Home » Library » Modern Library » A Reply to Glenn Miller’s Response to New Testament Reliability and Bias

A Reply to Glenn Miller’s Response to New Testament Reliability and Bias


I am a firm believer in honest debate because it prevents us from slipping into uncritical positions. For this reason I have much to thank Glenn Miller for his remarks of my Critique of New Testament Reliability and “Bias” in NT Development. (See Glenn’s comments on my original essay.) In a first draft, without peer review or an editor it is difficult to spot weaknesses, strengths, and outright omissions and errors; Glenn’s response to my paper is invaluable for pointing out problem areas and his work is appreciated. Having said that, I would like to take the opportunity to explain and further develop areas which I did not anticipate would pose problems. I will also continue to emphasize my main points in order to accentuate what I think are the valid disagreements between Glenn’s inerrancy position and my own skeptical point-of-view vis. New Testament reliability and development. At issue here is whether or not the NT texts are “reliable” in the sense of accurately representing the actual words and deeds of Jesus of Nazareth. The liberal theologians at the end of the nineteenth-century answered “no” to this question and were disappointed to learn that the gospels were products of Christian preaching (or kerygma) rather than accurate history books. This conclusion goes without saying today, even among conservative scholars. Glenn’s inerrancy position, if I understand it correctly, is a rare exception to biblical scholarship (which is what makes him such an interesting gadfly!) In this reply, I will not dwell on the well-supported findings of the form-critics, instead I shall move us into contemporary scholarship and the “new quest” for the historical Jesus. In so doing I hope to make it abundantly clear to the reader that the real question is not how reliable the NT is (since that is a loaded theological question) but that ancient Christianity was extremely diverse and dynamic prior to an eventual orthodox hegemony.

Jesus Barabbas or Jesus of Nazareth?

The first point of clarification takes us to the Trial of Jesus and specifically to the release of Jesus Barabbas (Mk. 15:6-15; Mt. 27:15-26; Lk. 23:13-25; Jn. 18:39-40). Previously I mentioned these passages in a footnote and Glenn has strongly suggested that I go into them in more depth. He rightly points out that Barabbas (Bar Abba) is better translated “son of Abba” rather than “son of God.” The confusion stems from my consideration that Jesus of Nazareth and Jesus Barabbas may actually be one and the same man. There are several reasons why I suggest this thesis; Jesus Barabbas was a lestes (Jn. 18:40) a “robber” (if you were a Roman) or a revolutionary hero if you were a fellow messianic Jew. He was involved in some kind of insurrectionist struggle in Jerusalem (Mk. 15:7; Mt. 27:20) and if the narrative is sound on this point had many supporters in the crowd. This is a close parallel to Jesus of Nazareth. After Jesus enters Jerusalem in the manner of a davidic king to the ancient nationalistic cries of “Hosanna!” (Mk. 11:1-10 and parallels) he enters the temple to scout it out (periblepsamenos; “scrutinize everything;” Mk. 11:11). Only in Mark, the first gospel, do we see this maneuver on Jesus’ part; however if one wished to count troops, assess their positions, and so on, this otherwise cryptic passage may make more sense.

The next day Jesus engages in the Temple Cleansing and begins to assert himself there. The Zealots who were fighting the Romans (and so by extension the collaborating temple cultus) wanted nothing more than such a radical move of this sort. Jesus would have been an instant hero to the Zealots just for the sheer symbolic stab at authority that his actions represented; indeed the very logistics of such an undertaking as the Cleansing would have required much more than twelve disciples since Herod’s temple area was roughly the size of about thirty-five football fields. The High Priest kept a sizable body of troops in and around the temple complex who were especially ready for trouble during high holiday festivals such as Passover. Tens of thousands of pilgrims trekked in and out of the temple during this time. If, as Mark assures us, Jesus “began chasing the vendors and shoppers out of the temple area” and refused to allow anyone with containers to go into the temple (11:15-16) he would have needed considerable help. Jesus anticipates the coming revolt when the Zealots used the temple compound as a fortress against Titus’ army (Josephus, War, 6.4). Soon after the Temple Cleansing the High Priest acts and arrests Jesus who is eventually turned over to Pilate. Suddenly the narrative introduces the second insurrectionist in our story: Jesus Barabbas (Mk. 15:7 and parallels). Considering that Jesus of Nazareth just led at least a successful symbolic scuffle against the collaborating temple cultus he and Barabbas would have had much in common to talk about in the jail cell while awaiting their fate. Indeed, the night of Jesus’ arrest Luke tells us that he equipped his disciples with swords (Lk. 22:36-38) in anticipation of trouble. If John is to be our authority, several hundred Roman troops (speira) composed of the garrison stationed outside of Jerusalem (Josephus, Ant., 20.5.3) and the temple police actually confronted Jesus (Jn. 18:12). This sizable detachment must have anticipated that Jesus had many armed followers ready to fight further supporting the suggestion that Jesus led a much larger group. (Commentators point to this and related passages in support of Jesus’ “radical eschatology” where Jesus anticipates a great battle preceding God’s Imperial Rule or “Kingdom of God” but that is beyond the scope of this paper. What is agreed upon is that Jesus is not a pacifist and does display a show of force.)

Horsley (1987) and Sanders (1985) suggest that the cleansing incident was small in scope so as to not attract the temple police, yet drew sufficient attention to warrant eventual concern and arrest. This is perhaps too convenient and seems to defy common sense. If we are to accept uncritically on the one hand that the fickle crowd got out of control when choosing Barabbas over Jesus, why are we suddenly mitigating and controlling this violent crowd scene in the temple? In any case the changing of pagan currency into acceptable coinage and the business of the pigeon merchants were necessary and vital activities of the temple. Their booths were not like the Hollywood movies, confined to a few square meters so that the camera can capture them easily on film. During Passover tens of thousands of people flocked to Jerusalem and its temple necessitating a huge infrastructure to support them. Zealot uprisings were common during the high holidays and temple police were ready for trouble as soon as it erupted in order to prevent possible escalation and Roman involvement. The last thing the high priests wanted was for Roman involvement inside the Temple compound.

All of this brings us back to a few key points. Jesus of Nazareth and Jesus Barabbas were imprisoned for the same offense, sedition against Rome. Each had many followers that supported their efforts. Jesus of Nazareth considered himself a Bar Abba (Mk. 12:6; 13:32; Mt. 11:27; Lk. 10:22) and prays to God as Abba the Father. Jesus Bar Abba (Barabbas) shares similar ascribed prestige either through his physical father or (as I suggest here) if Jesus Bar Abba was Jesus of Nazareth he shared this relationship with God the Father. If one considers that Mark’s job in composing his gospel was to deemphasize Jesus’ political adventures and to instead interpret his life in a christological fashion, the Jesus Barabbas episode makes more sense. Mark’s early Nazarene community lived during the First Jewish Revolt of 70 CE. What better way to disassociate yourself and your savior from an embarrassing revolt than to introduce a proxy or a “changling” who could inherit Jesus’ political baggage from him? This would certainly go a long way toward explaining why Jesus came to be crucified like a common political prisoner.

I readily admit that this argument is very untenable and I only mention it to generate discussion and thought so that the reader may see an example of moving beyond a limited literal reading of the text. A Q scholar would ask me to explain how it is that the earlier sapiental layer seems to concentrate solely on Cynic wisdom, rather than the political adventures of Jesus. Indeed, if the Q hypothesis is sound as I shall later argue, my entire argument crumbles before my eyes! But this should not stop us from considering seemingly contradictory ideas, “trying them on for size” as it were in order to see which best fits the data.

Glenn’s trouble as to whether or not the paschal amnesty was a likely ocurrence seems a minor point (as does the historical figure of Pontius Pilate) so I’ll move on to the question of Jesus’ sedition and blasphemy.

Blasphemy and the Son of Adam

In my earlier critique I suggested in a footnote that Jesus was punished for sedition but was accused of blasphemy. Glenn writes that the text doesn’t support this statement. However I believe that the evidence does support my earlier suggestion. After Jesus was arrested and taken before the Council the high priest asks, “Are you the Anointed, the son of the Blessed One?” (Mk. 14:61 = Mt. 26:63) Jesus replies that he is and that the high priest will see him coming in the clouds of the sky. At once the Council concurs that this is blasphemy and sentences Jesus to death (katakreno, “to condemn” Mk. 14:64; cf. Lev. 24:16).

Matthew tones this down somewhat, perhaps aware that if the Council really did “concur in the death penalty,” as Mark writes, they would have had no need to go to Pilate with the matter since they had jurisdiction to carry out their sentence (see Josephus, Antiq. 20.9.1 and S.G.F. Brandon, The Trial of Jesus of Nazareth, pp. 87-92). Sanders (1993) suggests that Jesus may have committed blasphemy but that Caiaphas felt it more politically expedient to turn Jesus over to Pilate on trumped up charges of sedition rather than risk a mob riot. This sounds like an ad hoc explantion to me, particularly in light of his other mitigation of the Temple Cleansing (see above), but perhaps it should be taken seriously.

It should be mentioned at this point that if Mark is telling his story accurately there would be no blasphemy at all. Proclaiming oneself a “son of Adam” (huion to anthropou) or Anointed (Christos) is no more blasphemous than if Jesus were to say he were an angel, King David or Moses resurrected. The OT precedence tells us that the phrase was used by God to address a prophet (see Ps. 8:5; Ezek. 2:1; Dan. 8:17). But by the time Mark wrote, a christological interpretation of “son of Adam” may have already begun to take on symbolic messianic connotations (Dan. 7:13, 22, 25; cf. 1 Enoch 46-71; 4 Ezra 13). The meaning of the phrase “son of Adam” or “man” (Aramaic, bar enas) is still debated since there does not seem to be a clear usage of it in the Synoptics or John’s Gospel. In other words, Mark’s usage of “son of man” tells us more about his eschatology than it does about a real historical event. It is clear however that Mark considers “Anointed” to be blasphemous and he reflects this belief in his story. If Jesus did declare himself to be the king of Israel, he would have put himself into secular hot water with the Romans, but certainly no religious offense is committed by this admission. The high priest’s dramatic behavior is an excellent literary plot device however, so we can forgive Mark’s artistic license for the sake of the overall story.

I think that if we were to delve into Augustine’s contrast of the Christian City of God versus non-Christian “evil” realms and the anti-Semitism which fomented in the early Church, we would be seriously off subject. While I would enjoy a lively discussion of how it came to be that Judas and the other disciples appear dark, dirty, and hook-nosed in Christian art while Jesus managed to become light-haired and fair, I agree that this is a discussion for another day. Let’s move on to the meat of our disagreement, which lies within the Synoptic Problem.

The Synoptic Problem

Central to our disagreement in NT development is the so-called Synoptic Problem and evidence of the layer within the canon which we call the Q gospel or the “Sayings Source.” Glenn maintains that there is “no textual data whatsoever that supports the belief of layers” within the NT texts (his emphasis). Aside from the poor word choice that wrongly equates scholarly consensus based on evidence with metaphysical belief based on faith, I feel that Glenn’s claim deserves to be taken seriously. If we are going to advance any theories at all about early Christian trajectories, the existence of Q is central to this consideration. I know that Glenn is uncomfortable deviating from the centuries-old orthodox position on this (despite his purported skepticism), but if we’re ever to get at the truth of the matter a brief excursion into the Synoptic Problem and the possible existence of Q is necessary. As we shall see later, Glenn’s claim that the Q hypothesis is flawed because there are no extant texts (“transitional forms”) to support it is now known to be wrong. We do have an extant Q-like document of the logoi sophon gattung: The Gospel of Thomas. The shared genre between Q and Thomas is very well supported and the Coptic Thomas confirmed what the form critics said about Q well before Thomas’ discovery in 1945.

The Synoptic Problem, simply put, is the close interrelations and dissimilarities of the first three gospels, Matthew, Mark, and Luke (or Luke-Acts). As soon as early scholars began to harmonize these three gospels (place them side-by-side in three rows in order to compare what each says about the same event in the narrative) they realized that there seems to be close literary agreement between them. This has led to the natural question of who came first and who relied upon whom when composing his own gospel. Did Mark really abbreviate Matthew as Saint Augustine suggested? Or did Matthew come later and have a copy of Mark at his elbow while composing his own gospel? These are the questions which scholars ask when attempting to reach a solution to the Synoptic Problem. There is one other aspect to the Synoptic problem which I have hinted at already, that being the mysterious Q gospel. There are about fifty instances in Matthew and Luke where each parallel the other very closely, word-for-word in some places and yet these parallels are absent from Mark. Form critics of the nineteenth-century suggested that both Matthew and Luke must have used a similar source and incorporated this source’s material into their own gospel. This source is called Q. The theory that Matthew and Luke used Mark and Q is called the Two-Source Hypothesis (pictured below). One should keep in mind that Q is a hypothesis, not an extant text. However, it is a very sound hypothesis and should be taken seriously; the evidence seems to support it very well as I shall argue. Before addressing Q however, I’d like to support the argument for the priority of Mark, since Glenn and I disagree on this point. For those interested in Glenn’s viewpoint, the best source to turn to is William Farmer’s The Synoptic Problem.

[Mark and Q | Matthew and Luke]

An Argument for the Priority of Mark

H. Holtzmann, in his work Die Synoptischen Evangelien, suggested that Mark should be considered the earliest of the four gospels. In the twentieth-century B.H. Streeter’s The Four Gospels further honed and refined these arguments even though his argument from arrangement has since been questioned. The Gospel of Mark itself still provides good internal evidence for its primitiveness and priority. Mark’s writing is awkward, brief, and choppy; his transitions in the narrative rushed. “Right away” (euthus) Jesus is driven into the wilderness (1:12) and “at once” he leaves the synagogue (1:29). Mark’s Jesus is a man in a hurry, dashing from one scene to the next. Mark’s Gospel is also the shortest, lacking resurrection appearance stories, the Sermon on the Mount (Mt. 5-7), Luke’s most famous parables (Lk. 18-19), and the infancy narratives, all of which took on greater theological significance as Christianity matured; Mark’s Christology is less developed as well. Why would Mark leave out such tremendous and theologically rich material from Matthew and Luke if he came after them? The argument for content demonstrates that out of Mark’s 661 verses all but 31 are paralleled between Matthew and Luke, 600 in Matthew alone so that Matthew reproduces about 90% of Mark in his gospel (Luke about 50%). Consider these three Synoptic parallels:

And whenever the Pharisees’ scholars saw him eating with sinners and toll collectors, they would question his disciples: “What’s he doing eating with toll collectors and sinners?” When Jesus overhears, he says to them: “Since when do the able-bodied need a doctor? It’s the sick who do.” (Mk. 2:16-17)

And whenever the Pharisees saw this, they would question his disciples: “Why does your teacher eat with toll collectors and sinners?” When Jesus overheard, he said, “Since when do the able-bodied need a doctor? It’s the sick who do.” (Mt. 9:11-12)

The Pharisees and their scholars would complain to his disciples: “Why do you people eat and drink with toll collectors and sinners?” In response, Jesus said to them: “Since when do the healthy need a doctor? It’s the sick who do.” (Lk. 5:30-31)

In places such as this where Matthew and Luke lean heavily on Mark’s narrative, they hesitate to change the pericopes. However Matthew and Luke do take the liberty of refining Mark’s grammar. In this parallel, Luke confidently has Jesus say “healthy” rather than “able-bodied”–a literary choice on Luke’s part since it does not change the meaning at all. Matthew and Luke both correct Mark’s awkward use of a double possessive by changing “Pharisees’ scholars” (grammateis ton Pharisaion) so that Matthew renders the simpler “Pharisees” and Luke compounds “Pharisees and their scholars.” There are places where Matthew cleans up after Mark to fix a theological mistake that he feels Mark made:

Jesus came from Nazareth, Galilee, and was baptized in the Jordan by John. And just as he got up out of the water, he saw the skies torn open and the spirit coming down toward him like a dove. There was a voice from the skies: “You are my favored son–I fully approve of you.” (Mk. 1:9-11)

Jesus comes from Galilee to John at the Jordan to get baptized by him. And John tried to stop him with these words: “I’m the one who needs to get baptized by you, yet you come to me?” In response, Jesus said to him, “Let it go for now. After all, in this way we are doing what is fitting and right.” Then John deferred to him. After Jesus had been baptized, he got right up out of the water, and–amazingly–the skies opened up, he saw God’s spirit coming down on him like a dove, perching on him, and–listen!–there was a voice from the skies, which said, “This is my favored son–I fully approve of him!” (Mt. 3:13-17)

God tells Jesus that “you are my favored son” and speaks to him in the first-person singular. Mark also tells us that the “tearing open” of the skies and the “voice” that Jesus hears is an internal experience that only “he saw” (eiden). The verb eiden is ambiguous and can refer to either a real voice and physical opening of the skies or a vision depending upon the context. If we were dealing with normal direct objects for the verb to distribute its action to (i.e., man, boat, sea) then the context would mean for this to be taken literally. However the two direct objects of eiden in 1:10, “skies” (tous ouranous) and “spirit” (to pneuma) are often apocalyptic images usually associated with a revelation or divine vision (Ezek. 1:1; Is. 63:14; Rev. 19:11). So it is not at all clear whether or not Mark means for us to take this as a literal external event or in the sense of divine revelation from God to Jesus alone. Matthew will have none of this ambiguity.

Matthew prefers that the post-baptism vision and God’s communication be an external physical event. He overrides Mark and writes that the skies physically opened up (anoigo oi ouranoi; cf. “mouth opens” 2 Cor. 6:11) and a voice proclaims to all that “this is my favored son” so that there can be no doubt that several witnesses were on hand who would know that Jesus is the Son of God. Matthew is also troubled by Mark’s carelessness in not seeing how Jesus did not need to be baptized for the remission of sin. How does a blameless lamb need the removal of sin? So Matthew inserts some additional narrative that subordinates John to Jesus and has John initially protest at performing the baptism. In this manner Matthew’s Jesus receives baptism for symbolic reasons only, not for any actual need to repent and turn to God. There are times when Matthew and Luke correct Mark’s many irritating and ungrammatical duplicate expressions. For instance, Mark writes “In the evening, at sundown, they would bring all the sick” (1:32). Matthew writes instead “In the evening, they brought many who were demon-possessed” (8:16) and Luke renders “As the sun was setting, all those who had sick people” (4:40). Matthew and Luke each take one part of Mark’s duplicate expression and use it in their own story.

Matthew betrays his knowledge of Mark in more obvious ways as well. Mark’s rendering of the Healing of the Paralytic (2:1-5 = Mt. 9:1-2) tells us that the door to Jesus’ house was so crowded that men bearing the paralytic had to dig through the mud roof in order to be able to lower the man down to Jesus. Matthew omits this detail but when Mark goes on to say that Jesus notices their trust (tan pesten auton) in 2:5, we understand the context of Jesus’ statement. Matthew however, having omitted Mark’s story, still retains Mark’s phraseology so that Matthew writes that Jesus “notices their trust” without providing the necessary context for us to understand the reason for that trust. Matthew is aware of Mark’s story but has forgotten that his readers are not aware of its details. Perhaps even more revealing is Matthew’s treatment of John the Baptist’s death (Mk. 6:17-29 = Mt. 14:3-12). In this story we see clearly how Matthew knows about Mark when writing his own gospel because he abbreviates too much of Mark’s story while showing knowledge of it. Mark tells us that King Herod fully empathizes with and respects the Baptist, yet his wife Herodias wishes the Baptist killed. Despite his intentions, Herod is tricked and is forced to execute the Baptist. Herod considers the Baptist upright and holy and so “grew regretful” (6:26) at having to perform the execution. Matthew decides to change the story and tells us instead that “Herod wanted to kill” the Baptist, not Herodias as Mark wrote (14:5). However, he betrays his knowledge of the Markan account because after Herod had the Baptist killed, he tells us that Herod was sad (14:9). This is confusing because nowhere in Matthew’s account of the story is there reason to think that Herod would be sad after having completed what he set out to do. There is another puzzling element to the Baptist’s execution which supports Markan priority. In 14:1 Matthew quite correctly refers to Herod as “tetrarch.” All NT writers, with the exception of Mark, use tetrarch for Herod’s title; Mark repeatedly uses the incorrect “king” (baseleus; see 6:14). But in one instance Matthew 14:9 retains Mark’s incorrect usage by also referring to Herod as “king.” Parker (1981) brushes aside Mt. 14:9 as mere “confused text,” a lapse on the part of Matthew. This explanation doesn’t seem to support the details of the narrative that have just been discussed however. It seems clear that Matthew is borrowing a Markan phrase because his version of the story does not indicate a reason for Herod’s regret and his readers are again left to wonder why Herod is sad. (This reminds me of an old detective movie where the suspect says, “I didn’t shoot him,” and the private-eye retorts “Shoot him? How did you know he was shot, since all I said was that he was murdered?”) In order to salvage this situation we could suggest that Matthew’s account contained more detail but was lost during a later transmission. But this is pure conjecture and an unsatisfactory answer. Other commentators, notably Taylor (The Gospel According to Saint Mark), have suggested a hypothetical “Urmarkus” that all three Synopticists relied upon. The Urmarkus is also not supported by the internal evidence and remains only an explanation based on conjecture. Once we realize that Matthew borrowed the story from Mark and changed the plot to suit his own needs, a great deal of confusion is lifted. It is only when we try to follow Augustine’s suggestion–that Matthew was written first and Mark was an abbreviation of it–that we run into trouble explaining these phenomena. Today, scholars overwhelmingly support the Markan priority hypothesis. This is not to suggest that Glenn disagrees with this scholarly consensus, but if I understand him correctly he argues for Matthean priority over Mark. Considering the above evidence that I went into in some detail (but certainly not exhaustive) this may place him into an awkward position. (See Robinson [1962] for an overview of Mark’s kerygma and history; Fitzmyer [1970], Neirynck [1976], Wood [1953], and Kümmel [1973] for support of Markan priority and refutations of Butler’s [1969] and Farmer’s [1976] views; and finally Mack [1993] for an excellent introduction of the Synoptic Problem and Q scholarship as it exists today.)

Why Even Bother with Biblical Criticism?

Before plunging into Q, I do wish to briefly defend the merits for biblical criticism. I am discouraged by Glenn’s polemics against the many scholars of the past two hundred years who have paved the way so that we may have this discussion now. Prior to the nineteenth century it was dangerous for a scholar to make observations about the Bible. Many, like Strauss, were fired from their professorships, excommunicated from the Church, or sent to prison like Thomas Woolston (who made the mistake of wondering aloud whether Jesus’ miracles really happened.) Others, like Aikenhead, were sentenced and hung for suggesting that Moses may not have written the Pentateuch (the first five books of the Hebrew Bible.) These were not men who, as Glenn puts it, engaged in “arbitrary assumptions” because they had an ax to grind or were out to destroy the Bible. They were men in search of the truth at the risk of their own lives. Glenn seems to disagree with the existence of Q (and all layers apparently except for an early Matthew; see below) If I understand him correctly, he maintains that the NT texts must be treated as historical documents and given the benefit of the doubt; since we cannot “prove” anything to the contrary we should just leave well enough alone. If Glenn is extending a theological interpretation in order to bolster a literalistic faith then perhaps to those hungry for such assurance this is good advice. But textual criticism does not attempt to destroy individual faith. It does not even concern itself with faith when studying these texts since those questions are left to the theologian. Christianity has always been based on the Risen Christ, not on so-called inerrant or historical NT texts (Patterson, 1995). In other words, if Q exists, Christian orthodoxy will not crumble tomorrow; on the contrary, our knowledge and appreciation of early Christianity will only improve. With this in mind, let’s turn to the existence of Q.

The Q Hypothesis

Q is the second source next to Mark in the two-source hypothesis. Q helps to explain the relationship between the Synopticists and how Matthew and Luke incorporated closely paralleled material into their gospels. This distinct literary tradition discernable within Matthew and Luke cannot be traced to Mark. The Q gospel is believed to be a once-existing collection of codified oral traditions based on the group of stories then-circulating in the early Jesus movements.

If we can find clear parallels in both Matthew and Luke that cannot be traced to Mark, then it becomes increasingly likely that at least an oral tradition familiar to both evangelists was available to them. However, what we find in the texts is not only a clear parallel, but instances where wording is too close to explain merely an oral tradition, but suggests instead a codified source. For instance examine this well-known Q pericope from Matthew and Luke:

You brood of vipers! Who warned you to flee from the coming wrath? Bear fruit worthy of repentance, and do not presume to say to yourselves, ‘We have Abraham as our father’; for I tell you, God is able from these stones to raise up children to Abraham. Even now is the axe laid to the root of the trees; every tree therefore that does not bear good fruit is cut down and thrown into the fire (Mt. 3:7-10).

You brood of vipers! Who warned you to flee from the coming wrath? Bear fruits worthy of repentance, and do not begin to say to yourselves, ‘We have Abraham as our father’; for I tell you, God is able from these stones to raise up children to Abraham. Even now is the axe laid to the root of the trees; every tree therefore that does not bear good fruit is cut down and thrown into the fire (Lk. 3:7-9).

Those words in bold type indicate variants between the two passages; all other words are the same between Matthew and Luke. Given that these variants are editorial or grammatical decisions on the part of the evangelists, these two passages are remarkably similar. Here are a few more sayings of Q found in Matthew with Luke’s variants placed in parentheses:

Why do you seek the speck that is in your brother’s eye, but do not notice the beam that is in your own eye? Or (How) will (can) you say to your brother,'(Brother) let me remove the speck from (that is in) your eye, and behold, there is a (when you yourself do not see) the beam that is in your own eye? You hypocrite, first remove the beam from your own eye, and then you will see clearly to remove the speck from (that is in) your brother’s eye (Mt. 7:3-5; Lk. 6:41-42).

For I myself am a man (set) under authority, with soldiers under me: and I say to one, ‘Go,’ and he goes; to another, ‘Come,’ and he comes; and to my slave, ‘Do this,’ and he does it. When Jesus heard (this) he marvelled (at him), and to those (turning) that followed him, he said “I tell you, I have not found such trust in Israel” (Mt. 8:9-10; Lk. 7:8-9).

And Jesus (he) answered them, “Go and tell John what you you hear and see (have seen and heard): the blind recover their sight, and the lame walk, lepers are cleansed, and the deaf hear, and the dead are raised up, the poor are given good news. And whoever is not offended by me is blessed (Mt. 11:4-6; Lk. 7:22-23).

What did you go out into the wilderness to see? A reed shaken by the wind? But what did you go out to see? A man clothed in luxurious clothing? Behold, those who wear luxurious garments (who are gorgeously apparelled and live in luxury) are in the houses of kings (royal palaces). But what did you go out to see? A prophet? Yes, I tell you, and more than a prophet. This is the one about whom it is written, ‘Behold, I send my messenger before your face, who will prepare your way before you.’ Amen. I tell you, among those born of women there has not arisen one (none is) greater than John; yet the least in the kingdom of the heavens (God) is greater than he” (Mt. 11:7-11; Lk. 7:24-28).

We piped to you, and you did not dance; we sang a dirge, and you did not mourn (weep). For John (the Baptist) has come (came) neither eating (bread) nor drinking (wine); and they (you) say, ‘He has a demon. The Son of man came (has come) eating and drinking; and they (you) say, ‘Behold, a glutton and a drunk, a friend of tax collectors and sinners!’ Yet Wisdom is vindicated by (all) her works (children) (Mt. 11:17-19; Lk. 7:32-35).

I follow you wherever you go. And Jesus says (said) to him, “Foxes have holes, and birds of the sky have nests; but the Son of man has nowhere to lay his head (Mt. 8:19-20; Lk. 9:57-58).

The harvest is large, but the workers are few; therefore beg the master of the harvest to send out workers into his harvest. (Go!) Behold, I am sending you as sheep among wolves (Mt. 9:38; Lk. 10:2-3).

Woe to you, Chorazin! woe to you, Bethsaida! for if the miracles done in you had been done in Tyre and Sidon, they would have repented long ago (sitting) in sackcloth and ashes. But I tell you it shall be more tolerable on the day of judgment for Tyre and Sidon than for you. And you, Capernaum, will you be exalted to heaven? You shall descend to Hades! (Mt. 11:21-23; Lk. 10:13-15).

Ask, and it will be given you; seek, and you will find; knock, and it will be opened to you. For every one who asks receives, and whoever seeks finds, and whoever knocks is admitted (Mt. 7:7-8; Lk. 11:9-10).

Well it is certainly not my intention to repeat all 50 or so Q pericopes here, only to show the harmony of a handful of them so that the reader may compare the Matthian and Lukan versions and see for themselves. (I recommend The Complete Gospels, HarperCollins, 1992, for the full Q text; also see the updates of the International Q Project published in the Journal of Biblical Literature every fall.) It quickly becomes apparent after examining Matthew and Luke in harmony that there is a large corpus of material that is almost word-for-word between them, yet is not found in Mark. This is the Q. No, it is not an external “hard” source and you will not find it in the archaeological record as Glenn rightly points out. But it is there nonetheless as you can see from the sampling of Q pericopes above. At this point the skeptic may ask why we need to jump to a Q hypothesis. Why not consider that Luke copied from Matthew, or Matthew from Luke instead? The second consideration, that Matthew borrowed his Q material from Luke, is no longer seriously considered today. Styler (1962) and Kümmel (1973) write for everyone in saying that Lukan priority can drop from consideration so I won’t beat that dead horse here either. However, the other consideration, that Luke copied from Matthew, is still widely claimed today. Let’s explore this other alternative. If we have good reason for doubting that Luke copied from Matthew, this argues ipso facto in favor of Q.

Did Luke Copy the Q Material from Matthew?

Streeter (1924) made an early argument against the possibility that Luke copied Matthew. He relates his encounter with one of the first scholars to compose a gospel harmony:

Sir John Hawkins once showed me a Greek New Testament in which he had indicated on the left-hand margin of Mark the exact point in the Markan outline at which Matthew has inserted each of the [Q] sayings in question, with, of course, the reference to chapter and verse, to identify it; on the right-hand margin he had similarly indicated the point where Luke inserts matter also found in Matthew. If Luke derived this material from Matthew, he must have gone through both Matthew and Mark so as to discriminate with meticulous precision between Markan and non-Markan material: he must then have proceeded with the utmost care to tear every little piece of non-Markan material he desired to use from the context of Mark in which it appeared in Matthew–in spite of the fact that contexts in Matthew are always exceedingly appropriate–in order to reinsert it into a different context of Mark having no special appropriateness. A theory which would make an author capable of such a proceeding would only be tenable if, on other grounds, we had reason to believe he was a crank.

This raises another obvious question, that being, if Luke knew about Matthew why would he shatter the beauty of the Sermon on the Mount (Mt. 5-7)? Luke places some of it in his own Plain, sprinkles some other of it throughout his gospel and ignores the rest. A careful historian such as Luke would find this practice beyond the pale. The implications of what Streeter mentions above raise another interesting point: Why is it that Luke never places his material that he has in common with Matthew at the same point in the Markan framework? Except for the baptism, temptation, and a few other passages, Luke would have had to face the monumental task of making sure his common material didn’t follow Matthew’s order, but that his own sources followed Mark’s order when Matthew’s didn’t! If Luke did copy Matthew, why didn’t he take over any of Matthew’s additions to Mark? (see Schmid [1930], Bradby [1956], and Downing [1964].) These questions have never been satisfactorily answered and therefore the current scholarly consensus is that direct dependence of Luke upon Matthew is improbable. (For good attempts at answering those questions see Argyle [1964] and Farmer [1975].) Bradby represents current scholarly consensus regarding the suggestion that Luke knew of and used Matthew in composing his own gospel and is worth quoting at length:

If St. Luke had Matthew as well as Mark before him when he wrote, we can picture him following Mark verbatim whenever he had him, and Matthew verbatim when had not Mark. But what, then, would happen when he had rival versions of an incident from Mark and from Matthew? What we should normally expect of any conscientious historian to do is to take the later and fuller version, in this case Matthew (since ex hypothesei Luke knew as well as we do that Matthew had incorporated Mark’s shorter Gospel into his longer one). We might, however, concede that for certain literary reasons he might follow sometimes one and sometimes the other. But if we find, as we have found in four important passages [Lk. 6:1-11; 8:4-15; 9:1-5; 9:18-27], that the later and fuller version is consistently spurned in favor of the earlier and shorter, and that there is not one clear instance in these sections of any non-Markan passage which Luke has derived from Matthew, we can hardly be blamed if we fall back, with relief, on the alternative hypothesis, that in many passages Luke has used Mark and in many others Luke and Matthew have each used a common source other than Mark, that is, Q.

The problems that erupt when we try to fit Matthean priority into the origins of the texts are just too great to surmount. We do disagree on the exact details and corpus of the Q layer, but not on its existence. The Q source and Markan priority is now so well-supported that the two-source hypothesis is considered de facto when discussing the Synoptics. This does not make it automatically correct. However the two-source hypothesis is the single best theory currently in use and has stood up very well to intense critical examination. If we are to entertain Matthean priority instead, the case needs to be made for why it is more tenable or accurate. This has not been done.

Matthean Priority: A House of Cards

Glenn revives the old Griesbach hypothesis which suggests that Matthew, not Mark, was the first written gospel and that Mark is a later redactor of Matthew. His timeline of textual tradition begins:

40 A.D. – Gospel of Matthew basically written

45 A.D. – Gospel of Mark written

54 A.D. – Gospel of Luke

The suggestion that Matthew is “basically written” and placed prior to Mark is an ace up the Griesbachian sleeve. This is so that when material is produced from Matthew that argues for redaction of Mark on Matthew’s part, the Griesbachians can claim that it belongs to a later layer of Matthew and not the original pre-Mark Matthew (or proto-Matthew) of 40 A.D. In so doing they can preserve the priority of Matthew (even if it is only a primitive layer) without having to awkwardly explain how such a mature theology could appear so early compared to a logoi sophon or Markan kerygma. Shaving the size of Matthew down to a proto-Matthew is a small price to pay for preserving the orthodox order and preventing the priority of Mark from taking hold. The problem with putting the theory ahead of the data is that extremely complex models involving conflated layers of proto-Matthews and proto-Lukes are necessary to make them hold (see Lindsey [1953] and more recently Parker [1981]). A Ptolemaic confusion results from making the data fit these models while a much simpler and elegant Q source suffices in explaining the data. Farmer’s (1976) solution to over-complexity seems very much akin to Glenn’s, which is to deny that Synoptic redactions and seams even exist and then tip-toe back to the glorious Matthean priority days in order to draw the desired results from this conclusion. But a poetic fate worse than refutation has struck Farmer since his decision to ignore the Q data: his colleagues have largely ignored him.

Glenn’s timeline is very convenient since he has already told us that it takes a minimum of five years for a later redactor to absorb and retransmit the tradition of an earlier writer; it just so happens that on Glenn’s timeline Mark follows Matthew by this handy five-year span. There is a good reason for this. He needs to support the idea that Paul quoted Luke as “Scripture” and so since Paul wrote in the mid first-century, this necessitates an earlier date for Luke. As just discussed a dependence of Matthew on Luke is so unlikely that no one even bothers to mention it anymore. Since Markan priority is rejected a priori, this leaves one last alternative, a dependence of Luke upon Matthew so that we now must push Matthew back prior to Luke at 40 A.D. Before this really gets out of hand, let’s examine why Glenn feels that he needs to place Luke and Matthew much earlier than scholarly consensus suggests.

Since Paul allegedly quoted Luke in his first letter to Timothy, Glenn dates Luke at 54 CE. Quite a house of cards is built up around this idea because Glenn goes on to say that Paul knew about Luke, considered it “Scripture,” and therefore Luke must pre-date Paul. He goes on to tell us that there couldn’t possibly be an earlier Q source because if Paul were quoting that instead of Luke, it would not have been lost, rather, considered authoritative and preserved in the canon. Since it was not preserved, Paul must have been quoting Luke or so the theory goes. This is a remarkable leap of faith which must be examined closely if for no other reason than to demonstrate the awkward interpretations that result from rejecting the priority of Mark and Q. The Timothy passage in question reads:

For the Scripture says, “You are not to muzzle the ox while it is treading out the grain;” and the workman deserves his pay (1 Tim. 5:18).

The Luke 10:7 parallel says “workers deserve their wages” which is also believed to be a Sayings pericope (Q 10:2-12). When Paul referred to “Scripture” he always meant what we now call the Old Testament which was to him the Greek Septuagint (e.g., Rom. 4:3; 9:17; Gal. 3:8, 22). Even if Paul had seen an early proto-gospel such as Q, he would never have considered such a text on par with the OT. If Paul were referring to an OT passage to make a contemporary theological point, he would be very clear in distinguishing the Word of God from his own (which is what the translators above also thought when they separated Deut. 25:4 from the Q pericope by placing the former in quotes). The Scriptures were used by Paul to refer to a specific passage (graphe) or the entire OT (haigraphai). No writer prior to the late second century refers to the NT writings as Scripture; Clement, Justin Marytr, and Polycarp refer to the OT when they speak of Scripture. It is not until we get to Theophilus of Antioch and Irenaeus (c. 180) that NT passages (including Paul’s letters) are treated as Scripture, i.e., the Word of God.

Despite this obvious problem, we have good reason to suspect that 1 Timothy is not genuinely Pauline anyway. Upon examing the three Pastoral Epistles (as the two Timothies and Titus are called since they are church manuals or tracts) we see that their style differs remarkably from Paul. Phrases such as “faithful is the saying” occur in these letters but nowhere else in Paul (1 Tim. 1:15; 3:1; 4:9; 2 Tim. 2:11; Tit. 3:8). If it were only a matter of counting phrases and words (which is not always predictable or factual) we could forgive this. But the Pastoral Epistles are filled with words, adverbs, conjunctions, and particles that betray a writer’s style far different from Paul’s. Hills (1991) while suggesting a possible genre for 1 John, doesn’t even qualify his reference to these letters as “deutero-Pauline.” Today, it goes without saying that the Pastoral Epistles are Paul-like, but certainly not Pauline. I refer to the late master Morton Scott Enslin (1938) for the final word on Timothy’s veracity:

It would be easy to [show Pauline dissimilarities in the Pastoral Epistles] but, as has been said in a different connection, “It is not necessary to use battering-rams on gates that stand open.” The whole situation is impossible, not to say absurd and grotesque, if we assume the historic Paul writing personal letters to the historic Timothy and Titus in real situations. But if we admit that we have here the work of a later writer seeking to meet the circumstances of his own day and wishing to gain for his instruction the authority of the now long-dead Paul, the absurdity is removed (pp. 299-307).

The letter that Glenn wishes to use to support an early Luke is actually a pseudipigraphical work–perhaps the production of an orthodox clergyman who realized that the second coming was long delayed and so desired to organize the early Church around less Parousiac expectations. Whoever the author was, he felt that his secular concerns would be treated with more respect if they came from the mouth of Paul than from his own. This technique was nothing new since it had already been employed by the gospel writers who have Jesus speaking authoritatively to theological points that arose in the evangelist’s day. No one can know for sure when 1 Timothy was written, but inspection of the problems addressed in the letter have led some commentators to place it no earlier than 100-120 CE. The fact that the author of Timothy seems to suggest that a Q pericope incorporated into Luke is on par with Scripture betrays it as a late second century work in my opinion.

We just have no support for Glenn’s suggestion that “virtually all scholarship, regardless of theological orientation accept Timothy as Pauline.” If we date Timothy at a terminus a quo of 100 CE, ten or twenty years separate Luke from it and perhaps 50 years from Luke’s sources to it. Therefore we have no need to jump to the conclusion that Timothy was written by Paul or that Matthew somehow precedes Mark. If one wishes to see that the entire NT canon is the inerrent and revealed Word of God, then strained interpretations such as these will obviously find their way to sympathetic ears. But if Glenn wishes to be taken seriously outside of that audience he may have to revise this position.

The Gattung of Q

Rudolf Bultmann in his History of the Synoptic Tradition first commented on the similarity between Q and wisdom literature (Proverbs, Job and Ecclesiastes from the Hebrew scriptures and perhaps Sirach and Wisdom of Solomon from the Apocrypha). Robinson (1971) took the next step in initiating a “systematic investigation” with the goal of determining the gattung or the genre of Q which he called the logoi sophon or “sayings of the sages” after Proverbs 22:17. His study of Q was a breakthrough because it connected Q to a newly-discovered text in Egypt called The Gospel of Thomas. Earlier fragments of Thomas (the Oxyrhynchus Papyri: POxy 1, 654, and 655) were orthographically dated to just prior to 200 CE and soon connected to the full Coptic version discovered in 1945 so that Thomas represented an example of very early Jesus sayings. When first discovered, Thomas was believed to be a purely Gnostic production of the third century and largely dependent upon the canonical gospels for its material. But Thomas’ literary genre, as Robinson showed, is much closer to Q than any other gattung, including those noncanonical gospels that did redact the Synoptics. In his introduction to Thomas in the Nag Hammadi Library in English, Helmut Koester states:

Direct dependence of The Gospel of Thomas upon another noncanonical gospel is very unlikely. More problematic is the relationship of The Gospel of Thomas to the canonical gospels. Whereas the latter all contain large segments of narrative materials, no traces of such materials are found in the former. Already this makes it unlikely that our document can be considered as an eclectic excerpt from the gospels of the New Testament. . . . The Gospel of Thomas is a closely related but independent collection of sayings. In its original form, it may well date from the first century (p. 125).

The Fellows of the Jesus Seminar agree with Koester; The Complete Gospels dates Thomas to “the latter decades of the first century CE” for several reasons. First, is the total absence of secondary accretions that one might expect if Thomas were a redactor of the Synoptics. In other “gnostic” gospels, this dependence is easily seen but not so with Thomas. Kloppenborg (1990) who has devoted a considerable amount of effort in understanding Q and Thomas notes that “most of the sayings in Thomas which have synoptic parallels occur in forms which are more primitive than their synoptic parallels” (p. 87). The collection of sayings in Thomas belong to the period when the Jesus movement was still appealing to the authority of individual apostles (like Mark’s gospel at 70 CE) rather than the twelve as a whole. Consider the following Thomas passage:

The disciples said to Jesus, “Tell us how our end will be.” Jesus said, “Have you discovered the beginning, then, that you are seeking after the end? (Thomas 18)

This is typical of the wisdom sayings in Thomas and suggests an early codification from the oral tradition which thrived in early Galilee, Samaria and Judaea. (This particular saying is a favorite of mine.) The genre that Thomas represents was incorporated into the Synoptics and not used after Matthew and Luke became popular. As Robinson pointed out, since Thomas belongs to the period of codified oral traditions in the mid to late first-century, it was assembled prior to the realization of the late second-century canon. All of this suggests that Thomas be placed at roughly the same period as the canonical gospels. There is considerable room for debate on the exact dating, particularly as Kloppenborg notes:

We know that even the canonical gospels, Matthew, Mark, Luke, and John, were not entirely stable until relatively late in the history of their transmission, so that one must frequently distinguish between earlier and later materials contained within them (p. 88).

Which suggests that Thomas may have endured this same flux in its development rendering a precise date impossible. Thomas scholarship has reached a date circa 70-100 CE today, which is probably not far off the mark since it could conceivably belong to Q at about 50 CE and certainly a terminus ad quem of 200 CE. Kloppenborg provides an excellent example of the primitiveness of Thomas. Luke’s incorporation of Q material is confusing in some places, particularly Luke 17:22-37: “for behold, the reign of God is entos hemon which is to say perhaps “inside of you” or “in the midst of you.” The former may be interpreted as a spiritual reality inside of each person while the latter takes on a communal tone. No one has been sure how early Christians understood this passage and Lukan (and Matthean parallels 24:23, 26-27, 37-39, 17-18; 10:39; 24:40-41, 28) inherit this phrase from their sayings sources. Thomas provides the answer by telling us that “the kingdom of the Father is spread out upon the earth, and people do not see it.” Thomas has provided us with an interpretation that suggests a present-tense physical manifestation of the kingdom (or imperial rule) of God.

Today, we must take seriously the suggestion that Thomas is a “sayings gospel with its own compositional integrity, generic identity, and transmissional history . . . situated at a particular juncture in early Christian history” (Cameron, p. 388). Part of the reason Thomas is only now being seen as an integral part of Christian origins is because it has taken scholars quite a long time to orient their thinking away from the embedded conventional picture that has formed uncritical portrayals in the past. To suggest, for instance, that some early Jesus followers did not see him as virgin-born conflicts with our popular myths and so meets stiff resistance. Consequently, because the logoi sophon genre does not seem to address Jesus’ infancy or passion narratives, some scholars uncritically reject it out of hand for this reason. What Koester (building on Bauer’s [1971] watershed work) demonstrated was that we must rethink this conventional portrayal which emerged from the evolved kerygma of the Synoptics and consider these other trajectories with equal weight. In other words, the fact that Matthew and Luke redact Q, should indicate that they considered the logoi sophon important enough to incorporate into their portrait of Jesus.

But a lot of work still remains to be done. We cannot over-speculate on what Thomas or Q may mean for early Christianity. However, we can acknowledge that a fresh and exciting portait of early Christianity is slowly emerging. This portrait is not “true” or “false” in the sense of their interpreting Jesus’ message; all philosophies of the early Jesus movements enjoyed equal validity. Just as many people in a large auditorium can come away with different understandings of what a speaker says, so did those in the early Jesus movements understand Jesus differently. It is only when we dogmatize our own understanding and presume that everyone else is wrong that Jesus’ teachings are truly lost.


Argyle, A.W. “Evidence for the View that St. Luke Used St. Matthew’s Gospel,” Journal of Biblical Literature 83: 390-96, 1964.

Bauer, Walter. Rechtgläubigkeit und Ketzerei im ältesten Christentum. (1934). Orthodoxy and Heresy in Earliest Christianity. Philadelphia Seminar on Christian Origins, trans. Robert A. Kraft and Gerhard Krodel, eds. Philadephia: Fortress Press, 1971.

Bradby, E.L. “In Defense of Q,” The Expository Times 68: 315-18, 1956.

Butler, B.C. “The Synoptic Problem,” A New Catholic Commentary on Holy Scripture. R.C. Fuller, L. Johnston, and C. Kearns, eds. New York: Nelson, 1969.

Cameron, Ron. “Gospel of Thomas and Christian Origins” The Future of Early Christianity. Birger A. Pearson, ed. Minneapolis: Fortress Press, 1991.

Complete Gospels. Robert J. Miller, ed. Sonoma, Calif.: Polebridge Press, 1992.

Downing, F. “Towards the Rehabilitation of Q,” New Testament Studies 11: 169-81, 1964.

Enslin, Morton Scott. Christian Beginnings. New York: Harper & Brothers Publishers, 1938.

Farmer, William R. “A Fresh Approach to Q,” Christianity, Judaism, and Other Greco-Roman Cults: Studies for Morton Smith at Sixty. Jacob Neusner, ed. Part One: New Testament, 39-50. Leiden: E.J. Brill, 1975.

__________. “A New Introduction to the Problem,” The Synoptic Problem: A Critical Appraisal. 2nd ed. Macon, Georgia: Mercer University Press, 1976.

Fitzmyer, Joseph A. “The Priority of Mark,” Jesus and Man’s Hope (Pittsburgh Theological Seminary) 1:134-147, 1970.

Hills, Julian V. “A Genre for 1 John.” The Future of Early Christianity. Birger A. Pearson, ed. Minneapolis: Fortress Press, 1991.

Horsley, R.A. Jesus and the Spiral of Violence: Popular Jewish Resistance in Roman Palestine. San Francisco: Harper & Row, 1987.

Kloppenborg, John S., et. al. Q Thomas Reader. Sonoma, Calif.: Polebridge Press, 1990.

Koester, Helmut. “Gnomai Diaphoroi: The Origin and Nature of Diversification in the History of Early Christianity,” and “One Jesus and Four Primitive Gospels,” Trajectories through Early Christianity. Philadelphia: Fortress Press, 1971.

Kümmel, Werner G. “Attempt at a Solution to the Synoptic Problem,” Introduction to the New Testament. Nashville: Abingdon Press, 1973.

Lindsey, Robert. “A Modified Two-Document Theory of Synoptic Dependence and Interdependence,” Novum Testamentum 6: 239-63, 1953.

Mack, Burton L. The Lost Gospel. New York: HarperSanFrancisco, 1993.

Nag Hammadi Library in English. James M. Robinson, ed. San Francisco: Harper & Row, 1988.

Neirynck, F. “Synoptic Problem,” The Interpreters Dictionary of the Bible. Nashville: Abingdon, 1976.

Parker, Pierson. “A Second Look at The Gospel Before Mark,” Journal of Biblical Literature 100: 395-405, 1981.

Patterson, S.J. “Yes Virginia, There is a Q,” Bible Review 11(5): 39-40, 1995.

Robinson, James M. “Logoi Sophon: On the Gattung of Q,” Trajectories through Early Christianity. Philadelphia: Fortress Press, 1971.

__________. The Problem of History in Mark. London: SCM Press Limited, 1962.

Sanders, E.P. Jesus and Judaism. Philadelphia: Fortress Press, 1985.

__________. The Historical Figure of Jesus. London: Penguin Press, 1993.

Schmid, Josef. Matthäus und Lukas: Eine Untersuchung des Verhältnisses ihrer Evangelien. Frieburg im Breisgau: Herder and Co., 1930.

Streeter, B.H. “The Document Q,” The Four Gospels: A Study of Origins. London: Macmillan, 1924.

Styler, G.M. “The Priority of Mark,” The Birth of the New Testament. C.F.D. Moule. New York: Harper and Row, 1962.

Wood, H.G. “The Priority of Mark,” The Expository Times 65:17-19, 1953.

all rights reserved