1. Introduction
In 1988 a substantial papyrus codex, containing among other material a bilingual glossary with numerous short extracts from the letters of Paul, was published as a monograph by Alfons Wouters.Footnote 1 This papyrus, known at that time as Chester Beatty AC 1499 and now as Chester Beatty BP xxi, has some significance for the Greek text of Paul: Wouters lists twenty-seven variant readings in its Greek text that do not appear in the apparatus of Tischendorf,Footnote 2 and the papyrus now figures in Nestle–Aland.Footnote 3 It is much more important for its Latin readings, which belong to the Vetus Latina tradition and include many interesting variants: Wouters lists sixty-seven variants in its Latin text not known from other sources,Footnote 4 and these readings have since been systematically entered into the Brepols Vetus Latina database.Footnote 5 The papyrus has been assigned by palaeographers to ca. ad 400Footnote 6 and is therefore among the earlier witnesses to Paul, particularly in Latin.
A few years after the original publication Wachtel and Witte published a re-edition of the papyrus’ Pauline material.Footnote 7 They did not do a full re-examination of the original and improved only a few readings.
The present re-examination presents a significant number of new readings and identifications, some of which affect words that had been accepted as unique variants in the Vetus Latina tradition. It also offers a new interpretation of the glossary's history.
The glossary is a peculiar work combining 444 entries apparently taken from various letters of Paul (268 from Second Corinthians, 89 from Galatians, 66 from Ephesians and 1 from Romans) with 84 entries from other sources. Chunks taken from particular letters, within which the entries follow the order in which those phrases occur in Paul, are combined in a surprising order, with later chapters preceding earlier ones and chunks from one book alternating with chunks from another; the non-Pauline material occurs mainly in two substantial sections but can also be found scattered throughout the work.
The format of the glossary is unusual. Most ancient bilingual glossaries, indeed most ancient bilingual texts, are written in narrow columns (one lemma per line), with one language in one column and the other in the other.Footnote 8 This ‘columnar’ layout made it easy for readers to find a particular entry in either language by skimming down the relevant column, but it required a lot of space, as lemmata are inevitably unequal in length and the columns have to be as wide as the longest lemma plus an intercolumnium: the columnar layout can easily result in leaving blank half the writing space on a papyrus. Occasionally, therefore, one finds papyri in a ‘post-columnar’ layout: the columns are run together to save space, so that Latin and Greek words simply alternate within long lines, with some kind of punctuation, spacing and/or indentation system to help readers keep track of where the entries begin.Footnote 9 The Chester Beatty papyrus is one of these: it uses a pair of short diagonal lines (″) to indicate where one entry ends and the next begins, and a double point (:) to separate lemma and gloss within an entry. When a single lemma has multiple glosses, double points are also used to separate the different glosses, but word division within a lemma or gloss is not indicated. A typical passage therefore looks like this (lines 1271–4):
Expanded into the usual columnar format and with diacritics and word division, the same passage would look like this:
Another peculiarity of the glossary is that it is inconsistent about which language comes first. Most of the time the Greek is the lemma and the Latin the gloss, as in the example above, but sometimes we find the reverse order, with the Latin coming first and the Greek following as a gloss. Sometimes the order reverses for just one or two entries, sometimes for much longer. Moreover the writer's use of punctuation is not completely consistent: sometimes the double point is used at the end of an entry and/or the double diagonal line is used within an entry. Therefore it is not always clear how the entries should be divided up, a fact that is particularly problematic when the text is difficult to read.
The sheets containing the glossary were once part of a papyrus codex that also contained a Greek grammar, a Latin model alphabet and a large number of blank pages. In view of the blank pages it is believed that the codex was bound as a blank notebook and then (partially) written, rather than being first written and then bound. It was still intact when discovered but unfortunately was dismembered into separate sheets by the Chester Beatty library prior to scholarly documentation, necessitating a reconstruction of the original order of the sheets.Footnote 10
2. The Creation and Transmission of the Glossary
Wouters envisaged the glossary as having no transmission history, with the scribe of this papyrus copying entries directly from a text of the letters of Paul.Footnote 11 But certain features of the glossary, to be discussed below, suggest that there were a number of intermediate steps between the ancient copy of Paul and this papyrus. One must therefore distinguish between what we can know about the scribe of the papyrus and what we can know about the original creator of the glossary.Footnote 12
The only element of the papyrus that can be attributed with complete certainty to the scribe is the handwriting. This handwriting is reasonably fluent in both languages, but Lowe and Wouters both believed that the scribe was more at home writing Greek than Latin.Footnote 13 The selection of material in the codex is probably also attributable to the scribe; its unusual grouping (a Greek grammar, a Latin alphabet and a bilingual glossary) evidently has a linguistic purpose, and while the grammar suggests someone learning Greek, the alphabet suggests someone learning Latin. Wouters proposed that the writer was a Greek speaker trying simultaneously to improve his knowledge of his own language and to learn a new one, but Wachtel and Witte suggested that the writer was a Copt who had already learned to write in Coptic and was now learning both Greek and Latin.Footnote 14
The Coptic hypothesis is persuasive on several grounds. It fits well with the scribe's greater facility with the Greek alphabet than with the Latin, since the Coptic alphabet is effectively Greek with a few extra letters. The use of a double point to divide a lemma from its gloss is a feature of Greek–Coptic as opposed to Greek–Greek or Greek–Latin glossaries,Footnote 15 and the closest known parallel to this glossary in terms of format is a trilingual language-learning text in Latin, Greek and Coptic.Footnote 16 That text has always been something of a puzzle, since it is not fully clear which language it was designed to teach. But perhaps the answer is that some Coptic speakers, like some English-speaking Classics students today, learned both Latin and Greek more or less simultaneously.
The original compilation of the glossary – probably not by this Coptic scribe – must have been a very different process from the one normally followed by ancient lexicographers. Most ancient glossaries have definitions that closely match their lemmata in both form and meaning, but this one uses much looser equivalents, often ones that do not really match when taken in isolation but are nevertheless equivalent in the context of the Greek text of Paul and a known Latin translation. For example, εὐφραίνων ‘making happy’ is paired with laetificet ‘would make happy’ in line 1237, and in lines 1248–9 ὑπερβαλλούσης ‘surpassing’ (genitive) is paired with excellentem ‘surpassing’ (accusative). The glossary must therefore have been compiled by taking two continuous texts, one in Latin and one in Greek, and trying to match them up.Footnote 17 Εὐφραίνων and laetificet are equivalent at 2 Corinthians 2.2, where the Greek reads εἰ γὰρ ἐγὼ λυπῶ ὑμᾶς, καὶ τίς ὁ εὐφραίνων με εἰ μὴ ὁ λυπούμενος ἐξ ἐμοῦ; and the Vulgate si enim ego contristo vos et quis est qui me laetificet nisi qui contristatur ex me? (RSV: ‘For if I cause you pain, who is there to make me glad but the one whom I have pained?’) Likewise ὑπερβαλλούσης and excellentem are equivalent at 2 Corinthians 3.10, where the Greek is καὶ γὰρ οὐ δεδόξασται τὸ δεδοξασμένον ἐν τούτῳ τῷ μέρει εἵνεκεν τῆς ὑπερβαλλούσης δόξης and the Vulgate nam nec glorificatum est quod claruit in hac parte propter excellentem gloriam. (RSV: ‘Indeed, in this case, what once had splendour has come to have no splendour at all, because of the splendour that surpasses it’.) The entries that do not come from Paul seem to have been taken from another bilingual source (now lost) in the same way, as they display a similar loose equivalence: for example certas ‘certain’ (feminine plural) and ῥητούς ‘definite’ (masculine plural) in line 1298.
Why, then, do some entries contain more than two equivalents, such as the glossing of ὑπομονῇ with in tolerantia, patientia in line 1271? Wouters suggested that the Latin version of Paul used by the compiler contained multiple variant readings and the compiler simply copied them all into the glossary.Footnote 18 He points out that two ninth-century Latin manuscripts of Paul, the Codex Boernerianus and the Codex Augiensis, both contain multiple Latin variants, and he discusses three entries with variants, in each of which this explanation works well because both variants are Latin ones attested in the Vetus Latina tradition. These are line 1271 (quoted above, from 2 Corinthians 1.6); line 1260, where αἰώνιον is glossed with inmensum and aeternum (2 Corinthians 4.17); and line 1357, where ἀνεθέμην is glossed with quievi and contuli (Galatians 1.16). But the ninth-century manuscripts with multiple variants are significantly later than the date of this papyrus, and there is no evidence for the existence of such texts of Paul in antiquity. Moreover the three examples discussed by Wouters are atypical, for in others this explanation works less well. Sometimes the doublet is on the Greek rather than the Latin side, as in lines 1355–6, where the Greek is εὐδόκησεν, εὐαρέστησεν and the Latin placuit (Galatians 1.15). Sometimes the extract does not come from the Bible, as in lines 1292–3, where καθοσιώσει is glossed with dicationi and devotioni. Literary texts rarely accumulated sets of synonyms beside the various words, so even though we do not know exactly what the source of this entry was, it is unlikely to have contained variant readings in the Latin.Footnote 19 Ancient glossaries, on the other hand, often accumulated additional equivalents, as collections of synonyms were more useful there than in a text. The easiest explanation for the multiple variants, therefore, is that the glossary had a transmission history as a glossary and that during that transmission additional variants were added to some entries.
The current layout of the text does not make additions feasible; there is absolutely no space to insert new material. But the situation would have been different if an earlier version of the glossary had a columnar layout with each entry beginning on a new line: such layout invites the addition of extra glosses because it has more blank space. And there is one passage in particular that can best be explained by the hypothesis that a variant was added while the glossary was in columnar format. This is lines 1306–7, containing entries from Galatians 3.11 and 5.1, in both cases with the Latin coming first. The papyrus reads mạṇịfeṣte : | [δ]ηλον ″ planẹ : c̣ọntinere : ανε̣χ̣εσθαι. In 3.11 the expected text is δῆλον and manifestum (‘Now it is evident that no man is justified before God by the law …’), and in 5.1 it is ἐνέχεσθε and contineri (‘… stand fast therefore, and do not submit again to a yoke of slavery.’). The papyrus’ ανε̣χ̣εσθαι is easy to explain as a misspelling of ἀνέχεσθε, a known variant in the Greek here, and continere is a plausible mistake for contineri, but what is one to make of plane?
Wouters, who did not see the double point between plane and continere, took plane continere together as the translation of ἀνέχεσθαι here, giving a unique Latin variant in 5.1. The punctuation, however, indicates that they should not be taken together. Moreover plane continere would be a very surprising variant here: there is nothing in the Greek that would encourage the addition of an extra word to the Latin, and Latin translations of the New Testament rarely introduce words that are completely unsupported by the Greek. It is more likely that plane was originally intended to be an additional gloss for δῆλονFootnote 20 – but in that case, why does the punctuation put it with continere rather than with δῆλον?
Perhaps the text originally looked like this:
Then someone attempted to add plane as an additional gloss for δῆλον, squeezing it in to the space available, which was not ideal. That resulted in a text that looked like this:
or perhaps like this:
The next copyist, as he changed the layout, put plane with what followed it rather than what preceded, and therefore ended up with manifeste : δῆλον ″ plane : continere : ἀνέχεσθαι.
If this explanation is correct, it follows that the text from which the glossary was originally compiled did not contain all the material currently in the glossary. It may have had some variants, but it probably contained only one continuous text in each language. What else can we know about the source text?
As observed above, it must have been bilingual. But it was probably not in columnar layout, for the Greek and Latin of the glossary entries do not always match perfectly. Sometimes more of the passage has been excerpted in one language than in the other (e.g. σωφρονοῦμεν paired with sobrii rather than sobrii sumus in lines 1206–7),Footnote 21 or words in scriptio continua have been misdivided (e.g. ἐν τῇ παρησίᾳ paired with in adventuti by misdivision of in adventu Titi in lines 1221–2),Footnote 22 or the compiler simply matched up the wrong part of the verse (e.g. συνισταν〈όν〉των paired with cumparantes rather than commendant in line 1199).Footnote 23 Such mistakes would not have occurred if the format of the original had given the compiler a clear indication of how to match up the two languages.
The compiler's source, therefore, was one or more texts with a non-literal translation not in columnar format. For the Pauline material this is what we would expect in any case, since ancient bilingual copies of biblical texts rarely use the columnar format:Footnote 24 it is easy to see how the compiler of the glossary could have worked from a facing-page translation of Paul, or even two separate copies, one in Greek and one in Latin. The difficulty comes with the non-Pauline material.
As Wouters observes,Footnote 25 most of this material must have been excerpted from continuous texts like the Pauline material, since it too has loose agreement between the two languages. Some of the non-Pauline entries are extensive enough, and distinctive enough, for one to be able to state with confidence that the text(s) from which they come are no longer extant: for example phrases such as se detulisse / ἑαυτοὺς παραγεγράφθαι (lines 1295–6) and conpetentem / προσελθόντων ἐντυχόντες (lines 1296–7). Some have a legal flavour, which led Wouters to suggest that their source was a juridical text:Footnote 26 iuxta iuris ordinem / κατὰ τὸν τοῦ δικαίου θεσμόν ‘according to the order/precept of justice’ (lines 1430–1), persolvere / τοῦ ταμίου λόγους εἰσφέρειν ‘bring in the accounts of the treasurer’ (1434–5), testibus / μαρτύρων ‘witnesses’ (1439), pensitationibus / τῶν εἰσφορῶν ‘payments’ (1440–1), iudicis auctoritate / τῆς δικαστικῆς προστάξεως ‘judicial authority’ (1441–2) and procummulgatam / πρυτανευθεῖσαν ‘publicly proposed’ (1444).
A freely translated juridical text from the fourth century would be unparalleled; at that period juridical texts were composed exclusively in Latin, and indeed one of the main reasons Greek speakers learned Latin was that they needed it to practise Roman law. We have just one translated juridical text from this period, a highly literal columnar translation made to help law students learn Latin.Footnote 27 Self-standing translations with different syntax in the two languages – the kind that must have been used to produce this glossary – do not emerge for legal texts until the sixth century. Moreover the words in the glossary are not technical legal terms belonging primarily in a juridical text: they are non-technical terms used by laymen to discuss legal matters, and in fact some are commoner in Christian writings than in juridical ones. Christian writers use legal language for a variety of reasons: when discussing Jewish law, when using legal metaphors, and when describing the trials of martyrs, for example.
Therefore the non-Pauline material in this glossary could well come from a Christian text. If so, the compiler may have been using just one source, one that included portions of Paul as well as non-Pauline material. Use of such a source rather than a copy of the Bible might also explain the bizarre ordering of the Pauline material: the source text could have included substantial quotations from Paul, arranged in an order that fitted the writer's argument rather than in the order in which he found them in the Bible. The compiler of the glossary would then have taken phrases in the order in which they appeared in his source.
In sharp contrast to juridical texts, Christian texts from the time of this papyrus often have the characteristics identified in the source texts for this glossary: circulating in both languages, with one being a loose translation of the other rather than a literal version in columnar format. In addition to the Bible itself, notable examples of such texts are the extensive records of multiple church councils,Footnote 28 the Acts of the Christian MartyrsFootnote 29 and the Shepherd of Hermas, which was composed in Greek in the second century and translated into Latin ca. 200 and again in the fifth century.Footnote 30 Too little survives of the source text of this glossary to allow us to determine its genre with confidence, but the extensive use of passages concerning fund-raising from 2 Corinthians suggests that the source text might have been a fund-raising document, circulated bilingually to maximise its reach.
Wouters’ explanation for the creation of the glossary, a single step from Pauline text to the preserved papyrus, requires a complicated, difficult and unlikely process: the compiler must have had access to multiple bilingual texts of the necessary type, both Pauline and juridical, and in creating the glossary must have moved back and forth between as well as within these texts. By contrast, the process suggested here is simpler and more plausible: the compiler worked from a single text (whether one bilingual edition or physically separate Greek and Latin versions), and probably went through it in order.
Unfortunately, if these arguments about the origin of the glossary are correct, its utility for reconstructing the text of Paul in either Latin or Greek is somewhat reduced. We have not a papyrus copied directly from a text of Paul, but instead a papyrus copied from a columnar glossary made from a text containing extensive quotations from Paul. Under those circumstances most of the deviations from the usual Pauline text probably arose during or after the composition of the source text, rather than being found in the copy of the Pauline letters used by the composer of that text. But on the other hand, the longer creation process means that the Pauline extracts in this papyrus probably come from an earlier text of Paul than was previously thought.
3. New Readings and Identifications
When re-examination changes the readings in a papyrus it is customary to produce a re-edition.Footnote 31 This papyrus, however, is of interest primarily for the specific individual readings where it differs from the mainstream textual tradition of Paul, and those readings have already been extracted from Wouters’ edition and absorbed by the scholarly literature on the New Testament and Vetus Latina. What scholars in these fields need, therefore, is not a re-edition of all 528 entries through which they can trawl to learn whether any of the important readings have changed, but rather a discussion of the particular passages in which the papyrus’ important readings need to be updated. That discussion is therefore provided here.
The Center for the Study of New Testament Manuscripts has recently made available a new set of digital images of this papyrus that are significantly more legible than the ones published by Wouters. These images have been important in allowing me to improve on earlier readings of the papyrus, and because they are freely available they can be used to verify most of the new readings.Footnote 32 However, I have also spent two days examining the original in the Chester Beatty Library, and some of the new readings come from that process rather than from the images.
Because Wachtel and Witte's work is not well known, this list of new readings includes ones where Wachtel and Witte have successfully corrected Wouters’ text. In other words, this section can function as an addendum to Wouters’ work, providing all the significant updates to the text since 1988. Readings of particular interest are presented first, in the order of the Biblical passages, and a complete list of all the new readings follows.
2 Corinthians 7.4? = Line 1286
Papyrus reading: παρακλησει : hortatione
Usual text: παρακλήσει and consolatione
RSV: ‘I have great confidence in you; I have great pride in you; I am filled with comfort.’
Wouters identifies this entry as coming from 7.13,Footnote 33 at which point hortatione would be a unique variant in the Latin (indeed it is now identified as such in the Vetus Latina database). But the entry could just as easily come from 7.4, which has the same usual readings in both languages but at which point hortatione would be partially paralleled by exhortatione in several other Vetus Latina versions (including the Codex Claromontanus, to which this papyrus is known to have a special affinityFootnote 34). Moreover, it is possible that this extract is not from Paul at all: the preceding entry in the glossary is from 2 Corinthians 1.17 and the following one not from the New Testament. Under those circumstances there is really not enough evidence to support the claim of a unique variant at 7.13.Footnote 35
2 Corinthians 7.14? = Line 1229
New reading: η επ̣ι̣ [:] q̣ụạẹ [ad]
Old reading: η̣ε̣ ̣ ̣ [ ̣] ̣ ̣ ̣ ̣[ ̣ ̣ ̣]
Usual text: ἡ ἐπὶ Τίτου (see below) and quae fuit ad Titum (but quae ad Titum fuit is a known variant)
RSV: ‘… but just as everything we said to you was true, so our boasting before Titus has proved true.’
Wouters did not identify the verse to which this entry pertained,Footnote 36 presumably because he could not be certain of reading even ηε. Now those letters are clear enough to be certain, and therefore the entry must be from 7.14 if the entries are in sequence here: the preceding entry is from 7.11 and the following from 7.15, and in that section of text only one passage can begin ηε. The correct reading of Paul is actually disputed here, and not all sources include the ἡ; Tischendorf for example prints ἡ καύχησις ἡμῶν ἐπὶ Τίτου, and there is a common variant ἡ πρὸς Τίτον that would also fail to match the papyrus. But ἡ ἐπὶ Τίτου has good manuscript support and is printed in the 28th edition of Nestle–Aland. In Latin the Codex Claromontanus, to which this papyrus has an especially close relationship, has quae ad Titum fuit.Footnote 37
2 Corinthians 9.2 = Lines 1184–5
New reading: [απο περυσι : ab a]nno | priore
Wachtel and Witte's reading: α̣[πο περυσι : a]ṇno priore
Wouters’ reading: [ ̣ ̣ ̣ απο περυσ?]ιν ″ | priore
Usual text: ἀπὸ πέρυσι and ab anno praeterito (but ab anno priore is a known variant)
RSV: ‘… I boast about you to the people of Macedonia, saying that Achaia has been ready since last year …’
As Wachtel and Witte already realised, the letters read by Wouters as ιν″ are really nno. This means both that the usual reading of the Greek, ἀπὸ πέρυσι, fits the lacuna perfectly without the need to assume that something else stood before it, and that the Latin was a full match for the Greek, not a partial match; there is space in the lacuna both for Greek ἀπό and for Latin ab. Wachtel and Witte believed that they could see the bottom of the initial α at the start of the lacuna,Footnote 38 but the lines in question are more likely to be the ε of κενωθῇ in the next line, since in this papyrus α is normally small and rounded.
2 Corinthians 10.16 = Line 1203
New reading: [υ]π̣ερεκεινα : ultra vos
Old reading: [υ]π̣ερ εκεινα : ụḷtra ẹos
Usual text: ὑπερέκεινα ὑμῶν and ultra vos
RSV: ‘… so that we may preach the gospel in lands beyond you, without boasting of work already done in another's field.’
Wouters did not identify the source of this quotation; Wachtel and Witte noticed the similarity to 10.16 but did not recheck the papyrus and were therefore uncertain of the identification.Footnote 39 With the new reading the correspondence with 10.16 is clear, although the two languages are not completely parallel since the Latin includes one more word of the text than the Greek.
2 Corinthians 12.7 = Line 1398
New reading: σ̣κολο̣ψ : stimụḷụs
Old reading: σ̣κολο̣ψ : sṭimeg ̣ ̣ ̣
Usual text: σκόλοψ and stimulus
RSV: ‘And to keep me from being too elated by the abundance of revelations, a thorn was given me in the flesh …’
Wouters believed that the usual Latin reading did not fit the traces on the papyrus,Footnote 40 but it does: the element that Wouters saw as the tail of g is actually the top of the second s of libentissime in the line below. A unique Latin variant has therefore disappeared here.
2 Corinthians 12.10 = Line 1400
New reading: ευδοκω̣ : bene opinor
Old reading: ευδοκω̣ : bene ṿịḍeor
Usual text: εὐδοκῶ and placeo mihi
RSV: ‘For the sake of Christ, then, I am content with weaknesses …’
Wouters was aware that his reading did not fit the traces well;Footnote 41 the new reading is clear on the original and on the new photographs. Both old and new Latin readings are unique variants.
2 Corinthians 12.13 = Lines 1400–1
New reading: κατα̣ναρκησα : | gravavo : egersio
Wouters’ reading: κατα̣ναρκησα : | gravaṿọ : egerf̣i
Wachtel and Witte's reading: κα̣τ̣α̣ν̣α̣ρκη̣σ̣ω̣ : | gravaṿọ (plus ‘egersi oder egerfi’ in n. 37)
Usual text: κατανάρκησα and gravavi in 12.13; καταναρκήσω and ero gravis (with a known variant gravabo) in 12.14
RSV (12.13 and 12.14): ‘For in what were you less favoured than the rest of the churches, except that I myself did not burden you? Forgive me this wrong! Here for the third time I am ready to come to you. And I will not be a burden …’
Wouters identified this entry as a quotation of 12.13 and saw gravavo as scribal error for gravavi. But Wachtel and Witte argued that gravavo must instead be gravabo and that the ending of the Greek looked more like ω than like α; they therefore identified this entry as coming from 12.14.Footnote 42 Their point about the Latin is not as strong as they claim it is, for although b is frequently written for v in this papyrus the reverse spelling is much rarer:Footnote 43 gravavo is unlikely to be an error for gravabo. Moreover the Latin in this glossary is not always a good match for the Greek, and therefore one should never force the Greek to fit the Latin. In this case it takes considerable forcing to make the Greek into a future, for the strokes that make up ω in Wachtel and Witte's reading are the ones that make up not α but σα in Wouters’ reading: καταναρκήσω is impossible, and the choice is between κατανάρκησα and καταναρκήω. As the Greek in this papyrus is usually correct, the latter reading is extremely unlikely, and therefore the extract must come from 12.13. In those circumstances, the best explanation of gravavo is a suggestion made informally by Hugh Houghton, that gravavo arose from miscopying of gravavi vos (vos being the next word here in the usual text).
Egersio is probably a misspelling of exerceo ‘drive on, keep at work’, presumably functioning as an additional gloss on κατανάρκησα. Neither Wouters nor Wachtel and Witte proposed plausible explanations for their readings.Footnote 44
2 Corinthians 12.20 = Line 1404
New reading: θ̣υμο̣ι : ịṇṛịṭạṭịọnes
Old reading: θ̣υμο̣ι̣ [ ± 10 ]nes
Usual text: θυμοί and animositates
RSV: ‘For I fear … that perhaps there may be quarrelling, jealousy, anger, selfishness, slander, gossip, conceit, and disorder.’
The Latin is a spelling variant of irritationes ‘irritations’; it seems to be a unique variant here, but the Codex Claromontanus has inritationes as a supplement to irae at Galatians 5.20.
Galatians 3.5 = Lines 1372–3
New reading: επιχορηγων [:] | tribuït
Old reading: επιχορηγων [:] | tribuet
Usual text: ἐπιχορηγῶν and tribuit
RSV: ‘Does he who supplies the Spirit to you and works miracles among you do so by works of the law, or by hearing with faith?’
The new reading matches the usual text exactly and eliminates what had been a unique variant in the Latin.
Galatians 5.4 and 5.12 = Lines 1307–8
New reading: distrueṭis : | κ̣ατηργηθη ″ inquietant : αναστατουντες
Wachtel and Witte's reading: distrueṭis : | [κ]ατηργηθησ̣ιν quietant : αν̣ασ̣τα̣τουντες
Wouters’ reading: distrueṭis : | [κ]ατηργηθησ̣ιν quietant : αν̣αντατουντες
Usual text: κατηργήθητε and evacuati estis at 5.4, ἀναστατοῦντες and conturbant at 5.12
RSV: ‘You are severed from Christ, you who would be justified by the law …’ (5.4) and ‘I wish those who unsettle you would mutilate themselves!’ (5.12)
The old reading quietant had the opposite meaning to the one required by both the context and the Greek; it arose from misreading the divider ″ as σ and consequently assigning in (letters that in this papyrus look exactly the same in both Latin and Greek script) to the wrong word.Footnote 45 In Latin the new reading is a unique and significant Vetus Latina variant. In Greek the new κατηργήθη, which if taken seriously would mean ‘he was parted’ (from καταργέω), is probably just missing the ending of the usual κατηργήθητε. Wouters’ αν̣αντατουντες, probably a typographical error, was corrected to αν̣ασ̣τα̣τουντες already by Wachtel and Witte, but actually none of the letters are doubtful.
Ephesians 1.22? = Lines 1377–8
Wachtel and Witte's reading: subiecit : | υπεταξεν
Wouters’ reading: subiecit : | υπεταξαν
Usual text: ὑπέταξεν and subiecit
RSV: ‘… and he has put all things under his feet and has made him the head over all things for the church …’
Wachtel and Witte are certainly right that the Greek is a singular rather than a plural.Footnote 46 It is less clear that they are right to reject Wouters’ identification of the entry as coming from Ephesians 1.22, where it matches the usual text perfectly in both languages. Admittedly this entry is followed and perhaps preceded by entries from Galatians 3.19, so if coming from Ephesians it is oddly placed, but it is difficult to accept Wachtel and Witte's alternative explanation, that this entry is an amplification of the previous one (διατάσσει : perordinat ‘ordains’). Such amplificatory entries do occur in this glossary,Footnote 47 but the connection with the previous entry is usually more obvious than it would be in this case.
Ephesians 3.8 = Line 1330
New reading: ανεξιχνιαστον : ịṇrep̣ṛensibilae
Wachtel and Witte's reading: ανεξιχνιαστον : ịṇrep̣ <re>ḥensibilae
Wouters’ reading: ανεξιχνιαστον : ̣ ̣re ̣(?)hensibilae
Usual text: ἀνεξιχνίαστον and (in)investigabiles
RSV: ‘To me, though I am the very least of all the saints, this grace was given, to preach to the Gentiles the unsearchable riches of Christ …’
Wouters, while declining to supplement the Latin in his edition, suggested in his commentary that it might be a misspelling of incomprehensibiles (attractive owing to an attested variant incompraehensibiles) or its synonyms indeprehensibiles or inapprehensibiles.Footnote 48 But none of these readings can be reconciled with the traces, which point rather to a misspelling of irreprehensibile (cf. Wachtel and Witte's version, with a different misspelling of the same word). As this word means ‘irreproachable’ it is a poor match for the Greek (which means ‘inscrutable’). The variant is unique in the Vetus Latina tradition.
Complete List of New Readings
The list below includes all places where Wouters’ text of the papyrus should be updated, whether because of my own work or because of Wachtel and Witte's. Passages discussed above are also included here for ease of reference. However, changes that pertain only to the addition or removal of dots, or to the punctuation of the original without changing the meaning, are not included.
1184–5 [ ̣ ̣ ̣ απο περυσ?]ιν ″ | priore Wouters (2 Cor 9.2): new reading [απο περυσι : ab a]nno | priore (see above)
1185 κα̣[υχημα : gloria]ṃur Wouters (2 Cor 9.3): new reading κε[νωθη : evacue]ṭur. The new reading is a different lemma from the same verse; like the original reading, it matches the usual text in both Greek and Latin.
1187 [προκαταρ]τισωσιν Wouters (2 Cor 9.5): new reading κ̣[αταρ]τισωσιν (usual text προκαταρτίσωσιν). Wouters’ supplement is far too long for the space available. Since καταρτίζω ‘prepare’ is considerably more common than προκαταρτίζω ‘prepare beforehand’, καταρτίσωσιν is a likely possibility, though I can find no actual parallels for that reading.
1203 [υ]π̣ερ εκεινα : ụḷtra ẹos Wouters (2 Cor 10.16): new reading [υ]π̣ερεκεινα : ultra vos (see above)
1224 ευ εξηρητον Wouters (not Pauline): new reading ευ εξηριτον (misspelling of εὖ ἐξαίρετον ‘well chosen’)
1229 η̣ε̣ ̣ ̣ [ ̣] ̣ ̣ ̣ ̣[ ̣ ̣ ̣] Wouters (2 Cor 7.14): new reading η επ̣ι̣ [:] q̣ụạẹ [ad] (see above)
1240 [ ̣ ̣ ̣ ̣ ̣ ̣ ̣ ̣] ̣ ̣ ̣[ ̣] absorbeatur Wouters (2 Cor 2.7): new reading [κα|τα]π̣ο̣θ̣η̣ absorbeatur (matching the usual text)
1261 τα δ : sed quae Wouters (2 Cor 4.18): new reading τα δε : sed que (matching the usual text except for a misspelling in the Latin)
1262 [in]|gemiscimus Wouters (2 Cor 5.2): new reading gemiscimus (usual text ingemescimus; this variant appears to be unique)
1267 suadamus Wouters (2 Cor 5.11): new reading suadamos (misspelling of unique Latin variant)
1268 misericorḍ[iarum?] Wouters (2 Cor 1.3): new reading misericorḍ[ie] (for misericordiae) (usual text misericordiarum, but misericordiae is a known variant)
1274–5 πρ̣[?]|ηχθην Wouters (not Pauline): new reading πρ̣[ο]|ηχθην (προήχθην, aorist passive of προάγω ‘lead on, induce, persuade’ (LSJ s.v. i.4), which can mean ‘be ready to do’ and would therefore be a good match for its Latin equivalent here, prumptus for promptus ‘quick to respond’ (OLD s.v. 4), if the source text had slightly different syntax in the two languages. The entry is unlikely to be from Paul because the ending -ήχθην, which is absolutely clear on the papyrus, occurs nowhere in the New Testament; nothing of which this word could be a corruption occurs near 2 Corinthians 1.8 or 1.9, the relevant section of text.)
1298–9 pensitaṭi[ ̣ ̣ ̣ ̣| ̣ ̣ ̣] ̣ ̣ φορο̣υ̣ α̣τελη Wouters (not Pauline): new reading pensitati[onis im|une]ṃ : φορο̣υ̣ α̣τελη (φόρου ἀτελῆ ‘not liable to tax’: cf. Codex Theodosianus 13.5.32 = p. 755.1)
1303 prius quam quod ″ παρ ο̣ Wouters (not Pauline?): Wachtel and Witte rightly object to Wouters’ identification of this entry as Galatians 1.9,Footnote 49 since παρ’ ὅ : id quod in 1413 must reflect the same passage of Galatians 1.9, and it is unlikely that the process by which this glossary was created could have resulted in two quotations of the same Greek passage with different Latin equivalents. (Compare Galatians 2.11, which occurs twice in this papyrus, at 1304 and 1364, but which has not only the same Latin but also the same mistake in the Greek in both places.)
1305 καρηργησαι Wouters (Gal 3.17): new reading καταργησαι (matching the usual text)
1307 planẹ c̣ọntinere Wouters (Gal 5.1): new reading planẹ : c̣ọntinere (see above)
1308 [κ]ατηργηθησ̣ιν quietant : αν̣αντατουντες Wouters (Gal 5.4, 5.12): new reading κ̣ατηργηθη ″ inquietant : αναστατουντες (see above)
1326 δογμα[σ]ιν Wouters (Eph 2.15): new reading δογμασ̣[ι]ν
1327 reconcilians Wouters (Eph 2.16): Wachtel and Witte's reading reconsilians is right (usual text reconciliet)
1330 ̣ ̣re ̣(?)hensibilae Wouters (Eph 3.8): new reading ịṇrep̣ṛensibilae (see above)
1338–9 ïn aṣṭ[u]|tia Wouters (Eph 4.14): new reading ïnaeq[ui]|tia (misspelling of usual text in nequitia). The old reading assumed a misunderstanding of the text, whereby astutia ‘craftiness’, which occurs later in the same verse, was matched to the wrong Greek.Footnote 50
1348–9 εὐτρα[ ± 2]|λολογεια Wouters (Eph 5.4): Wachtel and Witte suggest the supplement ευτρα[πε]λολογεια (usual text εὐτραπελία)Footnote 51
1354 α[φθ]εαρσια Wouters (Eph 6.24): this reading is correct and should not be replaced by Wachtel and Witte's α[φ]θ̣αρσια
1355 υπα̣ρχω̣ν των Wouters (Gal 1.14): new reading υπα̣ρχοντων (ὑπαρχόντων, usual text ὑπάρχων τῶν; this variant is not in Tischendorf's apparatus)
1360 [ ̣ ̣ ̣]ḍucṭọṣ Wouters (Gal 2.4): new reading [subin]ḍucṭọṣ (usual text subintroductos, but subinductos is a known variant)
1365 aedebat Wouters (Gal 2.12): Wachtel and Witte's reading aedabat is correct (misspelling of usual edebat)
1367 ων : cum sis Wouters (correctly, but with no identification): this pair can be found in Philemon 9, but Wachtel and Witte suggest that the source is Galatians 2.14 ὑπάρχων and cum … sis, where ὤν is a known variant in the Greek.
1371 inru[[ ̣ ̣]] Wouters (Gal 3.17?): new reading inruịṭ (usual text irritum facit)
1373 tribuet Wouters (Gal 3.5): new reading tribuït (see above)
1378 υπεταξαν Wouters (Eph 1.22): Wachtel and Witte's ὑπέταξεν is right (see above)
1381–2 a|gna : casta Wouters (correctly, but with no identification): this may be from Philippians 4.8 ἅγνα, where the Vulgate has sancta but the Vetus Latina casta
1383 obstruetur Wouters (2 Cor 11.10): new reading obstruïtur (usual text infringetur)
1391 οδοιπορειας Wouters (2 Cor 11.26): new reading οδοιπορειαις (usual reading ὁδοιπορίαις)
1398 sṭimeg ̣ ̣ Wouters (2 Cor 12.7): new reading stimụḷụs (see above)
1400 bene ṿịḍeor Wouters (2 Cor 12.10): new reading bene opinor (see above)
1401 egerf̣i Wouters (2 Cor 12.13): new reading egersio (see above)
1404 [ ± 10]nes Wouters (2 Cor 12.20): new reading ịṇṛịṭạṭịọnes (see above)
1405 diṣẹnsionẹṣ Wouters (2 Cor 12.20): new reading dịṣṣẹnsionẹṣ (matching the usual text)
1420 lberevẹr[ Wouters (not Pauline): new reading lbereverụ[
1421–2 [τα-?]|μιακας εισφορας επιπ ̣ ̣ ̣ ̣σκοντες Wouters (not Pauline): new reading τ̣[α]|μιακας εισφορας επιτινε̣σκοντες (‘increasing the taxes for the treasury’: επιτινε̣σκοντες is for ἐπιτεινέσκοντες)
1423 αποπληροιντο̣ Wouters (not Pauline): new reading αποπληροιντας (for ἀποπληροῦντας ‘satisfying’)
1436 queremoniae Wouters (not Pauline): new reading queremonias
