July 14, 2010
Rius-Camps on the PA
Review: J. Rius-Camps, The Pericope of the Adulteress Reconsidered, NTS no. 53, (Cambridge, 2007)
Rius-Camps: - on the PA:
Introduction - Rius-Camps' new theory
His Previous Paper - the Lukan Theory
The Non-Lukan Text - Codex Bezae (D)
Textual Evidence - unaccounted for...
Internal Evidence - Chiasm fatal to theory
Internal Evidence - 19th century ignorance
Internal Evidence - 20th century advances
Non-Johannine Passages - elsewhere in John...
Textual Evidence - mishandling variants
A Second Attempt at an Origin
J. Rius-Camps gives us a clear idea of his theory in the Synopsis:
"This article considers afresh the origin of the pericope of the adulteress, which is absent from some important manuscripts. Comparison of the witnesses to the text reveals that it has been preserved in two distinct forms, one (attested by Codex Bezae and the minuscules 2722 and 1071) that is Markan in style, and the other (attested by f 13) that reproduces the style of Luke.
The conclusion drawn is that the account was first composed by Mark (and placed after Mark 12.12) and subsequently adopted by Luke (after Luke 20.19). Because of the apparent moral leniency displayed by Jesus, the story would have been removed at an early date from both Gospels, and then later reinserted by some manuscripts but at different places."
The first thing that requires a remark is that this is a complete 'about-face' from Rius-Camps' position only a year earlier (2006), in which he gave a lecture arguing the Lukan origin of the Pericope Adulterae. He apparently reversed this position after reading Dr. Maurice Robinson's paper, ‘Preliminary Observations'...(FN 13, 2000):
"The starting point for a reconsideration of the hypothesis is the recent examination of the text of the PA by Robinson, which has identified at least 10 distinct text-forms..."
He semi-acknowledges this by calling the current paper a 'development' of his earlier position:
"This study ... develops the discussion ... in an earlier article (first given as a paper in the TC seminar of the SNTS General Meeting in Aberdeen, 2006.)"
He recaps the first article in three "points":
1. The (PA) is found today at John 7.53–8.11 ... There are many exegetes who maintain that the vocabulary and style of the PA do not correspond to the vocabulary and the style of the Gospel of John, although some continue to defend John’s authorship. 3"
3. See, for example, J. P. Heil, ‘The Story of the Adulteress (John 7,53–8,11) Reconsidered’, Bib 72 (1991) 182–92, and more recently M. A. Robinson, ‘Preliminary Observations...’, Filología Neotestamentaria 13 (2000) 35–59:
‘The present writer holds to the theory of Byzantine-priority and considers the PA original to John on internal, structural, and external text-critical grounds’ (36 n. 2).
The wording here is most unfortunate, implying possibly that the "some" who continue to defend John's authorship are also among the "many who maintain that the vocabulary and style...do not correspond to...the Gospel of John". In fact both Heil and Robinson (Rius-Camp's examples) disagree with this. They both find plenty of evidence for Johannine authorship, and so reject Lukan origin with supporting evidence.
Its clear that Robinson's arguments and evidence are weighty, since Rius-Camps has been forced to modify his position on Lukan origin after reading them (7 years later...).
Rius-Camps continues his recap:
2. A transitional statement (John 7.53–8.1) ...demonstrates that:
(a) the PA ... was originally attached to a gospel text...
(b) the characters of the preceding pericope are different from those in the PA,... ‘the scribes and the Pharisees’ (John 8.3) and ‘the elders’ (8.9)...
c) in the proposed location following Luke 20.19, ... ‘everyone went to his own home’ (John 7.53), while ‘Jesus went to the Mount of Olives’ (8.1).
3. From the transitional statement it further emerges that:
(a) the second attack [i.e., Luke 20:20-26] also took place in the Temple but on the following day ...(John 8.2a)
(b) once the PA was removed ... the 2nd attack (the 3rd one if the PA is included) would have taken place on the same day as the previous one according to what is read now in Luke at Luke 20.20.
If this summarizes Rius-Camps' earlier argument for Lukan authorship, one can understand why it was not very convincing. Luke 20:19 appears to be a plausible place for the PA, but there is no textual evidence whatever that the PA ever had this position in any copy of Luke at any time.
Speculative Literary Criticism...
It is now clear that despite appearances, Rius-Camps is not engaging in any form of textual criticism (i.e., restoration of the Gospel text using existing evidence). Instead he is engaging in speculative Literary Criticism, seeking to discover hypthetical forms and documents behind the Gospels as we have them, and as they are found in the most ancient evidence. His 'Conclusion' underlines the fact that Rius-Camps is operating outside the box of textual criticism:
"(1) The new hypothesis explored here may be stated as follows: the PA originally would have been part of the Gospel of Mark and would have been situated after the first attack by the High Priests, the scribes and the elders, questioning the authority of Jesus (Mark 11.27–12.12).
(2) Luke would have adopted it in his own work and would likewise have placed it after the first conflict of Jesus with the same Jewish leaders mentioned in Mark (Luke 20.1–19).
(3) Because of the moral strictness that prevailed at the end of the 1st century, the PA would have been eradicated together with the end of the preceding pericope from both Mark and Luke."
All of Rius-Camps' three stages are wholly fictional and take place before the existance of the earliest known MSS or fragment (i.e., 100 A.D.).
Authorship by Mark?
The first bit, authorship by Mark, is the 'new hypothesis', extending his theory of Lukan authorship, and it is meant to explain the non-Lukan form of the oldest known text(s) of the PA, i.e., Papias (c. 120 A.D.), Didymus (c. 340 A.D.), and Codex Bezae - D, d (5th cent.).
By contrast, the supposed 'Lukan' form identified by Rius-Camps, (Family 13) only exists in 9 miniscules, all late (10th -14th cent. A.D.), a text which can only be traced back with confidence to the 9th century.
The existing evidence of an earlier non-Lukan form seems to force Rius-Camps to acknowledge his 'Lukan' text as secondary to the one found in Codex D (now identified by him as PA Mark ):
"Luke would have adopted it in his own work..."
But in doing this, and attributing the Family 13 text to Luke himself, Rius-Camps tries to secure for this late 9th century text an existance in the First century and an authorship of the highest prestige, namely Luke himself. The Family 13 text now becomes identified as PA Luke.
Before we go any further however, we have to note a glaring omission: The Markan Origin of the PA is not a "new hypothesis" at all. This was first proposed by Hitzig among the Germans, restated by Holtzmann, and duly noted Meyer, long before 1875:
"Hitzig (on John Mark, p. 205 ff.) regards the evangelist Mark as the author, in whose Gospel it is said to have stood after xii. 17 (according to Holtzmann, in the primary Mark [i.e., "Ur-Mark"]).
- H. A. W. Meyer,
Critical and Exegetical Commentary on the NT,
Vol II John, (1875)
It is rather inexplicable that Rius-Camps would propose the idea, without knowing that it has been discussed regularly for over 100 years; and his failure to alert the reader of this and point to the relevant literature, is dumbfounding.
The existance of the early version in Codex Bezae (D), brought to Rius-Camps' attention in Dr. M. Robinson's (2000) article caused him to abandon the simple Lukan Origin Theory within a year (2007), and incorporate the theory of an earlier Markan Form for the PA, PA Mark.
But Rius-Camps only references Robinson superficially, and does not deal with his collation of 10 different forms for the PA, nor does he take seriously Dr. Robinson's opinion of its nature and origin. Rius-Camps essentially ignores the detailed discussion by Robinson, who is probably the world's leading living expert on the PA. Instead, he repeatedly quotes amateur textual critics like W. Willker, a chemist who posts textual-critical articles regularly on the internet, as if they had equal expertise and authority.
The ad hoc addition of a PA Mark to explain the ancient Bezae text and salvage whats left of his Lukan theory is simply clumsy and naive. Waffling over his own theory gives us no confidence whatever in any of his previous work.
The new 'Dual Authorship' theory appears to be a fantasy, having no real explanatory power to deal with the textual or internal evidence.
The simplification of the situation into two rival text-types is a serious error, since Rius-Camps has not chosen text-types appropriately nor informed the reader of the real historical situation.
(1) The Codex Bezae text is not a real text-type. The fact that its text is supported by two late miniscules (2722 [10th] and 1071 [12th]) does not by any means prove the circulation of a text-type. Most textual critics believe that these late manuscripts simply copied Codex Bezae here, in order to insert the story. They are not believed to be independent witnesses to this text. The only other support for the Bezae text is of course the Latin side of this bilingual MS (d), and a handful of Latin MSS with some similar readings.
(2) Von Soden (1911) found two different rival texts. One was the μ5 text, equal to the Textus Receptus version found in the majority of Byzantine MSS. The other was the μ6 text, essentially the Lectionary version of the story, found in almost equal numbers. A remaining large number of MSS contained a mixture of these two texts (μ7). These two texts competed throughout the Middle Ages, while the Bezae text was nowhere to be found, abandoned to obscurity.
(3) The passage is never found at Luke 20:19. The supposed 'Lukan' version in Family 13 is found only at Luke 21:38. This is just dismissed with the conjecture that it was uniformly added again later at the wrong place: '... communities that used Luke as their gospel would have reinserted it in the Gospel of Luke, but not in the place it originally occupied, after the first attack, but at the end of the long debate between the Jewish leaders against Jesus, after Luke 21.38.'
(4) The PA's position at John 7:52 dominates every textual transmission path that includes it. No explanation for this is given by Rius-Camps. He merely states '...the PA would have been inserted in different places of the Gospel of John or the Gospel of Luke.' ...but why? '...Most of the communities that decided to reinsert PA would have incorporated it in John as an illustration of the answer of Nicodemus to the Pharisees'... But if they knew it was authentic, why wouldn't they insert it where it belonged? If they didn't know it came from a gospel or where it belonged, why insert it at all?
There is no comparable case in any other NT variant. There is no example of inserting a story of unknown origin into any gospel 100 years after they were in circulation, nor any justification for it. The Ending of Mark is no example, because presumably in that case Mark's original ending was unfinished or lost, and the book required suitable completion.
Rius-Camps has both ignored the most common text entirely (μ5), and failed to account for the real historical competition between rival texts that actually occurred. His hypothetical theory of two 1st century texts, PA Mark and PA Luke completely fails to account for the texts as they really exist or the places where the passage is actually found.
Above all, Rius-Camps ignores the overwhelming textual evidence for its location in John, where it has been known to rest since the Jerome's day (c. 392 A.D.). There is no pre-400 evidence of the PA ever being located anywhere else, excepting a vague report by Eusebius about Papias in the early 2nd century. That reference may not even refer to the PA, and Eusebius may be inaccurate here, since apparently his own text omitted it entirely (cf. א/B).
In summary, although he spends some 7 pages (¶ 2. The Pericope, pg 9-16) discussing textual variants, Rius-Camps performs no textual criticism at all in regard to explaining the extant textual evidence itself as it stands. He is entirely caught up sifting for details he hopes can be made to fit his pre-100 A.D. theory of two "Ur-Gospels", Ur-Mark and Ur-Luke, each containing a different form of the PA.
Turning to the internal evidence for PA Mark and PA Luke, we find a similar and all too familiar failure.
Rius-Camps spends significant effort attempting to demonstrate that the PA has a Chiastic Structure, as follows:
|a) 8:2||time, location, characters|
|b) 8:3-6a||new chars., 2nd attack|
|c) 8:6b||reverse action of Moses, writing|
|d) 8:7||Pronouncement in contrast to Moses|
|e) 8:8||writing as Moses did 2nd time|
|e') 8:9a||Personal exodus for each accuser|
|d') 8:9b||Jesus alone with woman|
|c') 8:10||Jesus stands up again questions her|
|b') 8:11a||woman confirms accusers gone|
|a') 8:11b||final ammesty and commandment|
Ignoring the validity of this claim of a chiastic structure for the moment, Rius-Camps seems completely oblivious to the glaring fact that neither Mark nor Luke make use of such chiastic structures, and certainly not in the abstract, doubtful style illustrated here.
On the other hand, John does use chiastic structure extensively, and Rius-Camps seems to have no clue as to the devastating consequences a successful demonstration of chiastic structure would have on his own theory of PA Mark and PA Luke. At the same time, it would be yet one more striking piece of evidence for Johannine authorship.
As it turns out, it matters not that Rius-Camps' own attempt to find chiasm is unconvincing, because many far more convincing chiastic structures have already been discovered in John involving the PA.
Chiasm in John < - - Click here for more info.
Once again, all this effort by Rius-Camps seems to suggest a rush to press before considering the meaning or even direction his own investigations are taking. Sadly, this conclusion is consistent with his publishing timetable.
In spite of the extensive discussion of vocabulary, syntax and style, Rius-Camps has nothing substantial or new to offer in this area. Like so many other recent critics, he commits the following crippling errors, negating the value of all that effort:
(1) He ignores all the advances and developments of the last 100 years, and simply perpetuates 19th century methods and conclusions discredited a over century ago.
(2) He doesn't handle the textual variants objectively, but interprets them in favour of his assumptions, and ignores conflicting evidence.
Sadly, both faults are commonplace in Textual Criticism to this day, so the performance is not surprising.
Regarding Advances in the field, the reader can review these keypoints below:
(1) Up to the 1820s, "Internal Evidence" meant analysis for historical plausibility, and arguments were composed for and against the authenticity of the PA based on this.
(2) In the 1820s the focus shifted to authorship. Critics began to see that the rest of John and other gospels had equivalent problems regarding historical plausibility and verifiability, and so this previous line of inquiry was gradually abandoned in regard to the PA: Historical plausibility was recognised as irrelevant to the question of authorship and authenticity.
(3) From 1820-1880 Critics investigated vocabulary, syntax and style, and "Internal Evidence" began to be equated with this, rather than with historical questions. At the same time, the 'apologetic' approach was being abandoned in favour of a 'historical-critical method' that prefered skepticism to dogmatics.
(4) By 1850, critics were openly dismissing vocabulary arguments.
For instance, when S. Davidson (1848) presented a list of non-Johannine words in the PA, Tregelles dismissed them as unimportant:
"...the internal difficulties...if [the PA] had been sufficiently attested [by MS evidence], ...would not present anything insurmountable. ...Perhaps the difficulties have been over-estimated." .
Similarly, Alford's non-Johannine vocabulary list (GNT, 1863, pp.777f) was found nearly valueless by Nicolson :
"I feel bound to admit that the force of the internal evidence has been greatly overrated. ...Reviewing these 15 items...Only 3 are left... I think that these... do afford a presumption against Johannine authorship, though to each of them there is some sort of answer..."
(5) For most of the 19th century, real linguistic knowledge was lacking, and results were known to be unreliable. No statistical analysis of language was conceivable, and the key resources (such as the vast numbers of early papyri) had not yet been discovered and published. NT Greek was still viewed as just "classical Greek" containing religious language. Any oddities were assumed to be religious jargon or Semitic style. Opinions of 'style' were based on subjective personal impressions, instead of analysis.
(6) In 1882 Westcott & Hort publish their Greek NT. The Canons of TC had accumulated until the 1830s, accepted by rationalists favouring the historical-critical method. But they were not organized into a heirarchical system, and critics chose eclectically when principles or observations conflicted. Hort did little for 'Internal Evidence' except divide variants (and Canons) into two categories: (a) Transcriptional - errors copyists would have created, and (b) Intrinsic what the original authors would have written. It is clear that although he failed to describe and promote a heirarchical system, he in fact followed one of his own:
(1) Use only the oldest manuscripts (= the 'best witnesses'), namely א, B: i.e., rate them above all other textual evidence.
(2) Where א, B agree, follow them, with or without support.
(3) Where א, B disagree, prefer B.
(4) Prefer the Shorter Reading: Consistently, even if א, B stand alone. If B alone omits, then omit, and if א alone omits, consider it.
(5) Ignore Internal evidence, except to use it in support of the selected reading. That is, use all means to argue in favour of the reading already chosen.
Hort's 'Canons' naturally caused the deletion of the PA, with his chosen 'Textual evidence' overriding all Internal considerations. Hort did however also use some 'Internal' arguments to support this choice. He did not however offer an origin for the PA, but only classed it as non-Johannine. The Revised Version (1882) followed Hort, demoting it with brackets for the first time in a printed Bible.
(7) In the 1890s the Papyri are discovered in Egypt. The first large finds began in 1877, landing in Berlin, Oxford, and private collectors. In 1880 Graf acquired 10,000 more papyri and cloth, and in 1884 another hoard, which went to Vienna. Discoveries continued till 1894. Then Grenfell & Hunt found more in 1895/6. Until 1895 no real parallel texts to NT Greek were available. Some thought it was invented by the Holy Spirit. Beginning in 1898-1922, the first NT papyri were published in 15 volumes (P1 - P39)
(8) In 1895 A. Deissmann noticed that the papyri were quite similar to NT Greek. He concluded that NT Greek was the common language of the day. Deissmann published in 1908-23 (Eng. 1910/24). Huge advances began in NT Greek vocabulary and grammar. These too would negate mountains of previous guesswork. Still linguistic advance lagged woefully behind archaeological discoveries.
(9) 'Internal Evidence' was seen as even more precarious and subjective, and was further downgraded as too conjectural, and of far less significance than "hard" textual evidence.
(10) The first thorough examination of John's Style was published in 1905, Johannine Vocabulary by Edwin Abbott, followed by his Johannine Grammar (1906). The first attempt to describe the vocabulary of the Synoptics was Horae Synopticae (1909) by J. Hawkins. No credible assessment of Synoptic 'Style' was forthcoming until Nigel Turner's slim vol.IV of A Grammar of NT Greek (1976)!
All this new evidence demanded a complete overhaul of earlier 'Internal Evidence' arguments against the PA. But this sadly never happened.
(11) Next, Critics faced up to the 'Lukan' authorship theory. A closer examination by E.Buckley (1912) showed the evidence was conflicting and ambiguous. Similarities were divided among the Synoptics.
H. Cadbury (1917) attempted to shore up the case for Lukan authorship, but exposed a new problem: If the PA really originally belonged to Luke, then all early manuscripts were equally corrupt: If א, B incorrectly omitted it from Luke, then why believe them in omitting it from John? Lukan authorship would have a devastating effect on the credibility of א, B. But Cadbury's work had an even deeper methodological flaw: He doesn't list the many Johannine examples of words, grammar, and textual parallels found in the PA. He merely denies them: "No other NT writing has such close parallels as those given from Luke and Acts.". But the Johannine parallels do rival those in Luke/Acts, and it is also possible that those in Luke/Acts are a result of familiarity with John or his sources.
Ignoring this, H. McLachlan (1920) pressed on with the most thorough argument for Lukan authorship yet, also reconstructing his own text; - but few were convinced. It was too convenient, contrived, and too obviously extreme in its effort. Most dismissed the idea on textual grounds.
(12) In the 1890s another advance in Internal (Linguistic) criticism came. The Hebraists, Aramaists and Syriac experts got their hands on newly discovered early Syriac texts: SyrCur (1858), Dalman's Grammar (1894), SyrPal (1899). They began investigating and reconstructing Hebrew and Syriac versions of NT sayings. It would take another 20 years however, to work out key methodological problems in that task.
(13) Meanwhile, for the next 30 years, Greek NT scholars largely ignored Aramaic/Hebrew, in discussions of NT grammar and style. Maurice Casey describes the scene:
"Most [NT] scholars left Aramaic out. For example, ... Bultmann (1921) gives only occasional mention. ... Streeter (1924) virtually omitted the Aramaic dimension from Mark and Q: C.H.Turner (1927) omitted it from most of his studies of Markan style... Zerwick (1993) almost omitted Aramaic from a whole book on Markan style."
Yet there were some lasting results from attempts to reconstruct underlying Hebrew/Aramaic:
(a) Marshall (1891) only worked with single words. (Casey, ibid., p.8) Dalman (1898) unfortunately made the same mistake, using single words when what was needed was whole passages. That technique could never uncover written Aramaic sources (Casey, ibid., p.16,17). Wellhausen (1903-1911) also confined himself to single words, a very conjectural process (Casey, ibid., p.26). Torrey (1923-37) fell into the same error of working with one or two words. Other NT scholars made the same unconvincing suggestions, such as C.H. Dodd (Casey, ibid., p.27).
(b) S.R. Driver rightly demanded that Marshall deal with whole sentences. Meyer (1896) met the challenge by reconstructing whole Aramaic sentences, located in a context, showing phrases as they must appear in Aramaic: Only whole sentences can do that (Casey, ibid., p.12). Burney (1922) also offered complete Aramaic constructions, but suffered from errors in method. Finally, M. Black (1946) compiled the best of previous work, along with correct principles, and followed Driver in demanding whole sentences (Casey, ibid., p.30).
(c) This insistence on longer phrases, clauses and sentences for the purpose of establishing almost anything, from the text underlying a translation to an author's personal style, is probably the single most important breakthrough which experts in Hebrew, Aramaic and Syriac made. It forced a rapid improvement in methodology which removed many serious errors, conjectures, and unsupported claims.
Unfortunately, Greek NT scholars lagged behind, and did not pick up on this important requirement until years later. Regarding the PA, the same old vocabulary lists collected in the 19th century continue to be repeated as if they were credible evidence for an author's vocabulary, diction and style, to this very day.
If this new requirement for methodology weren't enough to invalidate the 'Internal Evidence' regarding the PA, the next development in this field kicked the whole foundation out from under it:
(14) In 1950, Richard Heard Exposed the true significance of E. Schweizer's (1939) work re: the PA.
"It has been shown, however, that there are a number of passages in John's gospel where the ‘Johannine’ characteristics of style, although not entirely absent, are relatively scarce:
These following passages are all narratives of a synoptic type and include:
the Miracle at Cana (2:1-11),
the Cleansing of the Temple (2:14-16),
the Healing of the Nobleman’s Son (4:46-53)’
the Anointing at Bethany (12:1-8) and the
the Triumphal Entry into Jerusalem (12:12-15);
It is at least possible that the evangelist was here using a written source or oral tradition that had become comparatively ‘fixed’ in form.' "
Thus any demonstration that the PA lacked significant Johannine features is completely undermined by the fact that so many other passages in John are similarly barren of the same features, but are not in dispute on textual or other grounds, real or alleged.
The final objection to Rius-Camps' Markan/Lukan theory is his handling of the textual variants of the passage. Instead of acknowledging the problem of 'Lukan' readings in Codex Bezae and 'Markan' readings in Family 13, he just dismisses them, without acknowledging their significance in terms of the state of either text.
"απο του νυν is found elsewhere only in Luke and almost always in the mouth of Jesus (Luke 1.48; 5.10 B; 12.52; 22.18, 69; Acts 18.6 B). However, since outside the NT it is a common time phrase that cannot be associated with any particular writer (in the lxx it appears 29 x [15x in the Apocrypha]), its presence here in Codex Bezae and its absence in ƒ13 cannot be taken as a typically Lukan mark." (p.17/395)
" 'with his finger wrote' (εγραψεν ƒ13 pc | εγραφεν K U L ƒ1 28. 118. 700. 1424mg pm) on the ground’. It looks like Luke has adopted the same phrasing as Mark, although changing the imperfect κατεγραφεν for the aorist of the simple verb."
How easily the ambiguous and problem readings are dismissed! A more unbiased presentation would note that neither of the chosen extant texts is sufficiently pure as an example of either Mark or Luke in style.