Links to other sites on the Web
Eusebius, in his book, The History of the Church, tells us that Papias, the second-century bishop of Hierapolis, wrote that “Matthew compiled the Sayings [of Jesus] in the Aramaic language and everyone translated them as well as he could.” This is about all we know of Papias because, like much that was written in antiquity, Papias’ book is lost. The quotation in Eusebius comes shortly after he has called Papias “a man of very small intelligence, to judge from his books,” (History of the Church 3.39, translated by G.A. Williamson, rev. and ed. by Andrew Louth, [London: Penguin Books, 1989]) but this seems to be Eusebius’ opinion of the man as a theologian rather than as an oral historian.
What has since cast doubt on Papias as a reliable witness is his use of the Greek word “logia” meaning “sayings” in his description of the Gospel of Matthew. This seems obviously wrong because what has come down to us as “Matthew” is mainly a life of Jesus and not just a collection of sayings attributed to him. It is true that our Gospel of Matthew contains many sayings that are not found in the Gospel of Mark (which otherwise shares most of the biographical details found in Matthew’s story), but Papias almost seems to be talking about a different book from the one we call Matthew. Besides that, many reputable biblical scholars tell us that Matthew and the other three gospels appear to have been composed in Greek, not Aramaic. (There may be some dispute about whether or not some passages in Syriac versions of the Gospels were at least partly translated from Aramaic rather than Greek, but it is likely that the Greek gospels were translated into Aramaic before being translated into Syriac.)
Even so, Papias, in spite of his apparent errors, is our source for the tradition that the narrative Gospel of Matthew is the first composed of the four canonical gospels (those gospels officially recognized by their inclusion in the New Testament). This view was challenged, finally, by a nineteenth-century German scholar who presented what is now called the Q hypothesis: Mark was the first narrative gospel, composed perhaps around 70 A.D.; Matthew was composed a bit later in the first century, using as two of its sources the Gospel of Mark and an unknown-and-now-lost collection of sayings that were dubbed Q from the German word Quelle meaning “source.” (Robert J. Miller, ed., The Complete Gospels: Annotated Scholars Version, A Polebridge Press Book [San Francisco: HarperSanFrancisco, 1994], 56.) Q was older than Mark, possibly going back to the time when Paul was preaching and writing or slightly earlier (in other words, earlier than 50 A.D.). Miller presents a version of Q (as have other scholars, most notably, Burton Mack). Q consists of the sayings (such as the Sermon on the Mount) that appear in the narrative gospels Matthew and Luke but which appear only rarely in Mark. Matthew and Luke do differ from each other somewhat in their versions of these sayings but not terribly much. This has suggested to many scholars that Matthew and Luke both copied from the same collection of sayings just as they obviously copied Mark. (Matthew and Luke both agree with Mark’s narration almost word for word except where they dress up Mark’s relatively plain and often poorer Greek grammar.)
What is most odd about Papias' statement is that he was almost certainly familiar with the narrative Gospel of Matthew that we know. Scholars agree that it had been written well before Papias lived (which was about 130 A.D.); yet he reported that Matthew wrote down the “Sayings” of Jesus rather than a narrative gospel. According to Eusebius, Papius was always eager, when he met elderly Christians, to ask what they could tell him about earlier Christian history. His source for the information about Matthew’s sayings collection (again, according to Eusebius) was either Aristion or John who were elderly Christians who may have met some of the early disciples. In regarding Papias as an unreliable, misinformed and untrustworthy dunce, we may be unnecessarily assuming that he was telling us that the sayings collection he attributed to Matthew is the same as the gospel by that name which we know from the New Testament. Instead, he simply may have faithfully reported what Aristion or John told him.
Other writers have been generous toward Papias, realizing that the sayings gospel he describes and the narrative gospel we know might not be the same. In their 1897 analysis of POxy 1, the first fragment of a sayings gospel to be discovered, Grenfell and Hunt write,
“… we may here have got for the first time a concrete example of what was meant by the Logia which Papias tells us were compiled by St. Matthew, and the [Grenfell and Hunt here writing in Greek: “Sayings of the Lord,” which is yet another lost sayings collection] upon which Papias himself wrote a commentary. The statement about St. Matthew (ap. Euseb. H.E. iii. 39), [Grenfell and Hunt quote, in Greek, the passage from Eusebius translated nearest the beginning of my first paragraph above], has always been taken as the starting-point in any discussion of the synoptic problem, but of the meaning of the word [logia] the most diverse views have been held. It is not of course at all likely that our fragment has any actual connexion either with the Hebrew Logia of St. Matthew or the [sayings of the Lord] of Papias. It contains nothing which suggests the one or the other, and probably many such collections were made. But it is difficult to imagine a title better suited to a series of sayings, each introduced by the phrase [Jesus said], than Logia; and the discovery strongly supports the view that in speaking of [logia] Papias and Eusebius intended some similar collection." (Grenfell, Bernard P., and Arthur S. Hunt, Sayings of Our Lord, Egypt Exploration Fund [London: Henry Frowde, 1897], p. 18.)
Their logic here is unassailable; Grenfell and Hunt knew that they had not discovered either of the two logia mentioned by Papias, but rather something merely resembling it in form, and they rightly guessed that other collections of this sort probably existed during the first centuries of the Christian era. Indeed, they never lived to find out that they had discovered sayings 26 through 33 from the Gospel of Thomas, a complete text of which was not found until 1945.
F.F. Bruce echoes Grenfell and Hunt saying that, most probably, “Logia” means “a collection of our Lord’s sayings”(F.F. Bruce, The New Testament Documents: Are They Reliable? [Grand Rapids, Mich.: The Inter-Varsity Press, 5th ed. 1943, 1960] p. 38. I am grateful to Professor Robert M. Price for pointing me toward F.F. Bruce’s contribution on this subject [Robert M. Price, personal communication, August 2, 2000].) Then Bruce says,
"It is difficult to avoid the conclusion that Papias was referring to just such a work as this when he said that Matthew compiled the Logia. …And when he adds that every man translated these Logia as best he could, this suggests that several Greek versions of them were current, which partly explains some of the differences in the sayings of Jesus common to the first and third Gospels [that is, Matthew and Luke, since Bruce, being an orthodox Christian scholar, puts the gospels in the traditional order]…. If we are right in naming the Matthean Logia as the source from which the ‘Q’ material was drawn [emphasis added] this compilation must have taken shape at an early point in primitive Christian history. …It may well have been in existence by AD 50. …The Gospel of Matthew seems to have appeared in the neighborhood of Syrian Antioch some time after AD 70. It represents the substance of the apostolic preaching as recorded by Mark, expanded by the incorporation of other narrative material, and combined with a Greek version of the Matthean Logia together with sayings of Jesus derived from other quarters."( Ibid., pp. 39-40)
Bruce further speculates that “Luke first enlarged his version of the Matthean Logia by adding the information he acquired from various sources” and this “was subsequently amplified by the insertion at appropriate blocks of material derived from Mark.”(Ibid., p. 43) Setting aside all other issues raised about dates, sources and reliability, in saying that “the Matthean Logia [was] the source from which the ‘Q’ material was drawn” by both Matthew and Luke, Bruce has not merely said that the “Matthean Logia” was the source of Q but seems to have implied that it could be identical to Q.
Admittedly, it is a leap to suggest that the logia now called Q was originally called something equivalent to “the Gospel of Matthew,” but if the Matthean Logia, as Bruce calls it, is Q, this does not contradict anything we now know about Q. The Matthean Logia-that is, Q-would later have been combined with the narrative Gospel of Mark by two different writers or sects, the first of whom authored the New Testament Gospel of Matthew and the second of whom authored Luke. Because narrative Matthew was written before Luke, it is not surprising— if Matthew incorporated much of the text of the earlier Matthean (or Q) sayings gospel—that the attribution to Matthew was taken over by the narrative gospel. From the point of view of its author or authors, the narrative gospel would have seemed the rightful heir to the title “Gospel of Matthew” because it was the first narrative gospel to incorporate the Matthean logia or sayings gospel. Attribution to the same apostle would have lent authority to the new text and been perfectly justified in the minds of first century writers who were entirely innocent of the modern concepts of plagiarism and literary fraud. All that mattered was that, after all, the new text did incorporate the earlier one.
The appropriation of the earlier gospel’s attribution to Matthew, however, made it necessary to resolve the confusion caused by two works with similar or identical titles. The most obvious solution was that the expanded, narrative gospel should entirely replace the Matthean Logia or Q. Before the earlier work fell completely out of use, it was used just once more as one of Luke’s sources, but then no one found any further need of it. (It has been suggested that obsolescence was also the fate intended for Mark after it was incorporated into the more elegantly written Matthew and Luke, but Mark was so popular that it continued to be copied, read and revered.)
While this hypothesis is supported by scant evidence, it is consistent with the evidence for Q and explains Papias’ seeming inconsistency in describing a text attributed to Matthew as a collection of sayings rather than the narrative gospel with which he must have been familiar. Also, while both Matthew and Luke would have read the Matthean Logia or Q Gospel in Greek rather than Aramaic, Papias’ and his source could have recalled a tradition that the Matthean Logia was originally composed in Aramaic (whether or not it actually was). “Everyone translated them as well as he could” implies, as Bruce notes, that a sayings gospel attributed to Matthew was circulating in a Greek translation—or several Greek translations—by the time that canonical Matthew and Luke were written. This might help to explain why, for instance, Luke uses neither half of Matthew 5:14 and yet the sayings from Matthew 5:14b (the second half of verse 5:14) through 5:15 are in the same order in Matthew as they are in Thomas 32-33:2-3. Did Luke know these sayings and not use them, or were they not included in his copy of the Matthean logia? Did the author of the New Testament’s Matthew get them from an early version of Thomas or was a version of the Matthean logia that was unknown to Luke used as a source for both Thomas and the New Testament’s Matthew? (Only half of the six distinct sayings found in Matthew 5:13-16 have parallels in any other text, and one of these three, the “city on a hill” saying, is only paralleled by Thomas.)
I feel that I would be remiss if I did not mention the hypothesis of Michael Donald Goulder, although I admit to knowing it only second-hand through John Shelby Spong who shares Goulder’s view but may or may not have communicated it accurately. As I understand it, Goulder believes that there was no Q, but that Luke instead copied from Matthew. This is supposed to explain the similarities between Luke and Matthew’s quotations of Jesus. A full discussion of this problem is beyond the scope of this paper, but it comes down to a question of why, if he knew New Testament Matthew, Luke changed the sayings of Jesus in the way that he did.
Many other scholars believe that Luke’s often simpler versions of utterances attributed to Jesus (such as the Lord’s Prayer and many passages from the Sermon on the Mount) suggest that Luke was more faithful to the original text he was copying than was Matthew whose wording of his quotations from Jesus often sound suspiciously like Matthew’s own language in his narrative. It is more as if Matthew edits to make the sayings of Jesus fit the theology he has added to his narrative—theology that differs from both Mark and Luke even though all three gospels are telling the same stories in otherwise identical words. (See preface to Goulder, Luke: A New Paradigm, Vol. 1. Journal for the Study of the New Testament Supplement Series, ed. David Hill, no. 20, Sheffield, U.K.: JSOT Press, 1989; and Spong, A New Chrisitanity for a New World, San Francisco: HarperSanFrancisco, 2002 (2001), p. 251 n. 12.)
As for Papias, I cannot escape the suspicion that, far from being mistaken about the provenance and characterization of Matthew, he was probably scrupulously faithful to his informant who, in turn, could have been correct in his knowledge of a sayings gospel attributed to Matthew. It is just possible that Papias unwittingly gave posterity our only clue to the identity of the Q Gospel.
Links to other sites on the Web
back to index
Reconstructing a Lost Gospel