Was there a “Septuagint Canon”?

By John D. Meade

In biblical and theological instruction and writing, it is common to refer to “the LXX” or “the Septuagint.” Old Testament / Hebrew Bible scholars refer to the LXX as the oldest translation of the Hebrew Scriptures, and scholars in New Testament and early Christian studies refer to the Septuagint as the text which the New Testament authors and early church fathers cited. How can professors in broader theological fields be expected to use the term any differently in their teaching and writing?

I want to provide brief commentary on this term, “the Septuagint,” for those teaching in biblical and Christian theological studies, so that we can provide an up-to-date description for our students on this very important piece of the history of the Bible.

Although the term “the Seventy(-Two)” referred originally to the Greek translation of the five books of the Torah/Law (e.g., Let. Aris.), it came to encompass the whole Old Testament by the time of Justin Martyr (d. ca. 165 CE). Often in the literature, though curiously, “the Septuagint” refers to a different canon of Scripture than the Hebrew canon1—that is, a wider canon including four or six extra books (Judith, Tobit, Wisdom, Sirach/Ecclesiasticus, and 1–2 Maccabees).

THE JEWISH ALEXANDRIAN CANON THEORY AND ITS REVISIONS

No doubt, this view arose due to the so-called “Jewish Alexandrian canon” theory. This theory attempted to explain how the Christian mega-codices of the fourth and fifth centuries CE came from Alexandrian Jewish groups and had more books than the narrower Palestinian Jewish canon of Josephus (Ag. Ap. 1.37–42).2 Thus, early Christianity inherited a wider Jewish canon containing some books in longer forms (e.g., Daniel) as well as new books (e.g., Wisdom of Solomon).

This theory has fallen on hard times due to the work of Albert Sundberg, who argued that the Jews of Alexandria and Palestine had the same canon, and it was Christians who formed their own wider Old Testament canon after they parted ways with Jews after 70 CE.3 However, scholars recently have attempted to construct revisions of the hypothesis, still locating the formation of the wider canon in Egypt in the Greek language.

One argument is that the Palestinian Nebiim (Prophets) was a collection of books formed in response to an earlier Alexandrian “biblical chronology” (Jewish history books + Judith and Tobit), which became the basis for the rest of the historical books in the LXX to which the poetic and prophetic books were added. The Christians inherited these lists from Alexandrian Jews and formed the Septuagint.4

Another argument attempts to revise the Jewish Alexandrian canon theory not by focusing on the external codices, but primarily by resting the argument for the Septuagint canon on internal criteria: (1) a Greek composition; (2) a coherent corpus (common terms used across books); and (3) a historical backdrop in the Jewish diaspora of Egypt.5

RESPONSE AND A WAY FORWARD

Although these revisions clarify the old hypothesis, one challenge remains: Neither Jews nor Christians know of this “Septuagint canon” from 1 BCE to 4 CE. Josephus claims that all Jews everywhere have “only 22 books” (Ag. Ap. 1.37–42). Philo of Alexandria cites as Scripture only from the books of the Law and the Prophets—not the deuterocanonical books. The New Testament authors and early Christian writers in the second century cite only books contained in the Hebrew canon.6 Furthermore, the early Christian canon lists from 2 CE and many lists from 4 CE closely cohere with the books of the Hebrew canon and do not include the deuterocanonical books; that is, the authors of these early lists do not know of a “Septuagint canon.”7

Even Cardinal Ximénes and the editors of the Complutensian Polyglot (published in 1517) specified the scope of the Septuagint with the rubric “Greek Translation of the LXX,” over books that had a Hebrew text. When the editors came to books without a Hebrew text, such as the deuterocanonical books, they applied a different rubric, “Greek Translation,” and did not specify the book as translated by the “LXX.” The editors of the Complutensian Polyglot would have thought that the Septuagint canon was equivalent to the Hebrew canon.

As professors teaching in biblical studies and Christian theology, we would do well by our students to put a moratorium on the term “Septuagint canon”—especially if by “Septuagint canon” one means a canon including the books of the Hebrew canon plus the six books of the Apocrypha. The three magisterial codices of the fourth and fifth centuries include Tobit, Judith, Wisdom of Solomon, and Sirach/Ecclesiasticus (including some of the books of Maccabees, but their inclusion is inconsistent across these MSS), and scholars have used these MSS as evidence for the wider “Septuagint canon” of the early church.

Since 1935, with the publication of Rahlfs’s Septuaginta (which includes books of the Hebrew canon + six deuterocanonical books + 3–4 Maccabees + Odes + Psalms of Solomon), defining the so-called Septuagint canon with wider boundaries than the Hebrew canon has been potentially confusing. The Jewish evidence from the Second Temple period (e.g., Philo, New Testament, Josephus) and the clear statements of church fathers (e.g., Melito of Sardis) furnish no evidence for a wider Septuagint canon among Jews or early Christians. Rather, the evidence conforms closely to the Hebrew canon. These sets of evidence should cause us to reconsider the evidence of the early codices and should force us to ask a different question: Are all books placed within a codex considered to be part of the canon? If the answer is “no,” then how are we to explain the data?

Fortunately, there is an answer at hand because several early Christians described books  that we call “apocrypha” as useful, neither canonical nor apocryphal (cf., e.g., Athanasius Ep. fest. 39). The ones who ordered Codex Sinaiticus, for example, probably would not have had a problem with placing the canonical books and the useful books between two covers; however, given their own canon lists, they probably would have maintained conceptual distinctions between these books. When talking about the Greek Scriptures, therefore, we should refer to the Hebrew canon in Greek dress, which early Christians simply called the Old Testament.

*This article was originally published in Didaktikos: Journal of Theological Education.


John D. Meade is Associate Professor of Old Testament at Phoenix Seminary. His recent projects include The Biblical Canon Lists from Early Christianity: Texts and Analysis (Oxford University Press, 2017; coauthored with Edmon L. Gallagher).


Before you go, check out the LXX Translation Ring on Logos Bible Software:

  1. Timothy Michael Law, When God Spoke Greek: The Septuagint and the Making of the Christian Bible (Oxford: Oxford University Press, 2013).
  2. Henry Barclay Swete, An Introduction to the Old Testament in Greek (Cambridge: Cambridge University Press, 1914; Peobody, MA: Hendrickson, 1989), 197.
  3. Albert C. Sundberg, Jr., The Old Testament of the Early Church (Cambridge: Harvard University Press, 1964).
  4. Philippe Guillaume, “New Light on the Nebiim from Alexandria: A Chronography to Replace the Deuteronomistic History,” Journal of Hebrew Scriptures 5 (2004).
  5. Jan Joosten, “The Origin of the Septuagint Canon,” in Die Septuaginta—Orte und Intentionen, eds. Siegried Kreuzer, Martin Meiser, and Marcus Sigismund (Tübingen: Mohr Siebeck, 2016), 688–99.
  6. Oskar Skarsaune, “The Question of Old Testament Canon and Text in the Early Greek Church,” in Hebrew Bible/Old Testament: The History of Its Interpretation, vol. 1, part 1, ed. Magne Sæbø (Göttingen: Vandenhoeck & Ruprecht, 1996), 443–50, esp. 445.
  7. Edmon L. Gallagher and John D. Meade, The Biblical Canon Lists from Early Christianity: Texts and Analysis (Oxford: Oxford University Press, 2017).
Share
Written by
Didaktikos

Didaktikos is a vocational journal for professors who teach in biblical studies, theology, and related disciplines—particularly at the graduate level and in service to the church. Didaktikos is published four times a year.

View all articles
18 comments
  • John,

    Excellent summary on the “Septuagint” Canon. I agree that one cannot find agreement on which books are to be in or out of the canon of the OG. I do consider that the two references to the OT canon mentioned by Jesus in Luke 24 does limit the books in question. There is also the parallel passages referring to Abel and Zechariah found in Matthew 23:35 and Luke 11:51 that appears to place a limit on the recognized books of the OT even by the time of Jesus. Thus, 3x of 4 references to the OT Canon are found in Luke. This does not surprise me in the least since he was most likely the “Man from Macedonia” who was in the vision of Paul at Troas in Acts 16 followed by the first of the “we” passages. Luke, most likely was “godfearer” or proselyte. He would have had access to the knowledge of Paul of the OT via the LXX. Luke 1-2 with the Songs of Elizabeth, Zechariah and Mary fit well with the use of the LXX.

    • Bryant,
      Thanks for your comment. If my argument is correct, then the NT authors aren’t thinking they are using the LXX or a LXX canon, but the Hebrew Scriptures in Greek dress. I’m not completely convinced those NT passages limit the books, but given the rest of the paucity of our evidence from this period, perhaps they corroborate the conclusion. Thanks again.

  • Thanks for the great article.

    FYI: the link to the video is broken.

    Best,
    Dale

  • Thanks for this article. Very interesting.

    I am not quite sure I understand your assertion that “The New Testament authors and early Christian writers in the second century cite only books contained in the Hebrew canon.”

    First, a glance through NA28’s appendix of NT use of the OT shows a long list of references from the six additional books. Certainly these are not all ‘citations’ in the strict sense, but 1) if any of them are then it would disprove your assertion, and 2) does an author have to ‘cite’ a text for him/her to assume it is canonical? Furthermore, does the citation of Enoch in Jude suggest that the author of Jude considered Enoch canonical? I am certain this is all familiar data to you, but your statement above confused me as it seemed to fly against all of this.

    Secondly, a search through http://www.biblindex.info shows that the additional six books are referenced over and over again by patristic writers.

    How would you explain this?

  • Jordan:
    Thank you for your comments and questions.
    (1) I’m using “citation” in the way you suspect. I’m not addressing the matter of allusion or other kinds of use because most scholars discern a difference between citation and allusion. In fact, if you follow up on the source I list in the footnote, Oskar Skarsaune makes my point for me. I encourage you to have a look at this. NA28 lists all kinds of allusions and echoes and such. But at no place does the NT cite from these six other books.
    (2) Remember, in this point, I’m talking about the NT and the second century fathers. Of course, the church fathers use and read all kinds of books per Biblindex. But I’m addressing the earliest period here because that’s when some suggest that early Christians either established the Septuagint canon for the church or received it from Alexandrian Jews. I’m still waiting for the evidence of earlier use.
    (3) The Jude citation of Enoch is interesting. Skarsaune does grant it as an exception as do I. However, as the post/article concerns the Septuagint canon, no ancient ever thought to include Enoch among these books either in their MSS or their canon lists. Tertullian speaks favorably about the book because of the Jude citation, but it’s never considered canon by the churches. This point answers your second question, I think. Christians cited all sorts of books (sometimes even as scripture) that they did not consider canon. Citation of authoritative literature as scripture is only one piece of evidence for a book’s canonicity; it is not decisive on its own. The converse is also true: 2-3 John are hardly cited in the first few centuries of the church’s history, but they do make many of the early canon lists. Thus an author’s canon list does not mirror his functional canon.

    Does this help?

    • Hi John,
      Thanks for taking the time to respond. Yes, that is helpful. When searching Biblindex for uses of any of the six books between AD 60-200, there are still some results, but it is limited and only an examination of the texts would clarify how the patristic writers are using the books in question.
      However, I am curious now as to how you are determining whether the earliest Christian writers considered a text to be canonical or not. As you point out, citation in the case of Jude’s use of Enoch and lack of citation of 2, 3 John do not determine canonicity. And yet, it seems to me that you are relying heavily for your argument on whether or not Philo or the patristics cited these books. (By the way, in another search on Biblindex, I noticed that Philo made fair use of some of the books in question. Granted, this doesn’t mean he considered them ‘canonical’, but it does beg the question as how one determines what Philo’s stance was towards these books.)
      Surely influence is measured by more than citations. At the moment, I am preparing a sermon on Isaiah 52:13 – 53:12. There are certainly a number of citations/quotations (not to mention allusions) of this passage in the NT, but the Isaianic passage has a much bigger influence on the paradigm of the first Christian writers than what citations alone suggest.
      I’m sure you are very busy, so I understand if you don’t have time to respond. The article has been helpful, and I am going to be more careful in how I define the LXX in the future.

  • Jordan, John,

    In follow up to your comments, I would say two things.

    First, If mere citation or allusion is what makes a book a part of the canon (whether Septuagint, Hebrew or NT makes no difference), then there is the problem with two quotes by Paul of Greek poets in Acts 17 and Titus 1. I doubt that any one would put the Greek poets as being part of the canon.

    Second, It appears that the authors of the NT, in this instance Jude, are referring to a particular quote or incident that is found in that particular book to be true, but not commenting on whether or not the rest of the book is true. It reminds of the quote, “A broken clock is right twice a day.”

    Furthermore, the OT acknowledges the source material that was used ot compile the information therein; yet that source material is not considered canonical.

    • Bryant,

      Yes, I would agree with all that. My question then is (which I was trying to put to John) is by what criteria are we discerning the stance of people like Philo and the patristics towards the six books above? It seemed that citations by these writers was being emphasized a fair bit in the post, but this seems circular if we are saying that other instance of citation do not imply acceptance within the canon.

      Canon lists are of course another and more explicit way of identifying what was considered canonical. On this point I know we have Josephus’ list, and if I remember correctly, the Muratonian Canon can be dated before 200. Other than that, I am not sure if we have any other lists that early.

  • Jordan and Bryant,
    You both are asking great questions, but I probably need to leave this as my last response. Let me see if I can summarize some threads.

    (1) Biblindex (as wonderful a tool as it is) does not index citations only. This is why the Skarsaune article is so helpful for the first and second century. Make sure to check it out. As a pertinent example, I examined all the references to Wisdom and Sirach in Philo according to Biblindex/Biblia Patristica. There was not a single citation, and I’m not even sure there were allusions to these books on the part of Philo. All references could be explained as thematic tropes or having no connection at all.

    (2) How do I determine canon? The first Christian canon list appears probably between 100-150 with the Bryennios List (see the Canon Lists book). Thus all we have before it are citations and material MSS, which are all open to interpretation. Citations can skew the picture. For example there is no citation of Genesis found among the Qumran materials but surely that book was scripture and rule for the community as its manuscript record seems to show. All I’m showing with the example of Philo in the article is that he only cites books from the Law and Prophets and from books of the much later Writings. He does not cite from all 22 books of Josephus but he only cites from books within this set, not the deuterocanonical books. This does not necessarily show his canon, but it’s all we have to go on. In any case, the proponent of the LXX canon has a lot of explaining to do in my estimation, for the most prolific Alexandrian Jew does not seem to recognize the six extra books, for he certainly does not cite them or probably even allude to them.

    If we couple this datum with the data of NT and second century church fathers, then a more firm conclusion could be made. Combine lack of citation with the second century lists that do not include the six extra books and we are on firmer ground for concluding that Jews and Christians did not recognize these books as part of their canon. Later use and clear statements about these books show Christians esteemed them and used them, but they did not reckon them as canon.

    (3) Yes, influence is measured by more than citations. Other uses such as allusions are important, and the evidence of MSS can also be an indicator for this early period. But we have to be careful in our interpretations because of the potential pitfalls we have been discussing.

    I hope this clarified some issues. We need to look at all evidence even as we recognize the potential pitfalls with the interpretation of each set.

    • John,

      Thanks again for taking the time to respond, and I know you will need to attend to other more important things.

      Your response and explanation is helpful, and I’m going to need chew on this some more.

  • Dear John, Jordan,

    Thanks for the interaction on this issue. A final thought regarding the “Septuagint Canon.”

    I have always thought that the issue of canonicity had something to do with inspiration as based on what Paul said in II Timothy 3:15-17; and, what Peter said in II Peter 1:20-21. It would seem that the recognition of what was canonical was intrinsic rather than extrinisic. It would seem that to do otherwise would be to place another authority over that of Scripture itself. I will not go there. Too many problems of external control, e.g. Council of Trent. The issue of final authority will not be settled any time soon; but we, at least, can narrow the playing field by relying on the Spirit of God to the witness that He has provided for us within us.

  • I trust the Holy Spirit to lead me into all truth. When Paul quoted poets, the intent of the quote was to bridge a truth to the reader by subverting the poetry. Not to give the poet or the entirety of the work credence.
    In contrast, when Peter and Jude quote Enoch and the author of Hebrews deliberately uses Septuagint citations they are writing theology. Asserting as fact the truth which if left to only the Hebrew cannon these theological truths would not exist. One example: Mr Luke in Acts 15:17 / Amos 9:12 – the rest (or remnant) of “men.” Hebrew – the remnant of “Edom.”
    The use of the Septuagint translation CHANGES EVERYTHING. Left as Edom it just sounds like a Zionist quote.
    Don’t have time to write about others.

  • I would like to mention three things, both I learned from Beckwith’s excellent book “the Old Testament Canon of the New Testament Church”.

    1) There’s no evidence anywhere of Jews having any debate on the books of the Apocrypha. Any idea of an open canon or an Alexandrian canon is only there to prop up Catholics and Eastern Orthodox or liberals who don’t like any idea of a settled biblical canon.

    2) Philo of Alexandria should know about the Alexandrian canon we would assume. He never quotes any Apocryphal book as Scripture.

    3) Melito of Sardis, in the 2nd century, asks the Palestinian church (primarily Jewish) what the OT canon was. This shows us a few things. First, he asked Palestine, not Alexandria. Secondly, it shows us he didn’t go down to the Septuagint store and just look at what the canon was. In other words, just saying “Septuagint” doesn’t give you a particular canon.

Written by Didaktikos
theLAB