Matt 7:15 δε

Very few witnesses (ℵ B Ω 565. 1424 pc), followed by Bover, Greeven, Lachmann, Merk, and Tischendorf (8th), omit δε after προσεχετε, while all others predating the 10th century retain it (including C E G K L M O S U V W X Δ Θ Π Σ Φ 047 0211 0281 Byz f1.13.35 33. 892. 1500. 2224*), which Soden, Tischendorf (7th), and Vogels support. Weiss (159) comments that "omission of δε through scribal error is of course very frequent . . . especially after προσεχετε, as in ℵ D in Luke 21:34, D in Matt 10:17 and 16:11, B D Δ Byz in Matt 6:1, and ℵ B in Matt 7:15, where only Tregelles in brackets retains it." Bloomfield (GNT, 1:53; Annotations, 6), in addition to mentioning that the conjunction was "lost by the carelessness of scribes, who often omit δε," later judges that it was "more probably removed by critics unable to follow up the thread of the connexion . . . , and who, observing its absence from the lectionaries, where from a lectio commencing with this verse the particle could not be used, accordingly expunged it." In addition to mere accident (-ΤΕ followed by ΔΕ) and possible lectionary influence, the omission could also have been facilitated by an early versional diglot that omitted the word, as almost certainly occurs with the omission of the same word after προσεχετε in Matt 10:17 (by D it sy-s sa-mss mae). Although far less likely, it is possible that some manuscripts omit the conjunction by assimilation to Matt 6:1, where δε is missing after προσεχετε in 3 of the 5 pre-10th-century manuscripts that omit it here (i.e., B Ω 565). For literature on the prevalence of omission (especially of smaller words/particles) over addition in the early period, see the note on Matt 1:22 του. See also the note on Matt 6:1 προσεχετε, where reasons are adduced why the δε in that case may not be primary.

Matt 7:10 εαν ιχθυν αιτηση

As in Matt 7:9 εαν αιτηση, a few manuscripts (with Bover, Greeven, Lachmann, Merk, Tischendorf [8th]) omit εαν and use the future indicative αιτησει (ℵ B C L Δ 0281 f1 33. 892 [f1: αιτηση]) in conformity with almost all manuscripts in Luke 11:11 (cf. the evidence presented in Matt 7:10 και)(so Wettstein, 1:340; Griesbach, Commentarius, 1:76; Meyer, 161) or with the same pattern in Luke 11:12 (so Soden, 1:1424). However, most witnesses (followed by Soden, Tischendorf [7th], and Vogels) use the conjunction εαν with either the aorist subjunctive αιτηση (so Soden and Vogels, following E G K M O S U V X Δ Θ Π Φ Ω [Byz] f35 565. 1500. 2224) or the future indicative αιτησει (so Tischendorf [7th], following L N W Σ 047 0211 f13 1424), the form of the latter easily arising from that of the former (cf. Weiss' comment on Matt 7:6 καταπατησωσιν), the change perhaps being exacerbated by the letter Η being easily confused for the letters ΕΙ in uncial script.

Matt 7:9 εαν αιτηση

Most witnesses (along with Soden, Tischendorf [7th ed.], and Vogels) have εαν (or αν) with the aorist subjunctive αιτηση (including E G K M Ν Ο S U V W X Π Σ Φ Ω 0281 [Byz ca. 1500 mss] f1.13.35 33. 565. 892. 2224) or with the future indicative αιτησει (ℵ1 L Δ 047 0211 1424), but at least five witnesses (ℵ* B C Θ 1500), followed by Bover, Greeven, Lachmann, Merk, and Tischendorf (8th ed.), both omit εαν and alter the verb to αιτησει either by (a) symmetrical conformation to the tense of the verb that follows (επιδωσει) or (b) harmonization to Luke 11:11 (so Soden, 2:18). Harmonization to Luke may also explain the omission of εστιν in B* L 565. 1424 al here in 7:9.
     In Matt 12:36 where most manuscripts have εαν λαλησωσιν (including E G K [L] M N S U V W X Y Γ Δ Π Ω 0250 [Byz 1540 mss] f1.13.35 565. 892. 1424; [Or]), probably symmetrical conformation to the tense of the following verb (αποδωσουσιν) is similarly to be blamed for the omission of εαν (ℵ B D [Byz 3 mss]) and the change to the future indicative λαλησουσιν (ℵ B C Θ [Byz 32 mss] 33), not incidentally involving the same witnesses as in 7:9.
     For similar alterations in this section by the same class of witnesses cf. Matt 7:6 καταπατησωσιν and Matt 7:10 εαν ιχθυν αιτηση.

Matt 6:16 οτι

As in Matt 6:5, after αμην λεγω υμιν the word οτι is omitted in some witnesses (ℵ B D 0233 0250 f1.13 565 700 al it) but retained in most (including E G K L M S U V W Γ Δ Θ Π Σ Φ Ω [Byz] 33 c g1 lat). See note on Matt 6:5 οτι, where also cases of both omission and addition of that word in a minority of witnesses throughout Matthew are noted. Reasons in favor of οτι here include: (1) the paucity and similarity of the witnesses that also omit the word in Matt 6:5; (2) the greater likelihood that an editor would harmonize all three occurrences of the saying that first appears in Matt 6:2 (αμην λεγω υμιν απεχουσιν τον μισθον αυτων) than disharmonize the second and third occurrences of the saying in Matt 6:5, 16 in most witnesses: αμην λεγω υμιν οτι απεχουσιν τον μισθον αυτων (cf. notes on Matt 6:4 αυτος, Matt 6:4 εν τω φανερω, Matt 6:5 αν); (3) the superfluousness of the word (it can be left out without harming the sense), coupled with the observation that scribes habitually omitted words, especially shorter ones, far more often than they added them (cf. note on Matt 1:22 tου and the literature cited there). Thus the reading contained in the consensus of most witnesses is to be preferred. On the omission of οτι in largely related witnesses and for similar reasons, see Matt 5:31 οτι.

Matt 6:5 οτι

The optional conjunction οτι appears after αμην λεγω υμιν in most witnesses (including E K L M S U W Δ Θ Π Σ Φ Ω 565 579 1241 1424 f; Or-pt Bas-pt) but not in a few others (ℵ Β D Z 047 f1.13 28 33 700 pc lat; Or-pt Bas-pt Chr). Griesbach (1:66–7) reasons, "Some omit οτι both here and in vs. 16, because it is not employed in vs. 2." Von Soden similarly, but printing οτι within brackets, writes, "In 6:5, 16 the omission of οτι after υμιν (cf. 6:2) is to be judged in the same way [as the omission of αν {6:5}, which is secondary, being conformed to 6:2]" (1:1424).
     The expression αμην λεγω υμιν appears 25 times in Matthew, seven uncontested with οτι following (18:13, 18; 19:23, 28; 21:31; 24:47; 26:21) and 13 uncontested without οτι (6:2; 8:10; 10:15, 42; 11:11; 18:3, 18; 21:21; 24:2; 25:12, 40, 45; 26:13). The three contested places where RP adds and NA27 omits οτι are: 6:5, 16; 23:36. The two contested places where RP omits and NA27 adds οτι are: 16:28; 24:34.
     In the present unit of variation the internal argument of Griesbach and von Soden is sensible. As in Matt 6:4 αυτος, Matt 6:4 εν τω φανερω, and Matt 6:5 αν, the less harmonized reading is generally preferable to the more harmonized one. In 6:16 the witnesses that omit οτι (ℵ B D 0233 0250 f1.13 565 700 al it) are similar to the ones that do so here. Thus in a few largely related manuscripts the second and third appearances of the expression in 6:5, 16 (αμην λεγω υμιν οτι απεχουσιν τον μισθον αυτων) were conformed to the first in 6:2 (αμην λεγω υμιν απεχουσιν τον μισθον αυτων), i.e., without οτι. This internal factor combined with the observation that οτι is completely unnecessary sufficiently explains why it came to be omitted here in some witnesses. Cf. also the note on Matt 5:31 οτι.

Other cases of omission of οτι in a minority of witnesses in Matthew:
5:32 (D 1506 it; Aug)
9:18 (ℵ D M f1.13 33 205 892 pc lat)
10:7 (B 1506 sy-s)
11:24 (ℵ*.2 33 pc; Ir-lat)
13:11 (M pc)
16:17 (B* 1424*)
16:18 (Θ 1 1506 1582 ff2 g1; Eus Chr Phot)
19:9 (B D Z 1424 pc it)
20:12 (ℵ B C2 D 085 f1 33vid 1424 lat sy-s.c.p; Or Chr)
21:16 (ℵ D S 983 1506 pc it vg-mss; Ir-lat)
21:43 (ℵ B* Θ 205 209 565 700 892 1506 pc)
23:36 (p77 ℵ B D L Θ f1.13 205 579 700 892* pc it)
26:29 (p45 ℵ B D Z Θ 0281 f1.13 33 205 892* al; Ir-lat)
26:65 (ℵ2 B C2 D L Z Θ 0281 33 700 892 pc latt; Or Cyr Chr)
26:72 (ℵ D pc)
26:75 (D pc it)
27:47 (ℵ D L Θ 33 700 892 1424 pc lat sy-s.p)
28:13 (33)  

Cases of addition of οτι in a minority of witnesses in Matthew:
6:2 (Zvid 157 pc h)
8:25 (Δ)
9:33 (V Θ 35 1424)
12:23 (D*)
14:4 (M pc)
16:28 (ℵ B L Θ f13 33 579 700 1424 al it)
17:11 (ℵ)
17:20 (C pc co; Or)
19:24 (ℵ C L M Z 0281 579 892 al aur)
21:11 (N)
23:39 (D Θ f1.13 205 it; Or)
24:2 (D sy-p)
24:5 (C* Φ pc f; Or)
24:34 (B D F L Θ f1.13 33 205 700 892 1424 al latt; Or)
26:64 (D pc bo)
27:63 (D 157 pc; Chr Or-lat)
27:64 (33 al a ff2 sy-s.p)

Matt 6:5 αν

After οπως a minority of witnesses omit the superfluous particle αν (ℵ Β D K L Z Π* Ω f1.13 33 1424 al; Or Chr), but most retain it (including E M S U W Δ Θ Π1 Σ Φ 28 565 700 1241). Griesbach (1:66) writes: "[The particle] αν either arose from the ΦΑΝωσι immediately following or was omitted on account of the word itself, whether by a scribe easily skipping over a particle introduced merely to take up space, it being useless, or through eagerness to avoid cacophony, or because in vs. 2 οπως and in vs. 16 the very same οπως φανωσι appear without αν. Therefore, arguments are not lacking with which αν may be defended." Von Soden (1:1424) similarly judges: "The omission of αν after οπως (cf. 6:2, 16) is secondary, since there is no conceivable reason for K [i.e., the Koine tradition] to have inserted αν only here."
      The construction οπως αν is relatively rare in the NT, being used only four times elsewhere (Luke 2:35; Acts 3:20; 15:17; Rom 3:4), two of which within LXX quotations. In Matt 6:16 only a few manuscripts (Δ pc) add αν after οπως (probably in conformation to 6:5), as similarly happens in 23:35 (ℵ2 C2 Mmg 33 pc). A number of the primary witnesses that omit αν here also omit it elsewhere in Matthew: 5:18 (B*); 5:26 (33 pc); 12:20 (L X pc); 12:50 (D pc); 24:34 (ℵ 1241 pc); 24:43 (D 33 pc). Furthermore, the particle αν is likely original in Matt 6:5 due to its being omitted by some for the following reasons: (1) accident, of which kind this especially characterizes the early period (cf. note on Matt 1:22 του), and aided by the possibility of initial homoeoteleuton error (αν . . . αν, producing οπως ανωσιν), which later was easily corrected by adding Φ in view of 6:2, 16; (2) it was thought superfluous; (3) conformation to 6:2, 16, where it is absent; (4) the rarity of the expression οπως αν (4x in the NT) yielded to the more general usage without αν (more than 50x in the NT); (5) the sound ΑΝΦΑΝ- was thought undesirable (cf. Griesbach above).

Matt 5:31 οτι

The conjunction οτι following ερρεθη δε came to be missing in a number of witnesses (ℵ B D L f1.13 33 565 700 892 1241 1424 pc lat; Chr) for any of several reasons: (1) scribal error; (2) superfluousness; (3) assimilation to the repeated nearby expression where no οτι follows ερρεθη (5:21, 27, 33, 38, 43). Griesbach (1:53) suggests transcriptional error and, rather significantly, appeals to the number of manuscripts in such cases: "Although the select manuscripts B C D 1. 33 with a few others and Chrysostom omit οτι, I rather prefer to retain it, since, with ος following, it most easily could have escaped the notice of the eyes of one or another scribe. In judging omissions of this kind, where that in broad daylight could have caused scribes to err, we think that some degree of weight ought to be assigned to the number of manuscripts." Bloomfield (GNT, 1:36) remarks that critics "cancelled the word as pleonastic and unnecessary," that "internal evidence is in favour of the word, from the greater probability of its being removed than inserted," and that in such cases "ancient Versions and Fathers are of little weight." Thus the presence of οτι, contained in most manuscripts (including E G K M S U V W Δ Θ Π Σ) and corroborated by reasonable internal evidence, is preferable to the omission. Cf. also the note on Matt 6:5 οτι.

Note: Bloomfield lists a number of passages, such as the following, where οτι is absent in particularly influential witnesses: Matt 9:18 (ℵ D f1.13 33 892 pc lat); 20:12 (ℵ B D 085 f1 1424 pc lat sy-s.c.p); 21:16 (ℵ D 983 1506 pc it vg-mss); 23:36 (P77 ℵ B D L Θ f1.13 205 579 700 892* pc); 26:29 (P45 ℵ B D Z Θ f1.13 33 205 892* al); Mark 1:27 (ℵ B L 33 f1 205 565*), 40 (D lat sy-p); 2:16 (ℵ B Cvid D L W Θ 33 1342 1424 2542 pc latt sa bo); 5:23 (D f13 a c e sy-p); 6:16 (ℵ B D L Θ f1 33 205 565 700 892 1342 pc latt sy-s.p), 18 (D 892 pc lat); 7:6 (ℵ B L Δ Θ 33 579 1342 pc sy-s.p); 8:24 (D W Θ f1.13 205 565 700 1424 2542 al latt co); 12:6 (L W Δ 1 33 788 892* al it), 29 (D W Θ f1 205 565 700 788 2542 al it sy-s); Luke 7:22 (ℵ B L W Θ Ξ Ψ f1.13 205 579 700 892 1342 1424 2542 pc it bo); 19:46 (ℵ B L Θ f1.13 892 2542 pc c l co); John 4:42 (B Ws pc b f r1 sy-c; Ir-lat); 7:31 (ℵ B D L T U X Δ f1 33 lat sy-c.p); 8:33 (f1 al b c ff2 l vg); 9:23 (D L a c e l); 13:33 (P66 ℵ* D W 579 1241 pc lat); 18:6 (ℵ A B D L N W Θ Ψ 0250 f1 33 565 al latt), etc.

Matt 3:10 δε και

     A few manuscripts (ℵ B C Ds M W Δ 0233 f1 700 pc co) omit και after δε either by accident or because it was considered redundant. Erasmus (2:19) notes the και is "not an idle conjunction, for it indicates with an increase in force that imminent danger approaches." Wettstein (1:265) follows Erasmus, and Griesbach (1:24–5), noting the same motivation behind the omission of και in Codex Bezae (D/05) at the parallel passage of Luke 3:9, says it was omitted because it appeared superfluous and that scribes failed to understand its force. Such especially applies to the versions, which in turn could have influenced the omission in some Greek copies. Fritzsche (131–2) rightly cautions that just because ηδη δε και was common among Greek writers does not mean that the και would have been spared from tampering had it for some reason been deemed unsuitable, and that at first glance δε pertains to the entire sentence to focus and multiply its greater intensity, and thus the και was thought unnecessary; yet the και indicates that the ax was even brought near to the trees, which, as Kühnöl (70) who says και was wrongly omitted notes, frequently refer to mankind in Scripture (Ps 1:3; Matt 7:17–19; John 15:1; Rom 11:17, etc.). Bloomfield (Annotations, 2–3) notes that the harshness of the construction lowers the possibility that και was introduced from harmonization to Luke 3:9, and that it was removed because it was thought to overload the construction, as was the και that is omitted in some manuscripts in Luke 5:35 (ℵ C F L M Θ f1.13 [and many late minuscules], etc.). Matthew is not unaccustomed to using δε και at the beginning of clauses (10:18 [versus Mark 13:9 and Luke 21:12]; 10:30 [versus Luke 12:7]; 18:17; 24:49 [versus Luke 12:45, which has τε και]; 25:22, 24 [versus Luke 19:18, 20], etc.). Moreover, examples may be multiplied where scribes, and especially the originators of the versions, opted to omit either δε or και when encountering the δε και combination (Matt 25:22 [δε omitted by ℵ B pc sa]; 26:35 [δε omitted by ℵ B C D L 067vid 0281 33 70 al lat sy mae bo-mss]; 27:41 [δε omitted by B K Θ f1.13 33 565 700 al lat sy-[s].p bo-pt; δε και omitted by ℵ A L W pc bo-pt]; Mark 14:31 [δε omitted by B f1 579 pc a c k sa-mss bo-ms]; Luke 6:6 [και omitted by ℵ B L X f1.13 33 pc it co sy-s.c]; 18:1 [και omitted by ℵ B L f13 pc a b c q co], etc.). See also Scholz (1:clxv), who similarly argues against the Alexandrian tampering of the intrinsically Lukan construction verb + δε και. Thus both internal and external evidence firmly support the authenticity of και. Cf. also the note on Matt 1:22 regarding scribal propensity toward omission over addition.

Matt 1:18 γαρ

After μνηστευθεισης a few manuscripts (p1 ℵ B C* Z f1 pc) omit γαρ, but the strong consensus of over 1600 manuscripts combined with Eusebius in the fourth century, W/032 around 400, and representatives from all the text-types (L Θ Δ Π f13 33 892 1241) with the possible exception of the Western indicate that the presence of the conjunction here is older than its absence. Bengel (Gnomon, 1:110) comments that 1:18 contains the introductory statement of the matters that follow, to which also the ουτως refers, and thus the conjunction γαρ (as in Heb 2:8) commences the handling of the subject, which corresponds with the introductory statement. Furthermore, Griesbach (1:15) and Bloomfield (GNT, 1:6) rightly judge that γαρ was omitted by copyists who did not perceive its force and in some versions because in translation it could be disregarded without harming the sense. Burgon (Traditional Text, 192–3) flatly says that the γαρ "is not represented here in so much as a single ancient Version" because it is untranslatable in this sense, and that had it been a scribal addition to the text "it could not by any possibility have universally established itself in copies of the Gospel" to the extent that it has done due to its untranslatable function and thus, to scribes, apparent superfluousness. On the grounds of its overwhelming and relatively equally early external attestation, superior internal character, and sufficient explanation of its absence in a minority of witnesses the presence of γαρ is to be preferred.