God Didn't Say That

Bible Translations and Mistranslations

Who are you calling a virgin?

The NAB’s decision to change “virgin” to “young woman” in Isaiah 7:14 has once again brought up the virgin birth, Mary, and the nature of prophesy, as well as the role of translation in accurately conveying the text of the Bible.

Most reports I’ve seen recently, though, confuse what are really three separate issues here.

Isaiah 7:14

The first issue is the text of Isaiah 7:14. The Hebrew there reads: “an alma … will bear a son and call him `Emmanuel.'” It has long been known that alma does not mean “virgin.” Rather, the Hebrew word applies to any young woman. So the English translation of that line should read along the lines of “a young woman … will bear a son…” (The evidence is widely known and readily available, including in my And God Said.)

Unfortunately, the Septuagint — the highly influential ancient Greek translation of the Old Testament — got the translation wrong here, translating the Hebrew alma as the Greek parthenos, which (probably) did mean “virgin.” It was an easy mistake to make, because most young women back then were virgins, and most virgins were young women. It would be like translating “teenager” as “high-school student” in a society where most teenagers were in fact in high school.

Based on this mistranslation, though, most modern translations — going back to the KJV and including the recently published NIV — translate “a virgin … will bear a son” here. (The NIV has a footnote, “or young woman.”) The new NAB (“NABRE”) is a notable exception. That version now has, “the young woman, pregnant and about to bear a son, shall name him Emmanuel.” Their choice to go with “young woman” reflects the correct understanding of the original Hebrew (though I do have problems with their phrasing of the rest of the line).

The Virgin Birth

Importantly, though, Isaiah 7:14 is not the description of “the virgin birth” of Jesus. Rather, we find the virgin birth first in Matthew 1:18-25, which brings us to the second issue.

As part of the description of Jesus’ birth, the text in Matthew cites Isaiah 7:14, noting that Jesus’ birth “fulfilled” (plirow) the phrophet Isaiah’s words (a point I return to below).

Matthew 1:18-25 only uses “virgin” (parthenos) in quoting Isaiah 7:14. But Matthew’s description of Jesus’ birth is nonetheless clear on the matter. The text uses the euphemisms “before [Mary and Joseph] came together [sunerchomai]” and “[Joseph] did not know [ginosko] her [Mary] until after she gave birth” to indicate that Mary was a virgin, and the text twice clarifies that the pregnancy was “from the Holy Spirit” [ek pneumatos agiou].

These combine to create a clear account: Jesus was born to a virgin.

The text in Luke 1:26-38 is similar in nature. Though again “virgin” is replaced with a euphemism (“Mary asked the angel, `how [is it possible that I will conceive] since I do not know [ginosko]” any men?), the text is clear, adding for emphasis that “with God nothing is impossible.”

The actual descriptions indicate a virgin birth, regardless of what the words in Isaiah 7:14 mean.


The third issue is how to reconcile the virgin birth with Isaiah 7:14, which is cited in Matthew 1:23.

The most straightforward way is to note that even though Isaiah 7:14 refers to a “young woman,” not a “virgin,” the text doesn’t say that she wasn’t a virgin. She could have been. (By comparison, the text also doesn’t say that the woman had long hair, but she might have.) In other words, Isaiah 7:14, even with the better understanding of the original text, doesn’t contradict anything in the NT.

The more nuanced way to reconcile the two texts is to recognize what the verb in Matthew 1:22, plirow, really indicates. Though the word is commonly translated “fulfill” (as in, “All this took place to fulfill what the Lord had said through the prophet [Isaiah]”), better is “match,” as I describe here (“What Happens to Prophecies in the New Testament?”). I won’t go through the whole explanation again, but for now I think it suffices to note that Matthew knew that the details in Isaiah 7:14 differed from those he was describing. After all, the name of the child in Isaiah 7:14 was Immanuel, not Jesus.

Either way (and even though it’s not really my place to say), I don’t see a huge theological problem here. And even if there were a problem, I would still be in favor of an accurate translation.


It seems pretty clear to me that Isaiah 7:14 mentions a pregnant woman (who, at least as far as translation can take us, may or may not have been a virgin) and that the NT refers to the virgin birth of Jesus. It seems equally clear that the lack of perfect harmony between the texts is in keeping with other kinds of prophesy in the NT.

Still, from the international stage (“traditionalists may see [the NABRE’s change from “virgin” to “young woman” in Isaiah 7:14] as a step away from the original meaning”) to local communities (“If the meaning of the language is changed to reflect that Mary may not have been a virgin, you’ve just denied the divinity of Christ”) the discussion seems skewed to me. It seems to start with theology, and then ask how the translations can be doctored to match that theology, while I think an accurate translation should stand on its own.

Or to put it another way, it seems to me that basing theology on a translation designed solely to support that theology is both bad translation technique and bad theology.


March 23, 2011 Posted by | Bible versions, translation practice, translation theory | , , , , , , , , , , , | 45 Comments

What Happens to Prophecies in the New Testament?

What happens to prophecies in the New Testament?

The obvious answer is that they come true, but I think a more careful look shows otherwise.

Matthew 1:18-22 / Isaiah 7:14

As an example of a prophecy apparently coming true, we might consider the first chapter of Matthew. The text starting around Matthew 1:18 deals with the virgin birth of Jesus, fulfilling the prophecy of virgin birth in Isaiah 7:14. The text even reads (Matthew 1:22; NRSV), “All this took place to fulfill what had been spoken by the Lord through the prophet.”

John 19:24 / Psalm 22:18

Similarly, according to John 19:24, the soldiers cast lots for Jesus’ tunic to fulfill the prophecy of Psalm 22:18, “…and for my clothing they cast lots.” (Matthew 27:35 has the same account, but not all manuscripts have the direct reference to Psalms there.)

Fulfillment of Prophecy

Both of these seem to be cases of prophecies coming true.

But the Greek word in each case is plirow. And while “fulfill” is one common translation of that verb, I don’t think it’s accurate.
Continue reading

October 19, 2010 Posted by | translation practice, translation theory | , , , , , , , , , , , , , , , | 21 Comments

Top Translation Traps: Myopic Translations

Sometimes it seems that translators look too closely at individual words, only asking “how do I say this ancient word in English?” rather than asking “how do I translate this text into English?” I think this flawed approach comes in part from ignorance, but also from the religious tradition that each word has meaning. So this is one way in which scientific translation can sometimes diverge from religious interpretation.

Getting it Right


As a simple example of a good translation that comes from looking beyond individual words, we can consider Numbers 24:5, which is about Jacob’s tents (“your tents, Jacob”). The first Hebrew word in that verse, ma, means “what” and the second word (tovu) means “were good.” But it’s wrong to translate “what good were your tents, Jacob?” Every translation that I know of gets this right with “how good are your tents…”

This is a case where a word normally has one translation (“what,” in our example) but certain circumstances call for another (“how,” here).


Matthew 1:18 is similar. It’s the first of several times we find the Greek phrase en gastri, “in the womb”: “Mary was found to be en gastri….” But it doesn’t mean that Mary was in the womb, because the next Greek word is echousa, “holding.” “Holding in the womb” is Greek for “pregnant,” or — as used to be common — “with child.”*

Again most translations get this right, correctly realizing that even though the Greek words for “in” and “womb” appear in the original, the English words “in” and “womb” have no place in the translation. To try to form a sentence with “in” and “womb” would be overly myopic, focusing too closely on the words and not on how they work together.

This phrase-level issue is pretty close to internal structure, which I discussed last week.

Getting in Wrong


One of the clearest ways in which translations are myopic is when it comes to light verbs like the Greek poieo. Acts 2:22 is a perfect example both of the problem and the difficulty of getting it right. There, God poieos three kinds of things: dunamis, teras, and simeion.

Translations such as (NRSV) “deeds of power, wonders, and signs that God did…” seem to ignore basic English grammar. We don’t say “do deeds of power” in English, or “do signs.” What seems to have happened is this: The translators looked at each word in isolation, myopically asking, “how do we say dunamis?” or, “how do we say poieo?” Once they had answers, they crammed them together.

Using “work” for poieo — as in the NAB, “mighty deeds, wonders, and signs, which God worked…” — doesn’t seem much better. “Working mighty deeds” similarly isn’t English.

We do have grammatical ways to express the same thing in English: “God’s wonders,” for example, or “God’s signs,” instead of “the wonders that God did/worked.” (I’m purposely ignoring dunamis for now.)

But the whole sentence makes that approach difficult, because “God poieod the three kinds of wonders through Jesus of Nazareth. Continuing the pattern we just tried, we would get “God’s wonders through Jesus,” but I don’t think that’s right, because the original Greek refers to how the wonders were performed, not what kind of wonders they were.

So we might try, “performed wonders,” which is at least grammatical in English. But we’ll run into trouble with “performed works,” which doesn’t make much sense, and we don’t have a translation for dunamis here.

Still, even without a successful translation (any ideas?) I think the concepts are clear. What we need here is the equivalent of “how” instead of “what” for ma, that is, a way of expressing in English what the Greek expresses very clearly. What we don’t want is what most translations offer: a translation that looks at each Greek word in isolation, renders it in English, and then hopes that those English words will make sense when put together.


Finally, to round things out, we can consider the Hebrew phrase eitz hasadeh. The words mean “tree of the field” (and this is how most translations render the phrase), but the phrase probably means “fruit tree.”


The lessons are clear, and, unfortunately, they invite a cliché in summary: Translators who focus myopically on the words risk missing the forest and seeing only trees.

[Posted at 33,000 feet on my way back from teaching in New Orleans.]

(*) UPDATE: J.K Gayle has posted some background on the Greek phrase “(holding) in the womb.”

March 15, 2010 Posted by | translation practice, translation theory, Translation Traps | , , , , , , | 3 Comments