Matt 6:1 ελεημοσυνην

Some witnesses (ℵ*.2 B D 0250 f1 892 pc lat Or-lat Hil Aug Hier) reflect that Jesus' first warning in chapter 6 was against doing one's "righteousness" (δικαιοσυνην) before men to receive their praise instead of one's "alms" (ελεημοσυνην), as most witnesses have (E K L M S U W Z Δ Θ Π Σ Ω 047 f13 33 565 1424 f k sy-p.h mae-1 Chr). In favor of ελεημοσυνην is Walter Nagel, "Gerechtigkeit -- oder Almosen? (Mt 6, 1)," Vigiliae Christianae 15:3 (1961): 141–45," whose main points are summarized below, followed by some further considerations:

1. Citing Bengel as an example, there has been a precedent to receive δικαιοσυνην based primarily on one's exegetical preference for it as the major subject of the Sermon on the Mount (cf. 5:6, 10, 20; 6:33).
2. A major theme of the Sermon on the Mount is not to be assumed from form-critical considerations, and if it were, it would be the "kingdom of heaven" and not "righteousness." Also, since it is doubtful that every original word has been preserved in the so-called great manuscripts of the fourth century, and since we are asking about the text of a much earlier time (second and third centuries), the few so-called great manuscripts should be interpreted in light of the tradition and not the other way around.
3. Structural analysis of Matt 6:1–18 indicates that Matthew introduces each warning with the second person plural (προσεχετε [6:1], προσευχησθε [6:5], νηστευητε [6:16]) and follows with the second person singular (ποιης [6:2], προσευχη [6:6], αλειψαι [6:17]). Thus structural analysis seems to require ελεημοσυνην in 6:1. [Note: almost all Greek witnesses in 6:5 do not have the second person plural but rather the second person singular προσευχη, which Nagel dismisses as secondary due to scribal conformation to the same form in 6:6].
4. The worth of the reading δικαιοσυνην shrivels in light of the ancient versions, since in support are only a few important Greek witnesses, part of the Old Latin tradition, and a single Old Syriac manuscript (sy-s). [Note: although cited in NA25 but not NA27, support of sy-s for δικαιοσυνην is unfounded, since it uses the same word (zdq) in 6:1 as it does in 6:2, 3, 4 for ελεημοσυνην. If anything, this circumstance rather suggests the support of sy-s for ελεημοσυνην!] For ελεημοσυνην are the Old Syriac tradition behind the Peshitta, the Arabic text of the Diatessaron ("alms"), the Curetonian Old Syriac ("gift"), the Nestorian form of the Syriac ("gift of mercy"), the Syriac-dependent Armenian version ("alms"), probably the Sahidic ("things of pity"; the Sahidic omits 6:1, but cf. mntna in 6:3), the Bohairic ("charity"), the sole African Old Latin manuscript (k/1) and the European Old Latin Brixianus (f/10) [both have eleemosyna and not justitia], Origen and Apollinarius are anything but complete (in fact, "For both of these church fathers ελεημοσυνη is the original text," p. 143). "And so of the Greek and extra-Greek material not much remains in support of δικαιοσυνη" (143).
5. On historical grounds, even the witnesses that have δικαιοσυνην do not support the sense of "righteousness," since in many places the LXX translates the Hebrew zdq with "alms," Köhler-Baumgartner suggests "kindness" as the trasnlation of zdq in Isaiah 56–63, in many places the Hillel school says the same (cf. Strack-Billerbeck), Jerome when he prepared this place in the Vulgate said that justitia (righteousness) meant "alms," and even Origen knows the reading δικαιοσυνην only in the sense of "alms." So even though the reading δικαιοσυνην is old, it is seen as a variation of ελεημοσυνην and not vice versa and thus has no intrinsic theological value, much less can it be seen as proof for an interplay with Pauline notions of the word.
6. Lexical considerations offer nothing conclusive in favor of either of the two words in question: "In Greek righteousness and alms are never confused or interchanged with each other. But this possibility due to pronunciation and spelling is clearly evident in the Western Aramaic and Syriac linguistic field" (144). E.g., Dalman's Aramaic-New Hebrew dictionary shows that through a small change "righteousness" (zdqa and zdquta) can be retained under the radical for "alms" (zdqta), and vice versa. In the earliest Syriac spelling or Estrangela, the similar interchange of alms (sdqta) and righteousness (sdiquta) occurs with merely the addition or subtraction of the two smallest letters of the alphabet, yod and waw. Perhaps it was a harmless word game, since the newly created word would have had the same meaning as the word that was changed (cf. sy-s, which uses the same word zdiqut in 6:1, 2, 3, 4). "Thus the reader could find here the well-known identification of alms and righteousness in the LXX and with Hillel certainly and the Judeo-Christian in general. On the same ground a Greek Judeo-Christian scribe unhesitatingly could have translated literally the wording found in sy-s and written δικαιοσυνη in place of the earlier ελεημοσυνη" (144).
7. "In order to avoid every misreading or mispronunciation a new root needed to be selected, in this case a derivative of the root jhb with the meaning 'offering,' which would no longer be deflected into a different word through small alterations." This word is found in sy-c, Ephrem, the first corrector of Sinaiticus, the Coptic, and, although younger than the two readings in question, it stands closer to "alms" and thus must be understood as a confirmation of that reading. "This judgment becomes sound through the observation that the most important Syriac witness presents this word in the plural (cf. Ephrem according to Burkitt: gifts vs. sy-c: auhbtkun). As the Arabic text of Tatian at Matt 6:1 likewise has a plural . . . , therefore Ephrem and Tatian confirm each other through these against the Greek wording with the opposite number and must therefore in this case be regarded as the oldest reading" (145). The plural is only sensible when alms or gifts is written or spoken, since righteousness is always only found in the singular in the OT passages. Sy-c has the singular because it followed sy-s. ℵ1 has the singular either in reference to sy-c or due to the common mistaking of ει and ι (itacism), thus writing δοσις instead of δοσεις. But also possible is an analogous occurrence of the alteration of οφειληματα into οφειλην in Didache 8:2 = Matt 6:12.

     Nagel's argument offers a plausible explanation for how a small minority of manuscripts that are often related could have come to read δικαιοσυνην. Below are some text-critical excerpts and further considerations for the present discussion.
     Erasmus (2:30), apparently unaware of the Greek textual support for δικαιοσυνην, notes that the Greek text has ελεημοσυνην, "not righteousness, in order to correspond with what follows," i.e., 6:2–4. He also notes that although Jerome and Augustine read "righteousness" in their Latin text, they nevertheless expound "alms." Bèze and Grotius both prefer δικαιοσυνην. Mill (14) explains: "I have no doubt but that the interpreter of Matthew thus [i.e., with δικαιοσυνην] rendered Tsedakah in this place; but ελεημοσυνην crept into the text from the margin of certain manuscripts"; and (Prolegomena:42 [§393]), "Righteousness specifically in this place is not strictly kind deeds to the poor, but evangelical righteousness in general, the kinds of which are alms, prayers, fasting, etc. Christ had said above that the righteousness which he examines from his own [children] surpasses the righteousness of the scribes and Pharisees," and furthermore the error occurred "since the word ελεημοσυνην occurs in v. 2 and in fact because of the particle ουν it appeared to refer to righteousness, and thinking of the first verse, a certain reader was prone to assign "goodwill," ελεομοσυνην, to δικαιοσυνην from this place of explanation [i.e., from 6:2]. And so it appears to have crept into the place of the other relatively more obscure word from the first centuries of the church."
     Wettstein (1:317) quotes Jerome (c. Pelag. II.), "Beware that you do not do your righteousness [justitiam], that is, alms [eleemosynam] . . . ," and also notes: "[Righteousness] is approved by Grotius, Mill, Bengel, and others, but they do not share the same opinion on the meaning of the word righteousness, as some hold that it may rightly mean everything--especially prayer and fasting--which the following passage is about, while others contend that here the words for righteousness and alms, according to the usage of the Hebrews, are synonyms. But I do not see how, by either (foregoing) method one decides, this [i.e., righteousness] can make sense in the passage. If by the former method, I would like to know the places where fasting and prayer are called righteousness: to live righteously is written δικαιοσυνην ποιειν, certainly not ποιειν την δικαιοσυνην αυτου. If by the latter [method], they will not easily persuade me that Matthew, where he writes about the same thing, referred to it with the same word three times but one time with another [word] which could not have failed to appear obscure; neither to the impartial judge should the authority of the Latins be greater than the testimony of the Greeks."
     Griesbach (1:59–64), who argues for δικαιοσυνην, sounds a familiar refrain: "For if there is a canon as true as it is certain, it is this: that reading is to be preferred which is harder, more obscure, more insolent, and approaching more nearly to the Hebrews' custom of speaking; surely δικαιοσυνη is better than the common ελεημοσυνη" (1:61). Speaking of the fact that the Latin translation of Origen's works often exhibit justitiam, Griesbach concedes, "I certainly grant that it could have been transcribed from the Latin version to which those translators had been accustomed" (1:61–2). "In addition, the gloss ελεημοσυνη obviously was carried over from vss. 2, 3, 4" (1:62).
     The argument that δικαιοσυνην is the harder reading is negated by the plausible reasons for its origin in a few manuscripts that are often related (cf. Nagel's article above). Once the error (or possibly the innocent play on words in the Syriac which influenced the Old Latin) entered the manuscript tradition, there would have been little reason for anyone to remove such a valuable Judeo-Christian gem in this place (cf. 5:20 and the importance scholars such as Mill and Bengel ascribe to the connection). In addition, one cannot discount the possibility that the similar endings of the two words (the last 6 letters of both are identical) could have created the initial introduction of δικαιοσυνην by accident, especially since that word appears 3 times in the previous chapter (5:6, 10, 20) and is 7 times more common in the NT than ελεημοσυνη (92x vs. 13x). The external support for ελεημοσυνην is overwhelming, including almost all Greek manuscripts, arguably the most important Old Latin witness (k/1), the Middle Egyptian (mae-1), and most versions directly or indirectly (cf. Nagel above). The patristic support for dικαιοσυνην is apparently entirely Latin and thus of no primary significance, especially since the origin of the Latin reading has been accounted for. Consequently, there is little reason to reject ελεημοσυνην as the original reading in Matt 6:1.