Geza Vermes On The Infancy Narratives

In a post late last year, I mentioned that I had ordered, but not yet read, Geza Vermes' book on the infancy narratives, The Nativity (New York: Doubleday, 2006). I recently finished reading it.

Vermes is an Oxford scholar, particularly known in the field of Jewish studies and for his work on the Dead Sea Scrolls, and he's a former Roman Catholic priest who left Catholicism around fifty years ago. The back flap of the book refers to him as "one of the world's leading authorities on Judaism in the age of Jesus...the first Professor of Jewish Studies at Oxford, where he is now Professor Emeritus".

He refers to his book as thorough (p. 16), one that takes into account "all the relevant information" from a large variety of fields (pp. 16-17), and "painstaking" (p. 145). But the book is only 172 pages long, has only two pages of endnotes (pp. 159-160), has a two-page bibliography characterized by liberal and moderate sources (pp. 161-162), makes many highly dubious assertions without supporting argumentation, makes little effort to interact with conservative scholarship, and doesn't break any significant new ground.

The book's prologue refers to the infancy narratives as only "very slightly" historical (p. xv). He later refers to the elements that have "a high degree of probability" as "the names and the place of residence of the child and the parents, but the date of birth could only be approximate, under Herod, and the locale controverted, Bethlehem according to tradition, but more likely Nazareth." (pp. 155-156)

The second chapter in the book is one of the most negative assessments of the conservative Christian view of the infancy narratives that I've ever seen. Vermes' background seems to be relevant here, given some of his comments about the Roman Catholic Church and its scholars. The remainder of the book isn't as bad, but the second chapter does give us a view of the mindset that's behind Vermes' more balanced comments elsewhere.

In that second chapter, we're told that "religious authority dislikes contradictions in its authoritative texts" (pp. 11-12). Thus, "efforts have been deployed from the early centuries of the Christian era by the official revisers and commentators of the Gospels to eliminate the manifest discrepancies between the infancy narratives of Matthew and Luke" (p. 12). While that sort of language might suggest widespread textual changes or scenarios from a Dan Brown novel, the example Vermes goes on to cite in the next sentence is Tatian's Diatessaron. But a gospel harmony like the Diatessaron (and many others have been produced and continue to be produced) doesn't "revise" the gospels or "eliminate" the differences among them. Gospel harmonies have circulated along with the gospels themselves. They coexist. And Tatian wasn't much of a "religious authority".

The reason why Vermes can criticize the differences between Matthew and Luke is because those differences were preserved by copyists and other "religious authorities" who were mostly honest in the transmission process. As Bart Ehrman notes, "It is probably safe to say that the copying of early Christian texts was by and large a 'conservative' process. The scribes - whether non-professional scribes in the early centuries or professional scribes of the Middle Ages - were intent on 'conserving' the textual tradition they were passing on. Their ultimate concern was not to modify the tradition, but to preserve it for themselves and for those who would follow them. Most scribes, no doubt, tried to do a faithful job in making sure that the text they reproduced was the same text they inherited." (Misquoting Jesus [San Francisco, California: HarperSanFrancisco, 2005], p. 177)

What's Vermes' objection, then? As I've noted elsewhere, harmonization is a common practice in historical research. The ancient Christians accurately preserved the text of the infancy narratives, and they offered some possible harmonizations of what they considered to be two historically credible accounts.

Later in the second chapter, Vermes criticizes "exegetes [who] have been happy to settle the problem of the virginal conception by simply calling it a miracle" (pp. 12-13). We're not told in what manner the virginal conception is a "problem" or why viewing it as a miracle is supposed to be unacceptable. The suggestion seems to be that belief in a virginal conception would be acceptable only if the event could be explained naturalistically. Why should we expect a naturalistic explanation? Vermes makes similar comments elsewhere. We're told that the "fabulous" elements of the infancy narratives "force" us to conclude that the accounts aren't historical (p. 155). Such assertions are repeated over and over by Vermes, but without any justification.

He goes on to criticize "self-appointed defenders of Gospel truth" who engage in "exegetical acrobatics" (p. 14). After mentioning C.E.B. Cranfield as a Protestant example, he comments that "For scholarly Catholic ecclesiastics equivocation seems to be the name of the game." (pp. 14-15) Who is he referring to? The examples he gives are John Meier and Raymond Brown, neither one a conservative. I would recommend that readers compare Brown's 752-page book on the infancy narratives to Vermes' 172-page book. The contrast is stark, not only in terms of size (Brown's book has larger pages and smaller print, in addition to a far larger number of pages), but also in terms of quality. Compare their arguments on the virginal conception, for example, the primary issue Vermes cites in his criticism of Brown. Brown was a liberal who was wrong on many points, but his work on the infancy narratives was much better than Vermes' book on the subject.

He later returns to his criticism of the Roman Catholic hierarchy in a discussion of Joseph Fitzmyer's treatment of the virginal conception in the gospel of Luke. Commenting on Fitzmyer's abandonment of the position that Luke's gospel doesn't refer to a virginal conception, Vermes writes, "Some may think that it was a pity, but quite understandable during the papacy of John Paul II." (p. 68)

What kind of scholarship does Vermes approve of, then? In the paragraph after his comments about Raymond Brown, he writes:

"Other Christian scholars have felt no reluctance to call a spade a spade. In the considered judgment of Rudolf Bultmann, one of the greatest New Testament exegetes of the last century, the original Semitic report of Matthew's Infancy Gospel contained nothing about the virgin birth. It was a motif unheard of in the Jewish environment of the age, he stated, and it was first added to the Gospel account in the course of its transformation in Hellenism. More recently, one of the most respected Jesus scholars, E.P. Sanders, also asserted without the slightest hesitation that the birth narratives are 'the clearest cases of invention' in the Gospels. As for the saying, 'I wouldn't put it past God to arrange a virgin birth if He wanted, but I very much doubt if He would,' it is attributed, genuine or apocryphal, to David Jenkins, the outspoken former Anglican bishop of Durham." (pp. 15-16)

Why should anybody find those claims by Bultmann, Sanders, and Jenkins convincing? Concerning whether Bultmann was "one of the greatest New Testament exegetes of the last century", see the discussions of his work in Richard Bauckham's Jesus And The Eyewitnesses (Grand Rapids, Michigan: Eerdmans, 2006) and Paul Eddy and Gregory Boyd's The Jesus Legend (Grand Rapids, Michigan: Baker Academic, 2007). Did Bultmann have a copy of an earlier version of Matthew's gospel that doesn't refer to a virgin birth? No. Do the early Christian and non-Christian sources tell us that there was such an "original Semitic report of Matthew's Infancy Gospel"? No, they don't. Vermes will later appeal to a series of dubious, highly speculative scenarios in which Matthew's sources and the author of Luke's gospel initially didn't believe in a virgin birth. Compare Vermes' assertions on such issues to the better arguments offered against Vermes' position by Raymond Brown and other less liberal scholars. Vermes' appeal to the speculations of Bultmann tells us something about his mindset.

He repeats many of the oldest and weakest objections to the infancy narratives without even acknowledging the counterarguments, much less refuting them. He'll sometimes acknowledge significant evidence supporting the historicity of the infancy narratives, but without following that evidence to its conclusion.

We're told that the authors of the two gospels probably had access to genealogical records related to Jesus, and that they probably consulted such records, but that they didn't intend to convey a historical account in their genealogies (pp. 35-37). We're not told why they would be seeking genealogical records to "prove" Davidic descent (p. 36) if "the aim pursued by Matthew and Luke in compiling their genealogies was doctrinal, and not historical" (pp. 35-36). Apparently, the gospel authors were writing in a fictional genre, yet they wanted to consult genealogical records in the process of writing that fiction in order to prove that Jesus fulfilled a prophecy that was expected to have a historical fulfillment. Vermes' mixture of fictional and nonfictional elements doesn't seem to make sense, and he makes little effort to clarify his train of thought.

The book repeatedly uses such shallow, and sometimes inconsistent, argumentation. Vermes repeatedly accuses Matthew of distorting the Old Testament texts he cites, on the one hand, while claiming, on the other hand, that Matthew made up stories about Jesus to fulfill those texts he distorted. For example:

"However, in the light of Matthew's previous record of manipulating the scriptural text, it is more likely that the Greek wording employed in this passage was his own handiwork. In that case one may infer that just as the parthenos-virgin formula of the Greek Isaiah 7:14 prompted the story of the virginal conception, the Hosea text thus understood supplied the inspiration for the story of the flight of the infant Jesus to Egypt. Jesus had to be transferred to Egypt in order to allow God to summon his Son from there to his home country." (pp. 107-108)

But, then, why did Matthew allegedly distort the text to begin with? If there was no Jewish expectation that the Messiah would go to and come out of Egypt, if the text of Hosea didn't refer to any such event, and if Matthew didn't think that such a thing actually happened in Jesus' life, then what was Matthew doing?

Vermes argues that the gospel authors were writing in a non-historical genre, yet he fails to demonstrate that the infancy narratives were initially read in that manner. To the contrary, he repeatedly acknowledges that the early sources took the narratives as historical accounts (pp. 91-92, 94-95). He'll often refer to how something in the infancy narratives "obviously" isn't historical (p. 92), but will make no effort to explain why the ancient sources, both Christian and non-Christian, interpreted the narratives differently than he does. He claimed that he would address "all the relevant information assembled from the parallel Jewish documents, biblical and postbiblical, and from the sources of classical literature and history" (pp. 16-17). But he has little to say about the large majority of the external evidence that runs contrary to his conclusions.

There are far too many other problems with the book for me to discuss them in detail. I'll give several more representative examples.

He assigns too much weight to the small minority of professing Christians in antiquity who denied the virgin birth, yet he doesn't attempt an explanation for why there was such widespread acceptance of the doctrine, including among churches around the world that had recently been in contact with the apostles. He mentions the rejection of the virgin birth among the Ebionites without acknowledging that other Ebionites accepted the doctrine (pp. 61-62).

In the process of arguing that Luke's gospel doesn't refer to a virginal conception, he dismisses Luke 3:23 with the comment that the passage "in the context also appears as another patent retouch made on second thoughts" (p. 69). He doesn't cite any manuscripts or other evidence to support his conclusion, but refers the reader back to a discussion on p. 33. What do we find there? We find another assertion without evidence (pp. 33-34), accompanied by a reference to further discussion in chapter 5 (the chapter from which I quoted at the beginning of this paragraph). On p. 33, Vermes refers to the comment in Luke 3:23 as "patently secondary". On p. 69, he refers to it as "another patent retouch made on second thoughts". I doubt that many people would expect a source such as a family genealogical record to discuss a virgin birth. But if Luke was using such a family record and knew that there was a virginal conception, then how would his mentioning of the virginal conception be "secondary" in a problematic manner? If we're supposed to believe that it was secondary in the sense that the original gospel of Luke didn't include it, then we'll need more than Vermes' cross-referenced assertions to justify that conclusion.

He misreads John 7:42 (p. 80) without interacting with the sort of data I discuss here.

He misrepresents what the Protevangelium Of James says about the census of Luke 2 (p. 85).

Like so many other critics of the infancy narratives, he speculates that vague similarities between the infancy narratives and some extra-Biblical literature of an unhistorical nature should be taken as evidence that the infancy narratives are similarly unhistorical. Extra-Biblical accounts that don't associate a star with somebody's birth are compared to Matthew 2 anyway if they involve some sort of "light portent" (p. 96). Thus, a Jewish tradition about Moses' house being filled with light when he was born is compared to Matthew's account of the star of Bethlehem (p. 96). Vermes will acknowledge that Jewish sources around the time of Jesus didn't associate stars with the births of significant figures, but he'll then go on to appeal to pagan sources in order to find some sort of precedent (p. 97). Vermes will look to Jewish or pagan sources, before or after the time of the gospels, pointing to either close or distant parallels, in order to provide a potential source from which the authors of the infancy narratives borrowed.

He comments that "Dreams seem to be essential in infancy tales" (p. 114), but Luke has none.

He repeats Raymond Brown's erroneous argument for discontinuity between the infancy narratives and the gospel accounts of Jesus' public ministry (pp. 132, 147-148, 152), an argument I've addressed elsewhere. He claims that Matthew's gospel is unaware of the relationship between John the Baptist and Jesus that's portrayed in Luke 1 (p. 132), but Matthew 3:14 suggests that John thought highly of Jesus prior to Jesus' baptism. Vermes says nothing about Matthew 3:14. Why should we accept his assertion that what Luke reports about the earlier life of John the Baptist was "totally unknown" to Matthew (p. 132)?

He accuses Jesus of being "haughty" in Luke 2:49 and other passages in the gospels in which He interacts with His family (p. 142). He makes no attempt to interact with any Christian defense of Jesus' behavior, including the explanations offered by the gospels themselves (John 7:5, for example).

Etc.

The front flap of the book refers to it as "A masterful work of biblical scholarship" that "gives readers a new and more powerful understanding of the events celebrated every Christmas season", and Vermes claims near the conclusion of the book to have presented a "painstaking" analysis (p. 145). Actually, the book seems to be aimed more at a general audience than a scholarly one, there isn't much about it that's new, and his analysis is far from painstaking. The book is another example of the liberal tendency to repeat bad arguments without much of an effort to interact with counterarguments that have been circulating for a long time.