The Septuagint in the Air

When the modern Christian reads the New Testament and encounters a quotation from the Old Testament — as it is written, the scripture says, the prophet declares — she usually pictures the writer reaching for a Hebrew Bible, opening it to the right passage, and copying out the verse. The picture is intuitive. The Old Testament is the Hebrew Bible. The New Testament writers were Jews, and Jews read Hebrew, and so the quotations must be coming from the Hebrew text. The verse on the page in the apostle’s hand and the verse on the page in the modern reader’s Bible must be, fundamentally, the same verse.

The picture is wrong almost everywhere it appears. The Bible most New Testament writers were actually reading — and the Bible most of their first-century audiences were actually hearing — wasn’t the Hebrew Bible. It was a Greek translation of the Hebrew Bible. The Old Testament of the New Testament is, much more often than not, the Greek Old Testament. Once this becomes audible, certain New Testament passages — passages that have produced some of the most weighted Christian theology of the last two thousand years — turn out to be carrying their weight on the Greek translation’s wording, and on a wording that sometimes diverges from the Hebrew text the modern reader’s Old Testament was translated from.

the sentences that don’t quite match

There are sentences in the New Testament that puzzle the careful modern reader because the quotation doesn’t quite match what she finds when she looks up the cited Old Testament verse.

The virgin will conceive and give birth to a son, and they will call him Immanuel (Matthew 1:23, NIV). Matthew is quoting Isaiah 7:14. The modern reader turns to Isaiah 7:14 in her NIV, ESV, NRSV, or any modern Old Testament. She sees a verse about a young woman — or, in the case of older translations following the Septuagint, a verse about a virgin. The Hebrew word Isaiah used doesn’t specifically mean virgin. Matthew’s verse does. Something has happened in between.

A body you prepared for me (Hebrews 10:5, paraphrase of NIV). The writer of Hebrews quotes Psalm 40:6 as words spoken by Christ at his incarnation. The modern reader turns to Psalm 40:6 in her Old Testament. She doesn’t find a body you prepared for me. She finds something like my ears you have opened (NIV) or you have given me an open ear (ESV). Ears, not body. Something has happened in between.

The rest of mankind may seek the Lord, even all the Gentiles who bear my name (Acts 15:17, NIV). James quotes Amos 9:11-12 to settle the Gentile-mission question at the Jerusalem Council. The modern reader turns to Amos 9:12 in her Old Testament. She doesn’t find the rest of mankind seeking the LORD. She finds something like that they may possess the remnant of Edom and all the nations that bear my name (NIV). Edom, not mankind. Possess, not seek. Something has happened in between.

What has happened in between, in every case, is the Greek translation. The verse Matthew, the writer of Hebrews, and James are quoting isn’t the verse in the modern reader’s Hebrew-based Old Testament. It’s the verse in the Greek translation of the Hebrew Old Testament that was the working Bible of first-century Greek-speaking Jews and Christians. To understand what the New Testament writers are doing with the Old Testament, the modern reader needs to know what that Greek translation was.

what the Septuagint is

The standard accessible treatment of this material is When God Spoke Greek: The Septuagint and the Making of the Christian Bible by Timothy Michael Law, a biblical scholar who completed his doctoral work at Oxford and has held research appointments there and at St Andrews. Oxford University Press published the book in 2013. Law’s argument, in summary, is that the Greek Old Testament wasn’t a secondary translation that early Christians grudgingly used because they couldn’t read Hebrew. It was the Bible of the Greek-speaking Jewish world for several centuries before Christ, the Bible most New Testament writers wrote with in hand, and the Bible the early church regarded as scripture for its first four hundred years. Only late — toward the end of the fourth century AD, when Jerome decided to translate the Old Testament from the Hebrew rather than the Greek — did Western Christianity begin to drift back to the Hebrew text. The reader who only knows the Hebrew-based Old Testament tradition is missing the Bible the apostles actually used.

The Septuagint’s origin story is well-known and partly legendary. In the third century BC, a Greek-speaking Jewish community had grown large and prosperous in Alexandria, Egypt — the new intellectual capital of the Hellenistic world. Within a generation or two of Alexander the Great’s conquests, many of these Jews were speaking Greek as their first language, not Hebrew. They needed a Bible they could actually read. A Jewish text called the Letter of Aristeas, probably composed in the second century BC — long after the events it claims to narrate, and regarded by scholars as a semi-legendary apologetic rather than a historical record — tells the story of how the translation came to be: Ptolemy II Philadelphus, king of Egypt, sponsored the project, and summoned seventy-two Jewish scholars (six from each of the twelve tribes) to Alexandria, where they completed the translation in seventy-two days. Later retellings embellished the story into a miracle: Philo has the translators, working independently, each arrive at a word-for-word identical version under divine inspiration, and later tradition added the detail of seventy-two scholars sealed in separate cells. The number seventy-two became seventy by the rounding habits of antiquity, and the translation acquired its name: the Septuaginta, the Seventy, abbreviated LXX.

The historical reality was less dramatic. The translation wasn’t produced in one heroic Alexandrian week. The Torah — Genesis through Deuteronomy — was probably translated in the third century BC. The Prophets and the Writings followed over the next two centuries, by different translators of varying skill, in different cities, with different translation philosophies. Some books are tight and literal — almost word-for-word reproductions of the Hebrew. Others are loose and idiomatic, more like paraphrases. A few, like the Greek Daniel, are substantially longer than the Hebrew Daniel and include material that doesn’t exist in the Hebrew at all. By the first century AD, when the New Testament writers were quoting their scriptures, they had in front of them a Greek Bible that was, in most places, a good translation of the Hebrew — and in a small but theologically significant number of places, a different reading than the Hebrew text now preserved in modern Old Testaments.

The Hebrew text most modern English Old Testaments translate from is called the Masoretic Text — the carefully preserved Hebrew tradition fixed by Jewish scribes called the Masoretes between roughly the seventh and tenth centuries AD. The Masoretic Text is a later tradition than the Septuagint, though it preserves an older Hebrew base. When the Septuagint and the Masoretic Text disagree, the disagreement is sometimes a translator’s interpretive choice (the LXX translator paraphrasing or smoothing the Hebrew), and sometimes a real textual difference — the Hebrew text the LXX translators were working from in the third or second century BC wasn’t always identical to the Hebrew text the Masoretes preserved a thousand years later. The Dead Sea Scrolls, discovered between 1947 and 1956, have confirmed that multiple Hebrew text-forms were in circulation in the centuries before Christ, and that some readings preserved only in the Septuagint actually do reflect a Hebrew text-form that existed in antiquity. The textual situation is older, messier, and more interesting than the standard Sunday-school picture admits.

What this means practically for the modern reader of the New Testament is this. When a New Testament writer quotes the Old Testament, the words on the page are almost always Greek words from the Greek translation. Sometimes the Greek matches the Hebrew exactly. Sometimes the Greek is a smoothing or paraphrase of the Hebrew. And sometimes the Greek says something the Hebrew doesn’t say — and the New Testament writer is making a theological point that depends specifically on the Greek wording. Three cases will show what this looks like in practice.

three places the Greek does the work

Therefore the Lord himself will give you a sign: The virgin will conceive and give birth to a son, and will call him Immanuel (Isaiah 7:14, NIV). The verse the modern Christian almost certainly knows by heart, because Matthew quotes it at the opening of his Gospel: All this took place to fulfill what the Lord had said through the prophet: “The virgin will conceive and give birth to a son, and they will call him Immanuel” (which means “God with us”) (Matthew 1:22-23, NIV). The virgin birth of Jesus is presented as the fulfillment of a prophecy spoken seven hundred years earlier.

The Hebrew of Isaiah 7:14 uses the word almah (עַלְמָה) — a young woman of marriageable age. It doesn’t specifically mean virgin. The Hebrew word that specifically means virgin is a different word, betulah (בְּתוּלָה), which Isaiah didn’t use. The Hebrew can carry the meaning virgin, but doesn’t specify it. The Septuagint translators, working in Alexandria in the second century BC — two centuries before the birth of Jesus, with no Christian agenda in view — translated almah with the Greek word parthenos (παρθένος). Parthenos, unlike almah, does mean virgin specifically. The translators may have made this choice because an unmarried young woman of marriageable age would be presumed a virgin in their cultural setting, or because they read Isaiah’s sign as something more dramatic than a routine pregnancy. Either way: the Greek translation locked the verse into the specifically-virgin reading two centuries before Matthew.

Matthew quoted the Greek. The whole virgin-birth-as-fulfilled-prophecy argument in Matthew 1:22-23 rests on the Septuagint’s parthenos. Without the LXX, Matthew has a young woman of marriageable age conceiving a son — which fits Mary, but doesn’t specifically rule out the normal way young women conceive sons. With the LXX, Matthew has Isaiah seven hundred years earlier specifying the virginal conception of the Messiah. The Christological weight of the verse — the weight that has carried two thousand years of Christmas sermons and Marian devotion — is real. It’s also riding, very specifically, on the Greek word the Alexandrian translators chose.

Sacrifice and offering you did not desire, but a body you prepared for me (Hebrews 10:5, NIV). The writer of Hebrews is making the central argument of his letter: that the sacrificial system of the old covenant has been fulfilled and ended in the once-for-all offering of Christ’s body on the cross. The argument turns, in Hebrews 10:5, on a quotation of Psalm 40 presented as words spoken by Christ at his incarnation. The body that was prepared is the body that was offered. The verse is Hebrews’ Christological pivot.

The Hebrew of Psalm 40:6 reads, literally, ears you have dug for me — a striking image of God hollowing out the psalmist’s ear-canals so he can hear and obey. Modern English Bibles based on the Hebrew handle the idiom variously: my ears you have opened (NIV), you have given me an open ear (ESV). The Hebrew is about hearing, about ears prepared for obedience. The Septuagint translator, faced with the unusual Hebrew expression, treated the ears as a part standing for the whole — the digging of ears as part of the larger work of preparing a body — and translated the line a body you have prepared for me (σῶμα δὲ κατηρτίσω μοι — sōma de katērtisō moi). Body, not ears. The Greek translation reframed an obedience metaphor into an incarnation statement.

The writer of Hebrews quoted the Greek. The Christological argument of Hebrews 10 — that Christ’s body was prepared, and that the same body was offered as the once-for-all sacrifice that ends the old sacrificial system — rests on the Greek translation. The Hebrew text’s ears you have dug for me wouldn’t have produced the argument. The Septuagint’s body you have prepared for me did.

The rest of mankind may seek the Lord, even all the Gentiles who bear my name (Acts 15:17, NIV). At the Jerusalem Council in Acts 15, the apostles are debating one of the most consequential questions in the church’s first decades: must Gentile converts become Jews — circumcised, Torah-observant, kosher — before they can be received as full members of the Christian community? The conservative party says yes. Paul and Barnabas say no. Finally James — Jesus’s brother and the head of the Jerusalem church — speaks. He cites Amos 9:11-12 to settle the question. The Gentile mission, he argues, is what the prophets had foretold.

The Hebrew of Amos 9:12 reads, in standard modern translations: that they may possess the remnant of Edom and all the nations that bear my name (Amos 9:12, paraphrase). The verse envisions Israel possessing Edom — the territory of Israel’s ancient cousin-nation, descendants of Esau. The Hebrew is a vision of political restoration: the restored Davidic kingdom will dominate Edom and the surrounding nations. The Septuagint reads differently: the rest of mankind may seek the Lord, even all the Gentiles who bear my name. Not possess Edom but the rest of mankind — humanity, not Edom. The Greek translator apparently read the Hebrew consonants aleph-dalet-mem not as Edom (אֱדוֹם) but as adam (אָדָם — humanity); the consonants are identical, only the vowels differ, and ancient Hebrew was written without vowels. The Greek also reframed the verb — not possess but seek — turning a vision of political dominion into a vision of universal religious search.

James quoted the Greek. That sentence — the rest of mankind may seek the Lord, even all the Gentiles who bear my name — read out at the council in Jerusalem in the late 40s AD, was the prophetic charter for the Gentile mission. The Hebrew text’s vision of Israel possessing Edom wouldn’t have produced the council’s decision. The Greek version’s vision of all humanity seeking the LORD did. One of the most consequential ecclesiological decisions in the church’s history — that Gentile converts could be received as Gentiles without first becoming Jews — was anchored, in the actual minutes of the Jerusalem Council, in a Septuagint reading that the Hebrew text doesn’t support.

what this means for reading the New Testament

The recovery this chapter is after isn’t a destabilizing one. The claim isn’t that the New Testament writers were sloppy quoters or that the Greek Old Testament is wrong or that the Hebrew Bible is wrong. The claim is that the Old Testament of the New Testament is, much more often than not, the Greek Old Testament — and that recognizing this changes how some of the New Testament’s most weighty arguments read.

What the modern Christian gains by hearing this is something her tradition has often had trouble holding together. The New Testament writers weren’t working from a fixed, settled, English-style Bible with one universally agreed wording for each verse. They were working in a textually complicated first-century environment in which the Hebrew scriptures had been translated into Greek, the Greek translation was in widespread use, the translation sometimes diverged from the Hebrew, and the apostles — speaking and writing in Greek to Greek-speaking audiences — quoted the Bible their audiences could read. When Matthew identified Isaiah’s parthenos with Mary, when the writer of Hebrews built his Christology on Psalm 40’s body prepared, when James settled the Gentile-mission question with Amos’s rest of mankind seeking the LORD, the apostolic writers were doing what reverent readers of scripture have always done: they were quoting the Bible they had. The Bible they had was the Greek Bible.

This isn’t a problem to be solved. It’s a feature of the textual world the New Testament emerged from. The early church recognized this without difficulty; for the first four centuries of the Christian movement, the Septuagint was the Christian Old Testament — Augustine, the Greek Fathers, the early ecumenical councils all used it as scripture. The Eastern Orthodox tradition uses it to this day. The shift to a Hebrew-based Old Testament in Western Christianity began with Jerome at the end of the fourth century, was completed by the Reformation, and produced the modern Protestant and Catholic Old Testaments that are translated from the Masoretic Hebrew text. None of this is wrong. The Hebrew text is the more linguistically original. But the Hebrew text isn’t, in most cases, the text the New Testament writers had in front of them.

For the modern reader, what this means in practice is small but real. When a New Testament passage quotes the Old Testament and the quoted wording doesn’t match what the modern reader finds when she looks up the cited verse in her Old Testament, she doesn’t have to assume a copying error or a hostile reinterpretation. The much likelier explanation is that the New Testament writer was quoting the Greek Bible his audience had, and that the Greek Bible read a little differently than the Hebrew Bible from which her modern Old Testament was translated. The Septuagint is what was in the air. The apostles breathed it. The first audiences heard it. The early church canonized it. And modern readers, working with Hebrew-based Old Testaments, can recover what the New Testament’s first readers heard by knowing that the Greek translation existed, that it was widely used, and that some of the most consequential New Testament arguments are doing their work on the Greek translation’s wording — not in spite of the wording, but because of it.