The Problem of the Virgin Birth Prophecy (2003)
Richard Carrier
In March 23 of 2003 I received the following question by email. I was encouraged to publish my answer online, since it would be of help to many others faced with the same question.[1] Some of my advice is applicable to many other questions of a similar nature, exemplifying the utility of the Secular Web Library as a research resource. The question went as follows:
Hello Mr. Carrier. I have a question about the Immanuel Prophecy (Isaiah 7:14). Did the Prophet Isaiah say that a “virgin” or a “young woman” would conceive? According to Tim Callahan in his book Bible Prophecy: Failure or Fulfillment? (cf. pp. 115-16), the Hebrew word almah, “meaning a young woman of marriageable age, was translated into Greek in the Septuagint as parthenos or ‘virgin.’ Had the Hebrew meant to say virgin it would have used bethulah, which means specifically a virgin.” Yet according to I. J. Mikulski, a Catholic priest, Isaiah “used the word almah that can be translated ‘young woman’ as a synonym for ‘virgin.’ In fact,” he says, “there is no instance in the Hebrew Sacred Writings (our Old Testament) where almah means a young woman who is not a virgin.” Mikulski says “Old Testament writers understood that and chose their words accordingly. Matthew also understood that, of course, when he quoted Isaiah’s word in Greek.” Thus, “He used the word parthenos that has the precise meaning ‘virgin’. That’s the word used in the Greek Septuagint version of the Sacred Writings (Old Testament) translated about 250 B.C. by Jewish scholars for the Greek-speaking Jewish community in Egypt. So,” according to Mikulski, “nearly three centuries before Mary gave birth to Jesus Christ, nearly three centuries before anyone had reason to question Catholic doctrine, the meaning of Isaiah’s words almah-parthenos-virgin was clear.” Between Callahan and Mikulski, who is right?
This is a very old debate. Callahan’s argument, for example, was long ago made by the infamous infidels Joseph Wheless and Joseph McCabe, whose writings are made available in the Secular Web Historical Library.[2] But their writings and methods are much out of date. We have several better discussions of the issue in the Secular Web Modern Library, which demonstrate that Callahan is not making anything up. For instance, the Hebrews did have a word that more clearly meant virgin: bethulah. So the choice of almah does count against Mikulski, with regard to the original Hebrew text of the bible. For three more modern analyses of the whole question of this virgin birth prophecy, see the relevant essays by James Still, Jim Lippard, and Farrell Till.[3]
However, it is important to point out first that the debate might be moot anyway. For the two options presented by Callahan and Mikulski do not exhaust all the possibilities, since Isaiah can be interpreted non-supernaturally even if he did mean virgin. After all, is it really unusual for a virgin to conceive? Say, on her wedding night? True, then she isn’t a virgin anymore. But she was until she conceived (literally, not at that very moment, but the Bible is rarely so precise). Since conception does not always occur the first time it would still be significant to say that a virgin conceived, meaning only that she conceived the first time she was with a man. In fact, this is the very conclusion reached by the renowned Catholic scholar and nativity expert Raymond Brown (whose own analysis of this question I will discuss further below).[4]
But that aside, in sum, the truth is more likely with Callahan. Here are the various reasons why:
1. The Textual Tradition is not Iron Clad
Mikulski has his history just a bit wrong, and the ground is shakier than he implies. First, the Septuagint translated around 250 B.C. was originally just the Torah. The book of Isaiah wasn’t in it. The translation of Isaiah into Greek was added to the Septuagint a century or so later (as with other OT books, including several that were not accepted into the Christian OT canon).[5] So Mikulski is wrong to assert that the translation to parthenos predates Matthew by three centuries (it probably predates Matthew by no more than one or two centuries, possibly less). Nor can Mikulski know that all Jewish scholars agreed on how Isaiah should be translated. We don’t know who added the extant translation of Isaiah to the Septuagint, or when, or where, or for what Jewish sect.
We also can’t be sure parthenos was the original reading. We have other pieces of the Septuagint among the Dead Sea Scrolls with variant readings not found in extant manuscripts of those same books. Since these are only small pieces, the fact that they contain otherwise unknown variants means there were probably variant readings for numerous other verses that are no longer attested in surviving manuscripts of the Septuagint. In other words, when Matthew wrote, not all copies of the Septuagint said the same thing. So alterations for sectarian reasons could have taken place between the first translation of Isaiah and Matthew’s reading of it. Since we don’t have any pre-Christian manuscripts with this verse in Greek, we can’t know for certain how common a reading parthenos was.
So, for example, if Matthew was relying on copies of Isaiah produced by Essenes, the Essenes could have altered the text for their own stylistic or sectarian reasons. Since Christians essentially chose which versions of the Septuagint would be preserved to the present day, we may have lost others that had different readings, and therefore we can’t be sure all Jewish scholars before Matthew were in agreement on whether parthenos was the most appropriate term.[6] We already know that Matthew took his verb from one of two variant readings for this verse.[7] So how many other variants were there that are now lost?
Of course, we can’t even be sure of the underlying Hebrew or Aramaic for the same reason. We know numerous variants existed in the Hebrew and Aramaic already in Matthew’s time, and many more remain today. So the argument cuts both ways. But the overall point remains: discussion of what words were where is always an uncertain business. This is aptly demonstrated by professor of Biblical History and Archaeology Gerald Larue here on the Secular Web.[8] Although this is, overall, a relatively minor point, it is significant enough to consider, contrary to Mikulski’s argument, that Jewish sectarianism could lie behind parthenos rather than unanimous Jewish agreement, and that the parthenos reading might have been a more recent development before Matthew’s time than Mikulski imagines.
2. The Greek Is Not So Definite
The Greek word parthenos carries a basic meaning of ‘girl’, hence it denotes ‘virgin’ only by implication. And in fact this word could also be used to refer to non-virgin women who weren’t married. Homer so uses it, and Homer was the standard textbook for learning Greek all throughout antiquity, so any writer of Greek would know of this word’s versatile and indefinite meaning.[9] So the Jewish translators need not have had virginity in mind, but youth. Still, this word carried a strong connotation of virginity, and there were Greek words that didn’t carry that connotation (like neanis). And Mikulski is right to point out that the choice to go with parthenos was made, presumably, by Jews. Even so, we can’t know what was in the mind of the scribe who chose that word. It is possible the Jewish translator of Isaiah wasn’t taking sides on whether ‘virgin’ was meant but was using a word that could mean either, and that only later did Christians take it as definitely meaning ‘virgin’.
Besides all that, the argument Mikulski uses works against him just as forcefully: for if the choice of parthenos over neanis by the Greek translator implies virginity, then the choice of almah over bethulah by the original author (presumably Isaiah–or, according to Christian belief, God Himself, speaking through his prophet) implies nonvirginity. Thus, even if some Jewish translator (speaking a Hebrew that is three hundred years newer than Isaiah’s) took this passage to be about a virgin, this does not make it any more likely that this passage originally meant a virgin. For contrary to Mikulski’s argument, a mistake by a Jew is still a mistake.
3. Evidence of OT Usage Is Uncertain in This Case
I am also skeptical of Mikulski’s usage argument. Being an experienced translator who has worked with five languages, I can confidently say it has always been nearly impossible to identify the exact parameters of denotation for every instance of a word in a surviving corpus. I’ll bet we have dozens if not hundreds of occasions where almah is used, in and out of the OT, where we can’t know if the denoted girl was a virgin or not. There is no way to make this determination in most cases of its use even within the OT.[10] Thus, we cannot assert too confidently that it never referred to nonvirgins. Many of the uses of the word even in the OT could refer to nonvirgins. We can’t pretend to know for sure.
This problem is compounded by the fact that the frequency of such an alternate usage would be too low for us to count on there being clear, extant examples. It would have been unusual for “a young woman of marriageable age” not to be a virgin. Fornication, even being raped in some cases, was a death penalty offense.[11] So not many such girls lived long enough to be called anything, much less almah. Divorce was also not generally an option for a woman. Only her husband could send her away, and only for sexual infidelity. By law, a man could not divorce a woman who was a virgin on their wedding night. Hence a woman’s father was expected to keep the bloody honeymoon sheets and show them to the whole community as proof of her virginity in order to block a divorce.[12] Women most often died before their husbands, and usually in labor: the mortality rate was probably around 10% for each birth. So young widows were rare, and virginal widows must have been incredibly rare (although not impossible: a husband who died before he made it to the bed would leave a girl a virgin and a widow). At any rate, a young nonvirgin, whether slut, whore, rape victim, divorcee, or widow, could have been called “a young woman of marriageable age,” hence almah. We just don’t know of many examples, so we don’t know if the word would have been used or refused in which cases.
So even if we find almah referring in some definite cases to actual virgins, it doesn’t follow that this was the only correct use of the word. It may have applied to any young unmarried girl, or to any young girl who married as a virgin. Nonvirgin young women would have been so rare that we’re unlikely to have many examples of them being referred to. Hence we can’t say what word would have been used for them. Yet it may have been almah, since it seems there was no better word to use. We can’t be sure it wasn’t. Moreover, the fact that the Hebrews saw a need to coin a word more definitely meaning ‘virgin’ (bethulah) implies that almah did not definitely mean virgin.
4. Where Is the Proper Method?
Both scholars seem to be deploying bad methodology. Callahan, like McKinsey,[13] doesn’t really tell readers how he knows what he knows. Nor does Mikulski. I personally happen to know that the argument comes from ancient Jewish polemics against Christianity. But I would expect Callahan to tell his readers that, identifying where the argument first appeared, and where else it has been argued that a reader can get more information from.
For example, the fact that this is one of the earliest Jewish polemics against Christianity adds a great deal of weight to Callahan’s argument against Mikulski. If even ancient Jews agreed with Callahan on the meaning of the Hebrew word, then that pretty much refutes Mikulski, who is a decidedly inferior expert–not being a native speaker of ancient Hebrew, much less a lifetime Hebrew scholar, like the rabbis of old (who were devoted to mastering and debating the Jewish meaning of what is unmistakably a Jewish book). Still, I want to know what the ancient Christian commentaries on the NT and Isaiah passages had to say about this, and what the Jewish Midrashic literature on Isaiah had to say. Neither Callahan nor Mikulski even seem to be interested in that.
5. Competent Authority Goes against Mikulski
They also should have consulted current scholarly commentaries in print. The best on this issue is Raymond Brown’s Birth of the Messiah. Like Mikulski, Brown is a devout Catholic. But he is an objective scholar, usually fair to all parties in any debate, and always erudite and cautious. He is internationally recognized as a leading, if not the leading expert on the Christian nativity accounts.
So I pulled him off my shelf and looked up the passage in question. And lo and behold, Brown tells us, with citations and quotations, that Callahan’s argument appeared first in a 2nd century Christian apologetic work: Justin’s Dialogue with Trypho (also known as Dialogue with a Jew). This proves, against Mikulski, that even ancient Jews didn’t believe almah meant only virgin, for Christians had to defend their reading of ‘virgin’ against Jewish critics, from the very earliest times (cf. Larue, above, for more on this point). Brown also relates some of the colorful history of the debate, like that fundamentalists once burned copies of the RSV translation of the OT because it had “young woman” in Isaiah, and Catholic bishops compelled Catholic translators of the NAB translation to go against their better judgment and put ‘virgin’ there. Thus Brown observes that many modern translations are the victim of ideological censorship (a common problem, and a main reason why if you haven’t read the Bible in the original languages, then you haven’t really read it).[14]
All in all, Brown’s detailed analysis only confirms Callahan’s point, not Mikulski’s. For instance, as Brown explains, Justin knew that Jews understood Isaiah to be referring to Hezekiah, son of Ahaz, and thus the Christians were “reinterpreting” a prophecy that had already been fulfilled. Brown also cites important scholarship on the meaning of almah and other details, making him an essential reference on this, if you want to explore the matter further. He surveys additional points and concludes that “Isa. 7:14 does not refer to a virginal conception in the distant future” but to “the imminent birth of a child, probably Davidic, but naturally conceived” (§ 5B2, i.e. p. 148). Since this comes from a renowned authority who is Catholic (and thus going against his biases), this conclusion carries special weight here.
To look at the question from a typical messianic Jew’s perspective, I always consult my copy of David Stern’s Jewish New Testament Commentary (1992). Stern is a messianic Jew who also believes Jesus was the messiah, and was virgin born. Consequently, he defends Mikulski’s position. But if you read Stern against Brown you will readily see that Stern’s analysis is off base and a bit naive and misleading.[15] For instance, he argues that since almah implied a good reputation, which required virginity, it thus entailed virginity. But this does not address widows (who had a good reputation even if not virgins), nor is implication the same as entailment. Stern is also not cognizant of the opinion of myself and Brown that within the natural ambiguity of the text in question virgins of good reputation can naturally conceive–not just in the sense that they can conceive on their wedding night (which is certainly a possible meaning of the Isaiah passage), but also, as Brown points out, in the sense that a man might in the future take as a bride someone who is now a virgin. On that reading, “a virgin will conceive” in the sense that someone who is now a virgin, at the time Isaiah spoke, would at some point marry and have a child (possibly even conceiving the moment she loses her virginity). Certainly that would be considered a fulfillment of the prophecy.
[1] This essay was subsequently revised in February of 2007, with an expansion of section 1, new and reorganized endnotes, and numerous minor changes throughout.
[2] Joseph Wheless, Forgery In Christianity (Chapter 2) and Joseph McCabe, The Story Of Religious Controversy (Chapter 14), both written in the early 20th century.
[3] James Still, “The Virgin Birth and Childhood Mysteries of Jesus“; Jim Lippard, “The Fabulous Prophecies Of The Messiah“; Farrell Till, “Prophecies: Imaginary and Unfulfilled.” Jewish apologists also have detailed rebuttals online at the Messiah Truth website: Does Isaiah 7:14 Foretell the Messiah? and Isaiah and His Sons.
[4] Raymond Brown, The Birth of the Messiah (New ed., 1999).
[5] This is explicitly stated by Josephus, Philo, and other ancient witnesses (cf. Josephus, Antiquities of the Jews 12.1-322 and Philo, On the Life of Moses 2.5.25-2.8.48), and implicitly confirmed by internal evidence. For details see “Septuagint” in the Oxford Dictionary of the Christian Church (ODCC), 3rd ed. (1997), pp.1483-84; as well as Joel Kalvesmaki, “The History of the Septuagint and its Terminology” (2005) and the relevant sections of “Bible Translations” in the Jewish Encyclopedia Online. The ODCC puts it succinctly: “internal evidence indicates that the LXX was really the work of a number of translators…that not all of it was translated at Alexandria, and that the work of translation extended over a considerable period.” See also the detailed discussion by Jennifer Dines, “The Septuagint: Some Current Research (2005).
[6] The fact that the Septuagint text survives to us primarily through Christian custody creates an additional problem that I will only mention in this endnote: we can’t be entirely certain that parthenos was not a Christian alteration of the Septuagint, instead of the original word chosen by the Jewish translators. We have enough fragments of other ancient translations of the OT to know that neanis was used more regularly in Isaiah 7:14, even in many cases by Christian translators. Although Justin’s “Jew” concedes that parthenos was in the Septuagint that he knew, this is still a Christian author putting words in a Jew’s mouth. So this is not proof that parthenos in Isaiah was not in fact a later Christian interpolation, seeking to normalize the text to Matthew’s. However, I have not studied the evidence on this question enough to pronounce a conclusion. I only note this as something scholars need to consider more carefully before assuming anything one way or the other.
[7] See textual note for Isaiah 7.14. And see Brown’s discussion in § 5B3 of Birth of the Messiah.
[8] Actually using the virgin controversy as an example: Chapter 32 – Texts, Manuscripts and Translations in Old Testament Life and Literature. Larue erroneously says Isaiah used bethulah. He must be confusing this with almah. His point is otherwise correct.
[9] Cf. Iliad 2.514. Also: Pindar, Pythian Odes 3.34; Sophocles, Trachiniae 1219; Aristophanes, Nubes 530. I’m not aware of any evidence that Koinê usage differed from Classical or Attic in allowing this connotation.
[10] There are only seven instances of its use that I know of: Gen. 24:43, Ex. 2:8, Psalms 68:25, Proverbs 30:19, Song of Solomon 1:3 and 6:8, and of course Isaiah 7:14. In none of these is “virgin” a required reading, except perhaps in the Song of Solomon 6:8, yet even there it’s unclear what category of attendant is meant (it might only have designated young female servants who did not perform sexual services).
[11] Deuteronomy 22:20-21. Even if a girl is raped, she merits the death penalty if she was engaged and no one heard her scream: Deut. 22:23-24. A virgin who is not engaged and then raped, even if she doesn’t scream, is not executed, but compelled to marry her rapist (and her father gets paid for his daughter’s sexual services): Deut. 22:28-29.
[12] Deut. 22:13-19 (where only cognates of bethulah indicate virginity) and Deut. 24:1-4.
[13] C. Dennis McKinsey, in The Encyclopedia of Biblical Errancy, discusses the virgin birth prophecy briefly on p. 159.
[14] For the curious: most Bible translations simply have “virgin” in Isaiah 7:14. But there are exceptions: e.g. The Message Bible has “a girl who is presently a virgin will get pregnant,” which implies Brown’s interpretation; the NLT offers “young woman” as a variant reading in a note; and the CEV translates it as “virgin” but then explains in a note the whole story of why this does “not imply a virgin birth.”
[15] See “How Do Missionaries Paint the Virgin Birth Into the Mouth of Rashi?” on the Outreach Judaism site. However, though typical, Stern is not the best advocate. Other messianic Jews make a more thoughtful case than Stern, which is fair enough to consider. If you want to explore the best contemporary Jewish arguments in favor of a virgin birth (and test them against the general points I have made), see the relevant material in Michael L. Brown, Answering Jewish Objections to Jesus III: Messianic Prophecy Objections (2003) and Daniel Gruber’s article “Modern Rabbis and the Virgin Birth of Messiah.”