Top Ten Surprises in the Septuagint

You may know that the Septuagint (abbreviated LXX by common agreement, with reference to the ancient legend of its drafting by seventy scholars) is the umbrella term for the Greek translation of the Old Testament. It was already complete in at least one version by the end of the third century BCE or so, thanks to the thorough Hellenization of the eastern Mediterranean world in the wake of Alexander’s conquests, and thus of the dispersed Jewish communities in those lands. The first edition generally goes by the name of the Old Greek, and revisions followed that go by the names of Aquila, Symmachus, Theodotion and Lucian in the centuries either side of the time of Christ, as well as Origen’s rendition in his Hexapla and others.

You may also know that the Septuagint is the most important source for the textual criticism of the Old Testament, being the oldest complete witness to the state of the Hebrew text lines a couple of centuries before Christ. When there are significant differences between the LXX and the Hebrew Masoretic Text lying behind our modern English versions, the sixty-four million-dollar question is often whether the LXX reflects roughly the same Hebrew text but paraphrases or amends it freely, or instead whether there is a quite different Hebrew text tradition lying behind the LXX that has since gone the way of the dinosaur.

The trend in biblical scholarship is tending towards the latter: an increased respect for the care and fidelity of the LXX translator in many cases, and the acceptance of the fact that that translator was looking at a Hebrew original rather different to the Masoretic tradition. In some cases this hypothesis has been buttressed by a Dead Sea Scroll (mostly Qumran) fragment that contains or indicates the kind of alternate reading the LXX uses. At times the Qumran scrolls approach the LXX in text-critical importance, but they are not as old as the textual ‘bifurcation point’ represented by the LXX, and for some Old Testament (OT) books they are very fragmentary and do not cover much of the book concerned. Have a look at some scroll images in the Dead Sea Scrolls Digital Library and you’ll see what I mean. The following (chosen from Wikimedia Commons to avoid violating anyone’s copyright) is the Community Rule from Qumran Cave 1, in rather good condition, but many of the scrolls are in little shreds, e.g. 1QSam (Samuel), which the Dead Sea Scrolls Bible of Abegg, Flint & Ulrich (p. 214) describes as having “a fragment roughly the size of a quarter…preserved at the same position on each of eight successive layers of leather.”

Dead_Sea_Scroll,_1Q Community Rule

So, back to the Septuagint. Let’s have ten of the most interesting and dramatic differences between the Septuagint and the Masoretic Text (MT):

  1. Perhaps most famously, LXX Jeremiah is about 1/8 shorter than the Masoretic version, a length difference partly composed of some larger chunks like Jer. 33:14-26 having no equivalent in LXX, and partly from accumulated smaller differences. It also has Jeremiah’s ‘oracles against the nations’ in the middle of the book rather than near the end, and these are in a different order. Jack Lundbom thinks that the Greek translator has made systematic visual errors in translation, leaving a superior MT, but most commentators would agree with text-critical guru Emanuel Tov, who thinks, given the support of two minor Qumran manuscripts of Jeremiah, there is a different Hebrew original text-type that was being used by the Greek translator. We probably have evidence here that this and the more familiar text type had diverged developmentally by the time this translation was made, and in fact, the MT version looks rather amplified compared to the Greek version.
  2. The LXX for Exodus 35–40 is briefer and differently ordered compared to MT. Once again, LXX seems to have captured a form of text that has not yet undergone its last revision to reach the form we find in MT and in our Bibles, where the completion of the tabernacle is more fully described in terms that match the instructions of Exodus 25–31.
  3. The LXX and MT represent two different versions of the David & Goliath story and events either side of it, seen in 1 Samuel 16–18. Tov (in paper cited at end, p. 4) and others think MT represents two versions of the story interwoven, and feel that the much briefer LXX version has fewer problems in terms of narrative coherence. For instance, the section after the victory over Goliath where Saul does not seem to know who David is, despite having him for an employee late in ch. 16, that is, 17:55-58, is not in LXX. Also, the saga surrounding consideration of David for marriage to two different daughters of Saul in chapter 18 is simpler in LXX: there is just one daughter, Melchol, and David marries her in reward for victory over the Philistines. No, I’m not sure what to make of that difference yet.
  4. The book of Joshua is rather different in LXX from MT, with length variations in both directions. And the important episode of covenant renewal found in our Bibles as Josh. 8:30-35 appears a little later in LXX, after our 9:1-2. The LXX shows its own internal divergences later in the book, with Codex Alexandrinus (A) and Vaticanus (B) reading like different editions in parts of the boundary lists in Joshua 15, 18–19, and then for the entire book of Judges!
  5. The LXX is slightly different from MT in Psalms, adding a 151th psalm and dividing certain psalms into two or fusing two into one in a few places. E.g. Psalm 147 is Psalms 146 & 147 in LXX. Some psalms also gain different titles, e.g. Psalms 145–148 in LXX are attributed to the prophetic pair of Haggai & Zechariah!
  6. In a case where LXX seems more clearly to represent a secondary development, the LXX translator of Job seemed to grow weary of the length of Job’s debate with his friends and abbreviated it somewhat.
  7. Similarly, the LXX version of books like Daniel and Esther reflects a tendency to expand on the original. A well-known example is where Daniel’s three friends, once thrown into the ‘fiery furnace’ in Daniel 3:22-23, sing a very involved praise song that shows dependence on Psalm 148 amongst other biblical texts.
  8. Ezra-Nehemiah show an interesting and complicated relationship to the LXX books of Esdras. Don’t even ask me to explain which Esdras pertains to which part of our Ezra-Nehemiah. It depends whether you’re looking at LXX, the Latin Vulgate, etc. etc. It confuses me every time. But as an example, the list of returnees from exile that already turns up twice in our Bible in Ezra 2 and Nehemiah 7 also turns up in a third copy in 1 Esdras 5, and each differs from the other in ways that witness textual development.
  9. Trust me, there are others, but I think I’ve made my point. These are some of the major ones.
  10. See #9.

What should we make of such differences? Which text is superior? It’s a mix, I think. Sometimes the LXX looks like a secondary development of MT and involves some rather free rewriting or paraphrasing of OT content. In such cases we might wish to lean toward the MT as our textual reference-point. Such a position has been taken quite strongly by some Christians, e.g. famously in the Helvetic Consensus (1675), which held the inspiration of the Hebrew text down to its vowels. But it isn’t so simple; at times the LXX seems to reflect an earlier stage of development of a book that continued to grow in Jewish tradition, as in the case of Jeremiah. And the New Testament writers are influenced by and often quote or allude to the LXX version of the OT wherever there is a difference to tell, including in instances where it makes an interpretive difference, e.g. in Hebrew 1:8-9 (check a commentary).

We will need to step lightly as we look for our textual reference points with regard to the Old Testament. These texts often were not created at a setting, but set more like jelly (jello for you Americans) in the fridge (refrigerator, for you more formal people), and sometimes the jelly was remoulded (remolded, for you Americans, and so forth) a little before it finally set. When God gave us his word, He didn’t hand it from the sky on golden plates. That’s the Mormon/LDS back-story, not the biblical one. God seems to prefer to get His hands dirty by involving human beings in His work – surprising and risky, but true. Call it grace. And humans, very smart ones and usually very spiritual ones, did a lot of revising and re-editing work on our scriptures before they got to us. That some of the textual differences we find when we compare text streams like MT and LXX are quite nuggety problems (which Goliath story should I credit?) is not surprising. It’s why if biblical criticism of the ‘higher’ kind was abolished tomorrow, it would spring back into being a few days afterwards. But it’s all a function of a God who, like a parent, lets his kids get involved in his work, even though it would be cleaner and simpler to do it all Himself. It’s the same as Christ builds His church. It’s counter-intuitive, but He seems determined to involve us. Call it grace.


Tov, E. “The Nature of the Large-Scale Differences between the LXX and the MT S T V, Compared with Similar Evidence in Other Sources.” In The Earliest Text of the Hebrew Bible: The Relationship between the Masoretic Text and the Hebrew Base of the Septuagint Reconsidered, edited by Adrian Schenker, 121–44. Atlanta, GA: Society of Biblical Literature, 2003.

Leave a Reply

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

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

Google photo

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

Twitter picture

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

Facebook photo

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

Connecting to %s