A Gospel Among the Scrolls?

036

On December 24, 1994, the Times of London ran a front-page story entitled “Oxford papyrus is ‘eyewitness record of the life of Christ.’” The article reported the claim that three papyrus fragments of Matthew’s Gospel in Magdalen College, Oxford, date to the mid-first century C.E. Instead of having been written a generation or more after Jesus’ death, as is—or was—the scholarly consensus, Matthew’s Gospel was written within a decade or so of the crucifixion by someone who was there at the time, so the article said. This, of course, would indeed be astounding and worthy of the treatment the Times gave it.

“Not since the discovery of the Dead Sea Scrolls in 1947,” the story continued, “has there been such a potentially important breakthrough in biblical scholarship.” The newspaper devoted nearly two full pages to the story, including an editorial that likened the alleged new discovery not only to finding the Dead Sea Scrolls, but also to Howard Carter’s discovery of Tutankhamun’s treasures and to Schliemann’s location of Troy. Two days later, the Times included an extended interview with Dr. Carsten Thiede of Paderborn, Germany, on whose scholarship this claim is based.

Thiede’s claims are threefold. First, Matthew’s Gospel is an eyewitness account of the life and teaching of Jesus—so today’s Bible scholars (who date it to about 80 C.E.) are wrong, and the early church fathers were right after all. Second, at the time Matthew was written the formal separation of Christianity and Judaism had not yet occurred. Third, “Recognition of Christ’s divinity was made before the end of classical Judaism in 70 C.E.”

038

All three claims made by Thiede are controversial, to say the least. In fact, they disagree with mainstream biblical scholarship.

Abraham Lincoln once said that the Times (founded in 1785) was one of the greatest powers in the world. While few would make that claim today, the Times is still one of the most influential newspapers in English-speaking circles. Until recently, it has rarely indulged in sensationalist reporting, so many readers have assumed that Thiede has uncovered genuinely new evidence for the early dating, eyewitness character and reliability of the Gospels.

Dr. Thiede’s views were reported on BBC radio national news bulletins that same day and were then picked up in the media in many parts of the world. On January 23, 1995, the international edition of Time magazine devoted a full page to the story with the headline “A Step Closer to Jesus?”

I was so surprised by these rash claims that, even though in the midst of preparations for a family Christmas, I faxed a letter to the editor of the Times in which I noted that, since initial publication of these fragments of Matthew in 1953 (they had been in the Magdalen library since 1901), several leading paleographers and New Testament scholars had dated them to the end of the second century. I also suggested, mistakenly, as it has turned out, that Thiede’s claims concerning the very early date and alleged eyewitness character of Matthew’s Gospel would not merit serious discussion by specialists. Thiede’s theory has attracted so much interest, however, that, like it or not, specialists have been forced to discuss it.

In my letter to the Times I also drew attention to Carsten Thiede’s involvement in another explosive contention. Thiede (who is director of a research center in Paderborn, Germany) has been the major defender and expositor of a claim made by the Spanish papyrologist Jose O’Callaghan that a fragment of the Gospel of Mark has been found among the Dead Sea Scrolls. This too had warranted a front-page story in the Times (16 March 1972).

In a series of publications in German beginning in 1984, Thiede has championed this theory of Mark among the Dead Sea Scrolls, but for nearly a decade his views attracted little interest.¹ Recently, however, interest has been fueled by two publications, one by Thiede (The Earliest Gospel Manuscript? The Qumran Fragment 7Q5 and Its Significance for New Testament Studies [Exeter: Paternoster, 1992]) and the other a book of scholarly papers given by Thiede and several well-known New Testament scholars at a symposium on this alleged fragment of Mark held at Eichstätt, Germany.²

Thiede is thus now proposing that we have first-century exemplars of both Matthew and Mark—Matthew from the Magdalen papyrus and Mark from the Qumran fragment.

If the Qumran fragment is really from Mark’s Gospel, then the scroll of which it was a part was probably hidden before the Roman Tenth Legion 039marched through the area in 68 C.E. on its way to Jerusalem. If so, then Mark’s Gospel also must have been written earlier than most scholars have supposed.

Here we will consider both of Thiede’s contentions—the Magdalen fragments of Matthew and the Qumran fragment allegedly of Mark. With the Magdalen fragments of Matthew, the question is their date. With the Qumran fragment, the question is whether or not it is really Mark.

First, the Magdalen Matthew (known to scholars as p64): It consists of three small papyrus scraps with letters on both sides.^a They were acquired on the antiquities market at Luxor, in Egypt, in 1901 by the Reverend Charles B. Huleatt, a former scholar of Magdalen College, who gave them to his college in that same year. In 1962, the distinguished Oxford papyrologist and paleographer C. H. Roberts recognized that two papyrus fragments in Barcelona (designated p67) containing other parts of the Gospel of Matthew were in the same hand and came from the same codex, or book, as the Magdalen papyrus. The same dating considerations therefore apply both to the Magdalen and the Barcelona fragments.

Incidentally, because the fragments come from a codex, rather than a scroll, an early dating would provide fresh evidence for early Christian preference for the codex rather than scrolls for their writings.

The Times article referred to a then-forthcoming scholarly article by Thiede that has since appeared in January 1995 in the Zeitschrift für Papyrologie und Epigraphik. Early copies of the journal were searched by scholars as if they were looking for gold nuggets. Unfortunately, the article turned out to be something of a damp squib. As is appropriate in an academic journal, it set out a number of proposals on technical points concerning the fragments, some of which are certainly correct. Most crucially, however, it contained a discrepancy between Thiede’s earlier claim as made in the Times and what he claimed in his scholarly exposition. In his scholarly exposition, Thiede concluded only that the Magdalen papyrus was late first century, the date most scholars assign to the composition of Matthew’s Gospel. To quote the conclusion of Thiede’s article:

“It may be argued that the Magdalen papyrus of Matthew could be redated from the late second to the late first century, some time after the destruction of the Temple in Jerusalem” (italics added).³

I am mystified by the discrepancy between the extravagant claims made in the Times (and echoed in the media worldwide) and the caution of the academic journal article, which did not even mention that Matthew’s Gospel offers eyewitness testimony to the life and teaching of Jesus. There is a world of difference between dating the Matthew papyrus to the mid-first century and to the late first century! The later date would not change any of the most widely held views about the origins of the Gospels; the earlier dating would change everything. In short, the sensational claim about the Magdalen papyrus seems simply to have evaporated into thin air. It is worth commenting on here only because the claim received such widespread publicity.

Even Thiede’s arguments that the Magdalen papyrus is late first century, rather than second century, are weak; remember, the issue at this point is the date of the manuscript, not the date of the composition of Matthew. The basic script genre in which it is written is known as “biblical majuscule” (majuscule simply means written in all capital letters). Although recent work has been done on the biblical majuscule style, partly in the light of recent discoveries, Thiede does not discuss it. He does not refer either to G. Cavallo’s magisterial 1967 study⁴ or to the splendid 1987 book by G. Cavallo and H. Maehler, Greek Bookhands of the Early Byzantine Period, A.D. 300–800.⁵ Both books note that the biblical majuscule style, which is not confined to biblical manuscripts, goes back to the end of the second century; it reaches its definitive form by the third century and the height of its development in the fourth century, the period that produced the famous biblical manuscripts Codex Vaticanus and Codex Sinaiticus. The two scholarly books listed above set forth a number of characteristics of both the formation and the shape of letters in this script in its “phase of greatest formal perfection.” Many of these characteristics are found in the Magdalen papyrus. I am baffled, as are other leading specialists, by Thiede’s failure to offer reasons for abandoning the consensus view that the Magdalen papyrus is an early example of the biblical majuscule style⁶ and his failure to discuss the Barcelona fragments from the same codex, which contain rather more verses of Matthew’s Gospel.

Thiede rests his case on a comparison with recently discovered examples of Greek handwriting from outside Egypt, where the Magdalen fragments came from—four fragments from the Dead Sea Scrolls (three from Cave 4 and one from Cave 7) and a fifth from the Greek minor prophets scroll from nearby Nahal Hever. The fragments from Qumran Cave 4 date to no later than 68 C.E., when the site was destroyed by the Roman legion on its way to Jerusalem. The same terminus post quem probably applies to the Cave 7 fragment as well. The example from Nahal Hever was deposited 040during the Second Jewish Revolt against Rome (132–135 C.E.), which is the latest date it could have been written. Thiede seeks to show that these are all in a style of handwriting similar to that of the Magdalen fragments. Unfortunately, there are serious problems with his case. Papyrologists insist that in dating manuscripts both the methods by which individual letters are formed and their shape be carefully considered.⁷ Thiede very largely ignores the former and does not offer a full comparison of the latter. While there are similarities between the shape of some of the letters in the Magdalen papyrus and their counterparts in the first century, there are also dissimilar letters that Thiede does not discuss.⁸

In short, there is no reason to abandon the generally held view that the Magdalen papyrus (and the Barcelona fragments) dates to the second century C.E.⁹

No one disputes that the Magdalen fragment and the Barcelona fragments are probably the earliest evidence we have for the text of Matthew’s Gospel. For this reason they will always be of special interest to scholars. It is to Carsten Thiede’s credit that he has redirected attention to these fascinating fragments. But it makes a great deal of difference whether they are mid-first century or late first century—as Thiede now claims—or sometime late in the second century, as most scholars previously thought.¹⁰

Now let us turn to the Thiede, or, more precisely, the O’Callaghan/Thiede claim to have found a fragment of the Gospel of Mark among the Dead Sea Scrolls.

The alleged fragment of Mark was found in Qumran Cave 7, which is unique among the 11 Qumran caves where inscriptional materials have been found. The documents from Cave 7 are all in Greek. But all 18 of them are tiny fragments with but a few, often illegible, letters on each fragment. Only two fragments—7Q1 and 7Q2—were identified in their initial publication in volume 3 of Discoveries in the Judaean Desert.¹¹ 7Q1 was identified as Exodus 28:4–7, and 7Q2 as Baruch 6:43–4 (Epistle of Jeremiah). Both are close to the text of the early Greek translation of the Hebrew Bible known as the Septuagint. Oxford don C.H. Roberts has dated fragments 7Q1–3 to about 100 B.C.E. and fragments 7Q4-18 to between 50 B.C.E. and 50 C.E. Some pottery remains found in the cave were dated to the first two centuries B.C.E. and prior to 68 C.E. As there is no clear evidence for the later use of Cave 7, most scholars have concluded that the scrolls to which the fragments belonged were all placed in Cave 7 before the arrival of the Romans in 68 C.E.

The fragment alleged to be from Mark’s Gospel is designated 7Q5. In 1972 Jose O’Callaghan claimed that 7Q5 contained parts of Mark 6:52–3. He later identified 7Q4 as 1 Timothy 3:16, 4:1, 3. With much less confidence, he suggested that seven other Cave 7 fragments were parts of New Testament writings: three of Mark and one each of Acts, Romans, James and 2 Peter.

Maurice Baillet, the editor of the original publication of the Cave 7 scrolls, rejected O’Callaghan’s identifications, as did several other respected scholars. A vigorous discussion followed.¹² All participants in the debate agree that unless 7Q5 is accepted as part of Mark 6:52–3 the other proposed links with New Testament writings are no more than interesting speculations.

7Q5 contains only 20 Greek letters on four or five lines. Of these 20 letters only 10 can be read with certainty, and there is only one complete word—kai (and). However, as Thiede himself has shown, the tiny size of the fragment and the small number of certain letters do not rule out the possibility of identification; papyrus fragments of comparable size have been successfully identified.¹³

Mark 6 recounts the feeding of the five thousand with but five loaves and two fishes. That night Jesus’ 041disciples encounter a dangerously high wind while in their boat on the Sea of Galilee. Frightened, they see Jesus walking on the water. When he gets into their boat, the wind stops. They are astounded. Then come verses 52–53, the verses from Mark 6 that are allegedly preserved in part on 7Q5. In the following English translation of Mark 6:52–53, I have placed in bold the letters that correspond approximately to the Greek letters preserved in 7Q5:

⁵² for they [the disciples] did not understand about the loaves, but their hearts were hardened.

⁵³ And when they had crossed over, they came to land at Genessaret and moored the boat.

One problem with the identification is that the Greek phrase here translated “to land” (epi ten gen) is missing from the Qumran text. We know this because there is not enough room for this phrase in the appropriate line. If we were to insert it in the appropriate place, line 4 would be nine letters too long. Yet the phrase translated “to land” is found in every ancient Greek manuscript of Mark and in all the early translations into other languages.¹⁴

Although the absence of the words “to land” is an embarrassment for the theory, it is not fatal; New Testament manuscripts occasionally contain readings not attested elsewhere. But this, alas, is not the only problem with the identification of this fragment as Mark 6:52–53.

In the end, the identification of 7Q5 as part of Mark 6:52–53 stands or falls on a particular reading of one damaged letter. In the two Greek transcriptions shown here, a dot beneath a space indicates that the letter above cannot be read with any confidence; a dot beneath a letter indicates a possible or probable reading. The transcription on the left is that of two Australian papyrologists, S. R. Pickering and R. R. E. Cook.¹⁵ The fuller transcription on the right is Thiede’s (the transcription of Jose O’Callaghan, who first proposed the identification with Mark’s Gospel, is similar to Thiede’s).

The critical letter is in line 2, the next to last letter on the right. Is this letter a Greek nu? The second letter from the left in line 4 is a clear nu. The nu in line 2 appears to be somewhat different. Thiede claims that the difference between what he takes to be a partial nu in line 2 and the clear nu in line 4 is not significant, for the scribe of 7Q5 did vary his letters slightly. Thiede points to the difference between the eta in line 4 and the eta in line 5.

But a simple test shows that this claim is fallacious. By using tracing paper on an enlarged clear photograph of 7Q5, one can compare the two etas; the difference is insignificant. If one then traces the clear nu in line 4 and tries to place it over the disputed damaged letter in line 2, it is immediately obvious that a nu simply will not fit there.

Unless the damaged letter in line 2 is nu, the fragment just won’t fit the text of Mark 6 at all.

Thiede quite properly insists that careful examination of the original is always preferable to photographs, even if they are infrared or enlargements. He has examined the original and on this basis concludes that the nu in line 2 is “highly possible.”¹⁶

Thiede, however, is not the only paleographer who has examined the original fragment closely, as well as the photographs. Several other experienced specialists have concluded that a nu in line 2 is either very unlikely or quite impossible.¹⁷ Pickering and Cook, for example, read an iota at this point; they stress the similarity of this damaged letter to the certain iota in line 3.

On the other hand, though most scholars have been skeptical, one of the most rigorous of specialists, Joseph A. Fitzmyer of Catholic University in Washington, D.C., states that “the issue cannot be simply dismissed,” adding, however, that O’Callaghan, both in his own writing and through Thiede’s writings, “has not yet proved his case.”¹⁸

On April 12, 1992, the Division of Identification and Forensic Science of the Israel National Police investigated 7Q5. With German television cameras rolling, a stereomicroscope was used to look closely at the disputed letter in line 2. In the photograph thus obtained, there are faint traces of what Thiede thinks is the top of the diagonal of a nu. These traces are not visible to the naked eye.

In May 1995 I showed the new enlarged photograph to T. C. Skeat, a very experienced papyrologist. He was certain that there simply isn’t room for a nu in this line. He also confirmed the judgment of S. R. Pickering and R. R. E. Cook that the next damaged letter looks very much like a damaged alpha—a further nail in the coffin of the theory that 7Q5 is part of Mark. R. G. Jenkins, who has been working on all the fragments from Cave 7 for some time, looked carefully at the original and the new photograph. He has reached the same conclusion: He thinks the faint traces that the stereomicroscope found may be no more than a shadow.

If 7Q5 is to be identified, however provisionally, 042as a fragment of Mark, then its presence at Qumran must be accounted for. Most specialists on Mark’s Gospel date it to just after the fall of Jerusalem in 70 C.E. Most still believe that it was written in Rome. O’Callaghan does not stop to consider how, apparently against all the odds, a fragment of Mark might have reached Qumran before 68 C.E. Thiede, however, notes that a damaged jar bearing the Hebrew inscription rwm’ (vocalized roma or ruma) twice on its neck has been found in Cave 7. Building on a cautious suggestion by Fitzmyer, the inscription might be an attempt at writing “Rome” in Hebrew letters. “Thus it would indicate the original ownership of the scrolls: they belonged to and came from the Christian community in Rome which supplied the ‘home communities’ in Palestine with material collected and copied in Rome or at any rate of Roman origin [such as the Gospel according to Mark?].”¹⁹ While it is possible that this inscription is an attempt to write “Rome” in Hebrew letters, I believe that the inscription is much more likely to be a proper name than a reference to the origin of the scrolls in Rome.

Thiede sketches several possible scenarios to explain how Christian scrolls in Greek may have found their way to Qumran. Although there are serious difficulties with them all, I do accept that it is just possible that this happened. Shortly before 68 C.E., when the settlement at Qumran was destroyed by the advancing Roman army, the sounds and smells of war were in the air. Perhaps the scroll of Mark was hidden in Cave 7 by someone without any link with the Qumran community. After all, Cave 7 was not far from important lines of communication.

A second possibility is that the copy of Mark was placed here later. Although there is no archaeological evidence indicating that Cave 7 was used after 68 C.E., there is no evidence that rules out the possibility. We know that during the Second Jewish Revolt, the so-called Bar-Kokhba Revolt (132–135 C.E.), Greek scrolls were hidden in caves above Nahal Hever, only about 30 miles south of Qumran. Perhaps, during that period, some Greek scrolls were hidden in Cave 7 also. Not all the mysteries of Cave 7 have yet been unravelled.

I believe that it is most unlikely that a copy of Mark found its way into Cave 7 in the first century. If it did, we would have to think again about many aspects of early Christianity. So the key question remains: Do the letters in 7Q5 fit Mark 6:52–53? If there is room for a nu in line 2, then this is possible. But at this crucial point we have only part of the vital letter. Most scholars will probably continue to believe that there just isn’t room for a full nu; therefore 7Q5 is most unlikely to be part of Mark’s Gospel.

On December 24, 1994, the Times of London ran a front-page story entitled “Oxford papyrus is ‘eyewitness record of the life of Christ.’” The article reported the claim that three papyrus fragments of Matthew’s Gospel in Magdalen College, Oxford, date to the mid-first century C.E. Instead of having been written a generation or more after Jesus’ death, as is—or was—the scholarly consensus, Matthew’s Gospel was written within a decade or so of the crucifixion by someone who was there at the time, so the article said. This, of course, would indeed be astounding and worthy of the treatment the […]