Kae Times

Hidden in Plain Sight

The Wrong Word in Isaiah

Believers and sceptics both fight over almah and parthenos. Both skip what surrounds the word in Isaiah 7:14: a frightened king and a child with a timetable. The doctrine did not need a mistranslation. It found a convenient word.

Iann Fox_

Leia em português · Lea en español · Lisez en français · Leggi in italiano · Auf Deutsch lesen

The word is four letters, and I copied it into my notebook this morning while the coffee went cold: ayin, lamed, mem, heh. Hebrew is written without vowels, so the three consonants at the front also belong to a young man in the First Book of Samuel, and the final letter turns him into a young woman. That is the whole of the lexical drama. Out of those four letters, read aloud in thousands of churches every December, an entire theory of how scripture works has been built, defended and attacked, almost always by people who have not looked at the verse around it.

Here is what most people are sure of. A prophet named Isaiah announced, some seven centuries before Jesus, that a virgin would conceive and bear a son called Immanuel. The Gospel of Matthew says so; the prophecy came true; the proof is in the book. The sceptic's version runs on the same rails in the opposite direction: the Hebrew only said "young woman," the Greek translators got it wrong, and a doctrine was born from a clerical error. I think both stories are wrong, and the second is wrong in the more interesting way, because it flatters the person who tells it.

Start with the Hebrew, since it comes first. The verse is Isaiah 7:14: hinneh ha-almah harah ve-yoledet ben. Something like "Behold, the almah is with child, or will be, and bears a son." The word almah occurs seven times in the Hebrew Bible. It is used of Rebekah at the well in Genesis, and of the sister of Moses, who is sent to fetch the baby's mother in Exodus. It turns up among the girls with tambourines in a psalm, in the Song of Songs, and once in Proverbs, in a line about the way of a man with a maid. It names a young woman of marriageable age. It does not assert anything about her sexual history, either way.

The sceptic's favourite follow-up is that Hebrew has a dedicated word for a virgin, betulah, and Isaiah did not use it. That is partly true and often overstated. Betulah is the word the law reaches for when virginity matters legally. But in Joel 1:8 a betulah mourns for the husband of her youth, which is not what a dictionary would predict. And Rebekah, called a betulah in Genesis 24:16, is called an almah a few verses later, in 24:43. So the two words are not a clean pair of opposites, and anyone who tells you the Hebrew obviously settles the matter in either direction has not read Genesis 24. A young woman in that society was presumed to be a virgin. The word did not need to say so, and it did not forbid it.

Now the Greek. The translation of the Hebrew scriptures begun in Alexandria in the third century BCE is called the Septuagint, and it was done in stages. The Torah came first. Isaiah came later, most probably in the second century BCE. The Great Isaiah Scroll from Qumran, copied in roughly that same century, reads the same word in this verse, so there is no sign the translator had a different Hebrew in front of him. He looked at ha-almah and wrote hē parthenos.

Was that a mistake? I do not think so, and this is where I part company with the tidy sceptical story. Parthenos in the Septuagint is looser than the English "virgin" it later became. In Genesis 34, Dinah has just been seized and violated by Shechem, and the Greek text still calls her a parthenos a verse later. The word shades toward "girl," "maiden," "young woman of good standing." A translator in Alexandria who chose parthenos for almah chose a respectable, ordinary equivalent. He may well have meant "the young woman." We cannot know; he left no note. What we can say is that the Greek was a defensible rendering of the Hebrew and had the unfortunate property of being open to a stronger reading. The word did not lie. It was merely hospitable.

Then comes Matthew, and the point people usually skip. Matthew 1:23 quotes the verse in Greek, in a form so close to the Septuagint that the dependence is plain: idou hē parthenos en gastri hexei kai texetai huion. He changes one verb. The Greek translation has "you will call his name Immanuel," and Matthew has "they will call." That is the document, and it is perfectly clear: Matthew read Isaiah in Greek, and the Greek handed him a parthenos. He did not always need the Greek, either. When he uses Hosea 11:1 for the child carried out of Egypt, he follows the Hebrew, "my son," where the Septuagint has "his children." He took what served.

Now look at where the quotation sits. Matthew does not begin with Isaiah and arrive at Mary. He tells the story first. In verse 18, before any prophet is named, Mary is found to be with child of the Holy Spirit; Joseph resolves to put her away quietly; an angel comes in a dream; and only then does Matthew step back to say that all this took place to fulfil what the Lord had spoken by the prophet. It is tempting to read that order as a confession: the story existed, and the verse was fitted over it afterward. I will not read it that way. That is the order in which Matthew chose to present the matter, and a story built out of a verse could be narrated in exactly the same sequence. The arrangement of a text is evidence about its author's intentions. It is not evidence about where a tradition began.

What is evidence is Luke. Here is my argument, and a reader can fight it: the word did not make the doctrine; the doctrine found the word. Luke tells a virginal conception in a form that differs from Matthew's in almost every detail. The annunciation is to Mary, not to Joseph. The angel is named, and Mary asks how this can be, since she has not known a man. And Luke never quotes Isaiah 7:14 at all. Mark, the earliest Gospel, says nothing of it. Paul, who wrote before any Gospel, says only that Jesus was "born of a woman." In John, people call Jesus the son of Joseph. So the evidence has this shape: two Gospels, differing nearly everywhere, agree on the bare fact, and only one of them reaches for the verse. If the story had been born from Isaiah, Luke is the author who ought to have needed it. That is probable, not proven. The two accounts could draw on a source we have lost, and Luke could have known a tradition that Matthew's circle shaped. But it is the more economical reading, and it moves the supposed mistranslation out of the position of cause and sets it where it belongs, as a convenience Matthew found at hand.

The early Christians and their critics understood that the question was the word, and they argued over it in the open. In his Dialogue with Trypho, written in the middle of the second century, Justin Martyr has his Jewish opponent object that the text says "young woman" and not "virgin" (chapter 43, and again in chapter 67), and Justin answers at length in chapter 71 that the Jewish translators had altered the scriptures. Trypho is a character in Justin's book, and whether any such conversation ever took place we do not know. What the passage proves is narrower and still useful: by Justin's day the objection was familiar enough that a Christian writer expected his readers to meet it.

Irenaeus reports something firmer, a matter of translations. In Against Heresies, book 3, chapter 21, written in the later second century, he names Theodotion of Ephesus and Aquila of Pontus, whom he calls proselytes to Judaism, as translators who rendered the word in Isaiah 7:14 as neanis, "young woman," and he says the Ebionites followed them. Irenaeus sets their reading against the older Greek of the seventy translators and treats it as the reading of his opponents. That tells me the two sides were not disagreeing about vocabulary. They disagreed about what scripture was for.

Which brings me to what I think everyone has been looking at wrongly. The question was never only which noun to choose; it was the whole sentence, and the situation it sits in. Take the little word at the front: ha-, the article. Isaiah does not say "a young woman." He says "the." Scholars read that article as pointing to a particular woman known to speaker and listener, or as a way of speaking about any young woman in the circumstance. I do not think it can settle anything alone. The next word, harah, is no more decisive. The same construction appears in Genesis 16:11, said to Hagar, who is already pregnant. It appears again in Judges 13:5, said to a woman who is not yet pregnant, and there it announces a conception to come. So the Hebrew can mean "is with child" or "will conceive," and Isaiah 7:14 has been read both ways by people who read Hebrew for a living.

The setting is less ambiguous. Around 734 BCE, Ahaz, king of Judah, was afraid of a coalition between Damascus and the northern kingdom of Israel. Isaiah finds him at the end of the conduit of the upper pool, which is where you go to see to your water when a siege is coming, and offers him a sign. The sign has a timetable. Before the child is old enough to refuse evil and choose good, the land whose two kings Ahaz dreads will be abandoned. The child is a clock. A sign addressed to a frightened man about the next few years makes sense of the whole passage; a sign about a birth seven hundred years off does not help Ahaz at all.

Who the woman was, I do not know, and neither does anyone else. Readers ancient and modern have proposed a woman of the court, a wife of the king, a wife of the prophet. In the chapter that follows, Isaiah goes in to a prophetess, she conceives and bears a son, and the boy is given a long and ominous name; some readers connect the two scenes and some do not. That is a proper dispute and I will not settle it here. What the verse in its own setting will not do is announce, to Ahaz, a virgin conceiving in the reign of Augustus.

So my verdict has two halves, and they do not meet in the middle. First, the quarrel over almah is a quarrel over the wrong target, and the plain sceptical story, "it was only a mistranslation," credits a clerical slip with a power it never had. Second, the believer's claim that Isaiah foretold a virgin birth is not something the verse can carry. Matthew did not misunderstand it. He re-used it, as readers of scripture in his world did constantly, with a Greek version that happened to bend his way.

What does the record not prove? It does not prove that Mary conceived without a man, and it does not prove that she did not. That is not a question a word can settle, and I have no interest in pretending a concordance can. A historian can say what the texts claim, when they were written, and which word they lean on. The rest belongs to the people who keep the feast.

I think about the translator in Alexandria sometimes, the one who wrote parthenos: a man whose name no one saved, working in a city whose great library is now a rumour. He could not have known that a single choice in a single line would be taken, two centuries on, for a hinge. He may have been tired. He may have thought it a perfectly good Greek word for a girl. The question the record cannot answer is the one I keep returning to. If a different man had been at the desk that day and had written neanis, would Matthew have found his verse somewhere else?