MAP — Modular Aggregation of Principles for Bible Translation

This compilation is presented without comment by the compiler. Dr. Bruce Metzger and Dr. Maurice Robinson, respectively representing an Alexandrian outlook and a Byzantine outlook, explain not to each other, but to us their readers, the internal and external evidence supporting their conclusions. You be the judge. The passages were chosen at random constrained only by where the scholars happen to have commented on the same passage. Four passages from each of Matthew, Mark, Luke, John, Acts give us a very interesting overview of two very different approaches to textual criticism.

Extracted text from ROBINSON_COMMENTARY_GNT_vs_METZGER_TEXTUAL_COMMENTARY.pdf:

The following table displays Maurice Robinsons Textual Commentary of the Greek New Testament, on the left, against Bruce Metzgers A Textual Commentary of the Greek New Testament for four variant sets in each of Matthew, Mark, Luke, John, and Acts. This is a randomly picked sample (subject to both authors having commented on the verses in question) to give the reader an opportunity to see two different approaches to weighing the external and internal evidence of the manuscript data. The reader needs to understand that Dr. Metzger is a champion of the Alexandrian manuscripts for presuppositional reasons, while Dr. Robinson is that for the Byzantine manuscripts based on the evidence. Bruce M. Metzger, A Textual Commentary on the Greek New Testament, Second Edition, 1994 Deutsche Bibelgesellschaft, Stuttgart. Used by permission; Maurice Robinson, Textual Commentary of the Greek New Testament, 2018 by Maurice Robinson. Used by permission. Compiled by Timothy Friberg, June 2018

ROBINSON TEXTUAL COMMENTARY GNT

METZGER TEXTUAL COMMENTARY GNT

MATTHEW 5.441 kai proseucesqe uper twn txt = B f1 pc k sys,c sa bopt Theoph Irlat vid Or Cyp 2 = eulogeite tou= katarwmenou= uma=, kalw= poieite tou= misousin uma=, kai proseucesqe uper twn ephreazontwn uma= kai

MATTHEW 5.44 (bis) umw/n kai. proseu,cesqe upe.r tw/n diwko,ntwn u`ma/j {A} Later witnesses enrich the text by incorporating clauses from the parallel account in Lk 6.27-28. If the clauses were originally present in Matthews account of the Sermon on the Mount, their omission in early representatives of the Alexandrian (a B), Western (itk Irenaeuslat Cyprian), Eastern (syrc, s), and Egyptian (copsa, bo) witnesses would be entirely unaccountable. The divergence of readings among the added clauses likewise speaks against their originality.

NA: But I myself say to pyou, plove pyour enemies; and ppray on behalf of the ones persecuting pyou. : But I myself say to pyou, plove pyour enemies; pbless the ones cursing pyou; pdo nobly to the ones hating pyou; and ppray on behalf of the ones despitefully using pyou and persecuting pyou.

The NA27 reading obviously presents a far shorter rendering of the passage than appears in the text. The primary question has to be whether it is more likely that such a substantial amount of material was omitted (either accidentally or intentionally), or whether a significant amount of material was deliberately inserted into the narrative. The parallel Lk 6.27-28 offers little help, since the order of the various elements differ dramatically when compared with what appears in the Matthean 5:44 reading (identical words underlined, even when the order differs): Mt = egw de legw umin: agapate tou= ecqrou= umwn, Lk = all umin legw toi= akouousin, agapate tou= ecqrou= umwn, Mt = eulogeite tou= katarwmenou= uma=, kalw= poieite toi= misousin uma=, kai Several minority support variations of the lengthy reading exist. These include the following: eulogeite tou= katarwmenou= umin, kalw= poieite tou= misousin uma=, kai proseucesqe uper twn ephreazontwn kai D* pc; eulogeite tou= katarwmenou= uma=, kalw= poieite tou= misousin uma=, kai proseucesqe uper twn ephreazontwn kai Dc pc Euspt; kalw= poieite tou= misousin uma=, kai proseucesqe uper twn ephreazontwn uma= kai 1230 1242* pc lat Euspt; eulogeite tou= katarwmenou= uma=, kai proseucesqe uper twn ephreazontwn uma= kai 1071 pc Cl Euspt; and eulogeite tou= katarwmenou= uma=, kalw= poieite tou= misousin uma=, proseucesqe uper twn ephreazontwn uma= kai W Euspt. 1

Any text enclosed in square brackets [ ] reflects the designation of such word(s) within the NA/UBS main text.

Lk = kalw= poieite toi= misousin uma=, eulogeite tou= katarwmenou= umin, Mt = proseucesqe uper twn ephreazontwn uma= kai diwkontwn uma=. Lk = proseucesqe uper twn ephreazontwn uma=. Aside from the differing introductions of each passage, that in Matthew not only has an opposite order of the second and third clauses from that of Luke, but other subtle differences (uma= for umin in the katarwmenou= clause; a kai preceding the proseucesqe clause that is not present in Lk; and the major inclusion of kai diwkontwn uma=, not even present in Lk). Had the NA27 text been original and the reading a harmonization, one would have expected the order of clauses to be identical, and without the minor alterations and major expansion found in the Matthean portion. In comparison, the overall NA27 main text reading appears to be a deliberate abridgment, totally removing the katarwmenou= and misousin clauses, and from the final portion eliminating the ephreazontwn clause (omissions indicated by brackets): Mt = egw de legw umin: agapate tou= ecqrou= umwn, [. . . ] kai proseucesqe uper twn

diwkontwn uma= [ . . . ].

It is not very likely that all these omissions could have resulted from accidental scribal error. 3 Equally, it is unlikely that the scribes involved with the reading would have intentionally harmonized in a quite non-parallel manner. Thus, a primary conclusion would appear to be intentional editorial alteration by the scribes or the scribal archetypes of the three named MSS primarily responsible for the NA 27 main text reading ( B f1). Given that a certain degree of versional support (primarily Syriac and Coptic) also exists for the NA 27 reading, this suggests that any such editorial alteration might relate to the time when recensional activity related to the production of such versions was active; this particularly when the same reading also appears in a very limited number of Greek and Latin fathers. Overall, there really is no good reason to reject the reading in this variant as an interpolation or harmonization, particularly in view of the extremely limited support among all other Greek MSS for the NA27 main text reading.

MATTHEW 8.284 Gadarhnwn = Gergeshnwn

txt = B C al sy

s,p,h

Epiph

MATTHEW 8.28 Gadarhnw/n {C} The healing of the demoniacs is recounted by all three Synoptic Gospels, and in each account there are three

One could suggest that (perhaps!) the archetype underlying the NA27 main text may have had the katarwmenou= and misousin clauses in the Lukan order in which case accidental omission of both clauses by haplograpy of kalw= kai might have occurred. Such would not, however, account for the omission of the ephreazontwn clause later in the verse. 3

Other readings include the harmonizing Gerashnwn, in 892c latt syhmg sa mae; and Garadhnwn, in D.

NA: And he having come into the other side, into the region of the Gadarenes, two demon-possessed ones met him, exiting out of the tombs, exceedingly fierce so that no one had strength to pass by through that road. : And he having come into the other side, into the region of the Gergesenes, two demon-possessed ones met him, exiting out of the tombs, exceedingly fierce so that no one had strength to pass by through that road.

The present unit and its Synoptic parallels reflects great confusion in relation to the name of the people and/or region involved. Even in the present unit, further differentiations of the term occur in isolated MSS. In fact, the situation becomes far more complicated when the main texts of NA27 and in the present unit are considered in relation to their parallel accounts elsewhere. The details can be shown clearly in the following table: PARALLEL LOCATION TABLE WITH VARIANTS Reference

NA27/WH

Mt 8.28

Gadarhnwn

Gergeshnwn

Witnesses:

B C Q al sys,p,h Epiph

LW (syhmg) bo Epiphmss

Mk 5.1

Gerashnwn

Gadarhnwn

Witnesses:

[2427vid]

latt

f1 f13

f13

syp,h

Gerashnwn

Lk 8.26 Witnesses:

Lk 8.37 Witnesses:

B D latt

syhmg

Gadarhnwn (sa)

Gerashnwn 75

B C* D 0279 579 pc latt (sa)

AWY

f13

Gadarhnwn

A W Y sy

Variants

Misspellings

Gerashnwn

Gazarhnwn * Garadhnwn D

892c latt

syhmg

sa mae

Gergeshnwn 2

Gergusthnwn W

L D Q 28 33 565 579 700 892 1241 1424 2542 al sys bo f1

Gergeshnwn L Q X f1 33 579 700* 1241 pc (bo) Epiph

Gergeshnwn *,c

Gergashnwn C2

LPQ 33 700* 1241 al (bo) f1 f13

Obviously, matters would be far simpler were the reading Gadarenes present in each case (although not so here in Mt) and if the NA27 reading were Gerasenes in each case (but again, not here in Mt). It should further be observed that the dominant Alexandrian/Caesarean reading in Mark and Luke is actually Gergesenes (paralleling the reading in Mt) but the critical editors followed Westcott and Hort in giving primary preference to B in those parallels. The shifting of support for the NA27 reading in each instance also should be considered: such is a sign of general textual instability among the MSS supporting the critical text. Not only is B the only witness common to all four units in the critical text, but the level of support for the critical text readings is only 3 MSS in Mt, Mk, and the first

principal variant readings referring to the place at which the miracle occurred: Gadarhnw/n( Gerashnw/n, and Gergeshnw/n. The evidence of the chief witnesses for the three accounts is as follows: Gadarhnw/n

Gerashnw/n

Mt 8.28

(a*) B Ctxt (D) Q syrs, p, h

it vg copsa syrhmg 2

Mk 5.1

A C K 13 syrp, h

a* B D it vg copsa

Lk 8.26

A K W Dgr Y 13 syrc, s, p, h

75 B D it vg copsa

Gerasa was a city of the Decapolis (modern Jerash in Transjordan) located more than thirty miles to the southeast of the Sea of Galilee and, as Origen perceived (Commentary on John, v, 41 (24)), is the least likely of the three places. Another Decapolitan city was Gadara, about five miles southeast of the Sea of Galilee (modern Um Qeis). Although Origen also objected to Gadara (which, he says, was read by a few manuscripts) because neither lake nor overhanging banks were there, Josephus (Life, IX, 42) refers to Gadara as possessing territory which lay on the frontiers of Tiberias (= the Sea of Galilee). That this territory reached to the Sea may be inferred from the fact that ancient coins bearing the name Gadara often portray a ship. Origen prefers Gergesa, not because it occurs in manuscripts he is silent about this but on the dubious basis of local tradition (it is the place from which, it is pointed out, the swine were cast down by the demons) and of the still more dubious basis of etymology (the meaning of Gergesa is dwelling of those that have driven away, and thus the name contains a prophetic reference to the conduct shown the Savior by the citizens of those places, who besought him to depart out of their territory). Of the several variant readings the Committee preferred Gadarhnw/n on the basis of (a) what was taken to be superior external attestation ((a*) B Ctxt (D) Q syrs, p, h geo1 mss known to Origen al), and (b) the probability that Gergeshnw/n is a correction, perhaps proposed originally by Origen,17 and that Gerashnw/n (which is supported only by versional evidence) is a scribal assimilation to the prevailing text of Mark (5.1) and/or Luke (8.26, 37).

occurrence in Lk (the second Lukan occurrence garnered support from 6 named MSS somewhat mysteriously, since the additional witnesses themselves vary from their testimony in the other related parallels). Given that the proper names involved are somewhat similar (a factor that readily could cause scribal confusion), the inconcinnity between what appears in Mt as opposed to the general unanimity in Mk and Lk (whether one follows , the NA27 main text, or the dominant Alexandrian/Caesarean variants) remains problematic, not only for interpretation but also in regard to questions involving textual transmission. The one certain element is that the distinctive reading in Mt (regardless of text followed) differs from the readings that distinguish the parallel accounts of Mk and Lk. Why such occurred remains a mystery, but from a text-critical standpoint (as opposed to interpretative or exegetical), the distinctions remain solid within the respective forms of text involved. As for the geographical locations cited within the Decapolis territory, the following data appear to be secure: Gerasa is a city about 30-35 miles southeast of the Sea of Galilee. It is unlikely that its territory (cwra, Mt 8.28; Mk 5.1; Lk 8.26) would have extended to the Sea itself. Gadara is a city about 6 miles southeast of the Sea of Galilee, and its territory did likely extend to the Sea itself, since boats have been depicted on its coins. Gergesa, while technically an unknown location, may well be identified with modern El Kursi, somewhat northward on the east shore of the Sea of Galilee, in an area where a steep bank, cliffs, and caves are present. Aside from more the radical form-critical assumptions of an original story originally set well inland and later modified to a seaside location, the Gergesene and Gadarene locations appear far more likely to represent the correct site of the event, with those specific readings also being those occurring in as well as in the more dominant Alexandrian/Caesarean witnesses. These considerations severely call into question the extremely limited support for the alternative forms that appear in the critical text (again due primarily to an undue reliance on Codex Vaticanus, the only common link for all the critical text readings, as noted above). Should one enquire as to why the MSS supporting the critical text would have favored the more difficult and well-nigh impossible reading involving the distant inland Gerasa, the answer is simple: Gerasa was a wellknown city within the Decapolis, and a small number of scribes who may have been unfamiliar with actual geography but were unaware of any Gergesene location could easily have altered the text in Mt from Gergesenes to Gadarenes and (perhaps due to similarity of sound) altered the Mk/Lk Gadarenes into Gerasenes. Regardless of possibilities, the reading is commended as best on both external MS support as well as internal geographical grounds, with the Alexandrian/Caesarean dominant reading holding a clear second place and the NA27 critical text reading being by far the least likely.

MATTHEW 11.15 wta = wta akouein

MATTHEW txt = B D 700 pc k sys

11.15 w=ta {B} In view of the frequent occurrence elsewhere of the fuller expression w=ta avkou,ein (Mk 4.9, 23; 7.16; NA: The one having ears, let him hear. Lk 8.8; 14.35), it was to be expected that copyists would add the infinitive here (and in 13.9 and 43). : The one having ears to hear, let him hear. If the word had been present in the original text, This identical variant unit occurs twice elsewhere in Matthew (Mt 13.9, 43) with varying degrees of support there is no reason why it should have been deleted among the minority MSS. When the same phrase occurs elsewhere in the Gospels, it is always with akouein present in such important witnesses as B D 700 al. (Mk 4.9; 7.16 in ;5 Lk 8.8; 14.35. So also Mk 4.23, which retains akouein while varying slightly (ei ti= ecei wta akouein akouetw). In addition, the phrase with akouein present occurs in some lectionary-related minority variants at Mt 25.29 (C3 H 892mg pc); Lk 12.21 (U f13 579 892c al); and 21.4 (G f13 892mg al). In Revelation, however, a related phrase without akouein present occurs 8x, the first 7x as o ecwn ou= akousatw (Rev 2.7, 11, 17, 29; 3.6, 13, 22), and the final occurrence (Rev 13.9) as ei ti= ecei ou= akousatw). Thus, only in Matthew does the NA27 main text read The one having ears, let him hear (o ecwn wta, akouetw). Such occurs 3x in that book, each time with the support for exclusion of akouein being minimal. All remaining MSS in those locations read with and include akouein in the phrase The one having ears to hear, let him hear. These NA27 Matthean variant units where akouein is excluded can be charted quite simply, demonstrating their scattered levels of support: Verse Mt 11.15 Mt 13.9 Mt 13.43

0242

700

Other pc k sys a e ff1 k sys a b e k vgst

From this chart it can be seen that the only consistent witness for exclusion of akouein is Codex Vaticanus; the remaining exclusionary witnesses differ in various degrees. Add to this the overall slim level of support for exclusion of akouein two occurrences having only 3 MSS and one having 5 MSS and the likelihood of Matthean originality for the exclusionary reading becomes extremely slim; this is even more particularly the case when the remaining Alexandrian witnesses do not support such exclusion, and even the fluctuating support of the Syriac and Latin witnesses demonstrate a similar degree of sporadic support. Put simply, there really is no reason to follow the fluctuating slim support for the NA 27 reading in any of these three variant units. Yet one cannot presume a threefold accidental omission of the same word in the same phrase. Either the exclusion was original to Matthew, was intentional in all three variant units, or was based upon some sort of accidental error.

Had the phrase without akouein been original to Matthew, one would have expected to see far more MSS and versional witnesses supporting the exclusionary phrasing rather than the limited and fluctuating support present among the existing data, perhaps also with patristic comment regarding the difference between Matthew and the parallel Synoptic narratives. To presume that all remaining witnesses deliberately chose to harmonize the Matthean phrasing to what appears in Mark and Luke would be to ignore the known tenacity of scribes in preserving the text that lay before them, to ignore the lack of scribal harmonization in numerous other places where similar differences occur in parallel accounts, and to ignore the fact of Matthew being the most popular Synoptic Gospel in the early church, which would suggest the Markan and Lukan accounts would have been more likely to have been harmonized to the Matthean form than vice versa. On the other hand, had there been an intentional editorial decision to remove akouein and that only in Matthew the overriding question would be why such might have been done. Exclusion neither adds nor subtracts anything significant from the general expression, and it would be quite unlikely that the omission of akouein from the similar phrasing in Revelation would have exercised any real influence on the matter. In addition, had this been a deliberate editorial decision on the part of some archetypal Alexandrian or Western scribe or scribes, one then is hard pressed to establish why the parallels in Mark and Luke remained untouched. So this scenario is really highly unlikely to have occurred. Alternatively, it is possible that akouein was omitted accidentally in one of the three Matthean units (likely in an earlier archetype) due to a skip in the common portion of text from akouein to akouetw. If such occurred in a single Matthean location by accident, it could be possible that a subsequent scribe or editor could have presumed such as a Matthean norm, and thereby would intentionally correct the remaining two occurrences in that book by similar exclusion. In any case, the paucity and fluctuation of support for the NA 27 main text reading in the three units under consideration weighs strongly against the originality of that phrasing, and continues to support the originality of the reading in each instance. MATTHEW 11.196 twn ergwn = twn teknwn

MATTHEW txt = B* W pc syp,h sams bo Jermss

NA: The Son of Man came eating and drinking, and they say, sBehold, a man gluttonous and a wine drinker; a friend of tax collectors and of sinners. And the wisdom was justified from her works. : The Son of Man came eating and drinking, and they say, sBehold, a man gluttonous and a wine drinker; a friend of tax collectors and of sinners. And the wisdom was justified from her children.

5 6

NA27 omits the entire sentence, following B L D* 0274 28 [2427] samss bopt. Another reading is pantwn twn ergwn, in f13 pc (k).

11.19 avpo. tw/n e;rgwn {B} The Committee regarded the reading te,knwn (widely supported by B2 C D K L X D Q P and most minuscules) as having originated in scribal harmonization with the Lukan parallel (7.35).20 The readings with pa,ntwn represent further assimilation to the passage in Luke.

Obviously the variant terms ergwn and teknwn do not resemble each other except for the genitive plural ending. In addition, there is no close connection to either word in any form in the near context: forms of teknon only appear many verses earlier and later (Mt 10.21; 15.26), and the closest context for any form of ergon occurs only at Mt 11.2, but not in a pertinent passage.7 On the other hand, the actual parallel, Lk 7.35, does read teknwn as with the reading in Mt, and the NA27 apparatus in Mt suggests harmonization to the Lukan parallel.8 Yet, had harmonization on the part of occurred in relation to teknwn, one would expect a similar assimilation of the Lukan pantwn9 but such is notably absent in Matthew. The Lukan parallel reads in apo twn teknwn auth= pantwn (with differing variants that harmonize to varying degrees with Mt). That parallel thus remains distinctly different from the Matthean reading.10 Had harmonization actually occurred, the absence of pantwn in Mt would be quite peculiar. Were the MSS somewhat evenly divided, it would seem that the principle favoring the non-harmonizing reading would apply but in fact, the NA27 text actually represents the easier reading since it makes explicit (wisdom justified by its works) that which is only implicit and in fact figurative in the text (wisdom justified by its children). Additionally, scribes and only three named MSS are involved at this point11 familiar with the emphasis in James regarding faith demonstrated by works could well have altered the reading in Matthew on that basis; in particular the comments of Jas 2.21, 25 was not Abraham/Rahab justified by works? (ouk ex ergwn edikaiwqh;) in both verses of which edikaiwqh is utilized in a connection favorable to ergwn.12 That the same three scribes did not alter the Lukan parallel is not significant, given their known inconsistencies as well as the tendency to alter the more popular Matthew in such matters. 13

Beyond these data, the NA27 phrase apo twn ergwn elsewhere occurs only at Heb 4.10 in a totally different context, while the phrase apo twn teknwn occurs only at the Lk 7.35 parallel. Variations of word order exist in the Lukan parallel unit, but teknwn remains present in each transpositional instance except for 2, which transposes with ergwn. 9 Although NA27 in Lk 7.35 places pantwn before twn teknwn auth= while reads twn teknwn auth= pantwn, this does not affect the issue of non-harmonization regarding pantwn in Matthew. The Lukan unit also has several minor variants that harmonize to varying degrees with Matthew, but these are not relevant to the present situation. 10 Some witnesses in Lk partially agree with the Matthean semi-harmonizing reading in f13 (pantwn twn ergwn). 11 Two MSS ( B) are Alexandrian, deriving from a common ancestral archetype; the third MS (W) is usually Byzantine in Matthew, but departs from that Textform in the present instance. 12 That the passages from James might appear to contradict Rom 4.2 where the same terms appear within a hypothetical conditional phrase in a different context (ei gar Abraam ex ergwn edikaiwqh excei kauchma) is not a primary issue at this point. 13 One can also compare the more remote parallels where forms of ergon and dikaiow appear together: Rom 3.20,28; 4.2; Gal 2.16; Jas 2.24. Cf. also the LXX, only Sirach 16.27 (kai ouk exelipon apo twn ergwn autwn) and Sirach 32.22 (kai apo twn teknwn sou fulaxai). Neither ek twn ergwn nor ek twn teknwn occur in the LXX. 7 8

MARK 1.414 [o] baptizwn = baptizwn

txt = L pc bo

NA: John came to be [the one] baptizing in the wilderness and proclaiming a baptism of repentance with reference to the remission of sins. : John came to be baptizing in the wilderness and proclaiming a baptism of repentance with reference to the remission of sins.

MARK 1.4 @o`# bapti,zwn

evn th|/ evrh,mw| kai, {C}

In view of the predominant usage in the Synoptic Gospels of referring to John as the Baptist (o baptisth,j occurs in Mk 6.25 and 8.28, as well as seven times in Matthew and three times in Luke), it is easier to account for the addition than for the deletion of the definite article before bapti,zwn. The omission of kai, in a few Alexandrian witnesses is the result of taking o bapti,zwn as a title.

The inclusion of the definite article here affects translation somewhat. The article preceding baptizwn elsewhere occurs only twice: Mk 6.14, where Iwannh= o baptizwn links the name with his function, John, the one baptizing; and Jn 1.33, where Jesus is denoted by function as o baptizwn en pneumati agiw). Had a title been intended, the phrase Iwannh= o baptisth= would have been more suitable (cf. Mt 3.1; 14.2; Lk 7.20, 33) but such does not occur at all in Mark. Further, baptizwn without a preceding article occurs nowhere else in Mark, and only 4x in the remainder of the NT all Johannine and in a context referencing or quoting John the Baptist (Jn 1.28, 31; 3.23; 10.40). Beyond this, the Mk 1.4 construction with egeneto preceding the participle appears awkward in context, and is otherwise unparalleled in the NT. It appears far more likely in the present variant unit that the 3 named MSS (perhaps reflecting a common archetype) were influenced by the direct parallel passage in Mt 3.1, paraginetai Iwannh= o baptisth=, which includes the article before the noun title. This is more particularly the case when the remaining Alexandrian MSS as well as all other MSS of other texttypes do not include the article in the Markan unit. The article in the present unit is superfluous, not reflective of Synoptic style, and weakly supported; there is no reason in the present unit to reject the reading. MARK 2:15-1615 oi grammatei= twn Farisaiwn = oi grammatei= kai oi Farisaioi

txt = B W 28 pc

NA: And it occurs, while he reclined in his house, also many tax collectors and sinners were reclining together with Jesus and his disciples; for they were many and they were following him. And the scribes of the Pharisees . . . : And it occurred, while he reclined in his house, also many tax collectors and sinners were reclining together with Jesus and his disciples, for they were many; and they followed him. And the scribes and the Pharisees . . .

MARK 2.15-16 auvtw|./ (16) kai. ivdo,ntej {C}

oi` grammatei/j tw/n Farisai,wn

The more unusual expression oi grammatei/j tw/n Farisai,wn is to be preferred, since the tendency of scribes would have been to insert kai, after oi grammatei/j under the influence of the common expression the scribes and the Pharisees. Since in the Gospels the verb

Also including o ante baptizwn within the longer variant unit cited in the NA27 apparatus, B 33 [2427]. Other readings include oi de grammatei= kai oi Farisaioi, in S 700 a c e ff2 samss; kai eiden oi grammatei= kai oi Farisaioi, in D r1; kai grammatei= twn Farisaiwn kai, in L 33 pc b bomss; oi grammatei= twn Farisaiwn kai, in D [2427]; and kai oi grammatei= twn Farisaiwn kai, in 88vid. 14 15

avkolouqei/n is used of Jesus disciples, never of those who The issue here is translational as well as exegetical. Does Mark mention two groups (the scribes and the were hostile to him, a full stop should follow auvtw|./ Pharisees) or a specific subgroup (the scribes belonging to the Pharisee party)? Certainly, the NA 27 type of reading occurs nowhere else, thus reflecting a hapax phraseology and a more difficult reading. The difficulty of such, however, should not be considered determinative, particularly when the reading in question is attested by only three named MSS reflecting the Alexandrian and Caesarean types but without the support of the remaining MSS pertaining to those types.

Unmindful of this usage, copyists transferred the stop to follow polloi, and inserted kai, before ivdo,ntej.

More significantly, even in the otherwise common linkage of scribes and Pharisees is wholly unique within Mark; beyond this, the phrase reflects the less frequent (5x) articular form of both terms16 (the more common non-articular linkage of the two terms is found elsewhere 11x only in Matthew and Luke).17 Although the major parties within normative Judaism of that era (Pharisees and Sadducees) each might have scribes pertinent to their group responsible not merely for copying the scriptures but to provide biblical interpretation in accord with the specific beliefs of their particular party such distinction is not otherwise claimed or maintained within the NT. Rather, the scribes and Pharisees tend to be grouped together in clear opposition to Christs claims and teachings without further distinction (in certain situations 18 such as the rejection of resurrection concepts Sadducees are specifically cited, but scribes of any type are not particularly associated with their position). Further, the context involving meal fellowship with tax-collectors and sinners could have offended either leading party of Judaism, so it would appear unlikely that scribes belonging to the Pharisaic party would be singled out specifically. Of greater text-critical importance, Mk 2.18 associates two disparate groups (the disciples of John the Baptist and those of the Pharisees) in regard to an unrelated question about fasting. One rightly could presume that if two such groups with little otherwise in common are linked in Mark, the previous incident could reflect a similar situation. There thus is no need for the scribes of Mk 2.16 particularly to be associated with the Pharisaic party. Rather, the articular distinction in Mk 2.18 (oi maqhtai Iwannou kai oi twn Farisaiwn), occurring twice in that verse, quite likely provided an impetus by attraction to the immediate context for the 3 named MSS in the present unit to read oi grammatei= twn Farisaiwn. Once more, it appears that the 3 named MSS supporting the NA27 main text of the present variant unit reflect a secondary scribal development, here based on a primitive or archetypal scribal error tied to the close-context influence of the twofold twn Farisaiwn that appears in the incident immediately following.

The articular linkage of scribes and Pharisees occurs elsewhere only at Mt 23.2; Lk 5.21; 6.7; 11.53; Jn 8.3. The non-articular linkage of scribes and Pharisees occurs at Mt 5.20; 12.38; 15.1; 23.13, 14, 15, 23, 25, 27, 29; and Lk 11.44. 18 Mk 12.18 represents the only occurrence of Sadducees in that book, specifically in relation to their rejection of the resurrection concept (so also in the parallels Mt 22.23; Lk 20.27). 16 17

MARK 2:1619 esqiei = esqiei kai pinei

txt = B D W pc it [2427]

NA: . . .having seen that he eats with the sinners and tax collectors, were saying to his disciples that, With the tax collectors and sinners he eats? : . . .having seen him eating with the tax collectors and sinners, were saying to his disciples, Why is it that with the tax collectors and sinners he eats and he drinks?

MARK 2.16 evsqi,ei {B} The addition of kai. pi,nei is a natural accretion inserted by copyists, perhaps under the influence of the parallel passage in Lk 5.30. The shorter reading, which is strongly supported by B D W al, was followed by Matthew, who added o dida,skaloj umw/n (Mt 9.11), an expression that, in turn, was adopted in Mk 2.16 by the scribes of C L D 13 al.

Continuing the same verse, the specifics of the scribal/Pharisaic complaint are made clear: while the NA 27 main text relates that Jesus eats with the tax collectors and sinners, the text shows the complaint extended to involve drinking with those rejected groups. Yet earlier in the verse it was noted only that they had seen Jesus eating with them, with no mention of drinking at that point. This alone could provide sufficient motivation for the three named MSS to exclude the phrase kai pinei from their text, assuming deliberate excision (the alternative would have been to insert kai pinei in the earlier portion of the verse but it would be far easier to align the two statements by simple omission rather than creative addition). On the other hand, the longer reading at the close of the verse also could have been omitted by haplography of -ei -ei, which may be a more likely situation, given that the 3 named MSS are somewhat disparate in regard to texttype (Alexandrian and Caesarean). Significantly, the parallel Mt 9.11 in both NA27 and shows the complaint to involve only eating (dia ti meta twn telwnwn kai amartwlwn esqiei o didaskalo= umwn), which also may have served as an impetus toward the shorter reading in order to harmonize to the Matthean account. 20

MARK 2:2221 apollutai kai oi askoi = ekceitai kai oi askoi apolountai

txt = 88 B 892 [2427]

NA: And no one throws new wine into old wineskins; but if not, the wine will burst the wineskins, and the wine is destroyed and the wineskins. But new wine is into current wineskins. : And no one throws new wine into old wineskins; but if not, the new wine bursts the wineskins, and the wine is poured out and the wineskins will be destroyed. But new wine has a throwing into current wineskins.

MARK 2.22 avpo,llutai

kai. oi` avskoi, {C}

The reading which best explains the origin of the others is that preserved in B 892 copbo. Since the pendant kai. oi avskoi, seems to require a verb, most witnesses moved avpo,llutai (making it plural) after oi avskoi,. Furthermore, under the influence of the parallels in Mt 9.17 and Lk 5.37, copyists introduced the verb evkcei/tai as more appropriate than avpo,llutai to describe what happens to wine.

Other readings include hsqien, in 88 D 892 c bopt; esqion (!), in Q; also within extended transpositions relating to the longer variant unit: ~ esqionta post amartwlwn, in A; ~ esqiei post telwnwn, in 579. The remaining NA27/ parallel in Lk 5.30 retains both terms in the complaint, but the tense differs from what appears in Mark (dia ti meta twn telwnwn kai amartwlwn esqiete kai pinete). Additionally, in Matthew esqiei is expanded by kai pinei in some less significant MSS (346 565 1506) and rephrased to harmonize with Luke in sys as esqiete kai pinete. 21 Other readings include kai oi askoi apolountai, in D it; ekceitai kai oi askoi, in L; , ekceitai kai oi askoi apolluntai, in W; and ekceitai kai oi askoi apoluntai, in Q. 19 20

This variant unit involves two issues, the presence or absence of ekceitai and the tense and location of apollumi. The reading is more complex and also more difficult, particularly in regard to the use of the future tense apolountai in regard to the wineskins that had already burst. Had the NA 27 reading been original, one is hard pressed to determine why the quite acceptable present passive apollutai would have been deliberately changed to a peculiar future tense as part of the editorial alteration in . On the other hand, if the reading were original, the simplification is quite natural: eliminate the problematic twofold statement by dropping eckeitai and replacing it with apollumi repositioned and with tense altered to that originally present in the now-displaced reading. Also pertinent are the parallel passages in Mt and Lk both of which support the reading in Mk, but with no evidence that would suggest harmonization by in Mark: Mk 2.22b Mt 9.17b Lk 5.37b

kai o oino= ekceitai kai oi askoi apolountai kai o oino= ekceitai kai oi askoi apolluntai kai auto= ekcuqhsetai kai oi askoi apolountai

In fact, if the entire passage is compared in all three Synoptic Gospels, the level of non-harmonizing variance among them is significantly high, regardless of whether the or NA27 text is followed. Yet when considering the text in Mark, the common element appearing in relation to the present variant unit in the three parallels is that each has some form of pouring out (ekcew) relating to the wine followed by some form of destruction (apollumi) relating to the wineskins and this without either parallel retaining identical wording. Such in itself strongly suggests originality of the Markan reading rather than any attempt at harmonizing to those Matthean and Lukan parallels from a presumed original NA 27 main text reading. A further consideration involves the three named MSS supporting the NA27 main text reading: although these witnesses all reflect a portion of the localized Alexandrian texttype, it is more significant that the remainder of the Alexandrian MSS do not concur with that reading, but instead support the reading of another point strongly in favor of originality here, since it would be quite out of character for those remaining Alexandrian MSS to increase the length of the reading as well as to complicate such by not harmonizing to one or the other of the Synoptic parallels in the process. Once again, the weight of evidence points to originality in the present variant unit.

LUKE 3.3222 Sala = Salmwn

Another reading is Salman, in f1 f13.

txt = 4 * B sys sa bomss

LUKE 3.32 Sala, {B}

NA: of Jesse of Obed of Boaz of Sala of Nahshon : of Jesse of Obed of Boaz of Salmon of Nahshon

In the LXX (Rahlfs), Ruth 4.20-21, reads Salman twice (so also the variant spelling of f 1 f13 in the present unit), while Codex Alexandrinus in Ruth reads Salmwn).23 Yet in 1Chr 2.11, the LXX (Rahlfs) has Salmwn twice, while Codex Alexandrinus there reads Salman.24 More importantly, the phrasing in Chronicles is a near-duplicate of the Matthean genealogical statement (LXX kai Naasswn egennhsen ton Salmwn kai Salmwn egennhsen ton Boo=; Mt 1.4-5 Naasswn de egennhsen ton Salmwn: Salmwn de egennhsen ton Boe=/Booz). Overall, the spelling of Salmwn, appearing in Mt in NA27/ without stated variant, coupled with the strong evidence of the LXX, strongly supports the spelling of Salmwn in the present Lk 3.32 unit as opposed to the mongrelized Sala of NA27 and the three named MSS supporting such. LUKE 10.3825 auton = auton ei= ton oikon auth=

txt = 45 75 B sa

NA: But as they proceeded, he himself entered into a certain village. But a certain woman, by name Martha, welcomed him. : But it happened as they proceeded, even he himself entered into a certain village. But a certain woman, by name Martha, welcomed him into her house.

The original reading appears to be Sala, (4 * B syrs, pal copsa, bomss eth), which copyists later assimilated to Salmw,n, the reading of both the Matthean parallel (Mt 1.45) and the Septuagint of 1 Chr 2.11, or to Salma,n, the reading of ms. B at Ru 4.20 f. (Salmw,n, ms. A). In view of the early tradition that Luke was a Syrian of Antioch it is perhaps significant that the form Sala, appears to embody a Syriac tradition (the Peshitta version of Ru 4.20 f. reads ).

LUKE 10.38 auvto,n {B} No motive is apparent for the deletion of the phrase [received him] into her house if it were present in the text originally. On the other hand, the bold and bare upede,xato auvto,n seems to call for some appropriate addition, which copyists supplied in various forms, some introducing oivki,an, others oi=kon, and each with or without auvth/j, eauth/j, or auvtoi/j.

Critical text advocates likely would consider the inclusion of the final phrase in this variant unit as an explanatory clarification, since without that phrase the passage seems to indicate only that Martha welcomed Jesus to the village, and perhaps the food preparation and serving might take place in an outdoor venue (not that this should matter in the overall context). The precise form upedexato otherwise occurs only at Lk 19.5 in the NT in relation to Zacchaeus. That phrase has no direct mention of a house, although the previous verse records Jesus saying shmeron gar en tw oikw sou dei me meinai. Elsewhere in the NT the verb upodecomai occurs 2x elsewhere: Ac 17.7 and Jas 2.25 both without mention of any house. Regarding the phrase ei= ton oikon, it occurs 15x elsewhere in Lk and 5x in Ac. Of these, the phrase immediately follows the finite verb 11x in Lk and 3x in Ac, while a pronoun (in each case auton, as in the present unit) The Cambridge LXX in Ruth shows as minor readings the Saalmon and Salmw, as well as the Latin Salam. More significantly, that apparatus shows the LXX main text Salman as read by B and two other LXX MSS; the majority read Salmwn (!), even though that spelling is not preferred by the editors. 24 The Cambridge LXX in 1Chr also shows Salaman as a minor reading. 25 Other readings include ei= thn oikian auth=, in 1 C2; and ei= thn oikian, in 3vid C L X 33 579 pc. 23

appears between the verb and ei= ton oikon only at Lk 16.27; 22.54. This at least suggests a Lukan stylistic possibility, even while the more frequent Lukan usage might not include a pronoun. Had the NA27 main text reading been original, a question would remain as to why such an unnecessary explanatory phrase would be thought necessary, particularly when the text without the phrase obviously was quite sensible (as witness the critical text acceptance of such). On the other hand, had the reading been original, even though exact haplographic omission due to a same beginning or ending may not have occurred, line skippage based on the similarity of auton and auth= very well may have been the proximate cause of omission (the 16 letters involved are easily within a possible line length, as shown by the hypothetical setting below):26 upedexatoauton eistonoikonautou kaithdehnadelfh Added to this consideration is the further issue regarding external evidence, since even the remainder of the Alexandrian witnesses (3vid C L X 33 579 pc)27 read the variant ei= thn oikian; such argues very strongly against any originality of the omitted phrase. Given that only the three named MSS omit the phrase, it remains far more likely that accidental omission as described above (either independently or due to a common archetype) is the most likely reason for the reading of the NA 27 main text, particularly when there was no compelling need to the phrase to be added (even in differing form) by the remaining majority of witnesses. 28 LUKE 11.1129 kai anti = mh anti

txt = 45 75 B pc McionE

NA: But whom of youp being the father, the son will ask for a fish, and instead of a fish he will give over to him a serpent?

LUKE 11.11 kai.

avnti. ivcqu,oj {C} The reading with kai, (45, 75 B copsa al) preserves a Semitism that most copyists replaced with mh,, the usual Greek interrogative particle

: But whom of youp being the father, the son will ask for bread, he will not give over to him a stone? or even a fish, not instead of a fish he will give over to him a serpent?

As a whole this verse has significant differences throughout between the NA 27 main text and that of . The present variant unit merely happens to be the only location within this verse where the NA 27 text is supported Many other arrangements could be postulated with similar results. Note that the earliest of these witnesses is the 4th-century and not 3 (dated VI/VII in NA27). 28 Also to be noted in this verse are the other differences between and NA27: egeneto de en; NA27 en de (11 MSS support); and kai auto=; NA27 auto=(7 MSS support). In each case the NA27 main text also has the support of the same 3 MSS as in the present unit. 29 Another reading is the conflation mh kai anti, in G pc. 26 27

by only three named MSS, while the remaining units possess additional levels of support. 30 To put all into perspective, the complete verse according to each edition reads as follows, with the differences underlined: 31 NA27: :

tivna de ex umwn ton patera aithsei o uio= tivna de umwn ton patera aithsei o uio= arton, mh liqon epidwsei autw;

NA27: :

icqun, kai anti icquo= ofin autw epidwsei; h# kai icqun, mh anti icquo= ofin epidwsei autw;

Compared to the more significant variants in this verse, the kai anti versus mh anti of the present unit appears almost minor. The parallel in Mt 7.10 offers little assistance, since the clause in question lacks anti in the context (mh ofin epidwsei autw). In addition, neither kai anti nor mh anti otherwise appear in the NT, nor is any major point of syntax at issue since either reading could be acceptable. It is quite possible that accidental omission of the lengthy 28-letter phrase (arton mh liqon epidwsei autw h kai) could have occurred by line skippage as shown below, followed by the NA27 kai replacing mh as a matter of later damage repair: !p!raaithseiouios artonmhliqonep idwseiautwhkai ixqunmhantiixqu os If such accidental omission did occur, it then becomes important that the lengthy omission occurs in only 4 named MSS (45 75 B 1241 pc), the primary 3 of which are those involved in the present units minor adjustment of mh versus kai. It thus seems clear that the Alexandrian archetype of the 3 named MSS was involved in an erroneous omission, with slight contextual repair to smooth out the resultant reading. This scenario appears even stronger when one notes that other key Alexandrian witnesses ( C L 33 Y) even with minor variation read with at this point. There thus is no reason to depart from a presumed originality of the reading in the present unit, nor in regard to the longer omission (which is discussed similarly under the 4 named MSS category). LUKE 13.2732 erei legwn umin, = erei, Legw umin

txt = 75c B 892 pc

LUKE 13.27 evrei/

le,gwn u`mi/n {C}

Other variant units involved in this verse are supported respectively by 4 and 6 named MSS, and these will be dealt with under their respective categories. The presence of ex in NA27 but not in is not mentioned in the NA27 apparatus for that unit; rather, the NA27 main text reading with ex present is claimed to be that of undivided . Von Sodens apparatus, however, specifically states that the absence of ex is the Byzantine (bold K) reading; also, Tischendorf7 says ex is included by ABCDKLMRX al50, but excluded by EFGHSUVGDL al pl a situation that suggests at least a divided Byzantine reading and likely a majority that omit ex. 32 Other readings include erei umin, in 579 pc lat syp sa bopt; and erei solus, in 1195 pc bopt. 30 31

NA: And he will utter, saying to youp, I do not know [youp], from where youp are. pStand apart from me, all workers of unrighteousness. : And he will utter, I say to youp, I do not know youp, from where youp are. pStand apart from me, all the workers of the unrighteousness.

The reading adopted by the Committee, though narrowly attested, seems to account best for the origin of the other readings. The awkwardness of the participle le,gwn (which probably represents the construction of the Hebrew infinitive absolute: he will indeed say to you) would have prompted copyists either to alter it to the indicative (le,gw) or to omit it as superfluous.

The NA27 main text reading here is not merely a solecism, but so grammatically problematic as to be well-nigh impossible.33 This is more clearly evidenced by the fact that other Alexandrian MSS either read with (L Y) or simply omit the problematic legwn ( 579) the latter further bolstered by similar omissions in the Latin and Coptic versions. Most likely in an archetype of 75 B 892 the final Nu of legwn was unintentionally added to legw due to attraction from the following umin, with the initial reading of 75* representing that scribes unintentional failure to represent that archetype, with correction following. This particular reading of NA27 exemplifies the principle that even when the more difficult reading is to be preferred, sometimes the difficulty of a reading is so great that such could not have been original. 34 It would appear that almost any research unencumbered by an undue preference for Codex Vaticanus should have no difficulty in rejecting this ungrammatical solecism as the product of scribal error. The reading on all counts is clearly original in the present unit.35 JOHN 6.1436 o$ epoihsen shmeion = o$ epoihsen shmeion o Ihsou=

txt = D W pc it vg sy cop Did st

s,c

lem

NA: Therefore the men, having seen what sign he did, were saying, This is truly the Prophet, the one coming into the world. : Therefore the men, having seen what sign Jesus did, were saying, This is truly the Prophet, the one coming into the world.

JOHN 6.14 o]

evpoi,hsen shmei/on {B}

Although the combination of B ita in support of a] shmei/a is impressive, the plural seemed to the Committee to have arisen from scribal assimilation to 2.23 and 6.2. The addition of o` VIhsou/j was made by copyists in the interest of clarity.

Appeals in various textual and other commentaries to a presumed Lukan semitism intended to represent the Hebrew infinitive absolute appear extremely strained, particularly since such otherwise does not occur in either the Lukan writings or the remainder of the NT. 34 In this regard, cf. the NIV which despite following the NA27 text (cf. Mounces NIV Interlinear) makes no attempt to render the problematic phrase, but instead reads But he will reply, I dont know you or where you come from. Similarly, Schwandts ESV Interlinear presents the NA27 reading, but translates the passage as though it read like , But he will say, I tell you, I do not know where you come from. 35 Such also appears to be the judgment of I. Howard Marshall, The Gospel of Luke: A Commentary on the Greek Text, NIGTC (Grand Rapids: Eerdmans, 1978), 566-567: It may simply be due to a scribal error for legw; so also Darrell L. Bock, Luke 9:51-24:53, BECNT (Grand Rapids: Baker, 1996), 1241-1242, who by a slight margin prefers the reading since it could be favored by internal evidence. So also Martin M. Culy, Mikeal C. Parsons, and Joshua J. Stigall, Luke: A Handbook on the Greek Text, BHGNT (Waco: Baylor University Press, 2010), 467: Given the overwhelming textual support for the [] reading . . . , however, we are likely dealing with an early scribal error. 36 Other readings include a$ epoihsen shmeia, in 75 B 091 pc a; and o$ epoihsen o Ihsou= shmeion, in 1424. 33

The issue here involves whether Jesus as the clausal subject is specifically identified. The meaning is conveyed identically in either variation, and Jesus clearly remains the referent. As a point of detail, the 3 named MSS that here underlie the NA27 main text are further supported in regard to the specific omission of o Ihsou= by 75 B 091 pc a; their actual differences involving the wording of the preceding three-word phrase. The NA27 omission of o Ihsou= in the present unit, there fore, is in actuality a matter of support by 6 named MSS rather than 3 named MSS. The present commentary deals with the number of supporting MSS for the entire phraseology involved in a particular NA27 variant unit, thus making the present unit with 3 named MSS the basis for categorization, even though when broken down into component parts the two-word exclusion of o Ihsou= actually has 6 named MSS in support. As an initial observation specific to the preceding phrase, occurrences of o$ epoihsen are rare, whether accompanied by a clausal subject or not. The phrase with a subject present occurs at Lk 6.3 (o$ epoihsen Dauid) and Ac 14.11 (o$ epoihsen o Paulo=); the phrase without a subject occurs only in the parallels Mt 26.13/Mk 14.9 (o@ epoihsen auth/). If the plural phrase a$ epoihsen is considered (the Alexandrian alternative of 75 B 091 in the present unit), such occurs 3x in in John (Jn 4.45; 11.45, 46) and elsewhere only at Mt 21.15. Of these, a stated subject appears only at Jn 11.45, 46, with that subject in fact being o Ihsou=. NA27 has one additional occurrence of a$ epoihsen at Jn 21.25, again with the stated subject being o Ihsou=. Extending the inquiry to any finite verb following the singular relative o@, the expression with a stated subject (articular or anarthrous) following the verb occurs at Mt 8.4 (o$ prosetaxen Mwsh=); Jn 4.5 (o$ edwken Iakwb); 6.22 (o$ enebhsan oi maqhtai); 14.26 (o$ pemyei o pathr); Ac 5.32 (o$ edwken o qeo=); and Ac 10.37 (o$ ekhruxen Iwannh=). Yet in contrast, 83 other NT occurrences find the singular relative o@ followed by a finite verb but lacking a stated subject this by far being the more familiar expression. The latter also is more frequent even in John, with 12 additional occurrences of the relative o$ plus finite verb but lacking a subject.37 Similarly, extending the inquiry to any finite verb following the plural relative a$, a stated subject (articular or anarthrous) elsewhere follows the verb at 1Cor 2.9 (o qeo=); 10.20 (ta eqnh); Col .3.6 (h orgh tou qeou); and Re 10.4 (ai epta brontai). Yet, again in contrast, the plural relative a$ followed by a finite verb but lacking a stated subject occurs 50 other times in the NT once more making the relative clause without a stated subject the far more common and more familiar form of expression. It thus well could be the case that the exclusion of o Ihsou= may have been due either to (1) accidental omission of the short 3-letter nomen sacrum phrase (ois2), just as with many other short words or phrases; or (2) scribal omission based on the more frequent and thus more familiar construction in which a relative clause followed

Jn 3.11 (2x), 32; 4.14, 22; 6.39; 8.38 (2x); 13.27; 16.17, 18; 17.2, 4; 19.22. Also cf. Jn 6.37 (o$ didwsin moi o pathr) and Jn 18.11 (o$ dedwken moi o pathr), both with an intervening dative pronoun.

by a finite verb lacks a stated subject; or (3) intentional omission with a design of textual or syntactical smoothing by eliminating otherwise obvious and therefore unnecessary terms. Although one cannot determine particular scribal motivation, the fact that only 3 named MSS (plus an additional 3 subsidiary MSS) omit o Ihsou= in the present unit while all other Greek manuscript witnesses include the proper name strongly suggests the originality of the proper name in this location; this is particularly the case when the text without the presence of the proper name is neither ambiguous nor confusing in regard to the person referenced. The investigation of this particular type of variant unit is not complete, however, since the identical unit recurs numerous times in the Gospels. In several Johannine instances, o Ihsou= is present in but absent in NA27, demonstrated by the following chart that helps to shed further light on the situation regarding this particular type of unit (instances involving epoihsen or possible haplography of -ois ois2 are specifically noted in the second column):38 Reference Jn 4.16 Jn 4.46 Jn 6.14 Jn 8.20 Jn 8.21 Jn 11.45 Jn 13.3 Jn 18.5

o Ihsou= ()

autois o is2 a epoihsen o is2

autois o is2

omit (NA27) B C* 33 pc a sa ac2 bomss 66 75 B C D L Ws 086 33 1241 al lat syc (75 ) (B) D W (091) pc it vgst sys,c cop Didlem (a) (not in NA27 apparatus) 39vid 66* 75 B D L T W pc b (e) ac2 pbo (not in NA27 apparatus) 66

vid

mss

66 75 B D L W 1241 pc lat pbo 60 B D pc it sys pbo

(5 MSS) (11 MSS) (6 MSS) (9 MSS) (8 MSS) (3 MSS)

Nor is this specific variant unit exclusive to John, but the identical unit also appears frequently in the remaining Gospels, almost exclusively with reading o Ihsou= and NA27 omitting such from its main text, often without mention in its apparatus:39 Reference Mt 4.12 Mt 4.23 Mt 8.3

o Ihsou= ()

omit (NA27) B C* D Z 33 700 1241 pc ff1 k vgst sys sa mae bopt Or (8 MSS) B (k) syc sa mae (1 MS) B C* f1 f13 33 892 al k samss bo Cyr (7 MSS) vid

A similar occurrence in Jn 8.9 (mono= o Ihsou=, o Ihsou= mono=, or o Ihsou= solus), where NA27 follows D 1 892 pc c vgst in reading only mono=, is not included in the chart; this because the NA27 text of the pericope adulterae is double bracketed by its editors as non-Johannine, and also because the NA27 apparatus is extremely limited and restricted when compared with the present writers full collation data for that pericope. Similar related but non-identical variant units in John that are not cited in the chart include the following: Jn 19.38, where reads o Ihsou= and NA27 reads autou; and Jn 19.39, where reads ton ihsoun and NA27 reads auton (so also Lk 22.63). 39 Elsewhere one finds omission of an anarthrous Ihsou= at Mt 9.12; 15.16; 16.20; 17.11; 17.20; 22.37; omission of anarthrous Ihsou at Mt 8.29; omission of Ihsou= from o de Ihsou= at Mt 24.2; Mk 7.27; replacement of o de Ihsou= by kai solus at Mk 1.41; Mk 5.19 (the latter not in the NA27 apparatus); tw Ihsou replaced by autw at Mk 10.52; omission of a phrase containing o Ihsou= at Mt 13.51. 38

,2 B sa (21 1 D 892 [k] sys,c bo) B D Q f1 f13 33 700 892* pc lat cop (a b ff2 sys,c) (not in NA27 apparatus) (not in NA27 apparatus) 1 B L Z 078* 0281 f 700 892* 1241 bo (not in NA27 apparatus)

Mt 12.25 Mt 14.14 Mt 14.22 Mt 14.25 Mt 18.2 Mt 26.38

autois o is2

Mk 5.13 Mk 6.34 Mk 8.1 Mk 8.17 Mk 11.14 Mk 11.15 Mk 12.41 Mk 14.22

B C L W D f1 28 (579) 892* 2542 [2427] (not in NA27 apparatus) (not in NA27 apparatus) 1 B D 892* [2427] aur i samss bo (not in NA27 apparatus) (not in NA27 apparatus) B L D Y 892 pc [2427] a k sams bo 1 B D W f13 565 [2427] it sys sa

Lk 7.22 Lk 8.38 Lk 9.43

(not in NA27 apparatus) (not in NA27 apparatus) 75 B D L X f1 579 700 1241 2542 pc lat sys,c cop

Lk 9.60 Lk 10.21 Lk 13.2 Lk 23.43 Lk 24.36

ois epoihsen o is2

(not in NA27 apparatus) B D X 1241 pc lat sys,c sa bopt 75 B L 070 1241 pc lat samss bopt 75 () B (C) L samss bopt 75 B D L 1241 it sys,c sa bomss 45

(5 MSS) (9 MSS)

(10 MSS)

(11 MSS)

(4 MSS)

(6 MSS) (6 MSS)

(11 MSS)

(7 MSS) (6 MSS) (5 MSS) (6 MSS)

Additional instances similar to the above recurring variant unit involve omission of titles such as o kurio=, o qeo=, as well as omission of additional other articular or anarthrous proper names; these are not cited in relation to the present discussion. In contrast, it should be noted that NA27 adds Ihsou= at Mk 16.19 (double-bracketed and therefore rejected),40 and also at Lk 5.34.41 Such addition is obviously the exception and not the usual practice of the predominantly Alexandrian scribes in such situations. Noteworthy in all these is the fact that, had the scribes added o Ihsou= or Ihsou= as a deliberate and systematic alteration, far more such additions should have appeared than presently might be claimed to exist.

The rejected inclusion of Ihsou= is supported by C* K L D f1 f13 33 565 579 892c 1241 1424 -844 -2211 al [2427] it vgcl sy cop Irlat.

The inclusion there is supported by 4 B C D L W X f1 f13 33 579 892 1241 2542 -844 -2211 al f syhmg cop.

In contrast, sporadic omission of o Ihsou= by a small minority of scribes, either accidentally (involving only a 3-letter nomen sacrum) or intentionally (for stylistic smoothing) easily could represent typical Alexandrian editorial practice. In essence, all the above-mentioned variant units reflect the identical internal considerations that affect the present unit, and their respective minority external support does not actually impact the overall situation; thus, each of the listed variants can be considered to have been addressed within the discussion of the present variant unit. JOHN 12.442 Iouda= o Iskariwth= ei%= [ek] twn maqhtwn autou = ei%= ek twn maqhtwn autou Iouda= Simwno= Iskariwth=

txt = 1241 -844 pc sy(s),p sa ac2

NA: But says Judas the Iscariot, one [out of] his disciples, the one being about to betray him . . . : Therefore says one out of his disciples, Judas of Simon Iscariot, the one being about to betray him . . .

The present variant unit involves transposition, whether the name Judas the Iscariot precedes or follows the descriptive phrase one of his disciples, as well as whether the patronymic of Simon or the definite article should also accompany the name. Although the reading of the main NA 27 text represents the editors preference,43 the apparatus reading omitting [ek] is more clearly representative of the Alexandrian archetype (supported by 66 75vid B L W 33 579 pc) and theoretically should have been the reading followed by the NA 27 editors.

JOHN 12.4 VIou,daj

o` VIskariw,thj ei-j @evk# tw/n maqhtw/n

auvtou/ The identification of Judas as Si,mwnoj (A K X D Q P Y 065 13 28 Byz) or as Si,mwn (1195 1242* 1344 2148 al) is a scribal accretion derived from 6.71. These same witnesses also smooth the sequence by placing the name after the indefinite ei-j evk tw/n maqhtw/n auvtou/ and before o` me,llwn auvto.n paradido,nai. It was thought best to retain the words ei-j evk, an expression that occurs in eleven other passages in the Fourth Gospel, but in view of the absence of evk in such early and noteworthy witnesses as 66, 75vid B L W 33, to enclose it within square brackets. (For the reading of D, see the comments on 6.71 and Mt 10.4.)

In view of that more obvious Alexandrian archetype reading (supported by 7 named MSS) without ek present, the 3 named MSS that support ek in the present NA27 main text either reflect an unconscious carry-over from the reading or an intentional alteration to the Alexandrian archetype for the purpose of stylistic smoothing. Apart from this factor, the remaining readings cited in the NA 27 apparatus represent a variety that differs from both the NA27 main text and . Nevertheless, the primary issues in this unit remain the basic phrase transposition and the matter of Simwno= versus the definite article o.

Other readings include Iouda= o Iskariwth= ei%= twn maqhtwn autou (omit ek) in 66 75vid B L W 33 579 pc; ek twn maqhtwn autou Iouda= Simwno= Iskariwth=, (omit ei%=)in K; ei%= twn maqhtwn autou Iouda= Simwno= Iskariwth= (omit ek) in Q; ei%= ek twn maqhtwn autou Iouda= o Iskariwth= (omit Simwno=) in f1 565; ei%= ek twn maqhtwn autou Iouda= Simwno= Iskariwtou, in Y; ei%= ek twn maqhtwn autou Iouda= apo Karuwtou, in D (omit Simwno=); and ei%= ek twn maqhtwn autou Iouda= Iskariwth=, (omit Simwno=), in vg. Immediately preceding is a variant unit in which MS 1241 and -844 (as a consistently cited 42

witness of the second order, here included in ) depart. If both units are considered together, the overall level of support drops to only the single MS . 43 NA27 Introduction, 6*

In regard to the latter issue, in both NA27 and the 3 remaining occurrences of Judas Iscariot in John (Jn 6.71; 13.2, 26)44 include Simwno=; therefore one would expect such to appear in the present unit, especially since there would be no good reason for departure from Johannine style at this point. The Alexandrian archetype (represented by the 7 other Alexandrian MSS cited above as well as the 3 named MSS that underlie the NA27 main text) easily could have omitted Simwno= by haplography (-= -=), the deficiency being repaired by insertion of the article preceding Iskariwth=. Yet elsewhere in John the definite article follows Iouda= only in the phrase ouk o Iskariwth= (Jn 14.22) or in relation to the descriptive phrase o paradidou= (Jn 18.2, 5); such makes the NA27 reading in the present unit incongruous with other Johannine usage. Also, when ei%= is present, John often places the proper name first (as with the NA27 text of the present unit):45 Jn 1.40 Jn 6.71 Jn 7.50 Jn 12.2 Jn 20.24

Andrea= o adelfo= Simwno= Petrou, ei%= ek twn duo Ioudan Simwno= Iscariwthn, . . . ei%= wn ek twn dwdeka Nikodhmo= . . . ei%= wn ex autwn o de Lazaro=, ei%= hn twn anakeimenwn Qwma= de, ei%= ek twn dwdeka

Most notably, Jn 6.71 also refers to Judas Iscariot, the construction being quite similar to that of the present unit. The structure of that prior occurrence well may have influenced the framers of the Alexandrian archetype in regard to the present Jn 12.4 unit.46 Thus, in regard to both the question of transposition and the replacement of Simwno= with the definite article, the case for the reading appears far stronger than that for the present NA27 main text or even the more strongly supported Alexandrian archetype reading.

Judas not Iscariot occurs in Jn 14.22, while Judas meaning the Iscariot but without that term occurs at Jn 13.29; 18.2, 3, 5. In John the practice is reverses at three points: Jn 6.8 (ei%= ek twn maqhtwn autou, Andrea= o adelfo= Simwno= Petrou); Jn 13.23 (ei%= twn maqhtwn . . . on hgapa o Ihsou=); Jn 18.26 (ei%= ek twn doulwn tou arcierew=, suggenh= wn). 46 Without ei%= present, the descriptive phrase precedes or follows the proper names of characters or locations equally. A proper name preceding the description appears at Jn 19.38 (Iwshf o apo Arimaqaia=, wn maqhth= tou Ihsou); Jn 19.39 (Nikodhmo= o elqwn pro= ton Ihsoun nukto=); Jn 11.2 (hn de Maria h aleiyasa ton kurion murw); Jn 11.49 (Kaiafa=, arciereu= wn tou eniautou ekeinou); while the description precedes the proper name at Jn 1.6 (egeneto anqrwpo= . . . onoma autw Iwannh=); Jn 1.41 (ton adelfon ton idion Simwna); Jn 3.1 (anqrwpo= ek twn Farisaiwn, Nikodhmo= onoma autou); Jn 4.5 (polin th= Samareia= legomenhn Sucar). 44 45

JOHN 13.2647 bayw to yimion kai dwsw autw = baya= to yimion epidwsw

txt = B C 1241 pc Or

NA: Jesus answers, That one he is to whom I myself will dip the morsel, and I will give to him. Having dipped therefore the morsel, [he takes and] he gives to Judas of Simon Iscariot. : Jesus answers, That one he is to whom I myself, having dipped the morsel, I will give upon. And having dipped in the morsel, he gives to Judas Iscariot, of Simon.

Among the 5 variant units cited in the NA27 apparatus within this verse (with numerous alternative readings among them), the present unit involves four issues: (1) the specific form of baptw, aorist participle or future active indicative; (2) the presence or absence of kai following yimion; (3) the simple or compound form of the future tense of didwmi; and (4) the presence or absence of autw. At various points the remaining alternative readings found in the apparatus lend partial support: (1) baya= in 33, bayw in L; (2) kai present in L; absent in A D K W f1 f13 33 565 pc; (3) dwsw in W; epidwsw in A D K f1 f13 33 565 pc; and (4) autw present in L 33; absent in A D K W f1 f13 565 pc. The NA27 main text reading, however, is limited to the three named MSS plus whatever pc here might entail, as cited above. Regarding each of the main points:

The verb baptw (not to be confused with baptizw, occurring 80x in the NT) only occurs 3x in the NT, with differing forms found in the remaining occurrences. 48 Whether the future indicative or the aorist participle occurs at this point perhaps might be a judgment call, except for the fact that in the very next sentence the same participial form in the compounded embaya= occurs (so also in Mt 26.23; cf. Mk 14.20, o embaptomeno=).49 Such would imply a similar participial form in the previous occurrence of the same verb. Secondly, it is quite possible that the 3 named MSS underlying the NA27 main text were influenced by the occurrence of other future active indicative forms in close context. In John the future active indicative occurs frequently (100x in ; 108x in NA27), with approximately half of these within the Upper Room discourse (51x in ; 52x in NA27).50 Noteworthy in this regard are the future active indicatives that appear in the present Jn 13 chapter, not only in the present verse 26 ( epidwsw, NA27 dwsw), but also in verses 21 (paradwsei), 32 (doxasei 2x), 33 (zhthsete), 36 (akolouqhsei=), 37 (qhsw), 38 (qhsei=). Of further note, the specific FAI-1S occurs 16x in the Upper Room discourse, with most of these appearing in chapters 13 and 14. These include epidwsw/dwsw of the present verse as well as Jn 13.37 (qhsw); 14.3 (etoimasw); 14.13 (poihsw); 14 (poihsw); 14.16 (erwthsw); 14.18 (afhsw); 14.21 (agaphsw, emfainisw); 14.30 (lalhsw). Such also readily could influence the scribes of the 3 named MSS

JOHN 13.26 ba,yw

to. ywmi,on kai. dw,sw auvtw|/ {C}

It is more likely that scribal alteration went from the simple (dw,sw) to the compound verb (evpidw,sw), which John uses nowhere else. Furthermore, the Semitic, paratactic style of two finite verbs connected by kai, is typically Johannine, whereas the omission of the conjunction and the hypotactic construction involving a participle (ba,yaj) has the appearance of being a stylistic modification introduced by copyists in the interest of elegance. Likewise, the redundant auvtw|/ after dw,sw, so characteristic of a primitive, Semitic style, would almost certainly be deleted by copyists. See also the comment on the following variant reading. 13.26 ba,yaj ou=n to. ywmi,on @lamba,nei kai.# di,dwsin {C} It is difficult to decide whether the words lamba,nei kai, were added by copyists to recall Jesus deliberate action at the Last Supper in taking bread (Mt 26.26; Mk 14.22; Lk 22.19; 1 Cor 11.23), or whether the words were omitted as irrelevant and unnecessary. In order to reflect the balance of both external attestation and transcriptional probabilities, a majority of the Committee decided to retain the words enclosed within square brackets.

underlying the NA27 main text or of an earlier archetype to adjust the parsing from an aorist participle to a FAI-1S in order to reflect a more dominant form of expression in this segment of text. 2. In regard to the inclusion or exclusion of kai, much depends on the first point noted above. If the aorist participle baya= is present, the conjunction (though remotely sensible) is superfluous; whereas if the future bayw is present, the conjunction becomes necessary in context for clear comprehension. It also is possible that the epi- prefix may have been misunderstood as a wrongly placed preposition, thereby provoking replacement with kai; this, however, appears highly unlikely. 3. Regarding the simple versus compounded form of didwmi, where the future active indicative is not otherwise in question, the NA27 form dwsw occurs elsewhere in John at Jn 4.14 (2x); 6.51 (2x ; 1x NA27). In contrast, the form epidwsw occurs nowhere else in the NT, although forms of epididwmi occur elsewhere in the NT 10x (7x in the Gospels).51 Yet the compounded epidwsw (I will hand over, deliver) carries a more specific and more appropriate meaning in a food-related context (here in regard to the morsel, to ywmion) than the simpler dwsw (I will give). A food-related context similarly appears in occurrences of the future tense of epididwmi elsewhere, where epidwsei is used to reflect whether a father would hand over to his son a stone, snake, or scorpion rather than bread, fish or an egg (Mt 7.910; Lk 11.11, 12); so also Lk 24.30, where Jesus, having taken the bread and broken it, hands it over (epedidou) to the two he met on the road to Emmaus; Lk 24.42, where the disciples hand over (epedwkan) to Jesus a piece of broiled fish.52 Symbolically, the use of the compound also would appear more appropriate by delivering (epididwmi) the morsel to the one who will betray (o paradidou=), particularly when the only occurrences of paradidwmi within the Upper Room discourse occur in the close context of Jn 13.2, 11, 21. Thus in both these areas (food-related context and symbolism) the reading has far more to commend it than the uncompounded dwsw of the three named MSS underlying the NA27 main text. Transcriptionally, the epi- prefix of epidwsw easily could have been lost in the 3 named MSS (4 including W) by haplography of i i, assuming inclusion of the conjunction (kai epidwsw), the skip occurring either independently in the MSS involved or due to a common archetype. Such likely explains the NA27 main text reading in this situation. Other readings include embaya= to yimion epidwsw, in A D K f1 f13 565 pc; baya= to yimion epidwsw autw, in 33; dwsw enbaya= to yimion, in W; and bayw to yimion kai dw autw, in L. Bayh/ occurs in Lk 16.24 and bebammenon in Rev 19.13. 49 Again, not to be confused with ebaptizonto (from baptizw). 50 The designation Upper Room discourse is considered here to encompass Jn 13-17, with the statements of Jesus commencing in that location, but with a process of journeying to Gethsemane beginning at the end of Jn 15 (egeiresqe, agwmen enteuqen), culminating at Jn 18.1. 51 Mt 7.9, 10; Lk 4.17; 11.11 (2x), 12; 24.30, 42; Ac 15.30; 27.15. Of these, the future active indicative epidwsei occurs at Mt 7.9, 10; Lk 11.11 (2x), 12. 52 The remaining NT occurrences of epididwmi relate to handing over a scroll (Lk 4.17) or a letter (Ac 15.30) or as a nautical term, giving way to the wind (Ac 27.15). 47 48

4. As to whether the indirect object (autw) should be present or absent, had autw originally been present, such theoretically might have been omitted by haplography of -w -w in the phrase (epi)dwsw autw. On the other hand, were the indirect object originally absent, it would be no surprise to find a limited number of scribes adding such to the phrasing, since the indirect object naturally might be expected. Of the remaining occurrences of epididwmi, the indirect object is autw at Mt 7.9, 10; Lk 4.17; 11.11 (2x), 12; 24.42; and autoi= at Lk 24.30, even while epididwmi appears without an indirect object at Ac 15.30; 27.15. Yet haplography is not of primary concern in the present instance, even though a similar instance of haplography was a point of appeal at the conclusion of the preceding point: the transmissional issue here is quite different, and of such importance that it should not be neglected (although often it is by modern textual critics): Textual omission due to haplography is more likely to have occurred when such is found in a limited number of MSS. When claimed to have occurred in an extremely large number of MSS, haplography is extremely unlikely, since a reasonable level of scribal correction of such an egregious error readily would be expected, thus resulting in the non-perpetuation and elimination of the error from the primary lines of transmission. Merely because some Byzantine readings have been claimed by adverse critics to have resulted from various forms of haplography (homoioteleuton, homoioarcton, or homoiomeson), such claims are unlikely on their face in view of overall transcriptional and transmissional processes as well as in view of known scribal proclivities. In contrast, omissions that could be due to haplography that appear in the critical text, supported only by a small number of preferred MSS, remain highly suspect as due to scribal error (individual or archetypal) in each instance. In each of the four areas involved in the present variant unit, the reading has been shown in each case to have a better claim to originality than the reading found in the 3 named MSS underlying the NA 27 main text (even with fluctuating minor support from alternative reading witnesses). Given also that the 3 named witnesses all reflect the same textual tradition and likely stem from a common archetype, it is likely that the reading of all four issues as noted above derive from that archetype rather than from independent alteration. JOHN 14.1553 thrhsete = thrhsate NA: If youp should love me, youp will keep my commands. : If youp should love me, pkeep my commands.

Another reading is thrhshte (V-AAS-2P), in 66 060 33 579 pc.

txt = B L pc cop Epiph

JOHN 14.15 thrh,sete {C} A majority of the Committee preferred the future tense

thrh,sete, read by B L Y 1010 1071 1195* 2148 al (and perhaps supported indirectly by witnesses that read the

aorist subjunctive thrh,shte( 66 060 33 al), instead of the imperative thrh,sate, which, though rather well

Given that a third-class condition is involved (introduced by ean), the verb agapate in the protasis (the ifclause) is understood as a present active subjunctive, although in other contexts different parsings could result.54 The present unit involves orthographic fluctuation in the apodosis (the then-clause), with three readings existing among the MS base.

supported (A D K W X D Q P 1 13 28 565 700 892 Byz), accords less well with evrwth,sw in the following verse.

Thus, the verb tense can be future (thrhsete, FAI-2P, youp will keep), as in NA27; an imperative (thrhsate, AAM-2P, pkeep), as in ; or, as found in a minor alternate variant with substantial Alexandrian support, a subjunctive (thrhshte, AAS-2P, youp should keep). Notably, all three readings in this variant unit differ by only a single letter a fact that suggests accidental error on the part of various scribes or their ultimate archetypes. Elsewhere within John, the verb threw occurs 18x, with 11x of these occurring in various forms within the Upper Room discourse (Jn 13-17). Of these, 2x are FAI (Jn 14.23 thrhsei; Jn 15.20b thrhsousin) and 1x is AAM (Jn 17.11 thrhson).55 Although these occurrences do not duplicate the readings of the present unit (one other instance, Jn 15.10a thrhshte (AAS-2P), duplicates the alternate variant reading found in the apparatus of the present unit), it is important to recognize the similarities of parsing, since the presence of such forms theoretically could have influenced various scribes in relation to the present unit. Even so, erroneous scribal alteration of single letters appears to represent a more proximate cause.56 The three other 3rd-class conditions that appear in the immediate context offer an informing insight: in each instance, the apodosis verb is FAI:57 Jn 14.3 (ean poreuqw, etoimasw); Jn 14.14 (ean ti aithshte [me] en tw onomati mou, egw poihsw); and Jn 14.23 (ean ti= aqgapa me, ton logon mou thrhsei). Most definitely, these close-context occurrences easily could have influenced a small number of scribes to conform the parsing of the apodosis in the present unit to a FAI form, particularly since the command form (AAM) does not otherwise occur within the Upper Room discourse, nor apparently anywhere else in the apodosis of a 3rd-class conditional clause introduced by ean. Thus the reading not only represents the rarer and therefore more difficult reading, but also the 3 named MSS underlying the NA 27 main text readily could have made the single-letter alteration of parsing to FAI due to attraction to the other FAI occurrences in the immediate context of this chapter within the Upper Room discourse. Once again, the reading is to be preferred on multiple counts.

Depending on context, agapate could represent either PAI-2P, PAM-2P or PAS-2P. Other forms of threw within the Upper Room discourse represent non-relevant parsings: Jn 14.21 thrwn (PAP-NSM); 14.24 threi (PAI-3S); 15.10a thrhshte (AAS-2P); 15.10b tethrhka (RAI-1S); 15.20a ethrhsan (AAI-3P); 17.6 tethrhkasin (RAI-3P); 17.12 ethroun (IAI-1S); 17.15 thrhsh= (AAS-2S). 56 The minor apparatus variant thrhshte may have been influenced by the two preceding verses, in each of which aithshte occurs. Such does not account, however, for the specific primary difference between NA27 and in the present unit. 57 Ean occurs 61x in John and 17x in the Upper Room discourse, but many of these are not within an if-then conditional context. 54 55

ACTS 1.558 en pneumati baptisqhsesqe agiw = baptisqhsesqe en pneumati agiw

txt = * B 81 pc

NA: Because John indeed baptized with water, but youp yourselves with the Spirit will be baptized with the Holy not after these many days. : Because John indeed baptized with water, but you yourselves will be baptized with the Holy Spirit not after these many days. p

The NA27 Greek text as it appears in the present unit is awkward and problematic, as demonstrated in the overly literal English rendering given above. In contrast, the reading follows the more usual Greek order, with the resultant text reading quite smoothly. Even modern translations that utilize the critical text avoid the difficulty by rendering the passage in a manner similar to (youp will be baptized with the Holy Spirit).59 That the NA27 form of the passage is unusual is clearly seen by a wildcard search sequence in the Gospels and Acts, locating any form of pneuma, followed by any non-articular intervening word, and followed by a form of agio=. Such occurs only 2x in the NT, and in each instance the intervening word is a form of eimi (Mt 1.20, ek pneumato= estin agiou; Lk 2.25, kai pneuma hn agion ep auton).60 In contrast, the reading represents by far the more usual structure, occurring in an identical construction when preceded by a form of baptizw as follows: Mt 3.11 Mk 1.8 Lk 3.16 Jn 1.33 Ac 11.16

baptisei baptisei uma= baptisei o baptizwn baptisqhsesqe

en en en en en

pneumati pneumati pneumati pneumati pneumati

agiw agiw agiw agiw agiw

Similarly, the phrase in question also occurs in numerous contexts without any form of baptizw present: Mk 12.36 Ro 9.1 Ro 14.17 Ro 15.16 1Co 12.3 2Co 6.6 1Th 1.5 1Pe 1.12 Jude 1.20

eipen en pneumati agiw summarturoush= moi . . . en pneumati agiw eirhnh kai cara en pneumati agiw hgiasmenh en pneumati agiw ei mh en pneumati agiw en pneumati agiw kai en pneumati agiw twn euaggelisamenwn uma= en pneumati agiw en pneumati agiw proseucomenoi

ACTS 1.5 evn

pneu,mati baptisqh,sesqe a`gi,w|

The great majority of witnesses read VIwa,nnhj me.n evba,ptisen u[dati( umei/j de. baptisqh,sesqe evn pneu,mati agi,w.| Several important witnesses have a different order of

words, involving chiasmus: thus * B 81 915 Didymus readevn pneu,mati baptisqh,sesqe agi,w|, and D Hilary Augustine read evn pneu,mati agi,w| baptisqh,sesqe. The chiastic order of words does not seem to be merely an Alexandrian refinement, for it is not confined to the Alexandrian text. The less elegant order in the great bulk of witnesses can be explained as a harmonization with the sequence of words in the parallel reported in the Synoptic Gospels, all of which place evn pneu,mati a`gi,w| after the word baptize (Mt 3.11; Mk 1.8; Lk 3.16). The envelope construction of the Alexandrian text (placing the verb between the noun and the adjective) may be an editorial refinement, or it may reproduce an emphasis intended by the author. A majority of the Committee preferred the Alexandrian text, considering the

weight of * B 81 915 Didymus to be superior to that of D (the evidence of Latin Fathers does not count for much on a point concerned with the presence or absence of the envelope construction in Greek).

Even the articular forms of the expression throughout the NT continue to follow the more usual order found in the reading: tou pneumato= tou agiou Mt 12.32; Lk 2.26; Ac 13.4; Heb 9.8 to pneuma to agion Mk 3.28; 13.11; Lk 3.22; Jn 14.26; Ac 1.16; 5.3; 8.18; 10.44, 47; 15.8; 19.6; 20.23, 28; 21.11; 28.25; Eph 4.30; Heb 3.7; 10.15 tw pneumati tw agiw Ac 7.51

11.15; 13.2;

There thus is no good or compelling reason to consider the reading of the 3 named MSS or their common archetype that underlie the NA27 main text as original in the present unit; not only does the awkwardness of the NA27 construction argue strongly against any Lukan originality for such, but the fact that even the remaining Alexandrian witnesses (74vid 2 A C Y 33 1739) join with all other MSS in maintaining the reading provides strong evidence for considering the NA 27 reading to be aberrant, and likely the result of some sort of archetypal scribal error, possibly an initial haplography of e e, skipping from de to baptisqhsesqe. This would then result in de en pneumati , for which the scribe upon recognizing the error would immediately correcting by inserting baptisqhsesqe. That scribe then would continue with agiw, without any real concern for the awkwardness of the resultant expression. Such would appear easily to account for the development of the NA27 main text reading in the 3 named MSS, or even independently in some otherwise unrelated minuscules (the pc also cited). On all accounts, the reading is strongly to be preferred. ACTS 4.461 [w=] ciliade= pente = wsei ciliade= pente

txt = B D 0165 pc

NA: But many of the ones having heard the word believed; and [the] number of the males became [as] five thousands. : But many of the ones having heard the word believed; and the number of the males became as if five thousands.

The issue here involves the matter of numerical approximation or even precision. Both the bracketed NA 27 reading as well as the reading state that the number of the males who had believed was approximately 5000 (literally, [w=] = [as], or wsei = as if). In contrast (as shown in the NA27 apparatus), the more dominant Alexandrian reading in the present unit (74 A 81 1175 pc) omits the word entirely, making the number to

ACTS 4.4 @w`j# It is difficult to decide whether the passage originally stated that the number of the believers was five thousand

(74 A 81 vg copsa, bo eth) and copyists added wj (B D 0165 1611) or wsei, (E P most minuscules), on the pattern of 2.41; or whether the qualifying word (which seems to be a favorite of Luke when referring to numbers) was dropped by scribes for whom the number 5000 had become a firmly fixed tradition.

Another reading is en pneumati agiw baptisqhsesqe, in D it. English translations all paraphrase the final clause (which is not involved in the current variant unit): NKJV, NASV, NRSV, ESV, not many days from now; RSV, before many days; NIV, in a few days; Weymouth, before many days have passed. 60 Outside of the Gospels and Acts, the only other occurrences of the wildcard sequence involve the quite unrelated and syntactically quite different 2Cor 7.1 (kaqariswmen eautou= apo panto= molusmou sarko= kai pneumato=, epitelounte=) and 1Pet 2.5 (oiko= pneumatiko=, ierateuma agion in ; NA27 includes ei= ante ierateuma). 61 Another reading is ciliade= solus, in 74 A 81 1175 pc vg. 58 59

seem specifically 5000. Apparently due to the joint testimony of B and D, the NA27 editors chose not to follow the dominant Alexandrian omission in their main text even though such joint testimony was not considered as strong a factor a few words previously when those same two MSS even joined by similarly excluded the bracketed [o] preceding ariqmo= apart from the more dominant Alexandrian support.62

To reflect the dubiety in the interpretation of the evidence the Committee preferred to retain w`j, which is supported by B and D, but to enclose the word within square brackets.

Had the omission of a term of approximation been original, one might suspect the addition of such in order to mollify the level of precision stated. As an example, at Mt 15.38, NA27 and read those eating were 4000 males, even while various other witnesses add terms of approximation (w= B Q f13 33 892 -2211, or wsei 579 1241 pc). Similarly, at Mk 6.44 (cited in SQE but not in the NA27 apparatus), NA27 and read those eating the bread were 5000 males, while w= is added by Q 565 700 2542 and wsei by f1 205 pc. There is no reason to suppose that a similar statement of precision necessarily should occur in Lk-Ac, particularly when the remaining 12 occurrences in those two books approximate by utilizing either w= (5x) or wsei (7x):63 Lk 2.37 Lk 8.42 Ac 1.15 Ac 13.20 Ac 19.34

w= w= w= w= w=

about 84 years) (about 12 years) (about 120) (about 450 years) (about upon 2 hours)

Lk 1.56 Lk 3.23 Lk 9.14 Lk 9.28 Lk 22.59 Ac 5.36 Ac 19.7

wsei wsei wsei wsei wsei wsei wsei

(about 3 months) (about 30 years) (about 5000)64 (about 8 days) (about 1 hour) (about 400) (about 12)

This list makes it clear that Luke uses either w= or wsei to indicate approximation in about equal proportion, with slightly more occurrences of w= in Acts and of wsei in Lk (but not enough to establish any pattern; the same applies elsewhere in the NT).65

The [o] ante ariqmo= is cited in the NA27 apparatus as omitted by B (D), where D merely alters the word order. All other witnesses include the article, including the same dominant group of Alexandrian MSS as in the present unit. 63 Additional instances of w= or wsei with various numerals do not imply approximation; these include examples such as w= e$n toutwn (Mt 6.29; Lk 12.27); w= ei%= twn profhtwn (Mk 6.15); w= ena twn misqiwn (Lk 15.19); and w= ena ekaston (1Th 2.11). 64 At Lk 9.14, w= is read only by D, while it-q, sys,c bo omit either wsei or w= entirely ( NA27 and jointly read wsei at this point). 65 Non-Lukan instances include Mt 14.21 (wsei, about 5000); Mk 8.9 (w=, about 4000); Jn 6.10 (wsei /w= NA27, about 5000); Jn 6.19 (w=, about 25 or 30 stadia); Jn 11.18 (w=, about from 15 stadia); Jn 19.39 (w=, about 100 pounds); and 2Pe 3.8 (w=, about 1000 years . . . about 1 day). There also exists a variant reading in B using w= at Ac 27.37 (about 76) where most MSS read were 276. 62

Particularly in view of the identical approximation involving the number 5000 using wsei in Lk 9.14, the case becomes particularly likely that identical language would have been utilized in the present unit. Given that only the three less-closely related named MSS (plus whatever unrelated minuscules might be signified by pc) read the bracketed [w=] in the present unit, whereas all other witnesses (beyond the dominant Alexandrians that simply omit either term) follow the Lukan pattern of approximation, particularly in relation to a previous use of wsei in relation to the round number 5000, it appears far more likely that the reading in the present unit represents the original Lukan form rather than either of the alternatives. The reading once more is to be preferred. ACTS 4.37 pro= tou= poda= = para tou= poda=

txt = E 36 pc p

NA: . . . one possessing for himself a field, having sold it, he brought the money and placed it toward the feet of the apostles. : . . . one possessing for himself a field, having sold it, he brought the money and placed it alongside the feet of the apostles.

The difference here involves the preposition describing the location in relation to the feet of the apostles. The accusative plural phrase tou= poda= occurs 51x in the NT (50x in NA27); 15x in Lk and 11x in Ac. When a preposition precedes, the far less frequent phrasing is pro= tou= poda=, occurring in only 3x elsewhere in the NT (Mk 5.22; 7.25; Rev 1.17) while the NA 27 main text has 3 additional occurrences (Lk 10.39; Jn 11.32; Ac 5.10). In contrast, the more frequently occurring phrase is para tou= poda=, occurring in 11x elsewhere in the NT (NA27 9x), all of these with one exception (Mt 15.30) occurring within Lk-Ac.

ACTS 4.37 pro.j

tou.j po,daj

The Textus Receptus, following 57, 74 A B D P Y and most minuscules, reads para. tou.j po,daj, whereas E 36 94 180 307 327 453 1884 al read pro.j tou.j po,daj. Since para. tou.j po,daj is the more urbane expression, and since there is no fluctuation of witnesses in 4.35 and 5.2, where para. tou.j po,daj appears, it is altogether probable that in 4.37 the original reading was pro.j tou.j po,daj, which scribes altered so as to bring it into harmony with the adjacent passages 4.35 and 5.2. It should also be observed that the same tendency to alter the less elegant

expression appears in 5.10, where pro,j ( A B D) is replaced in various witnesses by para, or evpi, or u`po,.

It appears quite clear from these data that Lukes sole preferred phrasing is para tou= poda=, and thus the reading of the 3 named MSS underlying the NA27 main text is an aberration likely the result of un

Robinson's Commentary of the GNT vs. Metzger's Textual Commentary

Attachments

0 comments