Thursday, April 9, 2009

Kloppenborg's stratification of Q


This article was originally published in the Journal of the Evangelical Theological Society, June 2003.


When John Kloppenborg (now John Kloppenborg Verbin) wrote The Formation of Q in 1987, it was impossible to foresee that this book would become a foundational study for some of the more radical presentations of Jesus as a non-apocalyptic Jewish Cynic.[2] Although Kloppenborg distances himself from the Jewish Cynic thesis, his work has been used not only to support that thesis but to revise Christian origins.[3] For example, Ron Cameron, discussing Kloppenborg’s work says that “Q demonstrates that there is no need to appeal to the crucified and risen Christ in order to imagine the origins of Christianity.”[4]
Kloppenborg himself believes that his hypothesis supports the idea that the Q community had a soteriology fundamentally different than the Pauline soteriology. Keylock summarizes the issue this way: “For Kloppenborg, Q represents a form of Christianity in Galilee that was ignorant of the Pauline tradition, [and] knew nothing of Jesus’ atoning death and resurrection…”[5]
Considering the implications of Kloppenborg’s hypothesis for the historical study of Jesus, this article will provide an analysis of Kloppenborg’s thesis as stated primarily in his book, The Formation of Q,[6] but also with reference to his later work, Excavating Q, published in 2000.[7]

Agreeing with Bultmann that Q was “a transitional stage between the un-messianic preaching of Jesus and the fully self-conscious kerygma of the Hellenistic churches,” [8] Kloppenborg insists that there are only two ways to account for this transition.[9] Either there were two kerygmas existing together in the same churches, or there were at least two kerygmas existing in different churches.[10] Since the core of Q’s proclamation centers on the parousia and not on the passion, Kloppenborg concludes that the earliest churches were proclaiming fundamentally different messages about Jesus, indeed, that Q “had an understanding of soteriology which was at variance with the passion kerygma.”[11]
Kloppenborg analyzes the arguments of scholars who propose that Q originally contained a passion narrative, found their arguments to be less than convincing and concluded that “The thesis of a Q passion account must accordingly be rejected.”[12]
Assuming that virtually all of Q can be reconstructed, Kloppenborg cites Arland Jacobson who argued that Q is pervaded with deuteronomistic theology.[13] According to Kloppenborg, this demonstrates that “Q was organized and redacted from a coherent theological perspective.”[14] Since Q contains both sapiential and apocalyptic forms, Kloppenborg asks whether Q underwent a “redactional intervention” such that one of these elements was formative and the other secondary.[15]
Assuming the answer is yes, Kloppenborg proposes a method to determine the principles used in the composition of clusters of Q sayings and their association into a whole.[16] He then meticulously analyzes the pericopae in Q separating the sapiential from the prophetic elements,[17] proposing that Q began as a sapiential document (often cited as Q1), was later revised with the addition of prophetic/judgment/apocalyptic passages (Q2), and was finally revised once more with the addition of narrative passages (Q3).


Kloppenborg’s case depends (1) on the assumption of Q’s existence, (2) on arguments that nearly all of Q can be recovered, and (3) on methodology used to separate sapiential from prophetic/judgment material. This critique will largely bypass the issue of Q’s existence, focusing instead on arguments Kloppenborg presents for the extent of Q and on his methodology for separating Q into strata.
1. The existence of Q. While some might argue that Kloppenborg’s theory is invalid because Q is “only” a hypothesis, or that Q did not exist, Kloppenborg would counter that all we have with regard to the synoptic problem are hypotheses and that the Two Document Hypothesis has been more convincing to most scholars than any other option.[18] His point is well taken. On the other hand, the increasing scholarly disagreement is such that the Two Document Hypothesis cannot be regarded as established fact.[19] Significant uncertainty remains.[20]
For those who hold the Two Document Hypothesis, this uncertainty alone is not a reason to discard Kloppenborg’s thesis. The real problem is the number of hypotheses Kloppenborg builds on top of this increasingly questionable foundation.[21]
2. The extent of Q. Kloppenborg argues at some length for the thesis that virtually all of Q has been reconstructed.[22] This is significant because if virtually all of Q has been reconstructed, it can be argued that Q did not contain a passion narrative and was indifferent to the Pauline kerygma.
On the other hand, if Q was originally more extensive than what can be reconstructed, it would be difficult to know whether Q ever contained the passion narrative or kerygma, and it would, therefore, be invalid to use Q as evidence for the circulation of two substantially different kerygmas in the early church. As Kloppenborg himself says, “The question of reconstruction is crucial if, for example there is any possibility that Q contained elements (such as extended narratives) it would require a radical reassessment of the character of Q”[23] (emphasis mine).
Kloppenborg cites two reasons scholars can be sure that virtually all of Q can be reconstructed. First, it can be assumed that Matthew and Luke would have treated Q as they did Mark.[24] In other words, just as Matthew and Luke included virtually all of Mark in their gospels, they can also be assumed to have included virtually all of Q.
There are, however, good reasons for questioning the validity of this assumption, not the least of which are Kloppenborg’s own arguments. For example, in Excavating Q, Kloppenborg argued against a theory of Wilhelm Bussman by saying Bussman’s hypothesis “rested on the assumption that each evangelist would always treat his sources in a uniform way.”[25] Kloppenborg disagrees, saying that this assumption is not supported by the way Matthew and Luke treated Mark.[26] Later in a different context, Kloppenborg notes: “We should not assume a priori, for example, that Luke would automatically treat a document lacking a clearly narrative structure in the same way the he treated a narrative document like Mark.”[27]
So while, in Excavating Q, Kloppenborg argues in two separate contexts that it cannot be assumed that Matthew and Luke would have treated Q as they did Mark, his entire thesis is based to a large extent on his assumption stated in The Formation of Q, that Matthew and Luke treated Q as they did Mark.
At the risk of creating a straw man one might assume that If pressed for an explanation on this apparent contradiction, Kloppenborg could possibly explain that on a micro level Matthew and Luke treat Mark differently than Q while on a macro level they treat Mark and Q the same by including virtually all of Mark and Q into their Gospels. Such an argument, however, would amount to special pleading since we do not know how much of Q they included, and it would not change the fact that it cannot simply be assumed that any author would use two sources the same way.
The second reason Kloppenborg cites for believing that virtually all of Q can be recovered is Kilpatrick’s argument that the disappearance of Q was explicable only on the assumption that it was almost completely absorbed in Matthew and Luke.[28] Some have objected, however, that by this logic Mark might have been expected to disappear also since virtually all of Mark is included in Matthew. In Excavating Q, Kloppenborg responds to this objection by citing Luhrmann who proposed that Mark’s survival and Q’s disappearance are simply accidents of history in that Mark’s gospel happened to be carried to Egypt where it was copied and Q was not. Further, responding to Dunn who argued that Q may have disappeared for theological reasons, Kloppenborg writes, “In fact we do not know why Q disappeared.”[29]
Kloppenborg is right, of course, but he seems unaware that in arguing that Q’s disappearance may have been an accident of history and by admitting that we frankly do not know why Q disappeared, he has contradicted one of his own primary reasons for assuming that we know the extent of Q, i.e. that the disappearance of Q “was explicable only on the assumption that it was almost completely absorbed in Matthew and Luke” (emphasis mine).[30] Therefore, the main reasons Kloppenborg proposes for assuming that virtually all of Q can be reconstructed have been successfully refuted by his own arguments.
3. A passion account? In The Formation of Q, Kloppenborg interacts with the arguments of Bundy,[31] Burkitt, [32] Hirsh,[33] Taylor and Schneider[34] who attempt to show that elements of the passion accounts in Mark and Luke (L) should be considered part of Q. After critiquing these arguments and finding them less than convincing, Kloppenborg concludes that “The thesis of a Q passion account must accordingly be rejected.”[35]
Finding flaws in an opponent’s arguments, however, is not the same as proving ones own position. For example, alien conspiracy theorists may find flaws in some of the government’s explanations for UFO’s, but they have not thereby proven that aliens exist. Similarly, the fact that Kloppenborg finds his opponents’ arguments unconvincing, does not mean he has thereby demonstrated that Q never had a passion account.[36]
The main evidence Kloppenborg provides for the lack of a passion account in Q is the assumption that virtually all of Q has been recovered, and that it is lacking a passion account. As seen above, however, Kloppenborg’s case for the extent of Q has been found to be invalid by his own arguments.[37]
Even if Q did not have a passion account, however, the assumption that the Q community did not, therefore, know of or believe in the passion and resurrection of Jesus is quite problematic.
(1) Philip Jenkins cites early mystery religions and quotations from Clement of Alexandria and Origen to point out that ancient religious groups were sometimes hesitant to spell out the full extent of their beliefs to the general public in writing. Jenkins argues that it is, therefore, invalid to assume that those who produced Q or the Gospel of Thomas did not believe in the resurrection of Jesus, for example, simply because they did not mention it in their writings.[38]
(2) Arland Hultgren argues that while references to the cross and resurrection are lacking in the papers of Martin Luther King, no one would suggest that King did not know of the crucifixion and resurrection of Jesus. Kloppenborg counters saying this is due to genre and that the Q people apparently had other legitimating strategies which did not include direct appeals to Jesus’ resurrection.”[39] But this sidesteps the issue. The fact is that while King had ample opportunity to use references to Jesus’ passion and resurrection, he apparently fails to do so. Likewise, even assuming that virtually all of Q has been recovered, the fact that it contains no passion narrative would not necessarily mean that the “Q community” knew nothing of, or was indifferent to the passion story.
The point is that Kloppenborg’s arguments that Q did not contain a passion narrative have been shown to be invalid. However, even if Q did not have a passion narrative, it would be invalid to cite that omission as evidence that the Q communities did not know or believe in the passion of Jesus.
4. Kerygma? As seen above, Kloppenborg followed Bultmann in assuming that Q was “a transitional stage between the un-messianic preaching of Jesus and the fully self-conscious kerygma of the Hellenistic churches.” [40] Since the core of Q’s proclamation centers on the parousia and not on the passion, Kloppenborg concludes that the earliest churches were proclaiming fundamentally different messages about Jesus.[41]
a. It is important to note that Kloppenborg did not demonstrate that Bultmann was right about Q being a transitional stage. He simply assumed that Bultmann was right and built on that assumption. Several recent studies have shown on historical grounds, however, that there is every reason to believe that Jesus’ preaching was in fact messianic.[42] If Jesus’ preaching was messianic, the basis for postulating a transition from un-messianic preaching to kerygma is moot, and one of the main reasons for assuming redaction in the first place is weakened or eliminated.[43]
b. If Q was originally longer than what is preserved by Matthew and Luke, there would be no way of knowing whether it originally contained references to the kerygma or not. Nevertheless, assuming, for the sake of argument, that Matthew and Luke preserved virtually all of Q, the fact that an early Christian community might produce a document focusing on Jesus’ teachings and second coming in no way demonstrates that this community did not believe in the kerygma.
Suppose, for example, that the only letter of Paul that survived was his “harsh letter,” reconstructed from Second Corinthians 10:1 - 13:10. Scholars might conclude from that reconstruction that Paul did not believe in Jesus’ atoning death or in salvation by grace through faith because he never mentions it. Such an assumption would of course be incorrect. [44]
This analogy is not perfect, of course, but it illustrates the point that to assume that we can know with any degree of confidence, what a community did not believe based on what they did not include in a single short document, is an argument from silence, and, when arguing against Hultgren, Kloppenborg himself made the point that, “Arguments from silence are precarious…All that we have to go on is what the text of Q itself offers.” [45]


1. Analytic tools. After discussing “models for redactional analysis” proposed by Luhrmann and Jacobson, including attention to “grammatical shifts, breaks in the train of thought, shifts in audience, [and] shifts in tradition or theology,”[46] Kloppenborg proposes two “tools” by which to analyze Q for redactional activity: “The first analytic tool is the determination of the compositional principles which guide the juxtaposition of originally independent sayings and groups of sayings. Naturally this presupposes and builds upon the results from form-critical analysis.[47]
It is important to note that this “first analytic tool” is not a method for determining whether various groups of sayings have been edited together. Instead, this “tool” appears to assume from the outset that that the various units of sayings in Q were in fact juxtaposed by one or more editors. This, however, is precisely the thesis that Kloppenborg is proposing to demonstrate.
The second analytic tool is that, “…redactional or compositional activity may be seen in insertions and glosses…Such insertions afford us a tool by which to stratify successive redactions of Q.”[48]
Although form critics have imagined for decades that they can separate insertions or glosses from original documents, the fact that they often significantly disagree among themselves about what is original and what is the result of redaction, would seem to indicate that their conclusions are somewhat less than certain. While there will always be disagreement regarding the degree of certainty involved in form critical conclusions, the point is that Kloppenborg’s assertion that redactional activity can be seen in insertions, is hardly an objective “tool.” This will be seen more clearly in the examples discussed later in this paper.
2. The relation of prophetic/judgment to sapiential. What this second “tool” apparently means in practice is that if a prophetic or judgment saying appears in an otherwise sapiential section, it can be regarded as evidence that the saying was a secondary redaction. Q 6:20b-49, Q 10:2-16, Q 12:2-12, Q 13:24-30[49] are among the passages Kloppenborg regards as sapiential but which contain prophetic or judgment elements --and in each case the prophetic/judgment parts are seen to be redactional. But as Collins notes:
There is no generic incompatibility, however, between these speeches and an apocalyptic worldview. Accordingly the sharp redaction-critical separation of the sapiential speeches from the announcement of judgment should be viewed with some suspicion and will need to be evaluated critically.[50]

Kloppenborg responds sharply to Collins’ criticism saying:
This is a straw man argument since I did not argue that wisdom was incompatible
with apocalyptic or prophecy and expressly rejected ‘generic purity’
arguments. The argument for the stratification of Q does not rest on
presumptions about the (in)compatibility of wisdom and apocalyptic…it depends on
literary not theological factors. The question is not whether wisdom and
apocalyptic, or wisdom and prophecy, can subsist in the same document; of course
they can and do in various documents. The question is, when diverse
elements subsist in a document, how does one understand the literary and generic
relationship among the various elements? Collin’s misunderstanding is
tediously repeated by others: Horsley 1994:736; Witherington 1994:216; D.
Allison 1997:4-5.[51]

The fact, however, that so many scholars have “misunderstood” Kloppenborg should cause him to wonder whether he has mis-communicated. Nevertheless, Kloppenborg’s first analytic tool was to determine “the compositional principles which guide the juxtaposition of originally independent sayings and groups of sayings.”[52] It is hard to avoid the conclusion that this principle assumes from the outset that prophetic and sapiential sayings were in fact originally independent and apparently somehow incompatible in the first place.
Further, as discussed below, the fact that forms of “prophetic judgment sayings” [53] and the motif of “imminence of judgment” [54]are two of the three indications that redaction has occurred makes it hard to understand how Kloppenborg can deny that he has no assumption of fundamental incompatibility. If there is nothing unusual or incompatible about prophetic elements in sapiential passages, why should the presence of these forms and motifs to be considered indications of redaction in the first place?
Finally, as seen above, in Formation of Q Kloppenborg argued that Matthew and Luke would have likely treated Q as they did Mark.[55] In Excavating Q, however, Kloppenborg argued the exact opposite.[56] In Formation of Q Kloppenborg argued that the disappearance of Q was explicable only on the basis that it was almost entirely absorbed by Matthew and Luke,[57] but in Excavating Q he argued that the reason for Q’s disappearance is not known.[58]
Using the same kind of assumptions found in The Formation of Q, a case could be made that the same person would not write such apparently blatant contradictions, and that it is more likely that an editor inserted the contradictory passages in response to attacks made against Kloppenborg’s original thesis.
While absurd, this illustration serves to emphasize the point that if a meticulous scholar like Kloppenborg could write what appears to be such flatly contradictory statements, how much more likely is it that ancient authors, who did not have the benefit of word processors, occasionally wrote things that appear to have “grammatical shifts, breaks in the train of thought, shifts in audience,”[59] etc.? Such subtle breaks or shifts do not necessarily indicate redaction any more than Kloppenborg’s apparent contradictions.
Kloppenborg’s methodological assumptions and analytical tools appear, therefore, to be subjective at best. And while Kloppenborg strongly denies any fundamental incompatibility between sapiential and prophetic, his first analytic tool and his assumptions about common features (below) seems to presuppose incompatibility.
3. Common features. After detailed discussion of judgment complexes in Q, Kloppenborg concludes that these sayings “reveal several common features which invite the conclusion that these four blocks belong to the same redactional stratum.” [60] The first common feature is projected audience. While the actual audience is the Q community, the projected audience is those who oppose the Q people.[61] The second common feature is form. Most common are “prophetic judgment sayings and apocalyptic words” which are stated in the form of chreia.[62] The third common feature is motif. Common motifs include the “imminence of judgment,” the parousia, and lack of repentance.[63]
While this method might sound good in theory, numerous problems are evident. (1) The issue of projected audience is not as certain as Kloppenborg might seem to imply. For example, in Horsley’s opinion,
…only one of the five clusters of sayings, Luke 11:14-16, 29-32, 39-52, appears
to be directed at the “outgroup” of this gene,,a,” or impenitent opponents as
the offensive or implied audience. All or most of the sets of sayings in
the other four complexes are addressed directly to the Q people themselves. [64]

(2) The separation of sapiential and prophetic elements into neat hermetically sealed units is not quite as clean-cut as Kloppenborg’s arguments seem to imply. In an otherwise generally positive assessment of Kloppenborg, Arland Jacobson noted,
It is striking that the layers are not as neatly distinct as they are made to seem. For example, the figure of Wisdom appears in the second rather than the first (“sapiential”) layer; arguably prophetic sayings occur in the sapiential layer (e.g., Q 6:23-23b; 12:11-12); and chria occur in the first recension (Q 9:57-62-the “best” chriae in Q 12:13-14).[65]

(3) Although Kloppenborg treats the projected audience, forms and motifs as if they were separate features, in many cases one is simply the necessary corollary of the other. Any passage which is prophetic in form might well contain the motif of imminent judgment and be directed to outsiders. To treat these as three separate pieces of evidence for redaction seems questionable at best.


Having separated the judgment sayings from the rest of Q, Kloppenborg points out that a significant part of what remains consists of sapiential themes, notwithstanding several important interpolations that are “controlled by motifs related to the coming of the Son of Man and the judgment of the impenitent.” Horsley comments, “In order to purify the stratum it must be purged of prophetic sayings, which are dismissed as later insertions.”[66] A few selected examples will be discussed below to show how this is done in practice.
1. Q 6:36-49. According to Kloppenborg Q 6:36-49 clearly belongs to the sapiential stratum because of its “predominantly sapiential idiom,” the fact that it contains a mild rebuke rather than a severe rebuke, and that “there is no compelling reason to suppose that 6:39-42 is formulated with outsiders and opponents in mind….”[67]
Kloppenborg divides this passage up into three sections: 6:36-38, 6:39-45, and 6:46-49 which has the effect of lessening the impact of the theme of judgment in the passage. Taken as a whole, however, this passage begins and ends with the theme of judgment. It begins with instruction to “Judge not, and you will not be judged; condemn not and you will not be condemned…” (Q6:37) and ends with a strong warning about the great ruin of the house--which could certainly be read as judgmental if not apocalyptic. The impression that these judgment sayings form an inclusio is confirmed by the fact that the body of the passage speaks of (1) the condemnation of those who judge the speck in another’s eye while maintaining the log in their own eye, (2) the rebuke, “You hypocrite, first take the log out of your own eye…” (Q 6:42), and (3) the question, “Why do you call me “Lord, Lord’ and not do what I tell you?” Some have understood this as an address to the eschatological judge[68]--a view Kloppenborg even admits is true for the Gospel of Matthew but, he says, does not fits Q’s respectful address. The passage could easily be read as directed to the “hypocrites” outside the community who were judging the Q people without first taking the log out of their own eye--the great ruin of their “house” is predicted.
Therefore, while the passage may be “predominantly sapiential in form,” (first stratum) it is arguably controlled by the motif of judgment and could easily be read as being directed toward those outside the community (second stratum). These factors would at least seem to demonstrate the subjective nature of Kloppenborg’s thesis, if not undermining it altogether.
2. Q 9:57-62. According to Kloppenborg, Q 9:57-62 consists of three chreiae[69] the first of which (Q 9:57-58) speaks of Jesus as the Son of Man.[70] Son of Man sayings are normally part of the second, or judgment, stratum, but that is problematic here since, according to Kloppenborg, Q 9:57-62 is clearly sapiential in nature. Kloppenborg solves this problem by proposing two reasons to assume that Q 9:57-58 was not originally part of the sapiential stratum. (1) It has a “differing logical structure and, (2) it circulated as an independent saying in the Gospel of Thomas 86.
The differing logical structure has to do with the fact that whereas the second (Q9:58-60) and third (Q9:61-62) chreia proclaim discipleship to be more important than family obligations or Elijah’s calling,[71] the first chreia (Q 9:57-58) doesn’t fit because “is in fact a statement about the Son of Man.”[72] Q 9:57-58 reads: “And as they were going along the road, someone said to him, I will follow you wherever you go.’ And Jesus said to him, ‘Foxes have holes, and birds of the sky have nests; but the Son of man has nowhere to lay his head,” (ERV).
It is probable, however, that only those who are already convinced that the Son of Man passage does not belong here, will see a different logical structure. Q 9:57-58 could easily be read to indicate that discipleship is more important than the security of having a home, in which case it fits the logical structure of Q 9:57-62 perfectly. However, even assuming for the sake of argument, that there is a break in the logic of this text, it could be argued that since the supposed editor of Q apparently didn’t notice any problem with the logic of this passage, it is equally possible that an original author saw no problem with the logic either, in which case there is no reason to assume redaction.
Kloppenborg’s second reason for proposing that Q 5 9:57-58 was not originally part of the sapiential stratum is that it “appears to have circulated as an independent saying” as evidenced by a similar passage in the Gospel of Thomas 86[73] which reads, “Jesus said [The foxes have their holes] and the birds have their nests, but the son of man has no place to lay his head and rest” (Gos. Thom. 86, cf. Mark 8:20; Luke 9:58).
Kloppenborg assumes, but does not demonstrate, that a saying in the Gospel of Thomas was originally independent. If the writer of the Gospel of Thomas borrowed from the canonical Gospels this argument becomes invalid.[74]
Even assuming that Q 9:57-58 was an independent saying, however, there is no reason to suppose that its inclusion in any document necessarily implies that someone has edited that document. It would be just as easy for the original author to have included the saying in Q as it would for an editor to have added it to a subsequent revision.
3. Q 10:2-16. Although Kloppenborg classifies Q 10:2-16 as a sapiential passage, he notes that the references to harvest in Q10:2 “is distinctive since it applies to missionary activity a metaphor usually found in the context of apocalyptic judgment.”[75] Further, the passage includes a very harsh element of judgment:[76]
I tell you it shall be more tolerable on that day for Sodom than for that city. Woe to you Chorazin! Woe to you, Bethsaida! For it the mighty works done in you had been done in ‘Tyre and Sidon, they would have repented long ago, sitting in sackcloth and ashes. But it will be more bearable in the judgment for Tyre and Sidon than for you. And you, Capernaum, will you be exalted to heaven? You shall be brought down to Hades! (Q 10:12-15).
Kloppenborg attempts to resolve the problem of having elements of judgment in this sapiential section by arguing that (1) the motif of harvest in 10:2 does not appear again in the rest of the passage,[77] (2) In 10:2 the Lord of the harvest is the sender, whereas 10:3 and 10:16 “imply a chain in which Jesus is the proximate sender;” (3) “10:2 exists independently in Gos. Thom 73,” (4) the harvest metaphor is also used of the ingathering of Israel alone[78] and (5) “the command to pray for more missionaries is not directed to those sent out in v. 3, ‘but to Christians who might be imagined to be gathered for prayer prior to the commissioning as in Acts 13:1-3,”[79] (6) the woes of Q 10:13-15 “are clearly secondary interpolations appended because of the mention of the rejection of missionaries in 10:10-11.”[80] The justification for this conclusion is that 10:12-15 makes “an unfavorable contrast of Israel’s fate with that of the Gentiles”[81] and that they “differ markedly from 10:4-11 in form, tone, implied audience and tradition-historical provenance.”[82] This, says Kloppenborg, is characteristic of the judgment stratum as seen in Q 7:1-10 and 11:31-32.[83]
This is a lot to unpack: (1) While the word “harvest” does not re-appear, the context of the passage makes it clear that the whole mission referred to in this passage is about harvest of people for the kingdom of God.
(2) Rather than seeing some kind of disjunction between God as the sender in 10:2 and Jesus as the “proximate sender” in 10:3, this passage may be read to mean that Jesus, as a divine messenger or prophet, sends out harvesters on behalf of the Father. In this reading there is no disjunction and no evidence of redaction.
(3) As argued above, even assuming that the Gospel of Thomas preserved Q 10:2 as an independent tradition and did not just borrow it from the Gospels, which is a big assumption, the original author of Q could just as easily have incorporated that tradition into Q as a later editor.
(4) That the “command to pray for more missionaries is not directed to those sent out in v. 3”[84] is only true assuming that verse 2 was originally independent. The context of 10:2 and 3 together imply that the command in verse 3 is directed to those in verse 2.
(5) To say 10:13-15 was appended because of the mention of missionaries in 10:10-11 is pure speculation, not evidence.
(6) The fact that 10:4-11 differs from 10:13-15 in tone, form, implied audience, etc. is simply not indication of redaction. a. 10:11 speaks of the rejected missionaries as shaking the dust off their feet against those who reject their message. This is a serious condemnation and leads very naturally into the condemnations of 10:13-15. b. To say that the implied audience has changed in Q 10:12-15 is to imply that a speaker or writer could not possibly address a friendly audience but include warnings toward opponents outside the community. If presidential speeches were evaluated by these criteria, one might conclude that warnings to foreign enemies which occur in speeches directed to “my fellow Americans,” were later redactions.


If Kloppenborg’s thesis about the stratification of Q is correct, early Christian communities proclaimed fundamentally different views of Jesus and of salvation. The Q community in particular was unaware of, or indifferent to, the proclamation of salvation by grace through faith in the atoning death of Christ. Further, as Kloppenborg notes:
…if Q’s silence concerning a salvific interpretation of Jesus’ fate makes
it difficult or impossible to conclude that the historical Jesus considered his
own death vicarious (as Dunn would have it), one might still wish to claim the
notion of Jesus’ death ‘for us’ (1 Cor 15:3) as a key Christian theologoumenon,
but it would be difficult to affirm any rootedness of this doctrine in the
historical Jesus[85]
One of the many problems with Kloppenborg’s hypothesis, however, is that it builds one hypothesis upon another upon another.[86] (1) Kloppenborg assumes the existence of Q[87] and while, for advocates of the Two Document Hypothesis, this alone should not cast doubt on Kloppenborg’s thesis, the important point is that scholarly doubt is increasing being expressed over the existence of Q and adding other hypotheses on top of this one rapidly decreases the plausibility of the theory.
(2) Assuming the existence of Q, Kloppenborg hypothesizes that virtually all of Q can be known and reconstructed. The primary reasons for this hypothesis, however, were shown to be invalid using Kloppenborg’s own arguments. Without this assumption, Kloppenborg’s case seems to be substantially weakened.
(3) Kloppenborg makes the assumptions that a) Jesus’ preaching was un-messianic, b) Q was a transitional stage between that un-messianic preaching and the Pauline kerygma and c) the only way to account for this transition was to postulate that the Q kerygma was different than the Pauline kerygma. As documented above, however, some scholars have shown on historical grounds that there is good reason to believe that Jesus preaching was messianic. If Jesus’ preaching was messianic, there would be less reason to suspect that the prophetic/apocalyptic or coming Son of Man sayings were redactional.
(4) Kloppenborg appears to assume that independent sayings were juxtaposed in subsequent editions of Q, and that they can now be separated into strata on the basis of form, motif and projected audience. It was shown, however, that in practice these tend to be very subjective criteria, that they should not be viewed as if they were three separate pieces of evidence, and that if similar arguments were applied to Kloppenborg’s own books, some of his work might be dismissed as later redaction by someone other than the author.
(5) Kloppenborg assumes that we can know what the Q community (another hypothesis) did not believe about Jesus based on what they did not include in their document, i.e. passion story and kerygma.[88] As shown above, however, this was an invalid assumption.
It must be concluded, therefore, that Kloppenborg, an obviously brilliant and meticulous scholar, has not demonstrated the stratification of Q, much less the presence of a competing soteriology in the early church. Ehrman sums the matter up well when he writes:
Let me repeat: Q is a source that we don’t have. To reconstruct what we think was in it is hypothetical enough. But at least in doing so we have some hard evidence, since we do have traditions that are verbatim the same in Matthew and Luke (but not found in Mark), and we have to account for them in some way. But to go further and insist that we know what was not in the source, for example, a Passion narrative, what its multiple editions were like, and which of these multiple editions was the earliest, and so on, really goes far beyond what we can know…[89]

[1] This chapter was originally published in the Journal of the Evangelical Theological Society 46/2 (June 2003) 217-232.
[2] Most notably John Dominic Crossan and Burton Mack. Although Kloppenborg distances himself from the Cynic view of Jesus, his discussion of it is quite sympathetic at many points. John Kloppenborg. Excavating Q, (Minneapolis : Fortress Press, 2000) 184-194, 426-442.
[3] See Burton Mack. Who Wrote the New Testament; the Making of the Christian Myth, (San Francisco : HarperSanFrancisco, 1995). For a review of this book see Dennis Ingolfsland, “A Review of Who Wrote the New Testament?” Bibliotheca Sacra 153 (1997) 205-221.
[4] Ron Cameron, “The Sayings Gospel Q and the quest of the historical Jesus: a response to John S. Kloppenborg,” Harvard Theological Review. 89/4 (October 1996) 351-354.
[5] Leslie R. Keylock, “Review of Excavating Q in Journal of the Evangelical Theological Society,” 45/2 (June 2002) 356.
[6] John Kloppenborg, The Formation of Q, (Harrisburg, PA : Trinity Press International, 1987).
[7] John Kloppenborg, Excavating Q. (Minneapolis : Fortress Press, 2000).
[8] Kloppenborg, Formation, 21.
[9] Ibid. 21.
[10] Ibid. 21-22.
[11] Ibid. 22.
[12] Ibid. 87.
[13] i.e. theology that pictures Israel as persistently disobedient and in need of repentance to avoid coming wrath. Kloppenborg, Formation 93.
[14] Kloppenborg, Formation 93. Richard Horsley would agree with this assessment, but argues persuasively that the genre of Q is more logoi propheton rather than logoi sophon. This being the case, it is possible to see Q as a whole rather than postulating that Q originally consisted of wisdom sayings into which prophetic sayings were extensively interpolated. See Richard Horsley. Logoi Propheton? Reflections on the Genre of Q.” In Future of Early Christianity, (Minneapolis :Fortress Press, 1991) 195-209.
[15] Kloppenborg, Formation 96.
[16] Ibid. 96-99.
[17] The prophetic, judgmental and apocalyptic elements are combined in the second stratum and are often used almost interchangeably both in Kloppenborg and in this paper, although some, including Kloppenborg, have questioned the use of the designation “apocalyptic” with reference to Q. See John S. Kloppenborg, “Symbolic Eschatology and the Apocalypticism of Q,” Harvard Theological Review, 80/3 (1987) 287-306.
[18] Kloppenborg, Excavating Q, 11, 54.
[19] See especially Mark Goodacre, The Case Against Q, (Harrisburg, PA : Trinity Press International, 2002). For Kloppenborg’s interaction with Michael Goulder on Q see Michael Goulder “Self-Contratiction in the IQP, Journal of Biblical Literature, 118/3 (Fall 1999) 506-517. Robert A. Derrenbacker and John S. Kloppenborg, “Self-Contradiction in the IQP?: A Reply to Michael Goulder,” Journal of Biblical Literature, 120/1 (2001) 57-76. Michael Goulder. “The Derrenbacker-Kloppenborg Defense,” Journal of Biblical Literature, 121/2 (2002) 331-336.
[20] Hodgson comments, “In a book with such an extensive history of Q research we miss some customary courtesies. For example, Kloppenborg might have given the work of Wm. Farmer and his students its due. They have revived the Griesbach hypothesis and, in essence, done away with Q. Farmer’s studies are an important line of research, are being carried forward in a new Society of Biblical Literature consultation, and, no matter how violently one disagrees with them, deserved attention.” Robert Hodgson, Jr., Review of “John S. Kloppenborg, The Formation of Q” in Biblica 70/2 (1989) 284. Ellis agrees: “Is it wise to pursue such a thesis with no recognition and, apparently no awareness of the currently widespread questioning and rejection of the Q hypothesis?” E. Earle Ellis, Review of The Formation of Q in the Southwestern Journal of Theology, 31/3 (Summer 1989) 66.
[21] Significantly adding to the instability of this foundation is the work of Mark Goodacre. The Case Against Q, (Harrisburg PA : Trinity Press International, 2002). Kloppenborg notes, “If his argument should be sustained, Q would become unnecessary and decades of Gospel research will have to be re-thought…The Case Against Q provides the most accessible and compelling defense to date on the theory of Gospel origins championed by James Ropes, Austin Farrer, and Michael Goulder” (Back cover of The Case against Q).
[22] Kloppenborg, Formation 80-88. Kloppenborg, Excavating Q (Minneapolis : Fortress Press, 2000) 91-104.
[23] Kloppenborg, Formation 42, cf. also 80. For an excellent discussion of Q as narrative, see Mark Goodacre, The Case Against Q, (Harrisburg, PA : Trinity Press International, 2002) 172-176.
[24] Kloppenborg, Formation 81-82.
[25] Kloppenborg, Excavating Q, (Minneapolis : Fortress Press, 2000) 61.
[26] Ibid. 61.
[27] Ibid. 89.
[28] Kloppenborg, Formation 81.
[29] Kloppenborg, Excavating Q 366-367.
[30] Kloppenborg, Formation 81.
[31] Observing that Luke elsewhere preferred Q to Mark, W.E. Bundy suggested that the passion accounts of Luke 22:25-27, 28-30 and 35-46 were part of Q. Kloppenborg, Formation 85.
[32] Burkitt proposed that Luke 22:24-30 “was a conflation of Mark 10:41-45 and Q”. Kloppenborg dismissed these passages as not directly dealing with the passion. Kloppenborg, Formation, 85.
[33] Kloppenborg cites several verses in the passion narrative which Hirsh believed to be part of Q, i.e. Luke 22:48, 62, 64, 69 and 24:47. While admitting that Luke 22:48 has been shown to be pre-Lucan, Kloppenborg observes that this doesn’t prove it came from Q. Kloppenborg noted that Luke 22:62 is “beset with textual difficulties” (some texts omit the verse) but “the statement that ‘Peter went out and wept bitterly’ is not a solid basis for a Q passion account” Kloppenborg, Formation 86-87. Luke 22.64 is also dismissed due to textual uncertainties--although one will look in vain at the UBS test for textual variants. The uncertainties consist in the fact that since Matthew 14:65 omits the phrase “and to cover his face and to strike him which is contained in Mark, and since the covering of Jesus’ face is necessary to make sense of Matthew’s question, “Who is it that struck you”, according to Kloppenborg, this suggests that both the face covering and the question were originally lacking in Mark and, therefore, also in Matthew. According to Kloppenborg, this means that “Luke 22.64 is due to redaction, not the influence of Q.” Kloppenborg, Formation 87, n. 1.
[34] Kloppenborg cites Taylor and Schneider who contend that Luke 22:69-70 come from another tradition. Kloppenborg responds saying this “is not evidence that this tradition was Q.” Kloppenborg, Formation 86.
[35] Kloppenborg, Formation 87. Elsewhere Kloppenborg speaks of “the liberation of Q from the servitude to the passion Kerygma…” Kloppenborg, Excavating Q 350.
[36] Kloppenborg uses this method several times in The Formation of Q, i.e., he analyzes opponents arguments in detail, dismisses them as less than convincing, and assumes that he has therefore proven his own position.
[37] It should be noted, however, that Q mentions the sign of Jonah. In the context of Matthew 12:40 this is explicitly tied to Jesus’ resurrection whereas in Luke 11:30 this is not explicitly stated. Kloppenborg assumes that Luke’s was closer to the original but for all we know, Luke’s audience may have been familiar with the tie between the story of Jonah and the resurrection so Luke felt no need to explain it. It is quite frankly, impossible to know whether this connection between Jonah and the resurrection was part of Q or not. This being the case, it is impossible to assert that Q did not have reference to the resurrection.
[38] Philip Jenkins. Hidden Gospels; How the Search for Jesus Lost its Way (New York : Oxford University Press, 2001), 73-78.

[39] Kloppenborg, Excavating, 375-376.
[40] Kloppenborg, Formation 21.
[41] Ibid. 22.
[42] See for example, Ben Witherington. The Christology of Jesus (Minneapolis : Fortress Press, 1990). N.T. Wright. Jesus and the Victory of God (Minneapolis : Fortress Press, 1996). Dennis Ingolfsland. “The Historical Jesus According to John Dominic Crossan’s First Strata Sources: A Critical Comment.” Journal of the Evangelical Theological Society 45/3 (September 2002) 405-414.
[43] As a whole, Q presents Jesus as the apocalyptic judge and coming Son of Man. If Jesus’ teaching was un-Messianic, it is easy to see why these elements would be viewed later editions.
[44] To carry the analogy of the harsh letter even further: Supposing that this was the only one of Paul’s letters that survived, one can imagine the critics arguing that the “gospel” this “harsh” letter proclaims is just the good news of the coming Jewish kingdom. We can be sure that it has nothing to do with Jesus’ atoning death since the idea of Jesus suffering for sins never occurs once in the harsh letter! Since Paul speaks of his own suffering and even hints at his own death and exaltation (being “caught up to the third heaven”), it would have been perfectly natural for Paul to have made a comparison of his own experiences with those of Jesus--if Paul was aware of Jesus resurrection and ascension. In fact the only hint of the resurrection at all is in 13:4 which says that “he was crucified in weakness, but lives by the power of God.” But this doesn’t actually say Jesus rose from the dead--just that he survived crucifixion! Maybe the later gospel writers misunderstood and turned this into a resurrection. Or since, this is the only hint of resurrection theology in the entire “harsh” letter, maybe it was added by a later editor--similar to the way the writer of Matthew made the Jonah saying of Q into a sign of the resurrection of Jesus. It might be concluded, therefore, that Paul knew nothing of the kerygma. For Paul, Jesus was a martyred man of God to whom one should be devoted as an example--certainly not the resurrected Son of God who dies for the sins of the world. If the “harsh” letter was the only evidence we had from the “Paul community” this conclusion might seem possible to some minds. But since we have other letters of Paul, we know that this hypothetical critical assessment of the harsh letter could not be farther from the truth. Such is the danger of taking one document--especially a hypothetical and possibly incomplete document--and drawing conclusions about what the people who produced this document could not have believed based on apparent omissions.
[45] Kloppenborg, Excavating Q 371.
[46] Kloppenborg, Formation 97.
[47] Ibid. 98.
[48] Ibid. 99.
[49] Ibid. 171.
[50] John J. Collins. “Wisdom, Apocalypticism, and Generic Compatibility” in In Search of Wisdom: Essays in Memory of John G. Gammie, ed. By Leo Perdue et al. (Westminster/John Knox Press) 185). For a similar assessment see also M. Eugene Boring. Review of The Formation of Q in the Journal of the American Academy of Religion, LVIII/2 (Summer 1990) 294.
[51] Kloppenborg, Excavating Q 145-146. See also 150-151 where the point is re-emphasized even more forcefully.
[52] Kloppenborg, Formation 98.
[53] Ibid. 168-169.
54] Ibid. 169.
[55] Ibid. 81-82.
[56] Kloppenborg, Excavating Q 89.
[57] Kloppenborg, Formation 81.
[58] Kloppenborg, Excavating Q 366-367.
[59] Kloppenborg, Formation. 97. Kloppenborg is favorably quoting Arland Jacobson, see Formation 98.
[60] Q 3:7-9, 16-17; 7:1-10, 18-35; 11:14-26, 29-33, 39-52; 12:39-59 and 17:23-35. Kloppenborg, Formation 166.
[61] Kloppenborg, Formation 167.
[62] Ibid. 168-169. Pronounced KRAY-ah. A “chreia “ is “a brief anecdote focused on reporting a memorable saying or action associated with a prominent person. The Greek word referred to "what is useful (or needed)," especially in confrontations with others (war, business, debate). In Greek rhetoric it designated a narrative statement or story that could be cited to make a point.” See for this definition and further information.
[63] Ibid. 169.
[64] Richard Horsely. “Logoi Propheton? Reflections on the Genre of Q.” In Future of Early Christianity, (Minneapolis :Fortress Press, 1991) 195-209. See also Richard Horsley, Sociology and the Jesus Movement, (New York : Continuum, 1994) 108-111. For Kloppenborg’s response to Horsley see John S. Kloppenborg. “The Formation of Q Revisited: A Response to Richard Horsley,” Society of Biblical Literature 1989 Seminar Papers ed. By David J. Lull, (Society of Biblical Literature, 1989) 204-215.
[65] Arland D. Jacobson. Review of The Formation of Q: Trajectories in Ancient Wisdom Collections” in the Journal of Biblical Literature, 108/1 (1989) 152. Tuckett’s observations may also be of interest here. Arguing against the theories of Vaage and Mack “that Q1 is to be seen as Cynic”, Tuckett notes “But according to Kloppenborg’s analysis, the chreiai in Q characterize ‘Q2, rather than ‘Q1’. Indeed they serve to distinguish the alleged layers. It is thus rather odd, to say the least to have a generic similarity between the cynic material and Q2, whilst asserting that the primary link between Q and cynic traditions is at the level of Q1. This problem can be alleviated in part by assigning some chreiai to Q1 rather than to Q2. However, this distinction in forms used in the two alleged layers was a part of the argument for distinguishing the layers in the first place; hence the transfer of the chreiai to Q1 undermines a further support in the argument for the existence of the strata themselves.” Christopher Tuckett, Q and the History of Early Christianity, (Peabody, MA : Hendrickson, 1996) 382.
[66] Richard Horsley. “Logoi Propheton? Reflections on the Genre of Q”, In The Future of Early Christianity. (Minneapolis : Fortress Press, 1991) 198.
[67] Kloppenborg, Formation 185-189.
[68] Ibid. 185-186, 188.
[69] Pronounced KRAY-eye.
[70] Kloppenborg, Formation. 190.
[71] Kloppenborg sees Q 9:61-62 as an allusion to the calling of Elijah in 1 Kings 19:19-21.
[72] Kloppenborg, Formation 191.
[73] Ibid. 191.
[74] For a brief argument against Thomas’ independence from the Gospels, see Mark Goodacre, The Case Against Q, (Harrisburg, PA : Trinity Press International, 2002) 147-150. For a more thorough case, Goodacre refers his readers to Christopher Tuckett, “Thomas and the Synoptics,” Novum Testamentum, (1998) 132-157.
[75] Kloppenborg, Formation 93.
[76] Ibid. 195.
[77] Ibid. 193.
[78] Ibid. 193.
[79] Ibid. 193.
[80] Ibid. 195.
[81] Ibid. 196.
[82] Ibid. 199.
[83] Ibid. 196-199.
[84] Ibid. 193.
[85] Kloppenborg, Excavating Q 362.
[86] Kloppenborg acknowledges this criticism, but his response to it, appears to my mind, to have completely sidestepped the issue. Kloppenborg, Excavating Q. 111.
[87] For some scholars, the existence of Q seems to have moved beyond the realm of hypothesis to being a dogma to be defended even at the cost of reason. Goodacre quotes Christopher Tuckett and Frans Neirynck who argued, in effect, that even if it was conceded that the minor agreements of Matthew and Luke against Mark did indeed demonstrate that Luke knew and occasionally used Matthew that would not prove that Q didn’t exist. But as Goodacre points out, “…the Q hypothesis is predicated on the assumption that Luke did not know Matthew…” Mark Goodacre, The Case Against Q, (Harrisburg, PA : Trinity Press International, 2002) 165-168.
[88] As Wilson notes, Kloppenborg’s “argument involves hypothesis upon hypothesis upon hypothesis, a house of cards which can easily tumble” S.G. Wilson, Review of The Formation of Q” in the University of Toronto Quarterly, 58/1 (Fall 1988) 227-228.
[89] Bart Ehrman, Jesus, Apocalyptic Prophet of the New Millennium. (New York : Oxford University Press) 133.