Ed. Note: The following is a collection of my contributions to various debates regarding Q, on Crosstalk (the pioneer email conference hosted by Harper San Francisco), its successor Crosstalk 2 (moderated by Jeffrey Gibson) & the Westar Egroup.  These have been conservatively cleaned up for publication while preserving the original ad hoc character of the discussion. Besides orthographical & grammatical corrections, the current edition has some added typographical polish and editorial clarification.  But other features characteristic of email have been retained -- e.g. common abbreviations or technical scholarly shorthand & Latinate transcription of Greek or Semitic characters. An alphabetical table of these items is provided in the appendix to this page.

The issues in dispute in the posts below are:

1.   Whether there is a solution to the synoptic problem that is simpler than Q.
2.   Whether Luke is independent of Matt.
3.   Whether Farrer's hypothesis that Luke used Matt is easily dismissed.
4.   Whether Q is easily dismissed if one assumes Luke used Matt.
5.   Whether Luke edited Mark & Matt in a similar way.
6.   Whether Luke's version of a Q saying is more primitive than Matt's.
7.   Whether Luke's Q sayings use wording typical of Matt.
8.   Whether Mark used Q.
9.   Whether GThom used Q. [see also letters 8 & 9 under Testing Thomas]
10. Whether Q sayings were formed like Cynic chreiai.
11. Whether Q tends to make sayings about God's kingdom refer to future events. 
12. Whether Q contained apocalyptic predictions.

I am indebted to the following partners in dialog for having provided the intellectual stimuli that prompted me to think these issues through:

Stevan L. Davies, Prof. of Religious Studies, College of Misericordia. Website: Thomas Homepage.
Gail Dawson.
Mark S. Goodacre, Department of Theology, U of Birmingham, UK. Website: NT Gateway.
Corey W. Liknes, instructor, Prairie Bible College, Alberta CA.
Dean Pielstick, business program coordinator for Northern Arizona U.

I have not space here to represent their arguments in full..  Messages dated after May 1999 are still found in the archives of Crosstalk 2 & Westar.

----

Mahlon H. Smith
Professor Emeritus
Department of Religion
Rutgers University
New Brunswick NJ

 

Date: Wed, 22 Dec 1999 01:32
From: Mahlon H. Smith
To: Westar Egroups
Subject: Q

Dean Pielstick wrote:

All of the sources that I have read about the Q hypothesis refer to the independence of Matthew and Luke. This seems to be critical to the existence of a real Q. However, these sources seem to gloss over the background regarding this independence. To me Mt/Lk don't read as being independent at all, and scholars seem to date them 5-10 years apart (most dating Matthew first). This seems to make it highly probable that the author of Luke would have had ample opportunity to have seen Matthew and Mark before or during his writing. Can anyone point me to better sources attesting to the independence of Matthew and Luke?

---

Dean et al:

This is good place to break my month long silence, since -- after years of questioning evidence of a written sayings source common to Matt & Luke other than Mark -- I have, over the past decade or so, emerged as a spokesman for (if not necessarily expert on) that hypothetical sayings gospel that we scholars call Q.

First, Dean & Brian you are not alone in challenging the 2 source hypothesis as a solution to the synoptic problem. There are many respectable scholars who ascribe to some alternative, primarily theories proposed by:

1. Augustine of Hippo (ca. 400 CE): Mark condensed Matt, Luke drew on both;
2. J. J. Griesbach (ca. 1790): Luke revised Matt, Mark produced a condensed version of both (à la Readers Digest);
3. A.M. Farrer (1955): Matt expanded Mark, Luke drew on both.

The last 2 are have the most adherents among current scholars who dissent from the majority that support the 2 source hypothesis. Wm. Farmer revived the Griesbach hypothesis in 1964. Current followers prefer to call this the 2 Gospel hypothesis. This is preferred by conservative scholars in the U.S. who put their historical faith in the 2nd c. patristic tradition that Matt was written first. But outside the U.S. the Farrer thesis has the most support among critics of the 2SH. British scholars Michael Goulder & Mark Goodacre are the primary current champions of this solution to the synoptic problem.

In my early days I was a Farrerite as I believe Daryl Schmidt was. And Mark Goodacre claims E. P. Sanders was (is?). Sanders himself shies from identifying himself with any synoptic theory, however, because as he pointed out in his book on the Synoptic Problem: none of the "solutions" to the synoptic problem that have been proposed thus far is itself without problems.

The question is which hypothesis causes the fewest problems in explaining the relationship of the synoptic texts. Which theory gives the most plausible explanation of the linguistic data that we find in the texts themselves.

There are 3 observations that continue to convince most synoptic specialists -- myself now included -- that the 2SH presents the fewest problems for interpreting the literary relationship of Matt, Mark & Luke:

1. Mark is the roughest of the gospels from a grammatical & rhetorical standpoint, Matt & Luke are both more polished & present fewer problems. This points to the priority of Mark.

2. Luke almost never uses any of Matt's grammatical & substantive improvements of Mark (other than omissions). This points to the independence of Matt & Luke in making revised expanded editions of Mark.

3. Where Matt & Luke share non-Markan material, Luke almost never puts it in Matt's narrative context. The only 2 exceptions to this are the prophetic warnings of JB & the temptation dialogue between J & the devil. But in both cases Mark has a shorter account, which would have led any editor to insert additional material at these points. But where Mark does not have any parallel (e.g., J's sermon) or even if he does (e.g., mission instructions, parable of the mustard, etc.) Luke does not follow Matt's narrative sequence in inserting non-Markan material into his gospel.

What this means from the standpoint of editorial theory is that if Luke used Matt he deconstructed Matthew's gospel, disassembling Matt's long well-organized speeches & recalling bits & pieces of them in random locations throughout his gospel. He would have acted as literary equivalent to the Sower, scattering to the wind the seed that Matt had gathered into neat stacks in his barn. For what purpose is unclear, since Luke's use of this non-Markan sayings material that he shares with Matt is less thematically organized than Matt's.

To appreciate the incongruity of such editorial activity on Luke's part, one has to work through a detailed comparison of Matt & Luke's texts. There have been 2 recent books that have tried to explain in detail how & why Luke edited Matt. From the Farrerite perspective there is M. Goulder's massive commentary Luke: A New Paradigm & from the Griesbachian perspective there is Beyond the Q Impasse by Farmer's group. Both, however, fail to present a rationale for the elaborate editorial gymnastics that they claim Luke went through (in the pre-computer age!) to rearrange non-Markan material he extracted from Matt that sounds convincing to most current synoptic scholars. Moreover, neither adequately explains why Luke would have dismantled a polished coherent work like Matt & followed the rough & problematic text of Mark instead.

To get a sampling of the editorial problems raised by the chief rivals of the 2SH I invite you to look at my on-line Synoptic Gospels Primer. 

To avoid such inexplicable problems other scholars have proposed more complicated source hypotheses that envision Luke as having access to other non-extant sources (proto-Luke or other sayings sources). Some recent advocates of the Griesbachian & Farrerite theories even identify this sayings source as Q (although the contents they propose for it differ from the non-Markan parallels in Matt & Luke that 2SH scholars have long called Q). For an easily accessible sampler of these variations see Stephen Carlson's Synoptic Problem Home Page. 

Therefore, for whatever problems & questions it may raise scholars are not yet ready to dispense with Q. If anything, Q is now more certain than it was a century ago.

Shalom!

Mahlon

Date: Sat, 11 Sep 1999 13:41
From: Mahlon H. Smith
To: Westar Egroup
Subject: Q

Gail Dawson wrote:

For a well-presented web page on the canonical status of Q (which is a revised edition of a paper presented at the Spring 1997 meeting of the Westar Institute by Mahlon Smith)    <SNIP>
It is a difficult issue, even to be able to read enough to begin to understand the complexity of the problem.

---

Thanks for the plug, Gail. I posted my Westar paper on the internet simply because of the volume of cyber-material & debate that I found that was critical of Q without facing the problems with the alternatives. It is gratifying when someone cites my work as a useful source.

It is relatively easy to be skeptical of a hypothetical source. But once one begins to weigh alternative explanations of the patterns of verbal agreement/disagreement among the synoptic gospels, the Q hypothesis begins to commend itself for its relative simplicity. Of course there are many places where appeal to Q does not explain all the features of non-Markan material common to Matt & Luke. But we need to remind ourselves that in reading any gospel we are working with a reconstructed text. So judgments about the content of sources always rest on which reconstruction is more historically plausible.

The thing that continues to persuade most synoptic scholars to prefer working with a non-extant Q is simply that other theories, such as the Griesbach-Farmer hypothesis (that Mark edited Matt & Luke) or the Farrer-Goulder hypothesis (that Luke used both Mark & Matt) prove to be far more complicated when it comes to explaining the differences between the synoptic texts. In theory, the alternatives to Q may sound simpler. But the difficulties become evident when they are applied to explain the texts themselves. On the basis of either major alternative to Q, Luke -- the most rational & systematic of the gospel writers -- would have adopted capricious, extremely complex editorial techniques. So relatively few synoptic scholars regard this as a more logical & preferable solution to the synoptic problem than wrestling with the vagaries of Q.

Shalom!

Mahlon

Mon, Dec 15 1997 11:30 
From: Mahlon H. Smith
To: Stevan L. Davies
CC: Crosstalk; Mark S. Goodacre
Subject: Farrer and GThomas 


Greetings Steve, Mark G. et al:

Can a former Farrer fan be admitted to the fray? Former, that is, in the sense of supporter of his position (which I was as a Drew seminarian in the early 60's). I still have the greatest admiration for his logical clarity & rhetorical skills, even though my own analysis of the texts has led me now to champion an opposing camp.

Stevan Davies wrote:

I have read as much of Austin Farrer as I can bear. You can too, via  Mark G.'s website.

The richness of Farrer's language and the ease with which he writes are not untypical of scholarship of yore, and a curse and damnation placed upon the scholarly writing of today. He's fun to read just for the prose.

---

Thanks, Steve, for such a quick response to Mark's posting of an article that every biblical scholar should read & now can, thanks to Mark's labors.

I was always puzzled why Farrer's argument made so little impact on North American synoptic scholarship. I used to think that it was the continental prejudice of the SBL. But I'm now convinced that Farrer's publishers & ultimately Farrer himself were to blame. If you want to bury an important thesis --- or at least impede its circulation in the market place --- publish it in a Festschrift or collection edited by someone else, where it will be weighted down by a dozen or so ho-hum theses until it sinks into academic oblivion. Farrer's thesis merited more detailed critical analysis than it initially got over here. Fortunately, Mark & others Brits have not let it die. Now that Mark has resurrected it & guided its ascent into cyberspace, no gospel scholar who owns a computer & a telephone line has any excuse for unfamiliarity with it. Pity that research advances in the field & the American liberation of gospel studies from the context of ecclesiastical theology now make so much of it seem dated.

Your opening line puzzles me a bit, Steve. Does your expression of de trop in reading Farrer mean "I can't bear to read any more of this!" or "There's too much there to digest & rebut in one cyber-note"? I would hope it's the latter. Farrer's prose is to be admired, to be sure. But even more admirable & worthy of imitation is his rigorous clarity of logic & gentlemanly grace in intellectual sparring. As a charter member of the oft-maligned but seldom fairly or accurately reviewed Jesus Seminar, I have often wished that more American gospel scholars would prove themselves worthy to carry Farrer's scholarly sandals (if not his theses).

The reason why Farrer's arguments cannot be dismissed as lightly as proponents of other theses might like is that Farrer built his thesis on empirical evidence & logic rather than traditional opinions, either ecclesiastical or academic. Long before Farmer, Farrer emerged as a Socrates in an arena of academic dogmas & challenged critical scholars to prove that there was any substance to the unseen Source that all professed but none (up to that time) dared reveal. Unlike the Griesbachians, Farrer is not easily dismissed as a traditional conservative, since his theory of gospel authorship allows for a substantial amount of synoptic material to be complete theological fabrications by Matthew or Luke. If two decades or more after Farrer's death, proponents of Q (& Thomas) finally produced tangible evidence to support a century & a half old hypothesis of a synoptic sayings source, Farrer himself cannot be faulted for not recognizing the existence of such.

Indeed, the real value of Farrer for the 21st century is in keeping supporters of Q honest. After more than two centuries it is relatively easy for critical scholars to challenge the hypothesis of Matthean primacy. But the fact that it is easy to convince even undergraduates of this, is apt to promote smugness & a false sense of security among us. The 2SH may still be the majority position among synoptic specialists, but the silent majority of non-specialists out there remains to be convinced. Markan priority does not prove Q. The only way to establish the probability of a documentary sayings source is to disarm Farrer's argument. You began the task, but have hardly mortally wounded it.

So, please forgive me for asking you to digest more Farrer than you may now care to stomach. Despite my own championship of Q in cyberspace, I'm going to act as devil's advocate to provoke you into refining your defense. As I've discovered in developing my Synoptic Primer, Farrer remains an agile & worthy opponent, whose position is not so easily undermined as your comments assume. If Mark G. jumps into the fray, I reserve the option to switch sides & defend a fellow Q-Thomas musketeer. D'accord?

---

Farrer wrote:

"For the hypothesis wholly depends on the incredibility of St. Luke's having read St. Matthew's book. That incredibility depends in turn on the supposition that St. Luke was essentially an adapter and compiler. We do not now, or ought not now, so to regard him. And being once rid of such a supposition, we can conceive well enough how St. Luke could have both read St. Matthew's book as it stands, and written the gospel he has left us. Then at one stroke the question is erased to which the Q hypothesis supplied an answer. For the hypothesis answered the question, "From what does the common non-Marcan material of Matthew and Luke derive, since neither had read the other?"

Davies rebuts:

 There's a pretty easy answer to this question. If we assume that Luke had read Matthew (and I do) then Luke can do one of two things. He can take Matthew and revise it to fit his own notions. Or he can take Matthew's sources and revise them to fit his own notions. Farrer concedes that Luke did that. He took Matthew's source Mark and revised it. One can therefore hypothesize that he also took Matthew's source Q and revised that. As Luke did not depend on Matthew for his Mark, he did not depend on Matthew for his Q. This possibility undermines completely Farrer's strong assertion that if Luke knew Matthew therefore the Q hypothesis necessarily fails.

---

Mahlon referees:

Touché! But no kill. You have merely parried Farrer's thrust by which he sought to disarm supporters of Q. You astutely use Farrer's own argument to show that it is "possible to hypothesize" that Luke knew Matthew & his source. But you are too over-confident in concluding that a hypothetical possibility "completely undermines" Farrer's position. For we have a text of Mark on which Luke can be shown to be literarily dependent. But we do not have such a text of Q (Documenta Q notwithstanding). Much of reconstructed Q is simply Lukan sayings paralleled in Matthew. To say that Luke knew this is tautologous. I think Mark & Goulder & other Farrerphiles would retort: What is to prove that this is a "source" & not just Luke's editing of Matthew? The force of Farrer's argument is that if Luke knew Matthew then it is not necessary to posit Q. Mere possibility does not have the weight of necessity. The reason for the ascendancy of the Q hypothesis in the skeptical world of biblical scholarship is not that it was thought possible but that it was thought necessary. A necessary object is one that is either empirically or logically indispensable. If Farrer can adequately explain the non-Markan Matthean parallels in Luke as Lukan redaction of Matthew, then Q is both logically & empirically dispensable. For the footing on which the reconstruction of Q as an independent source rests (the hypothesis that Luke sometimes preserves a more primitive form of the tradition than Matthew) would be an illusion. The simpler is not always the earlier. The possibility of a hypothetical source like Q becomes the preferred option only when details of the text cannot be adequately accounted for on the basis of empirical sources. So, the only way to "completely undermine Farrer's strong assertion" is to demonstrate that one cannot explain Lukan parallels to Matthew simply as the product of the editorial activities of Matthew or Luke. But this can only be done in detailed exegesis by scholars who are willing to test Farrer's hypothesis. Beware in granting Farrer that Luke had direct "knowledge" of Matthew, for this gives him an empirical razor that his heirs (like Mark G.) are prepared to wield like Ockham.

---

Farrer wrote:

"It is notorious that Q cannot be convincingly reconstructed. No one reconstruction, to say the least of it, is overwhelmingly evident, and no proposed reconstruction is very firmly patterned."

Davies counters:

This is not the case. The International Q Project has done it.  Kloppenborg has done it. There is a satisfyingly complete little book. Now, one doesn't have to like that little book or admit that it is satisfying, but the absence of same cannot any longer be an argument against Q.

---

Mahlon intervenes:

Brave try, but only a scratch. Despite my own great admiration for the scholarship of Jim Robinson, John Kloppenborg & others involved in the Q project I have to honestly admit that these have not yet become "overwhelmingly evident" to the average reader of the gospels. You & I & other 2 Sourcers might be convinced that these are valuable tools. But anyone who has sat in on the SBL Q section should be well aware that we are still far from the unanimity among Q scholars that Farrer is demanding. If you poll all gospel scholars in the SBL I'd dare say that the degree of consensus would probably be lower & if you took your poll into the pulpits & pews...well... You see what I mean?

A stronger parry of Farrer's thrust is to deny the premise. Where in the reconstruction of any biblical text has there ever been "one reconstruction" that is universally or at least "overwhelmingly" accepted by experts? What about the text critical history of the ending of Mark? Or John 8? Or the wording of any passage with textual variants? If there were overwhelming unanimity in these there would be no need for a critical edition of the biblical text or a textual apparatus. But who would ever think to suggest that the lack of unanimity regarding the details of a gospel text is a valid argument against the existence of such a text?

---

Back to Farrer:

"We have no reason to suppose documents of the Q type to have been plentiful... . No, in postulating Q we are postulating the unique, and that is to commit a prima facie offence against the principle of economy in explanation."

Davies thrusts:

Oops. Gospel of Thomas kills this argument dead.

---

Mahlon comments:

Nice touch. But watch out, he's still breathing! Thomas is evidence that sayings collections were used in early Christianity. But how early? Scholars like you, Steve, Marvin, Helmut & Dom have done an impressive job in getting skeptical non-Coptic specialists like me to recognize much of Thomas as a pre-synoptic source. Three decades ago, I would never have thought that I would ever accept Thomas as historical evidence of first-century Christianity. And there are many still today, like Michael Goulder, who simply dismiss Thomas out of hand. We've come far. Your website provides the single greatest boost in public knowledge & respect for the historical value of Thomas. But can we honestly say today that we represent a majority consensus among biblical scholars?

---

Farrer continued:

" ... I have yet to be convinced that there were such Christians, or that their existence in the first days was a psychological possibility."

Davies gloats:

Kills this one dead too. And if Thomas doesn't, Crossan, Mack, Borg, etc. have done so. Non-passion Christianity isn't a  non-starter.

---

No argument here. Farrer exposed himself to your thrust by this rare lapse into subjective opinion. In historical knowledge, doubts of a non-witness are not admissible as evidence of non-existence.

Hope I haven't given you more to read than you can bear. Its been good to resume our dialog after so many years.

Mahlon

Date: Fri, 28 Aug 1998 22:58
From: Mahlon H. Smith
To: Stevan Davies
CC: Crosstalk
Subject: Q Nomenclature

Stevan Davies commented on a post by Mark Goodacre:

I presume Mark G. means Luke used Matthew (as opposed to Luke knew of Matthew's existence). In that case I'd agree that it does not seem reasonable that Luke should be thought also to have used a hypothetical source for material available through Matthew. That's not impossible, but pretty unlikely and, if the proposition that Luke used Matthew had been accepted in scholarship all along, the hypothesis of Q possibly never would have arisen.

---

Possibly; even probably, if one means by Q not just Matthew's non-Markan sayings source but the hypothesis that the scope of that source was NMM // L.

Still I have to disagree that one can make Q vanish simply by producing a few minor agreements that may be interpreted as evidence that Luke knew & used the text of Matt as well as Mark. This would be an inevitable conclusion only if it could be shown that non-Markan material in Matt were inventions of Matt's own imagination & were known only by those who had read or heard quotations of Matt. M. Goulder was well aware of this & I think Farrer was too, when he insisted that the evangelists have to be regarded as creative theologians. For if Matt did not create his non-Markan sayings, they must have circulated prior to and independently of the version recorded in Matt's text. If Matt had access to these sayings from prior tradition so too did others, including possibly Luke. This possibility is magnified to probability, given the extent to which the Lukan version of non-Markan Matthean sayings vary from the text of Matt in both wording & narrative setting.
There are three plausible explanations of this phenomenon:

1. Luke has freely altered the text of Matt, because he deliberately rejects the Matthean Gestalt, but reshapes Matt's sayings to fit a vision of Jesus that is uniquely his own creation. OR

2. Luke reshapes Matt because he has good evidence that Matt had himself reshaped earlier tradition to fit Matt's unique image of Jesus & as a historian, Luke generally prefers to cite pre-Matthean versions of the logia rather than Matt's paraphrases. OR

3. Luke simply did not know Matt, but used material from a non-Markan sayings source that Matt had also used.

The third is the simplest solution from a redactional critical perspective, since it does not require one to account for every deviation of Luke from the text of Matt by speculating on his reasons for making editorial alterations. This is essentially what has commended it to the majority of modern synoptic scholars.

The first is the simplest solution from a source critical perspective, since it does not require one to account for the construction & disappearance of a text valued by Matt & Luke but apparently no other early Xn writer. This is essentially what has commended it to Farrerites & Griesbachians alike. And for those like Goulder, Goodacre, & Farmer's associates that is worth the time & effort to try (a) to prove that Luke knew & used Matt & (b) to develop detailed speculations about Luke's agenda & method in revising Matt.

But even if advocates of #1 are successful in showing that advocates of #3 are wrong in asserting Luke's total ignorance of Matt, they still have not dispensed with #2. And if non-Markan Matthean logia circulated independently of Matt (which would be the case if Matt did not create them de novo), then #2 is simpler than #1 from both a redactional & a source critical perspective. For if Mark is a source of Matt (as most synoptic scholars have thought for more than a century), it can be demonstrated that even if Luke knew Matt he regularly preferred to follow the earlier, less polished wording of Matt's source (Mark) than to adopt Matt's redactional improvements.

Hence, I predict that Q is not going to disappear "by its own weight" simply by hammering at the non-Markan minor agreements between Matt & Luke. In the long run the Q hypothesis might be strengthened by the recognition that Luke might have gotten from Matt some of the pericopes that now appear to be misfits in Q.

Shalom!

Mahlon

Thurs, Dec 18 1997 18:20
From: Mahlon H. Smith
To:  Mark S. Goodacre
CC: Crosstalk
Subject: Luke's editing of his sources  

Mark Goodacre wrote:

The way that Luke treats the Sermon on the Mount is similar to the way in which he treats Mark 4 (etc.).

---

Really, Mark? If you mean that he seems to condense passages & eliminate material in both places, you are right in general. But if you launch a detailed comparison Mark 4//Luke 8 on the one hand (as I do in my Synoptic Gospels Primer) and Matt 5-7//Luke 6:20-49 there are many obvious differences in the way Luke handles his "sources" (if one supposes he is editing Matt in the second case). 

Luke 8:4-18 not only keeps the Mark 4:2-24 structurally intact, it preserves the grammatical & logical features of the source text without correction, even though the hINA clause in Mark 4:12 (//Luke 8:8) contradicts Luke's own understanding of Jesus' use of parables as illustrations. Luke's redaction of Mark's text in this case is limited to (a) a change of setting & (b) elimination of some awkward wording. Here in editing Jesus' sayings in Mark Luke's editing is here very conservative. Even though Matthew has a parallel to this section Luke uses none of Matthew's improvements on Mark, which is why Michael Goulder says Luke has "put Matthew aside" at this point. I suspect that Luke's inattention to Matthean redaction is rather good evidence that he did not use Matthew at all. If he had used Matthew we might expect him to have left the parables of the mustard seed & leaven where Matthew put them. When Luke presents the parables of the mustard & leaven 5 chapters later he does not include any of Matthew's adjoining material in the chapter on parables. Moreover, he does not use any of the rhetorical details of the mustard seed parable that Matthew shares with Mark. This looks to me like Luke is using Matthew's source for this pair of parables rather than Matthew himself.

Luke 6:20-49 has only an approximate structural similarity to Matt 5-7.  The wording of many sayings (Blessed are you poor... etc.) differs significantly enough to say that if Luke was using Matt's Sermon on the Mount he has thoroughly reinterpreted it. The order of sayings in the two sermons is also not identical (love enemies & turn cheek are transposed). Sayings are found in Luke's sermon (woes & blind leading the blind) that are not in Matt's & most importantly Matthew's basic thesis (J comes to fulfill Torah) is missing. If Luke used the Sermon on the Mount he gutted it & focused it on a saying that was not primary in Matt (love enemies). Since this is not what he did to his source text in Mark 4, I suspect his sermon is like his version of the parables of the Mustard Seed & Leaven: a conservative reproduction of his sayings source. If this is the case, Luke's source cannot be Matthew, but is probably the same sayings source (Q) that Matthew edited in a completely different
way.

Is this circular reasoning? I call it just plain good sense. To recognize Luke's dependence on Q in these places makes things very simple. To argue for his dependence on Matthew necessitates a mountain of unprovable speculation on his motivation for moving, ignoring or eliminating mountains of material. I hope he had a Pentium 200! 

Peace, 


Mahlon

Thurs, Dec 18 1997 18:21
From: Mahlon H. Smith
To: Mark S. Goodacre
CC: Crosstalk
Subject: Re: Luke's editing of his sources

Stevan Davies wrote:

We may find that the more primitive forms are generally found in Luke. We may find that it does not usually happen that authors revert less primitive forms into more primitive forms. If these things are so, then we'd have good evidence that Luke did not use Matthew.

Mark Goodacre replied:
 
In principle this is laudable but in practice there are large difficulties. For Q has largely been constructed on the assumption  that Luke contains some more primitive forms of sayings than those in Matthew, and this observation then gets bound into the argument. Why believe in Q? Because sometimes Luke has the more primitive form of sayings and sometimes Matthew. How do you know what is more primitive? Look at Luke's form of Q sayings. This kind of circular logic is never far away in Gospel scholarship.

---

 
Come now, Mark. You know that "more primitive" generally means "less developed." In sayings it can mean a number of things:

1. Less polished -- i.e., rough grammar, awkward style, etc.: e.g., Mark's style & grammar is "more primitive" than Luke's, meaning it is more likely that Luke edited Mark than the reverse;

2. Less elaborate -- i.e., simpler construction, more consistent: e.g., Mark's version of JB's message is "more primitive" than Matthew's & Luke's because it (a) juxtaposes simple clauses & (b) is more concise;

3. Logically prior -- i.e., this presupposes that: e.g., Mark's narrative outline is "more primitive" than Luke's because Luke 4:23 presupposes cures at Capernaum in Mark 1-2, that Luke narrates later;

"More primitive" is not an arbitrary judgment. True, in sayings material it must be used with caution, since the simpler expression is not always (or even usually) the earliest written. As an erstwhile poet, I know too well that it takes work to get a good memorable concise line. Moreover, with respect to Q sayings -- pardon me NMM // L -- primitiveness is not always a criterion that favors the 2 Source hypothesis, since Luke's version of some pericopes are more polished & elaborate than Matthew's (e.g., love enemies, don't judge, centurion's servant, lost sheep, parable of pounds). In such cases Luke is probably editing his source. But what source? 

Here, fortunately we have a neutral witness: GThom. Thomas' form of some sayings is closest to Mark's (e.g., parable of Mustard seed), while Thomas has NM sayings that are only in Matthew (e.g., weeds & wheat) & others that are only in Luke (e.g., cast fire on earth). Either Thomas is an eclectic source that has culled sayings from all the gospels with no discernable pattern of preference for one or the other; or Thomas represents a record of oral tradition that is independent of the synoptics. To maintain the former one has to construct a redactional hypothesis of Thomas' editorial technique that makes the task of explaining Luke's alleged technique of editing Matthew a picnic. It is far simpler to concede Thomas' independence. 

In comparing synoptic/GThom parallels GThom has a "more primitive" form of a saying most of the time -- that is to say, it lacks rhetorical flourishes & other details that are inherent to the synoptic sayings (even in passages that are virtually identical in both Matt & Luke). In other sayings, GThom has similar wording to both M & L but is identical to neither. A good example is GThom 54: "Blessed are the poor, for yours is the kingdom to heaven." Note that the basic structure of the saying is more parallel to Luke than Matt except for the circumlocution for the name of God. Thus GThom confirms that in this case Luke preserves a more primitive form of the saying than Matt's.

Luke's form of the beatitudes is also "more primitive" than Matthew's because it is (a) shorter, & (b) more coherent (4 blessings of people in distress vs. 8 commending a variety of ethical virtues & 2 blessings of the oppressed). Yet Luke is not directly dependent on the form of these sayings which are scattered throughout GThom. So, in this case it is more probable that Luke is not editing Matt but is dependent on some sayings source similar to but not identical with GThom.

Is there anything circular in this logic?


Peace,


Mahlon

Date: Mon, 20 Apr 1998 15:08
From: Mahlon H. Smith
To: Mark S. Goodacre
CC: Crosstalk
Subject: Thomas an "offshoot" of Q?

Mark Goodacre wrote:

Actually, Q contains a good number of references to the speaker. I am not quite sure where Peter is seeing the contrast between Q and Thomas on this one. As I have pointed out before, Q is thicker with narrative interludes than is usually realized.
<SNIP>
Admittedly many are simply 'he said' etc., but often it is more than that and there is a good number of different speakers, John, Jesus, the devil, centurion, John's disciples, some, someone, another, crowds etc.

---

Thanks for the catalog of references to speakers in Q, Mark. It's always gratifying to see a scholar who is so objective that he can dispassionately examine the contents of a document that he doubts ever existed ;-). But the actual list is not quite as long as the 33 references you give. Note that 10:2, 11:29&39a, & 13:18 are scholars' reconstructions where there is no common Greek wording in Matt & Luke. And other instances you cite of references to speakers are not part of the narration of a compiler but are rather integral to the sayings themselves: i.e., 11:2&18. Moreover, the bulk of this type of narration is in pericopes that many Q advocates regularly assign to Q2 or Q3 (the Temptation, the Centurion at Capernaum, JB material & the Beelzebul controversy). As I recall Peter Kirby was replying to a claim that Thomas was an offshoot of Q1.

Mack gives only 14 narrator's references to a speaker in his reconstruction of Q1 (Lost Gospel 73-80):
1. These are the teachings of Jesus (a hypothetical incipit)
2. Seeing the crowds he said to his disciples (likewise a hypothetical introduction to the Q sermon in Luke 6:20ff)
3. 6 places in the trio of dialogs with erstwhile disciples (Luke 9:57ff).
4. The "he said" of Luke 10:2 (introducing the Q version of mission instructions)
5. 3 places in the dialog introducing the parable of barn builder (Luke 12:13ff) &
6. 2 "he said"s introducing the parables of mustard & leaven (Luke 13:18ff).

Of these note:
1. 1,2 & 4 are scholarly guesses unsupported by verbatim agreement between the text of Matthew & Luke;
2. 5 is a solely Lukan passage that was debatably derived from Q.
3. The third dialog with a would-be disciple (3) is found only in Luke & is not even found in all reconstructions of Q [I note that you do not count it].

Thus there are only 7 narrator's references to a speaker in Matt & Luke that are incontestably traceable to Q1. And apart from the pair of dialogs on discipleship these consist simply of 3 "he said"s. Of course, there may have been many more. The more Q passages one traces to Q1, the more narration there will be. (Because of parallels in GThom, I'd include some of the JB sayings in Matt 11 that Kloppenborg & Mack assign to Q2). But, still, compared with the regular identification of the speaker in virtually every aphorism in GThom, textual evidence for such phenomena in Q1 is slight indeed. I conclude that Q1 has the characteristics of a written collection while GThom, though written, preserves the monotonous marks introducing a transcription of originally independent oral chreiai.

---

I wrote:

The problem of proving GThom's dependence on Q is analogous to the problem of proving Luke's dependence on Matthew. In both cases the author of the allegedly dependent work would have had to have been a disciple of Derrida bent dismantling well-constructed compositions.

Mark G. commented:

The more that I see this objection to Luke's use of Matthew, the more implausible I find it. But I have written about this several times before on Crosstalk, and on the web site, and I don't want to bore people once again except to summarize:

---

You never bore people, Mark, at least not me. I'm quite familiar with your argument on this point. But as a good debater you know that it is always good tactics to hammer at a point that one perceives as a weakness in an opponent's position. You keep bringing up the MA's; so I'll keep bringing up the randomness of Luke's presentation of Q material vis-à-vis Matthew's.

When it comes to the act of composition, I think anybody should be able to grant that it is easier to envision an author acting like Matthew, combining material that was previously found in different sources (Mark & Q) or different places in the same source (Q) than to explain an editor's reasons for dismantling a well-built structure (like the sermon on the Mount), preserving the skeleton of the edifice but moving it to a different site & scattering some of the other pieces around the narrative landscape. But I was not intending to open the issue of Luke's alleged use of Matthew anew. I was merely using this as an analogy to argue that Thomas would have had to be even more of a deconstructionist than that if as Peter's correspondent alleges, GThom is an "offshoot" of Q.

Shalom!

Mahlon

Date: Fri, 05 Jun 1998 14:24
From: Mahlon H. Smith
To: Mark S.Goodacre
CC: Stevan Davies, Crosstalk
Subject: Matthew's sources

Pardon my belated intrusion in this stimulating debate on calculating literary dependence on distinctive language. But I have 2 cents to contribute that cannot wait until I've digested the whole volume of backed-up e-mail in my in-box.

Mark Goodacre wrote:

2. The calculation that the Lucan form of certain Q sayings is prior to the Matthean form is based partly on recognising distinctively Matthean language among such sayings.

---

This is true only if one is doing intellectual history. The notion that the Lukan version of a saying of Jesus is closer to the original Jesus sayings source can be traced to critical scholars who observed that the Matthean version often contained phrases or ideas that were elsewhere typical of Matthew. But for the past 30 years or so it has not been "based" even partly on this observation. Recent analysis of the original form of Q sayings has been influenced primarily by 2 other converging observations:

1. The first is based on study of the mechanics of orality. The Lukan form of Q sayings is generally "simpler" in logic, syntax & vocabulary than the Matthean parallel & hence more memorable orally. Thus, it is better credited to a residue of oral transmission than to literary redaction of Matthew. Where this is not the case, as in the pericope of the centurion's son or the parables of the pound & banquet, most Q proponents are prepared to grant the priority of the Matthean form in the sayings source.

2. This observation is confirmed by analysis of the gospel of Thomas. Luke's form of Q sayings is generally supported by Thomas against Matthew. Thomas is an excellent point of reference because it is primarily a compilation of oral units & it includes many pericopes that otherwise are in Matthew (e.g., parables of weeds, pearl, treasure) but not Luke. So Thomas cannot be plausibly held to have an anti-Matthean bias. The major flaw in Goulder's analysis is that he totally rejects the evidence of Thomas in accounting for the formation of the gospel tradition. As one who now admits to leaning towards Thomas' independence of the synoptics, you Mark should at least recognize the weight of Thomas as evidence of Luke's independence from Matthew.

Thus, the contemporary argument for the priority of the Lukan form of Q sayings is really "based" on evidence that has nothing to do with typical Matthean vocabulary. The results would be the same whether the wording peculiar to Matthew's version of Q sayings were used nowhere else in Matthew. The fact that Matthew echoes his special language elsewhere merely confirms a conclusion that now rests on a foundation of totally independent research. So Goulder's argument about the Matthean vocabulary fallacy doesn't disturb the Q hypothesis one wit. The progress in Q studies is analogous to developments in the physical sciences. The theories of evolution & relativity can be historically traced to the calculations of Darwin & Einstein. But they are no longer based simply on the initial observations of these pioneers.

---

Mark G wrote:

The difficulty with this is that Luke, as I have often claimed, does show signs of the retention of 'demonstrably redactional elements' of the kind that seem so noticeably lacking in Thomas. Take "hypocrites!" for example. Kevin Johnson was rightly pointing to the lack of this kind of thing in Thomas. Not so with Luke and Matthew: this phrase, characteristic of Matthew's redaction of Mark, crops up in Q, i.e. in Luke's use of Matthew. Likewise "O ye of little faith" and many more such. That is one of the reasons that I am inclining against the Thomas-dependence-on-the-Synoptics theory. Luke's use of Matthew gains strength from the presence of these Matthean redactional elements.

---

This is what I dub the real Matthean vocabulary fallacy, Mark. Just because vocabulary is typical of a person does not mean that that person is the source for occurrence of parallel wording in other works. My personal experience shows that the opposite is more often the case. Someone who repeatedly echoes pet phrases is more apt to have derived them from some other source than to have invented them. Here are a few test cases:

1. If you would compare the number of occurrences of academic theological jargon in modern scholarship, I bet you would find that terms like heilsgeschichte, Sitz im Leben, hermeneutics, eschatological, etc. occur less frequently in the works of the scholars who introduced them than in the works of their successors. I remember as a student in seminary showing off my comprehension of my wonderful new erudition by using such phrases more than my teachers. I have not conducted a scientific study of the matter but I suspect this is generally true in most peoples' experience. We all tend to repeat striking terminology that we pick up from someone else.

2. Reading my CrossTalk backlog of e-mail supports this suspicion. More than a month ago I intervened in the debate between Bob Schacht & Mike Grondin regarding historicity by claiming that a believer has to "bracket out" his personal theological convictions when making historical judgments if those judgments are to be accepted as anything more than faith statements. I used the phrase "bracket out" only once, but have counted at least 5 references to that phrase in the posts of Mike G. since then. Obviously my phrasing stuck in his mind & he has thought about it so much that he has used it over & over again. If one judged the source of that phrase merely on the basis of statistical frequency, one would conclude that Smith got the phrase from Grondin. But this stylistic judgment would be historically in error.

Hence, Matthew's frequent use of the term hypocrite is not good evidence that Matthew is the source of that term in Luke's version of the speck/timber aphorism. On the basis of logical analysis, Thomas' wording of this logion is more original than either of the synoptic versions. "You hypocrite" was obviously inserted as a rhetorical flourish. But on the basis of historical analysis the author who inserted it is more likely to have been the compiler of Q than Matthew.

Why? Not just because Matthew uses it more frequently but because Matthew regularly uses it in condemnations of the Pharisees (Matt 23) whereas the Lukan & Thomas parallels totally omit it. Luke's omission of "hypocrites" in these sayings is not plausibly credited to Lukan redaction, since he has no trouble putting this accusation on Jesus' lips in the special Lukan story of the woman with osteoporosis (Luke 13:15) & more significantly he uses it in his version of the Q/Thom saying on meteorological signs (Luke 12:56) whereas it is found in neither Thomas (91) nor Matthew (Matt 16:2-3).

Conclusion: "hypocrite" is a technical epithet introduced into the Jesus tradition by the author of Q in the speck/timber logion & independently echoed in different contexts by both Matthew & Luke (but not Thomas). If Thomas is not dependent of the synoptic versions of these sayings, then it is clear that Luke is not dependent on Matthew. Thus, the sole instance of a common use of "hypocrite" in a saying in Luke & Matthew is more easily credited to mutual dependence on a common source than to editorial fatigue by Luke who otherwise excised every instance of the term in Matt only to reintroduce it in 2 non-Matthean contexts.

So much for the use of "typically Matthean" vocabulary to determine literary dependence on sources.

Shalom!

Mahlon

Date: 18 Oct 2001 01:18
From: Mahlon H. Smith
To: Corey Liknes
CC: Crosstalk 2
Subject: Chreia Formation & Q

Corey Liknes wrote:

I have recently read an older article by Burton Mack in which he suggests that Cynic-like aphorisms associated with Jesus were elaborated upon or expanded by the Q community (or other Jesus communities I suppose) for the purpose of social formation. (Burton Mack, "Q and a Cynic-like Jesus" in Whose Historical Jesus? Michael Desjardins & William Arnal, Eds. Waterloo, ON: 1997).

I have two questions about this thesis. First, <SNIP>
I wonder where these original sayings came from. Gerald Downing and Robert Price point out the connections with cynical sayings of the time, but why would these sayings have to have been associated with the teachings of Jesus? In other words, did someone just compile a compendium of cynical sayings out of thin air as it were, and then apply the name "Jesus" to them? And what purpose would that serve the author?
<SNIP>
I guess I am just wondering how this concept of chreiai elaboration works. It seems to me that the elaboration cannot take place in a vacuum, but must instead be reliant to some degree on the original narrative context in which  these statements were made -- unless of course the statements were never made by Jesus at all.

---

I sympathize with your quandary Corey, since it is difficult to get Mack to commit himself on specifying what particular items in the sayings tradition are traceable to HJ. One of the reasons he left the JS was his lack of confidence in the project of distinguishing the voice of HJ from that of his tradants. Since he was more interested in the social formation of knowledge, he preferred to speak of communities rather than identifiable individuals generating/collecting sayings.

Without intending to champion Burt's reconstruction of Q or the social history of the early Jesus movement, however, I think a couple of clarifications will be help.

1. The Cynic hypothesis. Note that Mack postulates the sayings tradition as originating with a "Cynic-like Jesus." He doesn't identify HJ as a Cynic. But he calls attention to certain parallels to Hellenistic Cynic traditions in the type of rhetoric (social critique), forms (aphorism & chreia), & lifestyle (independence, itinerancy) found in the JS sayings tradition (particularly the Q stratum but also Thomas).  The fact that these Cynic-like elements in Q are diluted, modified & transformed in -- or many of those in GThom even excluded from -- the fully developed canonical gospel tradition, leads Burt to conclude that these elements must be from the original stages of the Jesus movement. Thus, Burt concludes that the founder of that movement (HJ) must also have been Cynic-like (not "cynical" in the popular sense of that word).

2. Social Formation. In The Lost Gospel Mack writes:

"If we ask about the character of the speaker of this kind of material, it has its nearest analogy in contemporary profiles of the Cynic-sage.  This is as close to the historical Jesus as Q allows us to get, but it is close enough for us to reconstruct a beginning of the movement that is both plausible and understandable. One should not underestimate the attraction of a Cynic-like sagery capable of enticing individuals into forming a discursive association."

Thus for Mack the analogy works at several levels. On the one hand HJ, like Diogenes, was remembered to have uttered several startling quips that challenged social conventions. These were largely unpremeditated, off-the-cuff responses to situations he encountered. The force of HJ's wit/radical wisdom in such situations. however, attracted followers who recalled these sayings & repeated them [analogous to Diogenes or Socrates].

HJ's original sayings were simple quips or aphorisms -- i.e., sharp pithy comments as distinct from common proverbs or maxims. In repeating them & transmitting them to others the people attracted to HJ had to contextualize them by creating a brief description of a situation that prompted them (since HJ did not himself describe the situation that prompted his response). This contextualizing creates the chreia. Once formulated, it provides "the official" account of what prompted the sage to say what he did.

Mack, however, remains reluctant to identify the setting of any chreia as an accurate recollection of the original occasion that actually provoked the bon mot simply because we have ample evidence in the gospels & in other chreia collections of the same aphorism with very different chreia settings or without any chreia frame at all. So Burt would deny your contention that the "elaboration" of the chreia "must...be reliant to some degree on the original narrative context."

The narrative context is precisely what is not original. Narrative contexts are always invented after the fact. How long after & by whom (an eye-witness or a later orator or scribe) is often impossible to tell. Even the ascription of a particular aphorism to a particular sage is often questionable since communities tend to credit their founder hero as the source of all the wisdom that that community embraces. With the notable exception of rabbinic tradition, oral communities generally resist complicating loyalties thru multiplying the names/voices of authoritative teachers. And they tend to preserve only those sayings that they find useful (hence the name chreia).

Once things get reduced to writing, a greater variety becomes possible. Chreiai can be strung together & edited into definitive biographies like the canonical gospels. But with the rise of scribalism one comes to a stage of social formation that is even further removed from the actual original social context of the radical itinerant "Cynic-like" sage -- in this case HJ.

Such are the observations/reasoning behind Mack's historical skepticism.

Shalom!

Mahlon

P.S. Criticism of the Cynic hypothesis (à la Horsley) or attempts to identify HJ as a millennial prophet (à la Allison) only serve to increase historical skepticism that the "Cynic-like" elements of the gospel sayings tradition can be traced to HJ or force modern interpreters to recontextualize these sayings by providing them with an apocalyptic narrative frame that these sayings do not have in Q or even our canonical gospel texts.