A Gateway to the Research of the Jesus Seminar

 [Home] [About Site] [Complete Gospels] [Data Base] [Westar Institute
[Profiles] [Publications] [Reaction] [ Search ] [What's New?] [Network]




Ed. Note: The following is a collection of my contributions to various debates regarding the Gospel of Thomas, on Crosstalk, the pioneer email conference hosted by Harper San Francisco, & its successor Crosstalk 2, moderated by Jeffrey Gibson.  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.   The identity Thomas.
2.   Whether Thom 12 is reliable evidence of HJ.
3.   Whether Thom 12 is evidence of an early date for GThom.
4.   Whether all works that promote James are to be dated early.
5.   Whether synoptic gospels are older than GThom.
6.   Whether GThom is indirectly dependent on synoptic gospels.
7.   Whether the form of a saying in GThom is earlier than synoptic parallels.
8.   Whether GThom is derived from Q [see also 9 under Questions about Q].
9.   Whether GThom is a later gnosticized copy of Q.
10. Whether GThom is closer to Luke or to Matt.
11. Whether GThom is an eclectic selection of sayings from Matt & Luke
12. Why GThom has parallels to sayings from all sources of canonical gospels.
13. Whether GThom is independent of the synoptics.
14. Whether parallels to synoptic sayings in GThom are random.
15. Whether the compilers of GThom were "scatter-brained".
16. Whether GThom parallels to canonical sayings are composites.
17. Whether GThom show any sign of literary dependence on synoptic gospels.
18. Whether GThom is derived from any written source or oral tradition.
19. Whether Thom 100 urges people to pay taxes.
20. In what type of setting was GThom composed.
21. Responsibility for new look at GThom.

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

Andrew Bernhard, graduate student in History at U of Oxford. Website: Gospels.net.
E. Bruce Brooks, Prof. of Chinese, U of Massachusetts at Amherst.
Stephen C. Carlson, patent lawyer. Website: Synoptic Problem HomePage
James R. Covey, systems developer, Cochran Interactive.
Stevan L. Davies, Prof. of Religious Studies, College of Misericordia. Website: Thomas Homepage.
Mark S. Goodacre, Department of Theology, U of Birmingham, UK. Website: NT Gateway.
Michael W. Grondin, computer programmer. Website: Codex II Student Resource Center.
Antonio Jerez, journalist, Sweden.
Jack Kilmon, paleographer & former student of W. F. Albright. Website: The Scriptorium.
Peter Kirby. Website: Early Christian Writings.
Robert M. Schacht, Research specialist, Northern Arizona U.
Jim West, Adjunct Prof., Quartz Hill School of Theology. Website: Biblical Studies Resources.

I have not space here to represent their arguments in full..  Those who wish to follow the Crosstalk debate on Thomas can find it in a less edited format in Andrew Bernhard's catalog of Additional Information on GThom. Posts dated after August 1998 are found in the archives of Crosstalk 2.


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





Date: Wed, 10 Dec 1997 03:04
From: Mahlon H. Smith
To: Jack Kilmon 
CC: Crosstalk
Subject: GThom, Logion 12

Jack Kilmon wrote:

   Prof. Smith:
    As one who supports a very early and independent GThomas, I would like to know more about the JS  rating of black for Logion 12.  The "Five Gospels" say little except for the supposition that conflict over leadership arose only after the death of Jesus and that "this saying looks back on Jesus, rather than reflects events in his own lifetime."  But, of course, so do the canonical Gospels.  I think Logion 12 is genuinely Yeshuine from the standpoint of dissimilarity as well as historical verification. After all, James did assume leadership after the death of Jesus and it is reasonable to assume that leadership was bequeathed to him by his older brother.   Since this bequeathal saying is verified by the historicity of the post-passion N'tarim and in the early writings of Paul, I was quite surprised that it was designated black.  Is there a firmer foundation for this rating than what I read in The Five Gospels?



   Hi Jack!

First of all let me ask you to drop this Prof. stuff -- especially when it comes to the subject of the JS. 12 years at the Seminar's roundtable with scholars the like of Funk, Crossan, Borg, Chilton, et al has confirmed me as a dedicated intellectual egalitarian.   I have always lived by Matt 23:8-12. Even though I no longer believe that this can be reliably traced to Yeshu bar Yosef, it remains what Bob Funk would call "true fiction": i.e., the sort of thing the historical Jesus might have said (coherence), even if the character of the textual evidence does not allow a historian to claim he probably did. This is just a long-winded way of saying: please call me Mahlon.

Now to Thomas 12. I've been attracted to this saying ever since I read Thomas 30 years ago & always thought a good case could be made for it being genuine. It certainly isn't a gnostic creation & probably came from Jacobite circles in Palestinian Jewish Christianity. I recall voting pink (fits the environment but is only attested in one source). But I am fuzzy on the discussion of this saying & I have no notes about it to fall back on. Steve Patterson, who steered the discussion & voting process on GThom sayings, would know.  I think I can figure out why this one got black. But to do so I'll have to clarify the JS agenda & procedure. So bear with a bit of background.

The goal of the first 6 year phase of our work was to evaluate every saying that could have circulated independently of a narrative or text as an oral logion (more than 1500, counting parallels). This was a prodigious undertaking that did not allow equal weight or time to be given to every saying (with only 2 weekend sessions per year). Every multiply attested saying was automatically given some air time. But there were written papers on only those that merited detailed research (controversial passages, well-known pronouncements, hard sayings, etc.) to enable the Fellows to make an enlightened objective estimate of the passage's value as historically reliable data. Naturally, sayings that were the subjects of papers got the most discussion.

After completing multiply attested sayings we reviewed singly attested saying by gospel. Here priority for papers & discussion was given to sayings that the program steering committee of specialists in that gospel thought had at least a passing chance of being considered genuine. Evidently, Thomas 12 was not given top priority by the Thomas committee, so no paper was presented on it. Yet I do recall it getting some discussion. Though I'm fuzzy on the specifics in this particular case, the line of reasoning that led this saying to be voted black runs something like this:

   1. A singly attested saying can be traced to the pen of the scribe who wrote it (in this case "Thomas").

   2. Since all gospel texts are written after Jesus' death, the burden of proof is always on the person who claims that the saying goes back to Jesus himself. The mere claim or possibility that Jesus could have said something is not evidence that he actually did.

   3. The only way to prove a singly attested saying is not a scribal invention is to produce evidence that scribe probably did not invent it (dissimilarity of style, theme, etc.; awkwardness, difficulty in assimilating it).

   4. Even if a saying can be shown to have pre-existed a particular text it cannot be confidently credited to Jesus without producing evidence that it is not likely to have entered the tradition from some other source oral or written (common lore or some earlier text). Material that is dialectically difficult (i.e., contradicts common opinion) or socially embarrassing in a Christian context is easier to trace to Jesus than material that is not.

   5. After the fact records of predictions of things that actually happened are never admissible as historical evidence, because there is no way of distinguishing recollection from retrojection. Did Jesus anticipate his death? Possibly, even probably -- unless he was totally oblivious the the controversy swirling around him. Can one prove that gospel sayings anticipating Jesus death were uttered by him? No. For our only sources are scribes who were themselves in position to know what actually happened. Did the disciples anticipate Jesus death? Given the unanimous insistence of the canonical gospels that they did not, this seems improbable (if they did, one would have to account for the rise of a multiply attested tradition that they did not).

    6.Thomas 12 is not a historically reliable report of a dialog between Jesus & his disciples before his death because:
    a. The disciples anticipate Jesus departure (which in fact happened);
    b. Jesus anticipates the prominence of James (which in fact happened).

Thomas 12 could have been invented by any of several sources:
    a. the Jacobite wing of Syro-Palestinian Jewish Christians seeking to champion the primacy of their hero over closer companions of Jesus (Kephas, the sons of Zebedee).
    b. James himself seeking to validate his claim to leadership of the Jerusalem church after his brother's death;
    c. Jesus himself.
Of these the third is the least historically provable given multiply attested gospel references to tensions between Jesus & his brothers.

One can speculate on a reconciliation between Jesus & James prior to Jesus crucifixion. But this saying is not reliable evidence that this event  actually occurred.






Date: Wed, 06 Oct 1999 13:32
From: Mahlon H. Smith
To: Crosstalk 2
Subject: Dating GThom

Jon Peter wrote:

Would anyone care to list reasons why Logion 12 (reference to James as J's successor) 
is discounted as a basis for an earlier dating of GThom ?

Not I! That's another aspect of GThom that advocates of a late date for this collection either overlook or squirm to discount. If GThom was dependent for its material on the synoptics -- as advocates of canonical priority are inclined to claim -- then it would most likely have named Peter as J's successor la Matt 16:18.





Date: Sat, 09 Oct 1999 02:21
From: Mahlon H. Smith
To: Crosstalk2
Subject: Dating GThom (silly question!)

Antonio Jerez wrote:

May I come with a counter-question: does the fact that "gnostic" works like the Apocryphon of James, the Apocalypse of James I and II (all also found at Nag Hammadi) also mention James give us reason to suppose that they must also have been written in the 50s CE ?

Its not the mere mention of James that makes this logion early, Antonio. It is (1) its decidedly non-gnostic & thoroughly authentic Semitic flavor ("...Ya'akov ha Zedek for whose sake heaven & earth came into being") & (2) its presupposition of a leadership void & innocence of any claim of Petrine primacy. The pseudo-Jacobite literature of the 2nd & 3rd c. CE is totally different in style & content & has nothing to commend it as the product of a Jewish Xn party of the mid 1st c. CE. In fact the infancy gospel of James' claim that Mary was raised by priests in the temple shows such ignorance of Judaism as to make it ludicrous.







Date: Sun, 19 Apr 1998 01:21
From: Mahlon H. Smith
To: Peter Kirby
Cc: Crosstalk
Subject: Thomas an "offshoot" of Q?

Peter Kirby wrote:

Earl Doherty claims that the Gospel of Thomas is an "offshoot" of an early stage of Q. He says further that Thomas originally had no historical Jesus, and that a redactor simply went through and added "Jesus said" to the sayings. I am not aware of any arguments offered for these beliefs.  Indeed, I know of no other scholar who holds this position, so it is a bit difficult for me to refute. What follows are only my musings concerning  Earl's hypothesis on Thomas.  

Any good hypothesis must be falsifiable. There must be a way to either confirm or disconfirm a scientific suggestion. In order to test it, we need some predictions based on the hypothesis. 

What follows are some of the things that I came up with (in chemistry class), which may confirm Earl's theory if they turn out to be true. <SNIP>



Pretty good set of hypotheses. Here's a synopsis of my read of the evidence: 

>1. (Every Q//Thomas parallel is in Q1) & 
>2. (Every Q1 verse is in Q//Thomas). 

are really tautologous since one can simply define Q1 as Q sayings with Thomas parallels. I.e., even if the test is positive it proves nothing unless one has first established Q1 by independent criteria la Kloppenborg or Mack. 

>3. (There is a relationship between the order of sayings in Thomas and Q1). 

Textual evidence proves FALSE. Some sayings which are conjoined in Q are scattered in Thomas. Some sayings which are separate in Q are fused in Thomas. 

>4. (The later Thomasine redactions have a consistent theme). 

Sometimes true. Thomas likes statements about unity that can be interpreted as typically Thomasine redaction. 

>5. (The later Thomasine redactors insert second century Gnosticism). 

Depends on how one defines "2nd c. gnosticism." Thomas is strikingly devoid of Apocryphon of John stuff but some sayings are amenable to a gnosticism like that of the Dialogue of the Savior or Valentinus' Gospel of Truth. These are quite distinguishable from the Q parallels however. 

>6. (There is no sign of oral tradition behind the Thomasine redactions). 

Textual evidence proves FALSE. Thomas is much more oral than Q. Even Q1 has marks of literary composition (e.g., the sermon & couplets with similar themes) that are not preserved in Thomas. Q has no doublets (i.e., duplicated or echoed sayings) while Thomas has several. 

>7. (The Q//Thomas sayings have no reference to the speaker). & 
>8. (Generally, the sayings in Thomas have no reference to the speaker). 

??? The only reason Q sayings do not refer to the speaker is that Q is reconstructed by extraction from Matthew/Luke where references to the speaker are usually regarded as belonging to the redactional narrative frame. But Q1 must at least have had an incipit like GThom that credited the contents (rightly or wrongly) to Jesus & Q2 must have had some mechanism for distinguishing the oracles of JB from sayings of Jesus or else Matthew & Luke would not have both credited the opening Q sayings to JB. There also may have been Thomas-like "Jesus said" or "And he said" prefaces to a number of Q1 sayings that both Matthew & Luke dropped in favor of more refined prefaces. In any case the ascription to a speaker is secondary in any sayings collection (i.e., it is the product of a transmitter rather than the original speaker). So, its presence or absence does not prove very much. If anything GThom's monotonous repetition of "Jesus said" is a mark of orality, while the absence of the same in Q at every level is a sign of literary refinement. 

>9. (There are sayings that don't make sense in the mouth of Jesus). 

That's true of every level of the gospel tradition. Everybody then & now wants to use Jesus to endorse what one already believes whether he really did or not. So the presence of sayings that are better attributed to someone else proves nothing. What is striking however is that Thomas does not tend to repeat Q sayings that are dubiously ascribed to Jesus & vice-versa, so the Q//Thom is a pretty good filter for isolating primary Jesus material. I say pretty good because there are some singly attested Q or GThom sayings that are better ascribed to Jesus than to a later redactor. 

>10.(There is no narrative material in Thomas). 

??? Not true. It's just that the "narrative" (generally in dialogs) in GThom is more rudimentary (they said/he said) than the traces of narrative in Q, which Mark Goodacre has summarized. 

I'd propose 2 more hypotheses that can readily be falsified which if true would indicate that the text of GThom was derivative from Q1: 

11. Sayings that are random in Q are organized in Thomas

Patently FALSE. Precisely the opposite is the case. Q tends to use blocks of similar sayings (couplets, clusters, & even a sermon) while Thomas often scatters aphorisms with similar motifs that are linked in Q. 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 on dismantling well-constructed compositions.

12. Thomas presents more embellished versions of parallel sayings than Q

Again, demonstrably FALSE. Lining up the Q/Thomas parallels shows that Thomas regularly has the shorter, logically simpler form of the saying in question. The only way to maintain that Thom is an offshoot of Q is to hypothesize that it is only indirectly dependent & that the compiler of Thomas was citing Q sayings from a very faulty memory. The problem with this hypothesis is that it often can be demonstrated that the Thomasine version of a given saying is logically tighter than the Q version. In other words, if there is any direct relationship between GThom & Q it is more likely that Q was a literary offshoot of GThom 1 rather than vice-versa, because it is more polished. 






Date: Sun, 21 Jun 1998 05:16
From: Mahlon H. Smith 
To: Jim West Cc: Crosstalk 
Subject: Thomas/Q project 

Jim West wrote: 

Thanks to Bob [Schacht] for collating the results. It seems that the original contention, i.e., that Luke is more like Thomas than Matthew has in fact been demonstrated. What does this mean? I would like to suggest that Q and Thomas were very similar, and that the possibility exists that Thomas is simply a later copy of Q. What do I mean? Simply put, Thomas is Q redacted by a proto-gnostic. <SNIP>

That's an interesting twist on the usual conservative assumption of canonical priority over any non-canonical text, Jim. Maybe we need a 2nd Q/Thomas project to test your last thesis ;-)

But I am dubious that redaction (if any) was in this direction, simply on the basis of a compositional comparison of Q & Thomas. If Thomas was editing Q then he was even more a disciple of Derrida than Luke (i.e., acc. to those who claim that Luke redacted Matt). If Luke deconstructed Matt's elegant thematic speeches, he at least left some blocks of sayings intact. But if Thomas edited Q, he left none.  I stand to be corrected by any Thomas expert, but the only pair of Thomas sayings that appear in synoptic sequence that I can recall off the top of my head are Thom 65-66. And the synoptic source of these is not Q but Mark. A quick glance through Crossan's Sayings Parallels (Fortress, 1985) -- the workbook of the JS, which BTW gave priority to Matthew's sequence -- should be enough to show the improbability that Thomas redacted Q in any form. Here's a sample from the sermon:

( [ ] = sayings that are probably NOT Q)
Poor blessed: Matt 5:3 Luke 6:20 Thom 54
Sad blessed:  Matt 5:4   Luke 6:21b


[Meek blessed:] [Matt 5:5] 



Hungry blessed : Matt 5:6 Luke 6:21a Thom 69:2
[Merciful blessed :] [Matt 5:7]



[Pure blessed :] [Matt 5:8]



[Peacemakers:] [Matt 5:9]



[Persecuted :] [Matt 5:10]


Thom 69:1
You when hated: Matt 5:11 Luke 6:22 Thom 68
Like prophets: Matt 5:12 Luke 6:23


[Salt salted] Matt 5:13 Luke 14:34f Mark 9:50
[Light of world:] [Matt 5:14a]


Thom 24:3
[City on hill:] [Matt 5:14b]


Thom 32
Lamp & bushel: Matt 5:15 Luke 11:33 Thom 33:2
[Good works:] [Matt 5:16]



[Fulfill Torah:] [Matt 5:17]



Not one dot: Matt 5:18 Luke 16:17


(Since much of the rest of Matt 5-7 is not in either Luke or Thom, I will list only those passages that are probably Q).

Accuser: Matt 5:25f Luke 12:58f


Divorce: Matt 5:32 Luke 16:18


Other cheek: Matt 5:39 Luke 6:29


Beggars: Matt 5:42a Luke 6:30a


Borrowers: Matt 5:42b


 Thom 95
Love enemies: Matt 5:44 Luke 6:27f , 35a


Sons of God: Matt 5:45 Luke 6:35b


What credit: Matt 5:46f Luke 6:32ff


Like Father: Matt 5:48  Luke 6:36


[Left/right hand:] [ Matt 6:3]


Thom 62:2
"Lord's" prayer: Matt 6:9-13 Luke 11:2-4


Treasure/moths: Matt 6:19f Luke 12:33 Thom 76:2
Treasure/heart: Matt 6:21 Luke 12:34


Eye/lamp: Matt 6:22f Luke 11:34-6


2 Masters: Matt 6:24 Luke 16:13 Thom 47:2
Don't worry: Matt 6:25-33 Luke 12:22-31 Thom 36
Judgment/measure:  Matt 7:1f  Luke 6:37f


Speck/log: Matt 7:3-5 Luke 6:41f Thom 26
Ask/seek/knock: Matt 7:7f Luke 11:9f Thom 2, 92, 94
Fathers' gifts: Matt 7:9-11 Luke 11:11f


Golden rule: Matt 7:12 Luke 6:31 Thom 6:2b
Narrow door: Matt 7:13f Luke 13:24


Grapes/thorns:  Matt 7:16-20 Luke 6:43f Thom 45
Lord/Lord Matt 7:21 Luke 6:46


I don't know you:  Matt 7:22f Luke 13:26f


House on rock/sand: Matt 7:24-27 Luke 6:47-49



The only plausible reason for suggesting that Luke redacted the text of Matt is that he preserves some semblance of "Matthean" sequences of sayings. Thom 32-33, 68-69 approximate couplets found in Matt but not Luke & hence are by definition not Q. So where's the evidence that Thom edited Q?

Jim continued: 
The idea, then, that Q is hypothetical without substance because it has not been found "in writing" is no longer viable. Q exists in written form in Thomas!

Mahlon retorts:
NOT! But proto-Thomas may have been a source for Q.

Jim concluded:
Now all we need do to recover "pure Q" is remove the redactional layers of Thomas, and viola!  Q!

I comment:

Surely, you jest! But if we remove the redactional layers of Q, we might be left with a text that approximates proto-Thomas. And if this can be traced to Jesus' "twin" brother Judas, that's about as close to ipsissima verba Jesu as you can get ;-)  







Date: Mon, 22 Jun 1998 12:01
From: Mahlon H. Smith 
To: Mark S.Goodacre 
Cc: Crosstalk 
Subject: The Thomas/Q Hypothesis 

Mark Goodacre wrote: 

The possibility that Luke and Matthew precede Thomas is, of course, a quite reputable position and one that we should not write out of the equation before doing this kind of primary research.


I totally agree that no rational explanation of data should be excluded without examination of the evidence, Mark. But your sentence is loaded with rhetorical echoes of a beleaguered orthodoxy that for centuries dogmatically insisted that only canonical texts were authentic & historically reliable because of their antiquity. (I refer to "possibility," "of course," "quite reputable," etc.)

Historically, it can be demonstrated that the claim that Luke & Matt were composed before Thomas is based primarily on apologetics rather than an unbiased examination of the data. For the historical priority of Matt & Luke is what the orthodox fathers (e.g., Clement & Irenaeus) claimed in their struggles with Christian groups that favored GThom, like the monks of Chenoboskian. So Coptic Thomas had barely been discovered before it was characterized by modern orthodox-trained scholars as an obvious "gnostic" fabrication sprinkled with excerpts from the Coptic translations of the canonical gospels. The identification of fragments found at Oxyrhynchus as remnants of 3 copies of Greek GThom undercut that argument. But despite the dating of these papyri as early or earlier than any surviving gospel ms. except GJohn & the fact that this data has been available for almost half a century, most Christian scholars blithely go on assuming the priority of the synoptic gospels to GThom. Why? Because we never learned anything about GThom in Sunday school & it is not read in church. Since, as Christians, our knowledge of Matt & Luke preceded our knowledge of GThom, we are naturally inclined to favor the priority of Matt & Luke over GThom. Whether we admit it or not, we have all been programmed to favor automatically any argument that supports what we think we already know. So, we are all usually slow to accept new evidence that contradicts what we already believe & inclined to reject contradictory evidence even before we have examined it. That is why Steve Davies, Mike Grondin, Steve Patterson, & Marvin Meyer are still a minority among biblical scholars & why they have to traverse continents & cyberspace to win a few converts among those who are hard of heart & slow to believe, like the rest of us.

Therefore, I would suggest that it is not "reputable" scholarship to begin "primary research" with the thesis that "Matt & Luke precede Thomas." Really reputable research begins on a level playing-field in which all possibilities are initially equal & proceeds to weigh the probabilities of historical priority against a neutral examination of the data. Non?

I am sure that you, of all people, need no lecture about the dangers of basing historical judgments on unproven prejudices. And I can empathize with your reluctance to reject a traditional thesis without examination. For my own Johannine research was prompted by the cavalier treatment of the 4th gospel as "late & unhistorical" by my own teachers. But in the case of Thomas, it seems to me that the "possibility" of the priority of the synoptic gospels becomes less probable the more one compares compositional features & evaluates the implications of alternate redactional trajectories.

Based on the extant evidence, the easiest hypothesis to defend is GThom's independence of the synoptics: a conclusion that I recall you were preparing to embrace. Then, to establish relative dating of the traditions preserved in each trajectory, one has to ask whether the synoptic or Thomasine form of parallel passages is rhetorically more primitive. Admittedly, the results of a through comparison will not be uniform. But if the bulk of Thomas/synoptic parallels point towards GThom preserving more elementary logical constructions than the synoptics, as many of us who have already undertaken that review have concluded, then the "possibility" that the original core of GThom was recorded after the publication & acceptance of Matt & Luke by mainline urban churches becomes increasingly less plausible.







Date: Mon, 22 Jun 1998 13:20
From: Mahlon H. Smith
To: Mark S.Goodacre
Cc: Crosstalk
Subject: The Thomas/Q Hypothesis

Mark Goodacre wrote:

I believe that I said that I was unconvinced by those like Tuckett who argued a direct (literary?) dependence line. I want to spend more time researching Thomas before committing myself on this. I do not like to make a decision until I have at least attempted to investigate everything carefully from the beginning.

Dear Mark:

Thanks for the speedy & patient clarifications of my comment on your reply to Mike Grondin. Let me assure you that I never for a moment meant to accuse you personally of basing scholarship on apologetics. I confess I read your sentence out of context & it struck a nerve, since the JS is regularly accused of distorting history by basing its conclusions on a late gnostic text of Thomas. All I meant to emphasize was that the presumption of synoptic priority in general is the residue of religious bias, that can be traced to an uncritical assumption that canonical gospels preserve "apostolic" tradition while non-canonical texts do not. Thus, according to this logic, any canonical parallels in non-canonical texts would have to be derived from the canonical gospels. For this to be chronologicallly possible, the canonical texts would have to have been published prior to GThom.

Your hesitancy to accept Tuckett's arguments for GThom' dependence on synoptic material is a clear indication that YOU do not think like this. As much as I admire Tuckett's mastery in defending Q, I agree with you on this one.

As for your policy of reserving decision until you have had time to complete your own careful investigation of the evidence: that is your prerogative & duty as a reputable scholar. I realize it takes time. I did this for more than 20 years before I formulated my position on John. And it took me 5 years after joining the JS to reach the conclusion that GThom preserves pre-synoptic tradition. Those who base their historical judgments solely on the research of others do not deserve the name of scholar. And you most certainly do.







Date: Wed, 08 Mar 2000 01:18
From: "Mahlon H. Smith" 
To: Crosstalk2
Subject: Posteriority of Thomas

I wrote:

But I think anyone who is honestly interested in unbiased historical research will have to admit that it is often easier to explain the synoptic versions of parallel logia as redactional refinements of sayings in Thom than vice versa. For, like the parable of the wicked tenants, Thom's version of the saying is often less developed & more difficult than that in the synoptics.

Mark Goodacre replied:

Although I enjoyed this post very much (as I do all your posts, Mahlon), I am troubled by the assertion about what anyone "honestly interested in unbiased historical research" would admit. If I say, for example, that I find Thomas to be familiar with the Synoptics on given occasions it is because I think that the evidence is pointing in that direction and not because of a lack of honest interest in unbiased historical research. I suspect that others like Tuckett feel the same way. Perhaps we are unconscious of a prejudice that leads us to see the evidence in a particular way, but if so, it seems to me that the only way to counter this is to show how the evidence can be construed differently.

You are reading more into what I wrote than what I intended, Mark. I did not say that anyone who has honestly unbiased would conclude that Thomas was not familiar with the synoptics. That would be too much to expect of debate on any scholarly issue. Honest people who strive for unbiased results often do come to opposite conclusions in interpreting evidence. I never meant to impugn the integrity of Ron Price, yourself or Chris Tuckett or anyone else who disagrees with my interpretation of the evidence.

All I meant was that if one does not start with an a priori assumption of the priority of the synoptic texts, then one should be able to recognize (if not wholeheartedly grant) that in cases where we have parallel sayings in G Thom & the synoptics, the Thomasine form is often (but not always) "less developed" -- i.e., less verbose, less polished, less compositionally complex, etc. -- and "more difficult" -- i.e., presents more logical problems, especially for those of us who have been raised in relatively orthodox Xnity -- than the synoptic version of the same saying. This is precisely the case with the parable of the wicked tenants. That is a phenomenological observation not a hermeneutical or a value judgment.

How one explains those phenomena then all depends on one's hermeneutical orientation. Some may honestly think it probable that the author of GThom has deconstructed, distorted & otherwise altered sayings drawn from this or that synoptic gospel. Others may just as honestly think it more probable that synoptic authors have refined and/or expanded a saying like that in GThom to fit their own theological views.

I happen to favor the latter option because that is how I see the synoptic authors working with synoptic sources. Mark is, as most synoptic scholars admit, a rough & problematic gospel. Matthew, on the other hand, is more stylistically refined & regularly presents less problematic versions of difficult pericopes in Mark. That is a phenomenological observation. How one explains it depends on one's tradition & source history models. Some scholars still argue that Mark has abbreviated (& otherwise clumsily edited) Matt. My experience as a literary critic & editor makes me think that the opposite is more likely: Matt has improved on Mark. In expanding my horizon to include GThom I have simply applied the same logic to a similar range of textual phenomena. The synoptics present a more complex, better organized, less difficult collection of sayings of J than does GThom. Ergo, in case after case I am led to conclude that the Thomasine version is more likely to represent the earlier form of a particular saying.

You may think you have good reason to conclude the opposite. I accept that without impugning your motives for doing so. But since you introduced the subject of an "unconscious prejudice," let me just ask for some circumspect reflection by suggesting that you ask yourself the question that I had to ask myself in coming to terms with early difficulties with GThom. If GThom was a canonical biblical text, would I still be sure it was derived from the synoptic gospels?

If one's honest answer to that question is "probably" then I would suggest asking oneself a 2nd question: what is different about the case of Thom vis--vis the synoptics from Mark vis--vis Matt & Luke that I am led to contrary conclusions regarding the priority of rough or polished texts in each case. If one's honest answer is "maybe not" then I think one has to face the possibility that one's unconscious reasons for considering the synoptics historically prior did involve a "canonical bias" just as I did. Where one goes after that is one's own business.






Date: Thu, 25 Jun 1998 15:00
From: Mahlon H. Smith
To: James R. Covey
Cc: Crosstalk
Subject: The Thomas/Q Hypothesis

James R. Covey wrote:

It's interesting in the context of the previous discussion that there seems to be more M material than L material in Thomas! A word of caution to those who would generalize that Thomas is more Lukan than Matthean.

Good observation & caveat, Jim. The claim that Thom is usually closer to Luke than Matt is an observation that is valid only in comparing the parallels to pericopes common to both Matt & Luke (=Q), not to the special M or L material. The fact that Thom does have a liberal dose of special M material is even more a warning to those who hold that Thom has an anti-Matthean bias. This has very definite implications for hypotheses regarding synoptic-Thom relations & it does pose serious difficulties for those who claim that Thom is a conflation of Matt & Luke.

If the compiler of GThom got his synoptic parallels from Matt & Luke, then he generally preferred Lukan wording to parallel passages, but had no problem echoing Matt if Luke did not have a parallel & included little material that he found only in Luke. Is this a reasonable pattern of redactional activity? One could always view the compiler(s) of GThom as esoteric freak(s) who is/are more schizoid than reasonable. But that is no explanation of this clear pattern of agreement. For an author to echo the Lukan form of Q passages he would obviously have to be fairly familiar with Luke & prefer it to Matt. If this were the case with Thom one would expect him also to prefer special Lukan material rather than special Matt. But that is demonstrably NOT the case. Ergo, from a redaction critical perspective, the hypothesis that GThom is dependent on Luke & Matt is highly implausible if not sheer nonsense.

As demonstration of this: compare Lukan parallels to Mark & Matt. Whether Luke knew Matt or his sayings source (Q) does not affect the model. It is demonstrable that Luke generally prefers Markan wording in a triple tradition pericope. Practically everybody admits that extensive synoptic agreement across the board presupposes Mark as the mediating text. When Luke presents non-Markan material from Matt he generally does NOT use wording & themes that are characteristic of Matt (in either special Matt material or Matthean parallels to Mark). Whether one explains this by arguing that Luke has deliberately censored Matt or that he uses Matt's source rather than Matt itself one has a relatively consistent pattern of Luke's general redactional practice: he does not present material with the Matthean flavor (whatever the source).

The only major exceptions to this are in the Markan-Q doublet sections, where Luke surprisingly seems to reverse his non-Matthean orientation by presenting a version of a pericope that is obviously closer to Matt than Mark (e.g., oracles of JB, temptation, parable of mustard, Beelzebul discourse, sign of Jonah -- all of which have been subject of debate on CrossTalk). To explain these pericopes there are only really two options. Luke got these pericopes from Matt or Matt's source. Those who see GMatt as Luke's non-Markan source (Farrer/Griesbach et al) have to argue that Luke was an eclectic editor, who sometimes for (no clear reason) abandoned his general preference for Mark to echo passages scattered randomly throughout Matt. Those who claim that Luke used Matt's sayings source (Q) argue that Luke acted consistently as an editor, generally preferring Mark UNLESS he knew of a parallel in Q. That is to say: Luke generally prefers Q to Mark & both of these to Matt (if he knew Matt). Since Luke is obviously not schizoid but the most rational of gospel writers, the latter option is still favored by most gospel scholars despite problems posed by the non-Markan "Minor Agreements", etc.

Thom, however, includes many Matthean passages that even Luke had no use for (e.g., parables of weeds, treasure, pearl) & excludes most of the major special Lukan material (e.g., parables of Samaritan, prodigal, unjust steward). Since, this is directly contrary to Thom's tendency to prefer Lukan forms of shared passages, the hypothesis that Thom edited Luke & Matt is self-contradictory. For a redactional hypothesis to be operable it has to present a model that makes at least some rational sense out of the content of texts, not just declare a priori that author X used sources Y & Z, whether the evidence points that way or not.

In the case of GThom, the lack of a consistent redactional model indicates Thomas is most probably independent of Luke & Matt.







Date: Fri, 26 Jun 1998 02:35
From: Mahlon H. Smith
To: E. Bruce Brooks
Cc: Crosstalk
Subject: The Thomas/Q Hypothesis

E. Bruce Brooks wrote:

In Response To: Survey Results and Subsequent Comment


"Thomas is frequently eclectic with respect to Matthew and Luke." This, as far as it goes, would be consistent with the possibility that Thomas is later than, not a source for, Matthew/Luke. Given the known behavior of Matthew and Luke themselves with respect to the source Mark at least for those who accept this sequence), where the last of the three (Luke) will sometimes prefer the immediate predecessor Matthew and sometimes go back to the more remote predecessor Mark, an oscillation between sources must be acknowledged to be as much an option for the latest text as is a balanced mixing of the two. In view of the comments of critics of, say, Lukan tertiority - please use this word, and notify the OED scanning committee - on the authorial strategies proposed for Luke, even a more plausible option. I don't think there is anything in perhaps it is numbers for Options 2, 3, and 4 that would preclude an the inference that Thomas has drawn on Matthew and Luke. That possibility might then deserve separate study.

Of course, Bruce, any inference from a text is worthy of study, if for no other reason than to double-check whether it is plausible or not. I agree that the numbers for the voting do not "preclude" inferring that Thom used Matt & Luke. But, as I argued in my post yesterday, that inference results in the conclusion that the compiler(s) of GThom, unlike Luke, did not act as consistent redactors but rather selected material in a scatter-brained way & deliberately distorted it. This, of course, is the classic accusation leveled against the authors of non-canonical texts by heresiologists from Irenaeus & Epiphanius on. But it only makes sense if one assumes a priori that (a) canonical sources are historically prior to non-canonical & (b) anyone who deviates from canonical texts is devious, schizoid or both. One is always free to draw that conclusion, of course. But then one should not pretend that one is comparing texts to determine relative dating & dependence. For one has dogmatically predetermined the results. The only point of testing redactional models is to determine which is more reasonable. And to be reasonable a model has to suppose that redactors generally act in reasonable ways that can be verified by normal editorial practices.

The basic problems with every version of the hypothesis that Thom is dependent on Matt & Luke are:

(1) it has to conclude that the editor(s) of this text acted inconsistently: on the one hand, preferring the Lukan form of double-tradition (Q) passages; on the other, preferring Matthean pericopes in single-tradition (special M or L) passages. But an even more serious objection is that

(2) this hypothesis cannot explain why Thom generally does not reproduce Markan material even though this is where Matt & Luke are in greatest agreement. One cannot account for this simply by arguing that Thom did not know Mark, since in a few places (like the parable of the mustard seed) GThom is actually closer to Mark than to either Matt or Luke. And even if one granted that Thom did not have a copy of Mark handy, one would be hard pressed to explain why he omits Markan passages that were taken over by Matt & Luke. A particularly good example of this type of glaring omission is that difficult logion in Mark 4:11f (//Luke 8:8//Matt 13:11,13)

"To you has been given to know the secret of the K of G, but for those outside everything is in parables, so that seeing they may see but not perceive & hearing they may hear but not understand..."

Now there's a logion that the author of GThom 1 should have found right up his alley. If the writer who began "These are the secret sayings that the living Jesus spoke...Whoever discovers the interpretation of these sayings will not taste death" had really known Mark 4:11f or a synoptic parallel he probably would have made it the very next saying in his work & issued an invitation to readers like this: "Here's a real brain-twister, guys & gals. Let's see what any of you can make of it! Solve it & you're guaranteed immortality!"

There, now, Bruce: I haven't "precluded an inference that Thomas has drawn on Matthew and Luke." I not only entertained the possibility but I made it the focus of a "separate study." I weighed it in the balance & found it seriously lacking in substance.







Date: Wed, 08 Jul 1998 05:29
From: Mahlon H. Smith
To: E Bruce Brooks
Cc: Crosstalk
Subject: The Thomas/Q Hypothesis

I wrote:

I agree that the numbers for the voting do not "preclude" inferring that Thom used Matt & Luke. But, as I argued in my post yesterday, that inference results in the conclusion that the compiler(s) of GThom, unlike Luke, did not act as consistent redactors but rather selected material in a scatter-brained way & deliberately distorted it.

Bruce objected:

I don't know, Mahlon, this seems a bit pejorative to me. Scatter-brained? I think we need to be able to assume that authors of texts had something in mind, even if we can't always figure out what it was (writing a text is no light undertaking, in the pre-computer age).

You miss my pun, Bruce. If the compilers of GThom used Matt & Luke, they deliberately selected thematically organized clusters of sayings in these gospels, deconstructed them into their constituent aphorisms & deliberately scattered the elements throughout their recompiled text so that adjacent sayings would not be linked either by form or motif. E.g., the beatitudes. If the authors of GThom had "something in mind" in re-arranging the material this way, it has not been evident to most later students of this collection. As far as I am aware Mike Grondin is the only Thomas expert who has discerned a plan in this gospel's construction.

Since the normal process of human memory is to draw links between phenomena that are similar, the compilers of GThom would have acted in an abnormal way. This is not impossible. But the only reason for assuming that they dismantled the Q blocks of sayings in Matt & Luke is the a priori assumption that they used Matt & Luke as sources. A far simpler & more normal explanation would be to assume that the compilers of GThom & Q worked independently, each drawing upon originally independent sayings that circulated randomly in the Jesus tradition & that the compiler of Q organized some of those sayings into thematically related clusters while the compilers of GThom did not. In other words, the scribes who recorded the Q version of J's sayings acted more as authors, while those who penned GThom acted more as students jotting down random notes.

This observation is confirmed by the fact that where there are parallel sayings clusters in Q & GThom, the Thomasine version is regularly stylistically simpler while the Q version is obviously more rhetorically sophisticated. In oral communication it is quite normal for people to remember only bits & pieces of well-formed speeches & to form simple catchword memory links. But if one is arguing for literary dependence of one author upon another & at the same time arguing that the more sophisticated text is prior (as is the case with those who argue that GThom is derived from the canonical gospels), then one cannot appeal to the normal eclectic functioning of oral mnemonics to explain the allegedly derivative text. If the editors of GThom used Matt & Luke they deliberately deconstructed & distorted many of its sayings, but in the process produced units that are generally rhetorically & logically simpler. At the same time the compilers of GThom included a lot of non-synoptic sayings that are quite sophisticated both logically & rhetorically. Thus, if they used canonical sources, their editorial style was self-contradictory: dismantling the literary structures of written texts while composing complex sayings from thin air. This is why the hypothesis of Thomasine dependence on canonical sources strikes most current Thomas scholars as bizarre.

Again Bruce:

In his paper on Thomas as an oracular compilation (see his website), Stevan Davies judges that Thos is, as a text and as a whole, unintelligible, not in terms of its use of other material, but simply as something to read.

"And they shall stand as a single one" (Thom 23:2). On the composition of GThom as a literary work Smith & Davies are unanimous. But you have introduced a qualifier that puts a negative spin on Steve's observation that that is not in his text. I seriously doubt that he would claim GThom was not unintelligible (= was intelligible) in its use of other material, if by "other material" one means written sources. In fact, I have never received more support from Davies than I have on this thread.

Bruce continued:

That strikes me as a valid hypothesis (though I am myself still inspired by the implied challenge to make sense of the text in some more conventional way). At any rate it is a nicer way of putting it than "scatterbrained." But whatever brain you attribute it to, several serious and repeated readers of the text have found that quality to be inherent in the text.

As one schooled in the synoptic gospels before reading GThom, I must admit that my initial impression of the latter text was that the mind of the author had some serious short-circuits. But after working through all 114 sayings I confess I have changed my mind. In some sayings (e.g., speck/timber, parable of the banquet) GThom's logic is clearer & more focused than the synoptic parallels. Thanks to long dialogs with two Steves (Patterson & Davies) & one Marvin (Meyer), I can even now see a logical mind at work in many uniquely Thomasine logia. It is precisely because I am now convinced that the compilers of GThom were sane that I have had to abandon any suspicion that they drew their material from the synoptic gospels.

Bruce objects to my rhetoric:

As for "deliberate distortion," again the terms are accusatory,...

Not at all. If one takes the synoptic texts as prior, GThom's parallels have to be described as "distortions." If one assumes that the synoptic parallels in GThom are derived directly from the canonical gospels, then those "distortions" have to be regarded as "deliberate." I make neither assumption, therefore, I am not accusing the compilers of any devious agenda. It is those who claim Thomasine dependence on canonical texts who are forced to come to this conclusion.

Back to Bruce:

...but if in fact Thom is drawing on the Synoptics and not vice versa, then that is presumably precisely what he is up to - adding Jesus as a sanction to material more directly expressing Thomas's own system (whatever we can agree or disagree to call it), and thus, if you insist, at least sometimes distorting the meaning those Jesus sayings had in their original Jesus-movement context. In support of this possibility,...

See what I mean? Even to entertain the possibility of Thomasine dependence you are forced to this conclusion. The only thing that I am insisting is that people come clean on what their presuppositions really are. If you are genuinely agnostic & merely testing alternate possibilities, the point of my rhetoric was maieutic: helping you see where the thesis of Thomasine dependence would lead you. But if you are really committed to defending the historical priority of canonical texts then my rhetoric was designed to force you to admit it. In either case, no malice was intended either to you or to the compilers of GThom.

Bruce resumes:

I think it is not out of the question that the Thomasites were trying to steal some steam from the highly successful Jesuine movement.

Agreed. But as Steve Davies argued, that goes for every gospel writer. Matthew steals Jesus' steam to power his personal thesis that the Torah has to be fulfilled down to the last detail. Luke harnesses the power of Jesus' name to promote his vision of universal peace & reconciliation between Jew & Gentile. Mark co-opts Jesus' "son of man" idiom to further his own apocalyptic interpretations of Daniel. GJohn has at least two agendas, the one interpreting every deed of Jesus as a messianic sign, the other making Jesus the divine mouth-piece for the author's own mystical soteriology. The question is not whether the compilers of GThom plundered the Jesus tradition to promote their own convictions (don't we all?), but whether the tradition plundered was the written texts of the 4 canonical gospels. It is this thesis that unbiased close comparison of the texts is bound to reveal is out of the question.

Bruce objects:

Again, I would tend to resist the imputation that all hypotheses of Thomasine dependence are "a priori" or otherwise flawed in their origins.

I did not claim that such a hypothesis was flawed. I was merely making a hermeneutical observation as forcefully as I could. Any theory of Thomasine dependence on the synoptic gospels is inevitably grounded in an a priori assumption that the synoptics are prior. In other words, this is a canonical bias brought to the analysis of the texts, not a conclusion based on impartial weighing of the ms. & literary evidence. Moreover, whether one is aware of it or not, this canonical bias is the direct product the 2nd-4th c. religious pogrom in which the "holy fathers" of the "catholic" church deliberately excluded & suppressed all early texts that did not support their own theological & ecclesiastical agendas. If this were not so, we would not have had to wait almost 2 millennia to find a single buried copy of GThom that escaped the bonfires of the ecclesiastical censors. Admittedly, this is strong language that pious ears may not like to hear. But it is the balanced judgment of an objective historian who also happens to be an ordained minister in a church that is heir to orthodox patristic tradition, not invective from a muckraking infidel or crypto-gnostic.

Bruce clarified:

And, at worst, what if they are? Testing will show them as wrong, and there would be an end of it. My own suggestion, at any rate, was precisely prompted by the tendency of the evidence to at least allow that hypothesis. I don't own stock in any of the churches, nor am I co-author of any currently hot-selling Thomas translation. I couldn't care less how this question comes out. I am philologically curious to see it come out, one way or another.

Fine! I have no objections to honest curiosity & I would never claim that just because I have become convinced that GThom is independent of the synoptics, anybody else has to accept this conclusion without examining the alternatives on one's own. But it is a fine line that separates weighing a possibility from defending a dogma. An honest test must weigh both the strengths & the weaknesses of every hypothesis before reaching a conclusion, not just make that hypothesis the unexamined basis for further speculation. From what I've read of your posts I trust your objectivity. So, I wish you Godspeed in your quest.

Bruce counsels:

Mahlon, I think you need to talk to Stevan [Davies] on the question of whether or not Thomas, as it stands, never mind its compositional processes, ends by being a normal text, capable of what we are accustomed to think of as a normal straight informational reading. Let me know how the encounter comes out. Then we can resume this discussion.

Thanks for the advice, Bruce. I am sure it is well-meant. But I regret to point out that it shows that as a new-comer to this list you are not yet very familiar with the personae behind these posts. Steve & I are old acquaintances of the same generation who have fought in the same lists both on CrossTalk & before that in the Jesus Seminar, sometimes as allies, at other times as friendly adversaries. And I have long been a vocal champion of his Thomas website. In the process we have gained a sense of & respect for each other's positions.

I may be wrong, but I don't think I have to talk to Steve to predict confidently that he would say that GThom reads like a normal collection of wisdom sayings, with compositional links (if any) those of oral catch-words rather than an overall grand plan. It does not read like a unified text. So those who expect to find a unifying grand plan in GThom (as one can in the narrative gospels) are doomed to conclude that it is incomprehensible. I am also rather confident that Steve would go on to say that the randomness & disjointedness of GThom's logia are evidence that it is not derived from the texts of the narrative gospels. Have I misrepresented you, Steve?

Bruce commented on my argument that GThom does not present parallels to much Markan material (particularly Mark 4:11):

Here Mahlon is so confident of being able to read the mind of the Thomas compiler that he is prepared to explain the things he left out of his sources. The rest of us are, more humbly, trying to make sense out of what he put in. After all, the amount of Synoptic material in Thomas is tiny with respect to the total Synoptic material. Some arbitrary omissions are almost mandatory. No?

You misread my mind, Bruce. I am hardly confident that I can discern the reasoning the compilers of GThom left material out, since I too am still trying to fathom why they put some stuff in. When it comes to interpreting the intentions of another author -- including you--, I'm as humble as anybody. So let's not get into a competition regarding who is more humble than whom.

What I am confident of, however, is my knowledge of what is in the synoptic gospels, since I've been studying them from an academic perspective for over 40 years & have been teaching them for the past 30. The point of my introducing Mark into the equation is that most of the material in Mark is also in Matt & Luke. So the frequency of the appearance of Markan material in GThom is a good objective test of whether the compilers of that text were deriving their material from both Matt & Luke.

If GThom was dependent on Matt & Luke, the author was bound to see that some material is common to both gospels & one could expect to find a higher proportion of parallels to this shared material than to sayings that were unique to Matt & Luke. That is the case. But the shared material in Matt & Luke is from 2 separate sources: Mark (triple tradition) & Q (double tradition). If the compilers of GThom had only Matt & Luke, they could not distinguish these sources. So, we would expect that, if they were concentrating on the material common to Matt & Luke, the percentages of Markan material in GThom would also be substantially greater than single tradition material. But that is NOT the case. In sorting out the synoptic parallels in GThom by source, the most parallels are to Q, the next most to special Matt, the next to special Luke, & the fewest to Mark. This is statistically surprising & not what one would expect to find if GThom was really dependent on Matt & Luke.

Add to that another phenomenon: in some of the sayings where GThom does parallel a saying that is in all 3 synoptics, the form in GThom is closer to Mark than to Matt & Luke (e.g., parable of the mustard). So if one is arguing Thomasine dependence on the synoptics, one cannot claim that Matt & Luke as the only sources. But if the compilers of GThom knew Mark as well as Matt & Luke, it is doubly mystifying why there are so few parallels to Mark in GThom. For then the authors would have deliberately discounted most of the triple tradition.

Thus, the point of my argument regarding Mark was not to claim definitive knowledge of the redactional prejudices of the authors of GThom, but to illustrate the theoretical problems raised by the hypothesis that GThom is based on the synoptics. The more one assumes this the less one is able to explain what is in GThom.

But if one assumes that GThom is dependent on oral tradition rather than any written text, then it is fairly easy to account for the randomness of the parallels in GThom & the synoptics. Moreover, this assumption allows the material in GThom to shed light on the composition of the synoptics. For many of the sayings that we previously thought were unique to Matt or Luke -- and therefore possibly fabrications of this or that author -- can now be proven to be derived from common oral tradition even though they were not recorded by either Mark or Q. Follow me?

Bruce responded to my calling attention to Mark 4:11::

But let's consider it. Mark 4:11 is one of the two most terrible sayings in that book: it asserts an intentional obscurity in Jesus' parables, with the intention that those hearing them should not understand them, and be accordingly.

I wouldn't say that because Mark doesn't. Granted Mark 4:11 is a very puzzling & troubling saying, if one takes it literally. That is probably why Matt revised it & Luke dropped the ending. But Mark does not claim that those who don't understand Jesus' parables are "damned to hell forever." For Mark previously has Jesus proclaim every human sin & blasphemy will be forgiven except the sin against the HS (Mark 3:28-29) & this spin on that saying is unique to Mark. It is Matt, rather than Mark, who delights in consigning people to hell (or at least the outer darkness). And the compilers of GThom had no problem including variants of such Matthean sayings (e.g., parables of weeds & banquet).

My point was precisely that Mark 4:11's emphasis on intentional obscurity, the mystery of God's kingdom, that is plain to insiders but hidden from outsiders, is a theme that the compilers of GThom probably would have been attracted to had they known it. For there are several Thomasine sayings that use similar themes: e.g., GThom 1-3, 5.

Bruce continued:

I hope to be forgiven if, in charity to the intent of the rest of the book I see in this a late addition to Mark.

I'm sure you'll be forgiven, but I hope you will forgive me for saying I think you are wrong here. There is no textual basis for assuming Mark 4:11 is a late addition to Mark, especially since Matt & Luke try to emend it. In fact the motifs of the mystery of God's kingdom & the lack of comprehension by "outsiders" are rather central to Mark's narrative agenda.

Back to Bruce:

Be that as it may, would it really have served Thomas's purpose? As I read the opening of Thomas, and Mahlon's perhaps lighthearted paraphrase also supports this, the compiler is eager to tempt people to try to solve the riddle of the secret sayings, and promises them eternal life (still a strong drawing card; visit your local New Age bookstore) if they succeed. The hiddenness or secrecy of the sayings is more of an attractant (YOU HAVE TO BE PERCEPTIVE to understand this) than a repellent (YOU CAN NEVER understand this, and will burn in hell accordingly).

I beg to differ, but again I must insist that it is not the purpose of Mark 4:11 to consign anyone to Hell. Quite the opposite. This logion confidently assures Jesus people that they have the secret of the K of G even though outsiders do not. This type of assurance is more comforting than GThom 3 that concludes with the warning: "If you do not know yourselves then you are in poverty & you are poverty."

Besides, GThom rarely has verbatim parallels to synoptic sayings. So if he altered the thrust of other synoptic sayings, I'm sure he would have had no problem in reworking Mk 4:11 -- that is, if he knew it.

Bruce concluded:

That quizzical but open spirit, and the thrust of Mark 4:11f, are to my (simple naive reader's) sensibility almost totally opposite. So I have no problem at all with Thomas's omission of Mark 4:11f. Neither, obviously, did the Thomas compiler. Case closed.

Ah, but Bruce, I perceive that you are neither simple nor nave. In fact you are a rather cunning fox in sheep's clothes ;-) Your logic betrays you. Is yours the judgment of a "humble" quester or the confident voice of someone who is pretending to know why an author "left out" a certain passage? Instead of testing the possibility that GThom might be dependent on the synoptics you are here baldly asserting this as a fact & proclaiming it as obvious.

To "omit" Mark 4:11, the compiler of GThom would have had to have known it. Such omission is not at all obvious. What is obvious is that GThom does not contain a parallel to Mark 4:11 or practically all of Mark. A far simpler & more obvious conclusion from this observation is that GThom did not have access to this material, not that he had no problem omitting it. As a teacher I cannot assume that a student knows something if (s)he doesn't cite it. (If I did I'd have to give everyone an A).

But if the compilers of GThom did not have literary access to Markan material, then they also did not have access to either Matt or Luke, since practically all of Mark is paralleled in these gospels. If that is the case, then the synoptic parallels in GThom are not evidence that GThom is dependent on the synoptics. You cannot simply dismiss this case by declaring it closed. But, if you were a fair judge, you might toss the arguments for Thomasine dependence on the synoptics out of court for lack of evidence.







Date: Fri, 26 Jun 1998 14:55
From: Mahlon H. Smith
To: Mark S.Goodacre
Cc: Crosstalk
Subject: The Thomas/Q Hypothesis

Mark Goodacre wrote:

Steve [Davies]'s answer (it seems to me) presupposes a great pool of oral tradition from which Thomas selected, a pool that contained (what we call) Q, M, L, Mk, Mk-Q overlap, MattR of Mk, LukeR of Mk and Johannine material.

I have always assumed that Christian origins were somewhat more diverse than this pool option, and that each evangelist will have had access to only certain strands of oral tradition, and that this is one of the reasons that there are diversities among the different kinds of material. Surely the evangelists (including Thomas) were spread widely across the ancient world. Has it not been an axiom of critical study of the New Testament that one of the reasons for differences between the Gospels is different places of origin, with accesses to different materials?

It has, Mark, particularly because of the impact of the Robinson-Koester trajectory model on research in early Xn sources during the last 3rd of this century. The trajectory model presupposes a point of origin from which lines disperse in all directions (like the big bang). But the trajectory model has not totally replaced the syncretism model that dominated the study of Christian origins for the first half of this century. It has also been an axiom of textual criticism for two centuries that scribes are as much influenced by oral memory as by a written text, so they are always prone to alter a text they are copying in the direction of wording they recall orally. Such secondary orality leads to the harmonization of texts, usually in the direction of Matt. This is not what we find in GThom, however. Thom will have a Matthean word (e.g., "heaven") in a logion that otherwise is closer to Luke. The only way the syncretistic model can explain a phenomenon that is a hybrid not identical to any other is to presuppose an oral rather than a textual milieu. Christians generally syncretize the gospels' birth stories precisely because they are working from oral memory rather than texts. In this case Luke & Matt represent divergent trajectories in the primitive Jesus tradition, but both strands have come to feed the common pool of Christian knowledge. For scholars like you & me to try to get students to separate this pool back into separate strands we have to counteract intellectual inertia.

In the case of Thomas, however, we are not dealing with the pooling of oral memory, but, rather, with a written text (and, from the ms. evidence, an early text at that). And the question is one of the relationship of that text to other texts. Thus, one has to construct a rather complex hybrid of trajectory & syncretistic paradigms in order to support the view that the tradition fossilized in GThom is derived from a conflation of Matt & Luke. Assuming the authenticity of a logion, one hypothesizes oral divergence or editorial emendation to account for variant forms of the saying in Matt & Luke. Then one has to hypothesize some sort of social milieu like relatively orthodox urban Christian communities that accepted both Matt & Luke in order to account for an oral pool that would be able to account for the kind of eclectic logia forms one finds in GThom. The author of GThom would have to be a member of such a community in order to syncretize Matthean & Lukan versions of Jesus sayings. (For it is unlikely that he was an outsider who ordered personal copies of Matt & Luke from Amazon.com via the Imperial internet or got it from listening to a gnostic mystagogue). Then, in order to account for an empire-wide attempt to discredit this "new" collection, one has to hypothesize that this member of a mainline church who wrote GThom (or his readers) either became schismatic(s) or was/were excommunicated by the leaders of the community that shared the same synthesizing oral pool of information.

Quite frankly, I think a simple trajectory model hypothesizing GThom's priority to & independence of Matt & Luke is far simpler & more historically probable. GThom records one early strand of Jesus sayings tradition. Matt & Luke later record divergent versions of another. In the social formation of early Xnity, Matt & Luke were accepted into the common pool by larger & more influential urban churches, while GThom continued to be used by rural & itinerant types with an ascetic, mystical lifestyle. Now I ask you? Which of these groups was probably closer to the social context of HJ & his original disciples?

Mark G. wrote:

It may be that the standard model is wrong, and that we do need to go for a hypothesis of a homogeneous pool of oral tradition, to which Thomas is our best witness.

Mahlon counters:

There's nothing wrong with the "standard model" (i.e., divergent trajectories), just the a priori prejudices that the original pool was identical with this or that text. Neither the stream model nor the pool paradigm are exclusive. Streams diverge from a common source, but they all eventually feed common pools or flow into the ocean. If they don't, they become like GThom or the Colorado River, disappearing into the sands of a parched desert.

Mark asks:

I would be interested to know if others see this as a problem. Thomas, let us remind ourselves, has parallels to every strand of synoptic material as well as to John. Do others see this as a potential problem for the theory of Thomasine independence? And if so, is anyone able to articulate it better than I am?

Mahlon defers:

Don't look at me. I think John preserves some really early stuff: not only SG but light/dark motifs, etc. (compare Paul). But more on that another day.







Date: Sat, 27 Jun 1998 10:05
From: Mahlon H. Smith
To: Antonio Jerez
Cc: Crosstalk
Subject: The Thomas/Q Hypothesis

The problem with making sweeping generalizations is that they are easily turned against you.

I wrote:

Quite frankly, I think a simple trajectory model hypothesizing GThom's priority to & independence of Matt & Luke is far simpler & more historically probable. GThom records one early strand of Jesus sayings tradition. Matt & Luke later record divergent versions of another. In the social formation of early Xnity, Matt & Luke were accepted into the common pool by larger & more influential urban churches, while GThom continued to be used by rural & itinerant types with an ascetic, mystical lifestyle. Now I ask you? Which of these groups was probably closer to the social context of HJ & his original disciples?

Antonio replied:

How do we know that the Thomas people were "rural & itinerant types".

We don't "know" that for sure, Antonio. But it is an educated guess from what we do know about GThom for the sake of a source hypothesis. GThom was not found in a major Hellenistic urban center like Alexandria but well up-river near the village of Nag Hammadi near Chenoboskian where Pachomius had retreated from urban life to found monastic settlements. It is this contrast that I intended to convey by the word "rural." We also know that GThom was not adopted by major urban churches like Alexandria, Antioch, Rome, etc. It may have been compiled or flourished at Edessa, Syria which was itself something of a cultural backwater in the Hellenistic world. We also know that GThom was used widely enough that heresiologists from major Hellenistic urban centers like Alexandria explicitly excluded it & even eventually included it in their book-burning campaigns. Otherwise, there would have been no need for someone to bury it at Nag Hammadi. But you're right, GThom wasn't exclusively a "rural" gospel since at least 3 copies were found at Oxyrhynchus, which --- though not quite a metropolis comparable to Alexandria --- could not be described as a village.

Antonio continued:

A single saying like Thomas 14b is not much to base such a hypothesis on. The same saying is also found in all the Synoptics and we don't claim that the groups who produced these gospels were rural and itinerant types.

I did not mean to infer that GThom was the bible of migrant workers & hobos. But Thom 14 is not the only logion that favors a detached life-style: cf. Thom 42 (which Steve Patterson translates as "become itinerants"). And the very style of a random sayings collection is similar to the collections of chreiai used by cynics whose heroes were the hippies of the Hellenistic world.

I agree, that Matt & Luke were probably urban gospels. But if one is going to take seriously their descriptions of J & his disciples, their sources of authentic J tradition definitely were not urban. GThom draws on the same sources as Matt & Luke but lacks their polish & cannot be proven to be dependent on them. Again, my point was merely to sketch a contrast of social settings to form a source hypothesis to account for common material in the synoptics & GThom that avoids the convoluted logic needed to claim Thom's dependence on Matt & Luke. They both probably had direct access to oral tradition from HJ's itinerant teaching career in the small villages of Galilee. In the case of the synoptics, those scholars who are convinced that Matt & Luke used a common sayings source (Q) generally characterize it as a product of itinerant Jesus people (on the basis of the mission instructions & the condemnations of Capernaum, Chorazin, Jerusalem etc.). I simply was proposing that GThom fits well into that type of environment.

Again Antonio:

And if an "ascetic, mystical lifestyle" is supposed to be closer to the social context of HJ and his disciples, then how do you explain the tradition in the Synoptics that Jesus was called "a glutton and a drunkard". Doesn't sound like an ascetic lifestyle at all.

Agreed. The term ascetic is not apt for describing HJ. It is better used of JB. But HJ apparently attracted a number of JB's followers & probably himself joined JB's group for a time (witness: the baptism). His logia about "selling all" & mocking those who amass Mammon encourage a simple detached life-style. The main difference between HJ & JB is that former had nothing against attending parties thrown by secular types while the latter would rather keep himself alive by raiding bee-hives & eating grasshoppers. Apart from this I would not read too much into "glutton & drunk" characterization of HJ. After all, it is introduced as a slur from opponents not as an objective description. (You can read my analysis of this Q logion & my arguments for authenticity whenever Funk gets around to releasing Wit & Wisdom of Jesus).







Date: Sun, 28 Jun 1998 02:49
From: Mahlon H. Smith
To: Bob Schacht
Cc: Crosstalk
Subject: Thomas & gospel formation

Mike Grondin wrote:

Finally, I should be somewhat surprised if "the majority of NT scholars judge Thomas to be dependent on the synoptics". If they do, it must be largely on account of their assigning to it a later date than others (myself included) would want to assign to it.

Bob Schacht replied:

That's the key, Mike; CrossTalk isn't exactly a random unbiased sample of NT scholars. But because of the work of the JS, not to mention Steve Davies, et al., attitudes may be shifting. Also, Ryan may be working with scholarship more than 10 years old.

Thanks for giving the JS credit for shifting the attitude of current scholarship on GThom, Bob. I wish it were so. But I don't know how many scholars, other than the Fellows of the JS, are swayed by our "consensus." (Have to put that in quotes in view of recent CrossTalk accusations that we reported consensus when there was in fact polarization, which is the case with GThom: some Fellows giving it great weight & others none). The work of individual JS Fellows like Steve Patterson & Marvin Meyers deserves credit for getting GThom accepted as an independent source, along with that of sometime guest Fellow Steve Davies, whose pioneering website has probably done more to make the world aware of GThom than all the books & articles that have appeared in the past half-century.

Among the scholarly elite, however, the voice that has been the most influential in getting people to take GThom seriously is probably that critic of the JS, Helmut Koester (whose work BTW the JS Fellows value & cite). Koester begins his Ancient Christian Gospels with GThom even before introducing the synoptic sayings source (Q). As regards the question of order, it is noteworthy that, for form critical reasons, he puts "Dialogue Gospels" third (e.g., Dialogue of the Savior, the Apocryphon of James & NB the dialog portions of GJohn). Next he treats narrative "collections" (e.g., miracle catenae, Egerton 2, passion narrative & GPeter). Only then does he turn to the canonical gospels, treating GJohn first (because of the dating of ms. evidence). The book concludes with W. L. Peterson's article on the Diatesseron as a prototypical gospel harmony (a category that some CrossTalkers seem still disposed to put GThom in).

Koester's work, published in 1990, was hailed as a magisterial piece of scholarship. This influenced the editors of 5G to present a gospel chronology similar to Koester's in a simplified & graphic format that could be more easily grasped by the general public. For this we were pilloried not only by fundamentalists but by scholars, including Koester. Maybe in a few years, when critics have tired of venting their fury on the JS, we will be recognized as having made some small contribution to changing public attitudes regarding GThom & other non-canonical texts. But until the attitudes of the scholarly establishment change towards the JS, I'm afraid we cannot claim credit for changing their attitudes towards GThom or anything.







Date: Tue, 30 Jun 1998 13:18
From: Mahlon H. Smith
To: Antonio Jerez
Cc: Crosstalk
Subject: The Thomas/Q Hypothesis

Stevan Davies wrote:

It is just absurd to think of Thomas taking this phrase from this source and another phrase from that source and weaving them together to produce something that is essentially the same as the sources' versions' anyhow.

Antonio objected:

It may be an absurd way of doing things, but that's exactly a method that was much in vogue in the 2nd century. The gospel quotations from Justin Martyr shows it, just like Tatian's Diatesseron.

Not quite the same thing, Antonio. 2nd c. harmonies like those of Justin, Tatian & the author of GEgerton tended to reweave units from discrete sources into a seamless garment so that one can still identify which source the author was basically using at any point, not to pick random words from the dialog sections of separate narratives to create hybrid logia devoid of their original narrative context. The former process is logical given the synthesizing tendencies of the human mind. Steve is right, however, in characterizing the latter process as absurd, since it would require simultaneous use of deconstructive, reconstructive & selective mental software, a feat that even my new 233 MHz Pentium would find challenging.







Date: Sun, 12 Jul 1998 00:07
From: Mahlon H. Smith
To: Stephen C. Carlson
Cc: Crosstalk
Subject: The Thomas/Q Hypothesis & the Synoptic Problem

Bob Schacht replied to Stevan Davies:

If your argument for the similarity of Thomas and Luke boils down to "They both got it from oral tradition," should not the same considerations apply to the synoptic problem? Or at least to the (non-)existence of Q which resolves, on your brand of analysis, into a shared oral tradition rather than a supposedly written but unattested gospel that both Matthew and Luke drew on? If not, what am I missing here?

Stephen C. Carlson replied:

There are five basic reasons [for a literary relationship between the synoptics] that have been explained, of which the first two have been the most popular.

1. Verbatim agreement. For example, in one Q passage, Matthew and Luke agree for 61 out of 63 Greek words of a presumably Aramaic speech.

2. Extensive agreement in order and arrangement, featuring a modicum of creativity (e.g. agreements in topic arrangements of parables and miracle, which are probably not intended to merely be chronological).

3. Substantially similar selection of material, when it features a modicum of creativity.

4. Presence of editorial comments and redactional material in two texts.

5. A consistent literary pattern between three documents (e.g. the fundament synoptic fact that Mark is the middle term in the Synoptics).

Mahlon replies:

This is as lucid & succinct a summary as I've seen anywhere. Mind if I quote you in my classes from now on?

Stephen continued:

Of the first three classical proofs for literary interdependence, Thomas and the Synoptics flunk all three, although some agreements may have been obscured by the differences in language and genre. This leaves (4) redaction and (5) literary patterns. Furthermore, even if (4) and (5) turn up in favor of Thomasine dependence, the absence of (1), (2), and (3) suggest more remote possibilities, such as secondary orality, unavailability of the documents (from which the author of Thomas learned them) when Thomas was composed, etc.

Mahlon intervenes:

Amen! The main clause of this last sentence is just the admission that I've wanted to hear from advocates of the hypothesis that GThom was "dependent" on the synoptics. From this & Mark Goodacre's recent posts on this thread I think it should now be quite evident that all sides recognize that there is no strong evidence that the compilers of GThom used the texts of the canonical gospels as direct sources for their versions of sayings which parallel those in the NT. This is the fundamental concession that Steve Davies & I have been trying to win in these negotiations. The corollary is the admission that all the Thomasine sayings with synoptic parallels have clear signs of being transcripts of sayings drawn from the grab-bag of oral memory rather than snipped from a polished narrative.

The main difference between you & Mark G. on the one side and Steve [Davies] & I on the other is over the issue of whether the sayings that the compilers of GThom recalled can be indirectly traced to synoptic prototypes or are better interpreted as the product of oral processes that do not presuppose the priority of the canonical gospels.

By "priority" I do not mean the chronological dating of texts so much as logical priority of the oral rhetorical constructions preserved in these texts. I think we can all agree that, particularly in cultures where the means of communication are predominantly oral & the dissemination of written materials primitive & expensive, a later text can sometimes preserve the older form of a tradition better than an earlier one. So, rather than rehash old arguments, I think it would be more productive to focus this debate on evidence that points to either GThom's indirect familiarity with elements of the peculiarities of the synoptic texts or independence from them.

The case for dependence cannot be decided just by finding a few passages that can be interpreted as indirectly derived from synoptic models, Rather, it would have to be based on demonstration that this is a consistent pattern. For with documents, the case for dependence (direct or indirect) always has to be demonstrated not merely alleged as possible. [Otherwise, I could convict students of plagiarism just because I suspect them of cheating]. The basic presumption of source criticism must be the independence of texts that do not in some way give clear explicit or implicit signs of dependence.

Your answer to Bob gives clear reasons why literary dependence can be confidently argued in the case of the synoptics. But, as you of all people know very well, even here the evidence is not so clear that there is unanimity on the direction of dependence. In the case of a text like GThom, where direct literary dependence on the canonical gospels is not clear or even probable, the case for the direction of the relationship of the forms is even harder to establish.

At the current state of investigation, the only thing that gives automatic preference to an assumption that forms of sayings in GThom are a redaction of those in the synoptics is the canonical bias that we have all inherited from the orthodox Christian tradition. We need to recognize & bracket out this prejudice. The probabilities for dependence or independence have to be weighed on a case by case basis. And this Thomas/Q project on CrossTalk is an important step in that direction.

If there are significant cases where the forms in GThom can be shown to be both (a) free of elements characteristic of the synoptic gospels & (b) rhetorically & logically more primitive, however, then it has to be concluded that the case for GThom's dependence on synoptic prototypes is improbable, no matter how many other sayings may be open to such an interpretation.

Stephen argued:

I believe that the (commonly accepted) solution to the Synoptic Problem can help us identify prima facie redaction of the tradition.
Thus, when we compare Thomas to Mark/Luke parallels, a theory of Thomasine non-dependence predicts that this prima facie Lukan redaction won't occur in Thomas, while Thomasine dependence predicts that some of it will occur.

Mahlon interjects:

In theory this is so. But the validity of the predictions of such a theory depends totally on literary -- i.e., textual -- evidence. The theory takes no account of the impact of oral traditions. Thus, it is probative only if it can be shown that elements of "Lukan" redaction are the product of Luke's own distinctive literary creativity & not derived from pre-Lukan oral tradition.

This is particularly important in view of Luke's agenda that he states clearly in the prologue. He claims awareness of the existence of "many" texts that record earlier traditions (how many he has actually seen or read is another matter). But he bases his own authorial credentials on assurance that he has personally "followed all things closely from the first" [PARAKOLOUQHKOTI ANWQEN PASIN], which presumably means not only those texts but early oral traditions with which he was personally familiar: "just as [KAQWS] those who were from the first eye-witnesses & ministers of the word [hYPHRETAI TOU LOGOU] delivered to us [PAREDOSAN hHMIN].

This describes Luke's redactional agenda. He presumes to edit & correct the writings of others because he is confident that he personally has direct access to a pristine oral tradition from the earliest witnesses. Whether that was really the case is irrelevant. Since Luke stresses that he is writing "accurately" [AKRIBWS] in representing earlier tradition, one cannot automatically assume that every Lukan alteration of Markan or Matthean material is the product of Luke's personal logic. Indeed, since Luke's avowed rationale for writing is to produce a text that represents primitive oral tradition more accurately than others that were available, there is always the possibility that Luke learned at least some of those elements that are usually labeled as "Lukan" from the oral tradition available to him.

If these elements are paralleled in GThom, which everyone appears ready to concede is the product of orality of some kind, then there is good reason for maintaining that they are the residue of primary oral tradition & not evidence of Thomasine dependence on synoptic literary developments.

Back to Stephen:

This prima facie Lukan redaction shows up in Thomas.

Mahlon objects:

Things are not always as they appear on the surface. "Prima facie Lukan redaction" may actually be pre-Lukan forms that Luke learned from oral tradition.

Stephen continues:

In fact, some of these examples are bolstered when the apparent redaction of Mark is shown to be very Lukan stylistically.

Mahlon objects:

E.g.? Does "very Lukan" mean? idiosyncratic? Does "stylistically" mean displaying Luke's sophisticated vocabulary, grammar & literary style? Is it certain that "the apparent redaction" is not just a correction of Mark, restoring a saying to its more primitive form? I regularly edit my students papers to present a more precise wording of things I know they had heard or read. Is such "apparent redaction" evidence that I made everything up?

Back to Stephen:

Therefore, natural presumption arises that Thomas knew Luke in some manner, shifting the burden of proof to those who would argue otherwise.

Mahlon comments:

This presumption is "natural" only if one presumes that Luke is sui generis & invented "Lukan" style de novo. If there is not strong evidence that one text is literarily dependent upon another, then it seems to me more "natural" to presume that parallel elements are derived from a common source that each drew upon independently, especially when the common denominator is snippets of sayings material rather than connected narrative.

Stephen thrusts:

In law, the burden of proof is like a game of hot potato. Whoever is holding it when the music stops loses.

Mahlon parries:

So, who stopped the music?

Stephen serves for match point:

Because of the presence of Lukan redaction of Mark in Thomas, advocates of Thomasine non-dependence should not be allowed to benefit from the lack of evidence in their favor. They are still clutching the hot potato and can't get rid of it.

Mahlon returns service:

Really? I don't feel my fingers burning. In fact, in case you haven't noticed, I just tossed the potato back to you & Mark [Goodacre]. And my next reply to Mark (God willing) will give evidence of how hot it really is ;-)







Date: Sat, 18 Jul 1998 14:07
From: Mahlon H. Smith
To: Stevan Davies
Cc: Crosstalk
Subject: Tribute to Caesar Correction

I wrote:

The ambiguity of this pericope ["Pay Caesar what is Caesar's, pay God what is God's" (Thom 100 / Mark 12:17 & parallels)] may be what prompted Mark to interpret it as a trap & to make it impossible to read it as encouragement to avoid paying taxes.

Stevan Davies replied:

With the exception of yourself and some other JSem folks (first I ever heard of this line of thought was from [Stephen] Patterson), everybody from Mark to yours truly over 2,000 years or so has found it to be such an encouragement... and so not quite an "impossibility."

Steve clarified:

I'm objecting here to my erroneous reading "impossible to read it as encouragement to pay taxes" rather than your original. I don't understand the original, upon re-reading. Is it a triple negative construction?

Mahlon clarifies:

As Bob Funk & Tom Simms have often reminded me, my syntax (like Paul's) often gets convoluted & therefore my logic hard to follow (e.g., the present sentence). Accept my apologies. The basic thrust of my argument is three-fold:

1. The variant versions of this pericope in GThom & the synoptics is evidence that the central aphorism ("Return to Caesar & God...") circulated independent of any set narrative frame. Even the minimal prose narration needed to introduce J's bon mot could & did vary. Therefore, one should not presuppose that any of the narration depends on eyewitness observation of the original occasion in which HJ uttered it.

2. The Thomasine narration is simpler & probably earlier than the Markan, since it only presents the minimal information that is necessary to make sense out of J's saying:

a. J is shown a coin
b. J is confronted with the imperial demand for taxes.

To read more into the Thomasine pretext for J's pronouncement is to read it through Markan glasses which the author of GThom probably did not own. Thomas does not specify who showed the coin to Jesus, so any attempt to identify the speaker requires speculation. My facetious suggestions that this could have been either disciples or peasants was meant to demonstrate this & your arguments that it could not have been either only proves my point. The speakers & their motivation are not explicitly identified in GThom as they are in Mark. So the only reason to assume that the situation that prompted the logion in GThom was identical with that in Mark is that we had prior knowledge of Mark's interpretation. I was arguing that we need to avoid this type of circularity if we want to hear the logion as GThom readers, who did not own a copy of Mark or any of his redactors, might have read it.

3. Presupposing the priority of the Thomasine form of this logion, I tried to argue that the narrative features peculiar to Mark are the product of his own imagination & were probably the result of (a) deductive reasoning (like your earlier reply to me) & (b) Mark's prejudicial conviction that Pharisees were engaged in a conspiracy with Herodians to trap J.

Then I speculated on Mark's rationale for inventing this expanded narration:

a. The pre-Markan narration was so vague that the implications of the logion was not explicit. Note that Mark does not specify that it was a "gold" coin. So one cannot be certain that this element of GThom was original or universal wherever this chreia was repeated. And even if one assumes that a 1st c. person would automatically assume that the coinage was imperial (because of the tax question), the question of ownership of the coin is not self-evident. As far as I know, Roman rules of private wealth were not all that different than ours. So, if I am in possession of a coin, it is mine, not Bill Clinton's just because the department of the treasury that answers to his directives minted it. Therefore, Jesus' logion of returning things to their rightful owners does not give an unambiguous answer to the question of foreign taxation.

b. Mark's interpolation of J's question about whose image is on the coin & the emphatic response that it is Caesar's makes an explicit link between the coin in question & J's logion, so that it is (almost) impossible to interpret that logion as anything other than advice to pay one's taxes. This is in line with Mark's repeated claims those who thought J meant to spark a nationalistic revolt were dead wrong. It is in line with Paul's advice to Christians in Rome. But it is not necessarily what J had in mind when he told fellow Palestinian Jews "give Caesar what is his" since (a) the exact historical stimuli for this logion are not certain (different narrators introduced it differently) & (b) the logical conclusion of the logion gives primacy to giving God his due.

Josephus gives us a pretty good historical basis for assuming that many Jews would have interpreted God's claims as preempting any imperial demand for taxes. E.g., his summary of the rhetoric of Judah of Gamala & Zaddok the Pharisee its effect on public unrest :

"They said that the tax was nothing other than outright slavery & they called for the people to claim their freedom... Now as the men received all that was said with pleasure, this plot made great progress" (Ant. 18:4-6).

I seriously doubt that there is any historical precedent for your exegetical inference in your previous reply that J's counsel to give God what is God's was referred to paying the temple tax with silver Tyrian half-shekels. The first time I ever heard that was minutes ago when I read your note. Herakles may have been the model for the Hebrew Samson, but I hardly can imagine that any Jew would have thought that an image that had been identified with Antiochus indicated YHWH's ownership. Tyrian silver was the sole coinage acceptable for use in the temple of the 1st c. for a very simple reason: conservative economics. It was the Swiss Franc of the 1st c. world because its silver content was guaranteed to be precise & stable, while other currencies including those of the emperor himself were easily debased.

If your reservations about the historicity of the temple incident are based on a conviction that J advised fellow Jews to pay the half-shekel tax, please reconsider the historical evidence for the latter. I think you will have to admit that it is of dubious historical merit at best.






Date: Thu, 01 Oct 1998 02:28
From: Mahlon H. Smith
To: Stevan Davies
CC: Crosstalk
Subject: Thoughts of GosThom

Andrew Bernhard wrote:

Without going deep into detail (yet) can you list the arguments that you would make for Thomas' independence (that is, Thomas not depending on the synoptics).

Steve Davies replied:

Gee. I thought I had just a minute ago.

1. Thomas' sayings almost always lack the redactional material in the synoptics.

2. They are often in more primitive form than paralleled materials. (This second does not follow from the first, it is another proposition.)

3. Lists are a primitive form within the Christian tradition, taken over by narratives etc.

I think that's about it.

Mahlon butts in:

Not to question Steve's masterful summation of the arguments re GThom's independence of synoptic texts. But I think there is a fourth argument to support Thomasine independence that puts the dependence people on the defensive:

4. Randomness: GThom is regularly more disorganized than parallel synoptic material.

It is this phenomenon that ultimately convinced me (trained in literary criticism before I became a biblical scholar) that GThom was composed without dependence on the texts of Matthew, Mark, Luke or Q. Almost invariably Thomas publishes separately what the synoptics publish together, without imposing any discernible logical pattern of his own. If one presupposes that the editor of GThom got his material from the synoptics, then one has to suppose that he invented deconstructionism almost 2 millennia before Derrida.

E.g., compare the arrangement of parables in GMatt 13 with Luke & GThom

1. Sower: Matt 13:1-9 Luke 8:5-8 Thom 9
2. Weeds & wheat: Matt 13:24-30


Thom 57
3. Mustard Seed: Matt 13:31-32 Luke 13:18-19 Thom 20
4. Leaven: Matt 13:33 Luke 13:20-21 Thom 96
5. Buried treasure: Matt 13:44


Thom 109
6. Pearl: Matt 13:45-46


Thom 76
7. Net: Matt 13:47-48


Thom 8

If you add to this the observation that GThom's parable of the mustard seed is closer to Mark's version than to Matt or Luke, you get an even greater pattern of disorder, since Thom does not reproduce the elements of the Q wording in which both Matt & Luke agree. (Similar lists illustrating GThom's randomness could be constructed for the Beatitudes, etc.)

The human mind may recall sayings from oral memory eclectically. But memory itself depends on organization of data (catchwords, logical patterns). If a logical pattern based on catchwords has been already established in a written document, it is not likely to be totally ignored or forgotten by someone who really knew & used that text. For language itself is the primary human tool for organizing, preserving & using data. An editor might prefer a different organization because of a different dominant motif (hence, Matt modifies Mark's list of seed parables because he favors the "kingdom of heaven" motif). But collectors of previously written bons mots do not generally fragment organized speeches & pepper the fragments around their work without producing some discernible plan of their own. Those who deny Q by arguing that Luke redacted Matt have to justify Luke's "dismantling" of Matthean speeches by indicating Luke's reasons for moving a passage (e.g., the Lord's prayer) to a more appropriate location in Luke's outline. But there is simply no good reason for preferring the Thomasine order over the Matthean (or Lukan) even at the basic level of mnemonics.

Ergo, GThom was composed without reference to any synoptic gospel or source.







Date: Wed, 28 Oct 1998 09:24
From: Mahlon H. Smith
To: Bob Schacht
CC: Crosstalk
Subject: Thomas/Synoptic Parallels

Bob Schacht wrote:

H0 ('aitch-zero' i.e., null hypothesis): Thomas was formed by copying selected sayings from first one synoptic gospel, then another, interspersed with an eclectic assortment of quotes from unknown sources.

There's another applicable statistic here: if we consider the synoptic gospels collectively, we could do a Wald-Wolfowitz Runs test. That is, you take all the GThom sayings in order (all 114 or 156 or whatever of them). For each saying write "A" if it has no synoptic parallel, and "B" if it has any synoptic parallel. You then have a sequence of letters like


The idea of Wald-Wolfowitz is that if the order is random, then consecutive runs of the same letter should not be very long (the longer the run, the rarer it should be). This test can be found in standard statistics books such as Blalock.

Using this kind of notation with, say, I = no parallel, M = Markan parallel, N = Special Matthew parallel, L = Special Luke parallel, Q = Q parallel, then H0 above could be expressed something like this: GThomas =


I assume that special rules will be needed, e.g. triple tradition material will always be recorded as 'M', etc.

Dear Bob:

In the JS alphabet M = special Matt & K = Mark/par. One should also add a J for John. Using the marginal source/parallel table in 5G Thomas for distinct aphorisms rather than the standard calculation of Thomas' "sayings" (many of which are compounds), one gets the following pattern:


I did not double-check the published "source" list to see how close the parallels really are or whether a Q parallel is closer to QM or QL or, in the case of Mk-Q overlaps, which the Thomasine saying is closer to. (Many of the K parallels are the oft repeated "whoever has ears to hear" -- always in a context not found in Mark).

So this list is in need of fine tuning. But it is sufficient to make the point that other than independent sayings GThom rarely has 3 (& never 4) sayings in tandem with a parallel in the same "source." This would be true even if one removed the sayings special to GThom. That would produce this sequence:


Moreover, wherever there are tandem sayings with parallels in the same source, they are almost never from the same context in that source. This clearly demonstrates the randomness of sayings in GThom.

As for H0 (the null hypothesis: "Thomas was formed by copying selected sayings from first one synoptic gospel, then another..."): No one who has ever bothered to compare the wording of GThom sayings with canonical parallels could ever seriously propose this. Even a superficial reading of GThom proves that its sayings are not "copied" from any canonical text.







Date: Sat, 14 Nov 1998 02:36
From: Mahlon H. Smith
To: Mark S.Goodacre
CC: Crosstalk
Subject: Thomas Gospel & Q

Jeff Peterson wrote:

I was actually thinking of three kinds of relationship between Q and Thomas one finds suggested in the literature:
1) The generic relationship that Mark mentioned (Q and Thom as "Sayings Gospels," unlike the canonicals)
2) The common theological milieu to which these texts attest, an alternative to Pauline-Nicene crucified-and-resurrected Messiah Christianity
3) The literary relationship between Q or a proto-Q and Thomas or proto-Thomas.

Steve [Davies] contests whether the latter relationship has in fact been mooted in the following (written at 9:40 PM 11/12/98!):

"Q1 evolves into Q2, Thomas into Thomas2 which have now taken on entirely different ideological trajectories. Since I'd maintain that nobody has ever suggested such a thing I'd better insist that I'm not suggesting it, just exploring the logic of the position. It does have a nice symmetry to it, doesn't it? No evidence... but nice symmetry."

I agree with Steve about the problematic nature of the evidence for this proposal; here it is in print, 18 years after the remark in Trajectories that Mark [Goodacre] helpfully supplied (emphasis mine):

"The materials which the Gospel of Thomas and Q share must belong to a very early stage of the transmission of Jesus' sayings . . . Thus, the Gospel of Thomas is EITHER DEPENDENT UPON THE EARLIEST VERSION OF Q, or, more likely, shares with the author of Q one or several very early collections of Jesus' sayings. . . . The close relationship of the Gospel of Thomas to Q cannot be accidental. . . . The Gospel of Thomas is EITHER DEPENDENT UPON Q'S EARLIER VERSION or upon clusters of sayings employed in its composition" (Helmut Koester, Ancient Christian Gospels, 95, 150, cited by John Dominic Crossan, The Birth of Christianity, 247-248).

Jeff cites Koester as evidence of a "relationship between Q and Thomas one finds suggested in the [secondary] literature" in reply to Steve Davies speculation about independent parallel trajectories of Q1>Q2 and Thom1>Thom2 (which he claims "nobody has ever suggested").

Just a few comments & clarifications:

1. Jeff's highlighting of Koester's words in Ancient Christian Gospels tends to obscure Koester's own emphasis in this passage. So pardon a revised version (boldface mine but emphasis Koester's):

"Thus, the Gospel of Thomas is EITHER dependent upon THE EARLIEST version of, OR, MORE LIKELY, SHARES with the author of Q one or several VERY EARLY COLLECTIONS of Jesus' sayings" (p. 95).

"The close relationship of the Gospel of Thomas to Q cannot be accidental. . . . The Gospel of Thomas is EITHER dependent upon Q'S EARLIER version OR upon clusters of sayings EMPLOYED IN ITS COMPOSITION" (p. 150).

The fact that Koester presents Thomas before Q in his own chapter on "the collection of the sayings of Jesus" is clear evidence that he favors the second of these alternatives. The fact that parallel sayings that are conjoined in Q appear independently in random order in GThom is one of Koester's reasons for championing the priority of Thomas over Q. It represents a more "primitive" (i.e., less thematically organized) stage in the compilation of aphorisms ascribed to Jesu. No one has been more consistent than Koester in championing this position. The reason he mentions the *possibility* of GThom's "dependence" on a very early version of Q is due probably to two factors:

a. As an objective scholar he recognizes that "a large number of authors" including "respected scholars" like R. M. Grant, E. Haenchen, etc. have argued that GThom is dependent on canonical material (ACG p. 84). Thus, the real point of Koester's "either" in the rhetorical alternatives quoted above is that IF (for argument sake) GThom is dependent on canonical material, its source must be a very early version of the synoptic sayings source which was not identical with the version in Matt & Luke.

b. As a good dialectician, Koester knows that the best way to dispose of objections to one's own argument is to show the weakness in alternative positions. IF the only way a theory of GThom's dependence on material in the canonical gospels can be defended in detail is to admit that this "canonical source" is a version (much) earlier than the current canonical text, then the alternative that GThom was independent of the canonical logia & better represents the pre-canonical version of Jesu's words becomes all the "more likely." That is Koester's actual position. So his "suggestion" that GThom may be "dependent" on Q is a rhetorical straw man. Thus, to cite Koester to support a hypothesis that GThom was dependent on the synoptics is grasping at illusory straws.

2. Ditto Crossan. While the early Dom favored the priority of Q material, his work on the aphorisms of Jesus (In Fragments) convinced him otherwise. In analyzing the composition & rhetorical structure of parallel sayings in Q & GThom, Crossan repeatedly argued that GThom preserved the more primitive form of the saying. He ascribed the elements unique to the canonical (=Q) version to the process of hermeneutical development in paraphrasing & applying sayings to later literary/social contexts. Thus, when Crossan came to spelling out his own theory of the chronological stratification of early Christian sources in his big HJ (1992), he like Koester lists GThom1 before Q1. He describes GThom1 (source #5) thus (my emphasis):

"A serial collection of Jesus' sayings with LIMITED individual linkage by means of theme, word or expression... There may be at least two separate layers in it... The collection is INDEPENDENT of the intra-canonical gospels (Davies; Crossan 1985; but esp. Patterson)... it also emphasizes how much of this collection is very, very early." (HJ p. 427f).

Crossan describes Q (source #10) thus:

"A serial collection of Jesus' sayings but with MORE COMPOSITIONAL ORGANIZATION than the Gospel of Thomas... There may be three successive layers in its development" (p. 431).

But even the earliest of these layers (Kloppenborg's "sapiential" source) is more rhetorically complex & thus logically later than the parallel material in GThom.

Koester & Crossan base their arguments for the independence of GThom from Q or any canonical text largely on a neutral examination of the form & content of the sayings themselves. The details they focus on anyone with eyes to see should be able to see. By contrast those who argue for GThom's dependence on canonical works have to discount the structural & rhetorical differences between GThom & Q (even more the synoptics) which most of the time point to the primitive character of GThom. This I can only ascribe to an a priori canonical bias (pace Mark Goodacre).

What puzzles me is Steve Davies' remark that "no one ever suggested" independent parallel developments of the trajectories Q1 > Q2 & Thom1 > Thom2. As I read Koester & Crossan this is precisely what they suggest. Koester even suggests that Thom1 (=parallels to canonical sayings) is useful for constructing Q, since sayings in Thom echoed only by Matt or Luke are more plausibly traced to the same source (Q) than to claim that they were either fabricated by the synoptic author or that Matt & Luke used another sayings collection (Thom) whose structure, spin & other contents they ignore.







Date: Tue, 24 Nov 1998 23:36
From: Mahlon H. Smith
To: Stevan Davies
CC: Crosstalk
Subject: Who was Who? (was Didymos Judas Thomas)

Stevan Davies wrote:

Some guy had the nickname "twin." He had some other name too. Why not Judas? Jesus had a disciple with the nickname "twin."

Andrew Bernhard objected:

Wait, how do you know that Thomas is a nickname? Would it have to be a nickname or could it be his real name? It was never used as a surname in Aramaic, but was it in Greek?

Steve replied:

Dunno. Don't think so though.

Mahlon intervenes:

As usual, Steve's candor cuts through the BS. But lest someone mistake his expression of ignorance as a lack of learning, let me point out it is really due to lack of concrete evidence. But some rational analysis & judicious speculation are still in order.

1. QWMAS was an Aramaic name like KHFAS. GJohn supplies both with Greek translations; while the synoptics used the translation for Kephas (PETROS), they preferred the transliteration for Thomas. Why? To minimize readers' speculation over a sibling relationship? Why else would Greek-speaking Christians have preferred an alien name to its Greek equivalent, when they did not do this with either PETROS or CRISTOS?

2. Neither QWMAS or KHFAS is documented as a proper name in Aramaic prior to the association with a disciple of Jesus (not withstanding Tom Simms' intriguing suggestion of Thomas' ultimate derivation from Thothmses). So whether Thomas or Kephas could have been used as a proper name in Aramaic culture will probably never be known for certain. If, as Matt & GJohn indicate, "Kephas" was a nickname for someone whose given name was quite different (Shim'on bar Yonah -- or bar Yochanan, if GJohn is right), then by analogy it seems likely that QWMAS was also a nickname.

3. Could QWMAS be regarded as a proper name in Greek culture? Of course, just as PETROS or KHFAS was used by Paul as the name of the chief apostle to the circumcised. In fact, was even PAULOS a given name or a nickname (Latin Paulus = "Shorty" transliterated into Greek)? Where's the evidence that PAULOS was a given name in Greek culture prior to the 2nd c. CE? Yet a self-styled apostle to the uncircumcised regularly signed his correspondence that way. Whether any of these names were given proper names in Greek culture prior to the Christian missionaries who were popularly identified as such is historically doubtful. At least, I have never found evidence that they were. Has anyone else?

What was the real given name of the Christian missionary who identifies himself as as PAULOS? Was it Saul, as the author of Acts supposes? If so, Saul son-of Whom? Or does "Luke" invent that name since Paul himself claims to be of the tribe of Benjamin (Phil 3:5)?

4. The real identity of QWMAS seems to have been a total cipher for the writers of canonical Christian texts. The name apparently was included in Greek lists of the Twelve (Matt 10:3//Mark 3:18//Luke 6:15; John 20:4; Acts 1:13). But like BAR QOLOMAIOS, with whom his name was sometimes linked, his full proper identity seems to have been either forgotten or suppressed. "bar Tolmai" was a patronym (like Johnson) rather than a proper given name. So even if the group called him that (to distinguish him from others with the same first name?), his parents obviously gave him some other name.

5. Likewise QWMAS. Even if Galileans were in the habit (like the Latins)-- which they probably weren't -- of giving their children birth numerical names -- e.g., Secundus, Tertius, Quintus, Sextus, Septimus, Octavius, Decius, etc.), QWMAS could not have been the given name for one of a pair of twins, since it would have been equally descriptive of his identical (or fraternal) sibling. Ergo, QWMAS must have been a nickname, even if this shadowy Christian missionary was called & identified himself by this name.

6. As a descriptive name, "Thomas" must have been someone's double. The question is whose? To be called simply "the twin" by one's comrades is like being called "junior." It implies that one is a reflection of someone whom everyone recognizes is more prominent. The only reason I can imagine for the Greek Christian community conveniently "forgetting" to mention this twin's given name & that of his brother is that he was Yeshu's identical but less charismatic sibling (more likely Yehudah rather than Ya'akov, since the later was regularly named).

7. As Hellenists like Paul came to venerate Jesu as "son of God," it would still have been acceptable to recognize James as his (half?/step?) brother. But it would have been out of the question to publicly continue to identify Judas (or anyone else) as Jesu's twin. For then logically he would have had equal claim to all the theological & messianic claims being made for Jesu, like Castor & Pollux, the celestial twins of the constellation Gemini, who were equally sons of Jupiter (whose own name was a corruption of Deus Pater: "God the Father").

That's just my hunch. Anyone have a better one?






Key to Standard Abbreviations
5G The Five Gospels L material special to Luke
ACG Ancient Christian Gospels M material special to Matt
BTW by the way MattR Matthean redaction
CE common era NT New Testament
GJohn Gospel of John Q source of non-Markan sayings common to Matt & Luke 
GThom Gospel of Thomas Q1 earliest draft of Q
HJ the historical Jesus QL version of Q material in Luke
H0 null hypothesis QM version of Q material in Matt
J Jesus SG Signs Gospel (early stratum of GJohn)
JB John the Baptist Thom1 earliest draft of GThom
JS the Jesus Seminar Xn, Xnity Christian, Christianity
K of G kingdom of God ;-) winking smiley face (to indicate humor)

- This page was revised 21 February 2023 -


Website designed by Mahlon H. Smith
copyright 1997- 2023