking about the figure of Jeremiah and not the Greek text of Jeremiah. At best, this supports Hays’s point about the prevalence of “Jeremiah” (text[s], themes, theology) in the literary milieu of Second Temple Judaism (pp. 457-60), but at worst it is an equivocation in terms that adds little to his argument for the “use” of LXX Jeremiah in Luke. Again, Hays never defines what he means by “use,” which I think only adds confusion.

Summary and Response

Again, my objective here is not to “destroy arguments and every lofty opinion raised against the knowledge of the Septuagint.” But I want to point out what I might call problematic habits that tend to show up in NT scholarship when interacting with the Septuagint. Ironically, Hays cites a scholar who points out that usually, New Testament use of terms found in the LXX indicates a common literary milieu for Greek-speaking Jews, but not necessarily literary dependence (p. 458 n. 20). I fear that this observation applies globally to Hays’s thesis, as he ends up conflating textual influence with theological (or literary, thematic, etc.) influence. Carefully distinguishing these two is a crucial but too-often overlooked aspect of studies of the New Testament “use” of the Old Testament. (2)

Hays argues that the use of Jeremiah in Luke is “easily overlooked.” I agree, but for very different reasons. In fact, I found little compelling evidence that Luke is “using” LXX Jeremiah, as much as he is simply familiar with the message of the book generally. Still, I would say that Hays does a good job showing how Luke is influenced by Jeremiah in this general way, and that we can see that influence because of the concepts that both Luke and Jeremiah share (in Hebrew or Greek). Yet once you show how Luke is familiar with and employs the distinctive themes found in Jeremiah in Hebrew – which Hays does – it is not at all necessary to bring the Septuagint into the equation. Now, if you could show that Luke had picked up some phrase from LXX Jeremiah that somehow changed what its Hebrew source text said, then I think you could talk about Luke’s “use” of LXX Jeremiah.

Hays’ Response

First of all, I think you are missing my point. I am not trying the show that Luke is using LXX Jeremiah in distinction to MT Jeremiah. I assume that before I even start. That is, I’m starting with the assumption that background OT literary, theological and lexical influence in Luke (as in much of the rest of the NT) comes via the LXX. I’m not trying to establish this or defend it. This is a widely held view in NT studies. What I’m trying to establish is that the OT Prophets should not all be lumped together into an amorphous “common prophetic language” in discussing NT use of the OT. I’m arguing that in regard to the specific themes of the persecuted prophet and the destruction of Jerusalem, Luke is not influenced in his language and allusion by “the prophets” in general but by Jeremiah in particular. In conducting that study, one has to use the LXX. That these same themes show up in MT Jeremiah is irrelevant to my argument.

The lexical comparisons are likewise along these same lines. For example, I point out that the Greek word for “false prophet” shows up on the LXX only 10 times, 9 of which are in Jeremiah. Thus the use of this same term in Luke 6:26 “for that is how their ancestors treated the ‘false prophets’” is not just “common prophetic language” but more specifically language common to Jeremiah.

Arguments about lexical usage, dependence and allusion are by nature only convincing when viewed cumulatively. That is, many of the individual word arguments are hardly conclusive on their own (as you note) and can be challenged one by one. But cumulatively, tracked throughout Jeremiah, underscored and strengthened by the many times that same word of phrase only occurs in Jeremiah (and not in Isaiah, etc.) the argument gains strength.


I’m grateful for Dr. Hays’s work, and for his generous (and kind!) response to my criticism. It is always refreshing when scholarly interaction is civil and clarifying.


(1) J. Ziegler. Septuaginta: Vetus Testamentum Graecum Auctoritate Societatis Litterarum Gottingensis editum XV: Ieremias Baruch Threni Epistula Ieremiae. Göttingen: Vandenhoeck & Ruprecht, 1957.

(2) For a helpful overview of similar misuses of “the LXX” in NT scholarship, see pp. 39-44 in McLay, R. T. The Use of the Septuagint in New Testament Research. Grand Rapids, Mich.: Eerdmans, 2003.

One comment

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.